avr
/
avrdude
mirror of https://github.com/mariusgreuel/avrdude.git
2
0
Fork 0
Code Issues Projects Releases Wiki Activity
avrdude/src/stk500v2.c

4560 lines
129 KiB
C
Raw Normal View History

Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
/*
* avrdude - A Downloader/Uploader for AVR device programmers
* Copyright (C) 2005 Erik Walthinsen
* Copyright (C) 2002-2004 Brian S. Dean <bsd@bsdhome.com>
Replace David's email address by his full name in the copyrights. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@565 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-12 23:24:35 +00:00
* Copyright (C) 2006 David Moore
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
* Copyright (C) 2006,2007,2010 Joerg Wunsch <j@uriah.heep.sax.de>
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
Replace outdated FSF postal address by a reference to the GPL info on their website. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1107 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-11-20 14:03:50 +00:00
* along with this program. If not, see <http://www.gnu.org/licenses/>.
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
*/
Properly close commend in $ Id $. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@455 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 20:58:45 +00:00
/* $Id$ */
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
/* Based on Id: stk500.c,v 1.46 2004/12/22 01:52:45 bdean Exp */
/*
* avrdude interface for Atmel STK500V2 programmer
*
Contributed by dcm@mit.edu: add support for the AVRISP mkII device. (Savannah patch #4789.) * serial.h: Declare usb_serdev_frame device descriptor. * stk500v2.c: Implementation of the AVRISP mkII handling. * usb_libusb.c: Add USB handling for short-frame delimited AVRISP mkII USB protocol; add distinction of different devices in usbdev_open(). * jtagmkII.c: Tell usbdev_open() to search for the JTAG ICE mkII. * usbdevs.h: (New file.) * Makefile.am: Add usbdevs.h, as well as some other forgotten files "make distcheck" complained about. * avrdude.conf.in: Add more aliases for the AVRISP mkII. * avrdude.1: Document how to use the AVRISP mkII. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@564 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-12 23:13:50 +00:00
* As the AVRISP mkII device is basically an STK500v2 one that can
* only talk across USB, and that misses any kind of framing protocol,
* this is handled here as well.
*
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
* Note: most commands use the "universal command" feature of the
* programmer in a "pass through" mode, exceptions are "program
* enable", "paged read", and "paged write".
*
*/
#include "ac_cfg.h"
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
#include <math.h>
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <errno.h>
Implement extended addressing needed for the ATmega256x devices. Document ATmega256x support. Also document Solaris port defaults in avrdude.texi. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@584 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-05-25 14:38:08 +00:00
#include <limits.h>
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
#include <unistd.h>
#include <sys/time.h>
#include <time.h>
Something I always wanted to do: replace all those private "extern" declarations in each file by a central header file "avrdude.h". git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@721 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-01-24 21:07:54 +00:00
#include "avrdude.h"
Join the former "public" header files (avr.h avrpart.h pindefs.h serial.h fileio.h safemode.h update.h pgm_type.h config.h confwin.h lists.h) into a single header that can be included by anyone wanting to link against the library. Adapt everything to cope with this situation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1311 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-05-19 10:01:59 +00:00
#include "libavrdude.h"
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
#include "stk500_private.h" // temp until all code converted
Parser does not need to know all programmer types now, new programmers will update only the table in pgm_type.c. * config_gram.y, lexer.l: removed programmer type keywords, use now locate_programmer_type() function * pgm_type.[ch]: added new files for table of programmer types * main.c: allow list of programmer types by -c ?type * avrdude.conf.in: changed all type keywords to quoted strings * doc/avrdude.texi: changed description of type definition, list of valid types is now included from generated file * doc/Makefile.am: generate list of programmer types for doc * all programmers [hc]: add xxx_desc string for description of programmer git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1051 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-01-31 17:03:43 +00:00
#include "stk500v2.h"
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
#include "stk500v2_private.h"
Contributed by dcm@mit.edu: add support for the AVRISP mkII device. (Savannah patch #4789.) * serial.h: Declare usb_serdev_frame device descriptor. * stk500v2.c: Implementation of the AVRISP mkII handling. * usb_libusb.c: Add USB handling for short-frame delimited AVRISP mkII USB protocol; add distinction of different devices in usbdev_open(). * jtagmkII.c: Tell usbdev_open() to search for the JTAG ICE mkII. * usbdevs.h: (New file.) * Makefile.am: Add usbdevs.h, as well as some other forgotten files "make distcheck" complained about. * avrdude.conf.in: Add more aliases for the AVRISP mkII. * avrdude.1: Document how to use the AVRISP mkII. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@564 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-12 23:13:50 +00:00
#include "usbdevs.h"
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
/*
* We need to import enough from the JTAG ICE mkII definitions to be
* able to talk to the ICE, query some parameters etc. The macro
* JTAGMKII_PRIVATE_EXPORTED limits the amount of definitions that
* jtagmkII_private.h will export, so to avoid conflicts with those
* names that are identical to the STK500v2 ones.
*/
#include "jtagmkII.h" // public interfaces from jtagmkII.c
#define JTAGMKII_PRIVATE_EXPORTED
#include "jtagmkII_private.h"
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
#include "jtag3.h" // public interfaces from jtagmkII.c
#define JTAG3_PRIVATE_EXPORTED
#include "jtag3_private.h"
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
#define STK500V2_XTAL 7372800U
Submitted by Brett Hagman: Add support for the "Wiring" board/bootloader * wiring.c: New file. * wiring.h: (Ditto.) * Makefile.am: Add new files. * stk500v2_private.h: Reorganize so some functions and struct pdata are globally known. * stk500v2.c: (Ditto.) * stk500v2.h: (Ditto.) * lexer.l: Add new programmer keywords. * config_gram.y: (Ditto.) * avrdude.conf.in: Add "wiring" programmer entry. * avrdude.1: Document the new programmer. * doc/avrdude.texi: (Ditto.) * AUTHORS: Add Brett Hagman. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@987 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-08-26 20:22:09 +00:00
// Timeout (in seconds) for waiting for serial response
#define SERIAL_TIMEOUT 2
// Retry count
#define RETRIES 5
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
#define DEBUG(...) msg_trace2(__VA_ARGS__)
Define DEBUG and DEBUGRECV unconditionally, as MSG_TRACE2 messages. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1392 81a1dc3b-b13d-400b-aceb-764788c761c2
2016-09-14 06:17:59 +00:00
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
#define DEBUGRECV(...) msg_trace2(__VA_ARGS__)
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
enum hvmode
{
PPMODE, HVSPMODE
};
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
main.c, pgm.c, pgm.h: Add setup and teardown hooks to the programmer definition. If present, call the setup hook immediately after finding the respective programmer object, and schedule the teardown hook to be called upon exit. This allows the programmer implementation to dynamically allocate private programmer data. avr910.c, butterfly.c, jtagmkI.c, jtagmkII.c, stk500v2.c, usbasp.c, usbtiny.c: Convert static programmer data into dynamically allocated data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@764 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-11-07 20:36:12 +00:00
#define PDATA(pgm) ((struct pdata *)(pgm->cookie))
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
/*
* Data structure for displaying STK600 routing and socket cards.
*/
struct carddata
{
int id;
const char *name;
};
Minor cosmetic changes: STK500 firmware version numbers are M.NN, so always display the minor number as two digits. Examine the response to the sign-on command to see which programmer hardware we are talking to, and then restrict the STK500 topcard display to devices detected as STK500. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@614 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-08-18 21:23:34 +00:00
static const char *pgmname[] =
{
"unknown",
"STK500",
"AVRISP",
"AVRISP mkII",
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
"JTAG ICE mkII",
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
"STK600",
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
"JTAGICE3",
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
};
struct jtagispentry
{
unsigned char cmd;
unsigned short size;
#define SZ_READ_FLASH_EE USHRT_MAX
#define SZ_SPI_MULTI (USHRT_MAX - 1)
};
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
static const struct jtagispentry jtagispcmds[] = {
Implement and document HVSP and PP modes for the AVR Dragon. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@677 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-11-01 21:47:25 +00:00
/* generic */
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
{ CMD_SET_PARAMETER, 2 },
{ CMD_GET_PARAMETER, 3 },
{ CMD_OSCCAL, 2 },
{ CMD_LOAD_ADDRESS, 2 },
Implement and document HVSP and PP modes for the AVR Dragon. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@677 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-11-01 21:47:25 +00:00
/* ISP mode */
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
{ CMD_ENTER_PROGMODE_ISP, 2 },
{ CMD_LEAVE_PROGMODE_ISP, 2 },
{ CMD_CHIP_ERASE_ISP, 2 },
{ CMD_PROGRAM_FLASH_ISP, 2 },
{ CMD_READ_FLASH_ISP, SZ_READ_FLASH_EE },
{ CMD_PROGRAM_EEPROM_ISP, 2 },
{ CMD_READ_EEPROM_ISP, SZ_READ_FLASH_EE },
{ CMD_PROGRAM_FUSE_ISP, 3 },
{ CMD_READ_FUSE_ISP, 4 },
{ CMD_PROGRAM_LOCK_ISP, 3 },
{ CMD_READ_LOCK_ISP, 4 },
{ CMD_READ_SIGNATURE_ISP, 4 },
{ CMD_READ_OSCCAL_ISP, 4 },
Implement and document HVSP and PP modes for the AVR Dragon. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@677 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-11-01 21:47:25 +00:00
{ CMD_SPI_MULTI, SZ_SPI_MULTI },
/* all HV modes */
{ CMD_SET_CONTROL_STACK, 2 },
/* HVSP mode */
{ CMD_ENTER_PROGMODE_HVSP, 2 },
{ CMD_LEAVE_PROGMODE_HVSP, 2 },
{ CMD_CHIP_ERASE_HVSP, 2 },
{ CMD_PROGRAM_FLASH_HVSP, 2 },
{ CMD_READ_FLASH_HVSP, SZ_READ_FLASH_EE },
{ CMD_PROGRAM_EEPROM_HVSP, 2 },
{ CMD_READ_EEPROM_HVSP, SZ_READ_FLASH_EE },
{ CMD_PROGRAM_FUSE_HVSP, 2 },
{ CMD_READ_FUSE_HVSP, 3 },
{ CMD_PROGRAM_LOCK_HVSP, 2 },
{ CMD_READ_LOCK_HVSP, 3 },
{ CMD_READ_SIGNATURE_HVSP, 3 },
{ CMD_READ_OSCCAL_HVSP, 3 },
/* PP mode */
{ CMD_ENTER_PROGMODE_PP, 2 },
{ CMD_LEAVE_PROGMODE_PP, 2 },
{ CMD_CHIP_ERASE_PP, 2 },
{ CMD_PROGRAM_FLASH_PP, 2 },
{ CMD_READ_FLASH_PP, SZ_READ_FLASH_EE },
{ CMD_PROGRAM_EEPROM_PP, 2 },
{ CMD_READ_EEPROM_PP, SZ_READ_FLASH_EE },
{ CMD_PROGRAM_FUSE_PP, 2 },
{ CMD_READ_FUSE_PP, 3 },
{ CMD_PROGRAM_LOCK_PP, 2 },
{ CMD_READ_LOCK_PP, 3 },
{ CMD_READ_SIGNATURE_PP, 3 },
{ CMD_READ_OSCCAL_PP, 3 },
Minor cosmetic changes: STK500 firmware version numbers are M.NN, so always display the minor number as two digits. Examine the response to the sign-on command to see which programmer hardware we are talking to, and then restrict the STK500 topcard display to devices detected as STK500. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@614 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-08-18 21:23:34 +00:00
};
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
Update STK600 routing and socket card data from XML file. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@911 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-13 12:52:38 +00:00
/*
* From XML file:
<REVISION>
<RC_ID_MAJOR>0</RC_ID_MAJOR>
<RC_ID_MINOR>56</RC_ID_MINOR>
<EC_ID_MAJOR>0</EC_ID_MAJOR>
<EC_ID_MINOR>1</EC_ID_MINOR>
</REVISION>
*/
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
/*
* These two tables can be semi-automatically updated from
* targetboards.xml using tools/get-stk600-cards.xsl.
*/
static const struct carddata routing_cards[] =
{
{ 0x01, "STK600-RC020T-1" },
{ 0x03, "STK600-RC028T-3" },
{ 0x05, "STK600-RC040M-5" },
{ 0x08, "STK600-RC020T-8" },
{ 0x0A, "STK600-RC040M-4" },
{ 0x0C, "STK600-RC008T-2" },
{ 0x0D, "STK600-RC028M-6" },
{ 0x10, "STK600-RC064M-10" },
{ 0x11, "STK600-RC100M-11" },
{ 0x13, "STK600-RC100X-13" },
Update STK600 routing and socket card data from XML file. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@911 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-13 12:52:38 +00:00
{ 0x15, "STK600-RC044X-15" },
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
{ 0x18, "STK600-RC100M-18" },
{ 0x19, "STK600-RCPWM-19" },
Update STK600 routing and socket card data from XML file. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@911 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-13 12:52:38 +00:00
{ 0x1A, "STK600-RC064X-14" },
{ 0x1B, "STK600-RC032U-20" },
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
{ 0x1C, "STK600-RC014T-12" },
{ 0x1E, "STK600-RC064U-17" },
{ 0x1F, "STK600-RCuC3B0-21" },
{ 0x20, "STK600-RCPWM-22" },
{ 0x21, "STK600-RC020T-23" },
{ 0x22, "STK600-RC044M-24" },
{ 0x23, "STK600-RC044U-25" },
Update STK600 routing and socket card data from XML file. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@911 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-13 12:52:38 +00:00
{ 0x24, "STK600-RCPWM-26" },
{ 0x25, "STK600-RCuC3B48-27" },
{ 0x27, "STK600-RC032M-29" },
{ 0x28, "STK600-RC044M-30" },
{ 0x29, "STK600-RC044M-31" },
{ 0x2A, "STK600-RC014T-42" },
{ 0x2B, "STK600-RC020T-43" },
{ 0x30, "STK600-RCUC3A144-32" },
{ 0x34, "STK600-RCUC3L0-34" },
{ 0x38, "STK600-RCUC3C0-36" },
{ 0x3B, "STK600-RCUC3C0-37" },
{ 0x3E, "STK600-RCUC3A144-33" },
{ 0x46, "STK600-RCuC3A100-28" },
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
{ 0x55, "STK600-RC064M-9" },
Update STK600 routing and socket card data from XML file. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@911 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-13 12:52:38 +00:00
{ 0x88, "STK600-RCUC3C1-38" },
{ 0x8B, "STK600-RCUC3C1-39" },
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
{ 0xA0, "STK600-RC008T-7" },
Update STK600 routing and socket card data from XML file. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@911 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-13 12:52:38 +00:00
{ 0xB8, "STK600-RCUC3C2-40" },
{ 0xBB, "STK600-RCUC3C2-41" },
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
};
static const struct carddata socket_cards[] =
{
Update STK600 routing and socket card data from XML file. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@911 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-13 12:52:38 +00:00
{ 0x01, "STK600-TQFP48" },
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
{ 0x02, "STK600-TQFP32" },
{ 0x03, "STK600-TQFP100" },
{ 0x04, "STK600-SOIC" },
Update STK600 routing and socket card data from XML file. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@911 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-13 12:52:38 +00:00
{ 0x06, "STK600-TQFP144" },
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
{ 0x09, "STK600-TinyX3U" },
{ 0x0C, "STK600-TSSOP44" },
{ 0x0D, "STK600-TQFP44" },
{ 0x0E, "STK600-TQFP64-2" },
{ 0x0F, "STK600-ATMEGA2560" },
Update STK600 routing and socket card data from XML file. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@911 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-13 12:52:38 +00:00
{ 0x15, "STK600-MLF64" },
{ 0x16, "STK600-ATXMEGAT0" },
{ 0x18, "QT600-ATMEGA324-QM64" },
{ 0x19, "STK600-ATMEGA128RFA1" },
{ 0x1A, "QT600-ATTINY88-QT8" },
{ 0x1B, "QT600-ATXMEGA128A1-QT16" },
{ 0x1C, "QT600-AT32UC3L-QM64" },
{ 0x1D, "STK600-HVE2" },
{ 0x1E, "STK600-ATTINY10" },
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
{ 0x55, "STK600-TQFP64" },
{ 0x69, "STK600-uC3-144" },
Update STK600 routing and socket card data from XML file. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@911 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-13 12:52:38 +00:00
{ 0xF0, "STK600-ATXMEGA1281A1" },
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
{ 0xF1, "STK600-DIP" },
};
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_getparm(const PROGRAMMER *pgm, unsigned char parm, unsigned char *value);
static int stk500v2_setparm(const PROGRAMMER *pgm, unsigned char parm, unsigned char value);
static int stk500v2_getparm2(const PROGRAMMER *pgm, unsigned char parm, unsigned int *value);
static int stk500v2_setparm2(const PROGRAMMER *pgm, unsigned char parm, unsigned int value);
static int stk500v2_setparm_real(const PROGRAMMER *pgm, unsigned char parm, unsigned char value);
Revamp terminal output: progress bar, callback and stdout/stderr (#1132) * Print parms output to stdout * Flush terminal writes and other minor changes * Prepare terminal for periodic calls to programmer to reset bootloader WDT * Only show progress reports for memories > 32 bytes or on -vv * Freeze progress bar on serious error * Allow cached r/w byte routines to be used in pgm->read_byte and pgm->write_byte
2022-10-23 20:56:45 +00:00
static void stk500v2_print_parms1(const PROGRAMMER *pgm, const char *p, FILE *fp);
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_paged_load(const PROGRAMMER *pgm, const AVRPART *p, const AVRMEM *m,
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
unsigned int page_size,
unsigned int addr, unsigned int n_bytes);
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_paged_write(const PROGRAMMER *pgm, const AVRPART *p, const AVRMEM *m,
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
unsigned int page_size,
unsigned int addr, unsigned int n_bytes);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
Fix all bugs in the stk500pp implementation. Eliminate pagebuf, and use stack-allocated buffers instead. The pagesize of all current AVRs is at most 256 only anyway, and this is unlikely to change with the STK500v2 protocol. The previous pagebuf implementation suffered from some possible buffer overrun. In stk500pp_write_page(), do always write full pages, rather than attempting to write a partial last page which did not get written at all. Fill the remaining bytes with 0xff. For (paged) write operations, correctly synthesize the mode byte. This mode byte is very different from the ISP mode byte (sigh). In stk500pp_read_byte(), when performing read operations on paged memory, start reading at the previous page boundary rather than the current address. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@589 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-17 21:25:59 +00:00
static unsigned int stk500v2_mode_for_pagesize(unsigned int pagesize);
Contributed by dcm@mit.edu: add support for the AVRISP mkII device. (Savannah patch #4789.) * serial.h: Declare usb_serdev_frame device descriptor. * stk500v2.c: Implementation of the AVRISP mkII handling. * usb_libusb.c: Add USB handling for short-frame delimited AVRISP mkII USB protocol; add distinction of different devices in usbdev_open(). * jtagmkII.c: Tell usbdev_open() to search for the JTAG ICE mkII. * usbdevs.h: (New file.) * Makefile.am: Add usbdevs.h, as well as some other forgotten files "make distcheck" complained about. * avrdude.conf.in: Add more aliases for the AVRISP mkII. * avrdude.1: Document how to use the AVRISP mkII. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@564 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-12 23:13:50 +00:00
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static double stk500v2_sck_to_us(const PROGRAMMER *pgm, unsigned char dur);
static int stk500v2_set_sck_period_mk2(const PROGRAMMER *pgm, double v);
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk600_set_sck_period(const PROGRAMMER *pgm, double v);
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static void stk600_setup_xprog(PROGRAMMER *pgm);
static void stk600_setup_isp(PROGRAMMER *pgm);
static int stk600_xprog_program_enable(const PROGRAMMER *pgm, const AVRPART *p);
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
Submitted by Brett Hagman: Add support for the "Wiring" board/bootloader * wiring.c: New file. * wiring.h: (Ditto.) * Makefile.am: Add new files. * stk500v2_private.h: Reorganize so some functions and struct pdata are globally known. * stk500v2.c: (Ditto.) * stk500v2.h: (Ditto.) * lexer.l: Add new programmer keywords. * config_gram.y: (Ditto.) * avrdude.conf.in: Add "wiring" programmer entry. * avrdude.1: Document the new programmer. * doc/avrdude.texi: (Ditto.) * AUTHORS: Add Brett Hagman. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@987 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-08-26 20:22:09 +00:00
void stk500v2_setup(PROGRAMMER * pgm)
main.c, pgm.c, pgm.h: Add setup and teardown hooks to the programmer definition. If present, call the setup hook immediately after finding the respective programmer object, and schedule the teardown hook to be called upon exit. This allows the programmer implementation to dynamically allocate private programmer data. avr910.c, butterfly.c, jtagmkI.c, jtagmkII.c, stk500v2.c, usbasp.c, usbtiny.c: Convert static programmer data into dynamically allocated data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@764 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-11-07 20:36:12 +00:00
{
if ((pgm->cookie = malloc(sizeof(struct pdata))) == 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("out of memory allocating private data\n");
main.c, pgm.c, pgm.h: Add setup and teardown hooks to the programmer definition. If present, call the setup hook immediately after finding the respective programmer object, and schedule the teardown hook to be called upon exit. This allows the programmer implementation to dynamically allocate private programmer data. avr910.c, butterfly.c, jtagmkI.c, jtagmkII.c, stk500v2.c, usbasp.c, usbtiny.c: Convert static programmer data into dynamically allocated data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@764 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-11-07 20:36:12 +00:00
exit(1);
}
memset(pgm->cookie, 0, sizeof(struct pdata));
PDATA(pgm)->command_sequence = 1;
* stk500v2_private.h (struct pdata): add boot_start * stk500v2.c: For the "flash" pseudo-memory of Xmega devices, distinguish addresses between "application" and "boot" area. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1081 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-04-19 13:32:34 +00:00
PDATA(pgm)->boot_start = ULONG_MAX;
main.c, pgm.c, pgm.h: Add setup and teardown hooks to the programmer definition. If present, call the setup hook immediately after finding the respective programmer object, and schedule the teardown hook to be called upon exit. This allows the programmer implementation to dynamically allocate private programmer data. avr910.c, butterfly.c, jtagmkI.c, jtagmkII.c, stk500v2.c, usbasp.c, usbtiny.c: Convert static programmer data into dynamically allocated data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@764 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-11-07 20:36:12 +00:00
}
Fix a serious memory corruption problem resulting out of the chaining of both, the stk500v2 and the jtagmkII programmers for some programming hardware (JTAG ICE mkII and AVR Dragon running in ISP, HVSP or PP mode), where both programmers have to maintain their private programmer data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@834 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-07-06 22:10:20 +00:00
static void stk500v2_jtagmkII_setup(PROGRAMMER * pgm)
{
void *mycookie, *theircookie;
if ((pgm->cookie = malloc(sizeof(struct pdata))) == 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("out of memory allocating private data\n");
Fix a serious memory corruption problem resulting out of the chaining of both, the stk500v2 and the jtagmkII programmers for some programming hardware (JTAG ICE mkII and AVR Dragon running in ISP, HVSP or PP mode), where both programmers have to maintain their private programmer data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@834 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-07-06 22:10:20 +00:00
exit(1);
}
memset(pgm->cookie, 0, sizeof(struct pdata));
PDATA(pgm)->command_sequence = 1;
/*
* Now, have the JTAG ICE mkII backend allocate its own private
* data. Store our own cookie in a safe place for the time being.
*/
mycookie = pgm->cookie;
jtagmkII_setup(pgm);
theircookie = pgm->cookie;
pgm->cookie = mycookie;
PDATA(pgm)->chained_pdata = theircookie;
}
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
static void stk500v2_jtag3_setup(PROGRAMMER * pgm)
{
void *mycookie, *theircookie;
if ((pgm->cookie = malloc(sizeof(struct pdata))) == 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("out of memory allocating private data\n");
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
exit(1);
}
memset(pgm->cookie, 0, sizeof(struct pdata));
PDATA(pgm)->command_sequence = 1;
/*
* Now, have the JTAGICE3 backend allocate its own private
* data. Store our own cookie in a safe place for the time being.
*/
mycookie = pgm->cookie;
jtag3_setup(pgm);
theircookie = pgm->cookie;
pgm->cookie = mycookie;
PDATA(pgm)->chained_pdata = theircookie;
}
Submitted by Brett Hagman: Add support for the "Wiring" board/bootloader * wiring.c: New file. * wiring.h: (Ditto.) * Makefile.am: Add new files. * stk500v2_private.h: Reorganize so some functions and struct pdata are globally known. * stk500v2.c: (Ditto.) * stk500v2.h: (Ditto.) * lexer.l: Add new programmer keywords. * config_gram.y: (Ditto.) * avrdude.conf.in: Add "wiring" programmer entry. * avrdude.1: Document the new programmer. * doc/avrdude.texi: (Ditto.) * AUTHORS: Add Brett Hagman. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@987 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-08-26 20:22:09 +00:00
void stk500v2_teardown(PROGRAMMER * pgm)
main.c, pgm.c, pgm.h: Add setup and teardown hooks to the programmer definition. If present, call the setup hook immediately after finding the respective programmer object, and schedule the teardown hook to be called upon exit. This allows the programmer implementation to dynamically allocate private programmer data. avr910.c, butterfly.c, jtagmkI.c, jtagmkII.c, stk500v2.c, usbasp.c, usbtiny.c: Convert static programmer data into dynamically allocated data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@764 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-11-07 20:36:12 +00:00
{
free(pgm->cookie);
}
Fix a serious memory corruption problem resulting out of the chaining of both, the stk500v2 and the jtagmkII programmers for some programming hardware (JTAG ICE mkII and AVR Dragon running in ISP, HVSP or PP mode), where both programmers have to maintain their private programmer data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@834 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-07-06 22:10:20 +00:00
static void stk500v2_jtagmkII_teardown(PROGRAMMER * pgm)
{
void *mycookie;
patch #9819: Address several leaks in SVN rev 1429 Submitted by Adrian Klieber git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1435 81a1dc3b-b13d-400b-aceb-764788c761c2
2020-03-14 22:34:45 +00:00
free(PDATA(pgm)->flash_pagecache);
free(PDATA(pgm)->eeprom_pagecache);
Fix a serious memory corruption problem resulting out of the chaining of both, the stk500v2 and the jtagmkII programmers for some programming hardware (JTAG ICE mkII and AVR Dragon running in ISP, HVSP or PP mode), where both programmers have to maintain their private programmer data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@834 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-07-06 22:10:20 +00:00
mycookie = pgm->cookie;
pgm->cookie = PDATA(pgm)->chained_pdata;
jtagmkII_teardown(pgm);
free(mycookie);
}
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
static void stk500v2_jtag3_teardown(PROGRAMMER * pgm)
{
void *mycookie;
mycookie = pgm->cookie;
pgm->cookie = PDATA(pgm)->chained_pdata;
jtag3_teardown(pgm);
free(mycookie);
}
main.c, pgm.c, pgm.h: Add setup and teardown hooks to the programmer definition. If present, call the setup hook immediately after finding the respective programmer object, and schedule the teardown hook to be called upon exit. This allows the programmer implementation to dynamically allocate private programmer data. avr910.c, butterfly.c, jtagmkI.c, jtagmkII.c, stk500v2.c, usbasp.c, usbtiny.c: Convert static programmer data into dynamically allocated data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@764 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-11-07 20:36:12 +00:00
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
static unsigned short
b2_to_u16(unsigned char *b)
{
unsigned short l;
l = b[0];
l += (unsigned)b[1] << 8;
return l;
}
Hide stk500v2_set_sck_period_mk2() behind an #if defined(HAVE_LIBUSB) as it is only used there (for the AVRISP mkII). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@590 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-19 21:19:42 +00:00
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_send_mk2(const PROGRAMMER *pgm, unsigned char *data, size_t len) {
Make the code compile warning-free: - declare a dummy "struct timezone" for some Win32 systems (MinGW) - fix a few printf() argument types - get rid off the prevailing "all filedescriptors are of type int" attitude The last item required a large sweep across the code, in order to replace all "int fd"s by "struct filedescriptor *fd"s, and pass pointers (as we cannot pass a union directly). In return, the code is now supposed to be fully 64-bit safe, rather than relying on a 64-bit pointer being converted to a (32-bit) int and back to a pointer as we did previously. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@694 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-12-11 12:47:35 +00:00
if (serial_send(&pgm->fd, data, len) != 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unable to send command to serial port\n");
Replace all occurences of exit() in potential library code by appropriate return values. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1301 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-05-16 15:52:25 +00:00
return -1;
Contributed by dcm@mit.edu: add support for the AVRISP mkII device. (Savannah patch #4789.) * serial.h: Declare usb_serdev_frame device descriptor. * stk500v2.c: Implementation of the AVRISP mkII handling. * usb_libusb.c: Add USB handling for short-frame delimited AVRISP mkII USB protocol; add distinction of different devices in usbdev_open(). * jtagmkII.c: Tell usbdev_open() to search for the JTAG ICE mkII. * usbdevs.h: (New file.) * Makefile.am: Add usbdevs.h, as well as some other forgotten files "make distcheck" complained about. * avrdude.conf.in: Add more aliases for the AVRISP mkII. * avrdude.1: Document how to use the AVRISP mkII. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@564 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-12 23:13:50 +00:00
}
return 0;
}
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
static unsigned short get_jtagisp_return_size(unsigned char cmd)
{
Resolve signed/unsigned comparisons in stk500.c and stk500v2.c
2022-07-24 18:41:42 +00:00
for (size_t i = 0; i < sizeof jtagispcmds / sizeof jtagispcmds[0]; i++)
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
if (jtagispcmds[i].cmd == cmd)
return jtagispcmds[i].size;
return 0;
}
/*
* Send the data as a JTAG ICE mkII encapsulated ISP packet.
* Unlike what AVR067 says, the packet gets a length of our
* response buffer prepended, and replies with RSP_SPI_DATA
* if successful.
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_jtagmkII_send(const PROGRAMMER *pgm, unsigned char *data, size_t len) {
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
unsigned char *cmdbuf;
int rv;
unsigned short sz;
sz = get_jtagisp_return_size(data[0]);
if (sz == 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unsupported encapsulated ISP command: %#x\n", data[0]);
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
return -1;
}
if (sz == SZ_READ_FLASH_EE) {
/*
* For CMND_READ_FLASH_ISP and CMND_READ_EEPROM_ISP, extract the
* size of the return data from the request. Note that the
* request itself has the size in big endian format, while we are
* supposed to deliver it in little endian.
*/
sz = 3 + (data[1] << 8) + data[2];
} else if (sz == SZ_SPI_MULTI) {
/*
* CMND_SPI_MULTI has the Rx size encoded in its 3rd byte.
*/
sz = 3 + data[2];
}
if ((cmdbuf = malloc(len + 3)) == NULL) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("out of memory for command packet\n");
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
exit(1);
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
PROGRAMMER *pgmcp = pgm_dup(pgm);
pgmcp->cookie = PDATA(pgm)->chained_pdata;
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
cmdbuf[0] = CMND_ISP_PACKET;
cmdbuf[1] = sz & 0xff;
cmdbuf[2] = (sz >> 8) & 0xff;
memcpy(cmdbuf + 3, data, len);
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
rv = jtagmkII_send(pgmcp, cmdbuf, len + 3);
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
free(cmdbuf);
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
pgm_free(pgmcp);
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
return rv;
}
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
/*
* Send the data as a JTAGICE3 encapsulated ISP packet.
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_jtag3_send(const PROGRAMMER *pgm, unsigned char *data, size_t len) {
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
unsigned char *cmdbuf;
int rv;
if ((cmdbuf = malloc(len + 1)) == NULL) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("out of memory for command packet\n");
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
exit(1);
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
PROGRAMMER *pgmcp = pgm_dup(pgm);
pgmcp->cookie = PDATA(pgm)->chained_pdata;
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
cmdbuf[0] = SCOPE_AVR_ISP;
memcpy(cmdbuf + 1, data, len);
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
rv = jtag3_send(pgmcp, cmdbuf, len + 1);
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
free(cmdbuf);
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
pgm_free(pgmcp);
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
return rv;
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_send(const PROGRAMMER *pgm, unsigned char *data, size_t len) {
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
unsigned char buf[275 + 6]; // max MESSAGE_BODY of 275 bytes, 6 bytes overhead
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
if (PDATA(pgm)->pgmtype == PGMTYPE_AVRISP_MKII ||
PDATA(pgm)->pgmtype == PGMTYPE_STK600)
Contributed by dcm@mit.edu: add support for the AVRISP mkII device. (Savannah patch #4789.) * serial.h: Declare usb_serdev_frame device descriptor. * stk500v2.c: Implementation of the AVRISP mkII handling. * usb_libusb.c: Add USB handling for short-frame delimited AVRISP mkII USB protocol; add distinction of different devices in usbdev_open(). * jtagmkII.c: Tell usbdev_open() to search for the JTAG ICE mkII. * usbdevs.h: (New file.) * Makefile.am: Add usbdevs.h, as well as some other forgotten files "make distcheck" complained about. * avrdude.conf.in: Add more aliases for the AVRISP mkII. * avrdude.1: Document how to use the AVRISP mkII. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@564 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-12 23:13:50 +00:00
return stk500v2_send_mk2(pgm, data, len);
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
else if (PDATA(pgm)->pgmtype == PGMTYPE_JTAGICE_MKII)
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
return stk500v2_jtagmkII_send(pgm, data, len);
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
else if (PDATA(pgm)->pgmtype == PGMTYPE_JTAGICE3)
return stk500v2_jtag3_send(pgm, data, len);
Contributed by dcm@mit.edu: add support for the AVRISP mkII device. (Savannah patch #4789.) * serial.h: Declare usb_serdev_frame device descriptor. * stk500v2.c: Implementation of the AVRISP mkII handling. * usb_libusb.c: Add USB handling for short-frame delimited AVRISP mkII USB protocol; add distinction of different devices in usbdev_open(). * jtagmkII.c: Tell usbdev_open() to search for the JTAG ICE mkII. * usbdevs.h: (New file.) * Makefile.am: Add usbdevs.h, as well as some other forgotten files "make distcheck" complained about. * avrdude.conf.in: Add more aliases for the AVRISP mkII. * avrdude.1: Document how to use the AVRISP mkII. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@564 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-12 23:13:50 +00:00
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
buf[0] = MESSAGE_START;
main.c, pgm.c, pgm.h: Add setup and teardown hooks to the programmer definition. If present, call the setup hook immediately after finding the respective programmer object, and schedule the teardown hook to be called upon exit. This allows the programmer implementation to dynamically allocate private programmer data. avr910.c, butterfly.c, jtagmkI.c, jtagmkII.c, stk500v2.c, usbasp.c, usbtiny.c: Convert static programmer data into dynamically allocated data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@764 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-11-07 20:36:12 +00:00
buf[1] = PDATA(pgm)->command_sequence;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
buf[2] = len / 256;
buf[3] = len % 256;
buf[4] = TOKEN;
memcpy(buf+5, data, len);
// calculate the XOR checksum
buf[5+len] = 0;
Resolve signed/unsigned comparisons in stk500.c and stk500v2.c
2022-07-24 18:41:42 +00:00
for (size_t i=0; i<5+len; i++)
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
buf[5+len] ^= buf[i];
DEBUG("STK500V2: stk500v2_send(");
Resolve signed/unsigned comparisons in stk500.c and stk500v2.c
2022-07-24 18:41:42 +00:00
for (size_t i=0; i<len+6; i++)
DEBUG("0x%02x ", buf[i]);
DEBUG(", %d)\n", (int) len+6);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
Make the code compile warning-free: - declare a dummy "struct timezone" for some Win32 systems (MinGW) - fix a few printf() argument types - get rid off the prevailing "all filedescriptors are of type int" attitude The last item required a large sweep across the code, in order to replace all "int fd"s by "struct filedescriptor *fd"s, and pass pointers (as we cannot pass a union directly). In return, the code is now supposed to be fully 64-bit safe, rather than relying on a 64-bit pointer being converted to a (32-bit) int and back to a pointer as we did previously. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@694 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-12-11 12:47:35 +00:00
if (serial_send(&pgm->fd, buf, len+6) != 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unable to send command to serial port\n");
Replace all occurences of exit() in potential library code by appropriate return values. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1301 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-05-16 15:52:25 +00:00
return -1;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
}
return 0;
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
int stk500v2_drain(const PROGRAMMER *pgm, int display) {
Make the code compile warning-free: - declare a dummy "struct timezone" for some Win32 systems (MinGW) - fix a few printf() argument types - get rid off the prevailing "all filedescriptors are of type int" attitude The last item required a large sweep across the code, in order to replace all "int fd"s by "struct filedescriptor *fd"s, and pass pointers (as we cannot pass a union directly). In return, the code is now supposed to be fully 64-bit safe, rather than relying on a 64-bit pointer being converted to a (32-bit) int and back to a pointer as we did previously. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@694 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-12-11 12:47:35 +00:00
return serial_drain(&pgm->fd, display);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_recv_mk2(const PROGRAMMER *pgm, unsigned char *msg,
Contributed by dcm@mit.edu: add support for the AVRISP mkII device. (Savannah patch #4789.) * serial.h: Declare usb_serdev_frame device descriptor. * stk500v2.c: Implementation of the AVRISP mkII handling. * usb_libusb.c: Add USB handling for short-frame delimited AVRISP mkII USB protocol; add distinction of different devices in usbdev_open(). * jtagmkII.c: Tell usbdev_open() to search for the JTAG ICE mkII. * usbdevs.h: (New file.) * Makefile.am: Add usbdevs.h, as well as some other forgotten files "make distcheck" complained about. * avrdude.conf.in: Add more aliases for the AVRISP mkII. * avrdude.1: Document how to use the AVRISP mkII. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@564 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-12 23:13:50 +00:00
size_t maxsize)
{
int rv;
Make the code compile warning-free: - declare a dummy "struct timezone" for some Win32 systems (MinGW) - fix a few printf() argument types - get rid off the prevailing "all filedescriptors are of type int" attitude The last item required a large sweep across the code, in order to replace all "int fd"s by "struct filedescriptor *fd"s, and pass pointers (as we cannot pass a union directly). In return, the code is now supposed to be fully 64-bit safe, rather than relying on a 64-bit pointer being converted to a (32-bit) int and back to a pointer as we did previously. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@694 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-12-11 12:47:35 +00:00
rv = serial_recv(&pgm->fd, msg, maxsize);
Contributed by dcm@mit.edu: add support for the AVRISP mkII device. (Savannah patch #4789.) * serial.h: Declare usb_serdev_frame device descriptor. * stk500v2.c: Implementation of the AVRISP mkII handling. * usb_libusb.c: Add USB handling for short-frame delimited AVRISP mkII USB protocol; add distinction of different devices in usbdev_open(). * jtagmkII.c: Tell usbdev_open() to search for the JTAG ICE mkII. * usbdevs.h: (New file.) * Makefile.am: Add usbdevs.h, as well as some other forgotten files "make distcheck" complained about. * avrdude.conf.in: Add more aliases for the AVRISP mkII. * avrdude.1: Document how to use the AVRISP mkII. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@564 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-12 23:13:50 +00:00
if (rv < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unable to receive from USB\n");
Contributed by dcm@mit.edu: add support for the AVRISP mkII device. (Savannah patch #4789.) * serial.h: Declare usb_serdev_frame device descriptor. * stk500v2.c: Implementation of the AVRISP mkII handling. * usb_libusb.c: Add USB handling for short-frame delimited AVRISP mkII USB protocol; add distinction of different devices in usbdev_open(). * jtagmkII.c: Tell usbdev_open() to search for the JTAG ICE mkII. * usbdevs.h: (New file.) * Makefile.am: Add usbdevs.h, as well as some other forgotten files "make distcheck" complained about. * avrdude.conf.in: Add more aliases for the AVRISP mkII. * avrdude.1: Document how to use the AVRISP mkII. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@564 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-12 23:13:50 +00:00
return -1;
}
return rv;
}
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_jtagmkII_recv(const PROGRAMMER *pgm, unsigned char *msg,
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
size_t maxsize)
{
int rv;
unsigned char *jtagmsg;
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
PROGRAMMER *pgmcp = pgm_dup(pgm);
pgmcp->cookie = PDATA(pgm)->chained_pdata;
rv = jtagmkII_recv(pgmcp, &jtagmsg);
pgm_free(pgmcp);
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
if (rv <= 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unable to receive\n");
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
return -1;
}
Resolve signed/unsigned comparisons in stk500.c and stk500v2.c
2022-07-24 18:41:42 +00:00
if ((size_t) rv - 1 > maxsize) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_warning("got %u bytes, have only room for %u bytes\n", (unsigned) rv - 1, (unsigned) maxsize);
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
rv = maxsize;
}
switch (jtagmsg[0]) {
case RSP_SPI_DATA:
break;
case RSP_FAILED:
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("receive failed\n");
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
return -1;
case RSP_ILLEGAL_MCU_STATE:
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("illegal MCU state\n");
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
return -1;
default:
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unknown status %d\n", jtagmsg[0]);
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
return -1;
}
memcpy(msg, jtagmsg + 1, rv - 1);
patch #9819: Address several leaks in SVN rev 1429 Submitted by Adrian Klieber git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1435 81a1dc3b-b13d-400b-aceb-764788c761c2
2020-03-14 22:34:45 +00:00
free(jtagmsg);
Submitted by John Voltz: add AVR053 oscillator calibration. * main.c: Add the -O option. * pgm.c: Add the hook for the perform_osccal() method. * pgm.h: (Ditto.) * stk500v2.c: Implement perform_osccal(). * avrdude.1: Document the -O option. * doc/avrdude.texi: (Ditto.) Partially closes bug #17487: AVR RC oscillator calibration routine not supported (feature request) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@671 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-09 14:34:24 +00:00
return rv;
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_jtag3_recv(const PROGRAMMER *pgm, unsigned char *msg,
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
size_t maxsize)
{
int rv;
unsigned char *jtagmsg;
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
PROGRAMMER *pgmcp = pgm_dup(pgm);
pgmcp->cookie = PDATA(pgm)->chained_pdata;
rv = jtag3_recv(pgmcp, &jtagmsg);
pgm_free(pgmcp);
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
if (rv <= 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unable to receive\n");
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
return -1;
}
Turn the usbpid parameter of the programmer into a list of PIDs. Make the JTAGICE3 programmer handle a list of PIDs, by trying each of them in sequence. Use a single, central jtag3_open_common() function to handle the common code of all jtag3_open_* functions. Centralize all USB VID/PID definitions in usbdevs.h. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1283 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-02-27 13:06:03 +00:00
/* Getting more data than expected is a normal case for the EDBG
implementation of JTAGICE3, as they always request a full 512
octets from the ICE. Thus, only complain at high verbose
levels. */
Resolve signed/unsigned comparisons in stk500.c and stk500v2.c
2022-07-24 18:41:42 +00:00
if ((size_t) rv - 1 > maxsize) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_debug("stk500v2_jtag3_recv(): got %u bytes, have only room for %u bytes\n", (unsigned) rv - 1, (unsigned) maxsize);
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
rv = maxsize;
}
if (jtagmsg[0] != SCOPE_AVR_ISP) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("message is not AVR ISP: 0x%02x\n", jtagmsg[0]);
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
free(jtagmsg);
return -1;
}
memcpy(msg, jtagmsg + 1, rv - 1);
free(jtagmsg);
return rv;
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_recv(const PROGRAMMER *pgm, unsigned char *msg, size_t maxsize) {
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
enum states { sINIT, sSTART, sSEQNUM, sSIZE1, sSIZE2, sTOKEN, sDATA, sCSUM, sDONE } state = sSTART;
stk500v2_recv(): Make length computation unsigned so it cannot accidentally become negative. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@767 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-01-26 08:01:51 +00:00
unsigned int msglen = 0;
unsigned int curlen = 0;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
int timeout = 0;
unsigned char c, checksum = 0;
bug #43626: Inconsistent timeouts in stk500v2 * stk500v2.c (stk500v2_recv): Add a reference to the bug report but don't change anything, lest to break it somehow git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1342 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-11-19 17:23:14 +00:00
/*
* The entire timeout handling here is not very consistent, see
*
* https://savannah.nongnu.org/bugs/index.php?43626
*/
Submitted by Brett Hagman: Add support for the "Wiring" board/bootloader * wiring.c: New file. * wiring.h: (Ditto.) * Makefile.am: Add new files. * stk500v2_private.h: Reorganize so some functions and struct pdata are globally known. * stk500v2.c: (Ditto.) * stk500v2.h: (Ditto.) * lexer.l: Add new programmer keywords. * config_gram.y: (Ditto.) * avrdude.conf.in: Add "wiring" programmer entry. * avrdude.1: Document the new programmer. * doc/avrdude.texi: (Ditto.) * AUTHORS: Add Brett Hagman. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@987 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-08-26 20:22:09 +00:00
long timeoutval = SERIAL_TIMEOUT; // seconds
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
double tstart, tnow;
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
if (PDATA(pgm)->pgmtype == PGMTYPE_AVRISP_MKII ||
PDATA(pgm)->pgmtype == PGMTYPE_STK600)
Contributed by dcm@mit.edu: add support for the AVRISP mkII device. (Savannah patch #4789.) * serial.h: Declare usb_serdev_frame device descriptor. * stk500v2.c: Implementation of the AVRISP mkII handling. * usb_libusb.c: Add USB handling for short-frame delimited AVRISP mkII USB protocol; add distinction of different devices in usbdev_open(). * jtagmkII.c: Tell usbdev_open() to search for the JTAG ICE mkII. * usbdevs.h: (New file.) * Makefile.am: Add usbdevs.h, as well as some other forgotten files "make distcheck" complained about. * avrdude.conf.in: Add more aliases for the AVRISP mkII. * avrdude.1: Document how to use the AVRISP mkII. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@564 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-12 23:13:50 +00:00
return stk500v2_recv_mk2(pgm, msg, maxsize);
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
else if (PDATA(pgm)->pgmtype == PGMTYPE_JTAGICE_MKII)
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
return stk500v2_jtagmkII_recv(pgm, msg, maxsize);
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
else if (PDATA(pgm)->pgmtype == PGMTYPE_JTAGICE3)
return stk500v2_jtag3_recv(pgm, msg, maxsize);
Contributed by dcm@mit.edu: add support for the AVRISP mkII device. (Savannah patch #4789.) * serial.h: Declare usb_serdev_frame device descriptor. * stk500v2.c: Implementation of the AVRISP mkII handling. * usb_libusb.c: Add USB handling for short-frame delimited AVRISP mkII USB protocol; add distinction of different devices in usbdev_open(). * jtagmkII.c: Tell usbdev_open() to search for the JTAG ICE mkII. * usbdevs.h: (New file.) * Makefile.am: Add usbdevs.h, as well as some other forgotten files "make distcheck" complained about. * avrdude.conf.in: Add more aliases for the AVRISP mkII. * avrdude.1: Document how to use the AVRISP mkII. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@564 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-12 23:13:50 +00:00
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
DEBUG("STK500V2: stk500v2_recv(): ");
Establish avr_*timestamp() routines for timing
2022-11-25 22:09:02 +00:00
tstart = avr_timestamp();
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
while ( (state != sDONE ) && (!timeout) ) {
Make the code compile warning-free: - declare a dummy "struct timezone" for some Win32 systems (MinGW) - fix a few printf() argument types - get rid off the prevailing "all filedescriptors are of type int" attitude The last item required a large sweep across the code, in order to replace all "int fd"s by "struct filedescriptor *fd"s, and pass pointers (as we cannot pass a union directly). In return, the code is now supposed to be fully 64-bit safe, rather than relying on a 64-bit pointer being converted to a (32-bit) int and back to a pointer as we did previously. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@694 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-12-11 12:47:35 +00:00
if (serial_recv(&pgm->fd, &c, 1) < 0)
Revert the serial_recv() timeout handling as it used to be before my yesterday's changes (i.e. before rev. 1.10 of ser_posix.c), that is, exit(1) in case of a timeout. Previously, the upper layers didn't see the timeout at all. Quite possible that some of these drivers could handle a timeout more intelligently though. At least for the rather sophisticated STK500v2 protocol, I think it should be possible to retry the request. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@460 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-11 17:09:22 +00:00
goto timedout;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
DEBUG("0x%02x ",c);
checksum ^= c;
switch (state) {
case sSTART:
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
DEBUGRECV("hoping for start token ...");
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
if (c == MESSAGE_START) {
DEBUGRECV("got it\n");
checksum = MESSAGE_START;
state = sSEQNUM;
} else
DEBUGRECV("sorry\n");
break;
case sSEQNUM:
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
DEBUGRECV("hoping for sequence ...\n");
main.c, pgm.c, pgm.h: Add setup and teardown hooks to the programmer definition. If present, call the setup hook immediately after finding the respective programmer object, and schedule the teardown hook to be called upon exit. This allows the programmer implementation to dynamically allocate private programmer data. avr910.c, butterfly.c, jtagmkI.c, jtagmkII.c, stk500v2.c, usbasp.c, usbtiny.c: Convert static programmer data into dynamically allocated data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@764 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-11-07 20:36:12 +00:00
if (c == PDATA(pgm)->command_sequence) {
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
DEBUGRECV("got it, incrementing\n");
state = sSIZE1;
main.c, pgm.c, pgm.h: Add setup and teardown hooks to the programmer definition. If present, call the setup hook immediately after finding the respective programmer object, and schedule the teardown hook to be called upon exit. This allows the programmer implementation to dynamically allocate private programmer data. avr910.c, butterfly.c, jtagmkI.c, jtagmkII.c, stk500v2.c, usbasp.c, usbtiny.c: Convert static programmer data into dynamically allocated data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@764 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-11-07 20:36:12 +00:00
PDATA(pgm)->command_sequence++;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
} else {
DEBUGRECV("sorry\n");
state = sSTART;
}
break;
case sSIZE1:
DEBUGRECV("hoping for size LSB\n");
stk500v2_recv(): Make length computation unsigned so it cannot accidentally become negative. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@767 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-01-26 08:01:51 +00:00
msglen = (unsigned)c * 256;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
state = sSIZE2;
break;
case sSIZE2:
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
DEBUGRECV("hoping for size MSB ...");
stk500v2_recv(): Make length computation unsigned so it cannot accidentally become negative. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@767 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-01-26 08:01:51 +00:00
msglen += (unsigned)c;
DEBUG(" msg is %u bytes\n",msglen);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
state = sTOKEN;
break;
case sTOKEN:
if (c == TOKEN) state = sDATA;
else state = sSTART;
break;
case sDATA:
if (curlen < maxsize) {
msg[curlen] = c;
} else {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("buffer too small, received %d byte into %u byte buffer\n",
curlen, (unsigned int) maxsize);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -2;
}
if ((curlen == 0) && (msg[0] == ANSWER_CKSUM_ERROR)) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("previous packet sent with wrong checksum\n");
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -3;
}
curlen++;
if (curlen == msglen) state = sCSUM;
break;
case sCSUM:
if (checksum == 0) {
state = sDONE;
} else {
state = sSTART;
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("wrong checksum\n");
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -4;
}
break;
default:
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unknown state\n");
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -5;
} /* switch */
Establish avr_*timestamp() routines for timing
2022-11-25 22:09:02 +00:00
tnow = avr_timestamp();
if (tnow-tstart > timeoutval) {
Revert the serial_recv() timeout handling as it used to be before my yesterday's changes (i.e. before rev. 1.10 of ser_posix.c), that is, exit(1) in case of a timeout. Previously, the upper layers didn't see the timeout at all. Quite possible that some of these drivers could handle a timeout more intelligently though. At least for the rather sophisticated STK500v2 protocol, I think it should be possible to retry the request. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@460 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-11 17:09:22 +00:00
timedout:
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("timeout\n");
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -1;
}
} /* while */
DEBUG("\n");
stk500v2_recv(): Make length computation unsigned so it cannot accidentally become negative. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@767 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-01-26 08:01:51 +00:00
return (int)(msglen+6);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
int stk500v2_getsync(const PROGRAMMER *pgm) {
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
int tries = 0;
unsigned char buf[1], resp[32];
int status;
DEBUG("STK500V2: stk500v2_getsync()\n");
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
if (PDATA(pgm)->pgmtype == PGMTYPE_JTAGICE_MKII ||
PDATA(pgm)->pgmtype == PGMTYPE_JTAGICE3)
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
return 0;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
retry:
tries++;
// send the sync command and see if we can get there
buf[0] = CMD_SIGN_ON;
stk500v2_send(pgm, buf, 1);
// try to get the response back and see where we got
status = stk500v2_recv(pgm, resp, sizeof(resp));
// if we got bytes returned, check to see what came back
if (status > 0) {
Minor cosmetic changes: STK500 firmware version numbers are M.NN, so always display the minor number as two digits. Examine the response to the sign-on command to see which programmer hardware we are talking to, and then restrict the STK500 topcard display to devices detected as STK500. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@614 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-08-18 21:23:34 +00:00
if ((resp[0] == CMD_SIGN_ON) && (resp[1] == STATUS_CMD_OK) &&
(status > 3)) {
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
// success!
Minor cosmetic changes: STK500 firmware version numbers are M.NN, so always display the minor number as two digits. Examine the response to the sign-on command to see which programmer hardware we are talking to, and then restrict the STK500 topcard display to devices detected as STK500. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@614 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-08-18 21:23:34 +00:00
unsigned int siglen = resp[2];
if (siglen >= strlen("STK500_2") &&
memcmp(resp + 3, "STK500_2", strlen("STK500_2")) == 0) {
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
PDATA(pgm)->pgmtype = PGMTYPE_STK500;
Minor cosmetic changes: STK500 firmware version numbers are M.NN, so always display the minor number as two digits. Examine the response to the sign-on command to see which programmer hardware we are talking to, and then restrict the STK500 topcard display to devices detected as STK500. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@614 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-08-18 21:23:34 +00:00
} else if (siglen >= strlen("AVRISP_2") &&
memcmp(resp + 3, "AVRISP_2", strlen("AVRISP_2")) == 0) {
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
PDATA(pgm)->pgmtype = PGMTYPE_AVRISP;
Minor cosmetic changes: STK500 firmware version numbers are M.NN, so always display the minor number as two digits. Examine the response to the sign-on command to see which programmer hardware we are talking to, and then restrict the STK500 topcard display to devices detected as STK500. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@614 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-08-18 21:23:34 +00:00
} else if (siglen >= strlen("AVRISP_MK2") &&
memcmp(resp + 3, "AVRISP_MK2", strlen("AVRISP_MK2")) == 0) {
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
PDATA(pgm)->pgmtype = PGMTYPE_AVRISP_MKII;
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
} else if (siglen >= strlen("STK600") &&
memcmp(resp + 3, "STK600", strlen("STK600")) == 0) {
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
PDATA(pgm)->pgmtype = PGMTYPE_STK600;
Minor cosmetic changes: STK500 firmware version numbers are M.NN, so always display the minor number as two digits. Examine the response to the sign-on command to see which programmer hardware we are talking to, and then restrict the STK500 topcard display to devices detected as STK500. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@614 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-08-18 21:23:34 +00:00
} else {
resp[siglen + 3] = 0;
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_notice("stk500v2_getsync(): got response from unknown "
"programmer %s, assuming STK500\n", resp + 3);
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
PDATA(pgm)->pgmtype = PGMTYPE_STK500;
Minor cosmetic changes: STK500 firmware version numbers are M.NN, so always display the minor number as two digits. Examine the response to the sign-on command to see which programmer hardware we are talking to, and then restrict the STK500 topcard display to devices detected as STK500. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@614 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-08-18 21:23:34 +00:00
}
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_debug("stk500v2_getsync(): found %s programmer\n", pgmname[PDATA(pgm)->pgmtype]);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return 0;
} else {
Submitted by Brett Hagman: Add support for the "Wiring" board/bootloader * wiring.c: New file. * wiring.h: (Ditto.) * Makefile.am: Add new files. * stk500v2_private.h: Reorganize so some functions and struct pdata are globally known. * stk500v2.c: (Ditto.) * stk500v2.h: (Ditto.) * lexer.l: Add new programmer keywords. * config_gram.y: (Ditto.) * avrdude.conf.in: Add "wiring" programmer entry. * avrdude.1: Document the new programmer. * doc/avrdude.texi: (Ditto.) * AUTHORS: Add Brett Hagman. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@987 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-08-26 20:22:09 +00:00
if (tries > RETRIES) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("cannot communicate with device: resp=0x%02x\n", resp[0]);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -6;
} else
goto retry;
}
// or if we got a timeout
} else if (status == -1) {
Submitted by Brett Hagman: Add support for the "Wiring" board/bootloader * wiring.c: New file. * wiring.h: (Ditto.) * Makefile.am: Add new files. * stk500v2_private.h: Reorganize so some functions and struct pdata are globally known. * stk500v2.c: (Ditto.) * stk500v2.h: (Ditto.) * lexer.l: Add new programmer keywords. * config_gram.y: (Ditto.) * avrdude.conf.in: Add "wiring" programmer entry. * avrdude.1: Document the new programmer. * doc/avrdude.texi: (Ditto.) * AUTHORS: Add Brett Hagman. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@987 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-08-26 20:22:09 +00:00
if (tries > RETRIES) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("timeout communicating with programmer\n");
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -1;
} else
goto retry;
// or any other error
} else {
Submitted by Brett Hagman: Add support for the "Wiring" board/bootloader * wiring.c: New file. * wiring.h: (Ditto.) * Makefile.am: Add new files. * stk500v2_private.h: Reorganize so some functions and struct pdata are globally known. * stk500v2.c: (Ditto.) * stk500v2.h: (Ditto.) * lexer.l: Add new programmer keywords. * config_gram.y: (Ditto.) * avrdude.conf.in: Add "wiring" programmer entry. * avrdude.1: Document the new programmer. * doc/avrdude.texi: (Ditto.) * AUTHORS: Add Brett Hagman. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@987 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-08-26 20:22:09 +00:00
if (tries > RETRIES) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unable to communicate with programmer (%d)\n", status);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
} else
goto retry;
}
return 0;
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_command(const PROGRAMMER *pgm, unsigned char *buf,
Submitted by John Voltz: add AVR053 oscillator calibration. * main.c: Add the -O option. * pgm.c: Add the hook for the perform_osccal() method. * pgm.h: (Ditto.) * stk500v2.c: Implement perform_osccal(). * avrdude.1: Document the -O option. * doc/avrdude.texi: (Ditto.) Partially closes bug #17487: AVR RC oscillator calibration routine not supported (feature request) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@671 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-09 14:34:24 +00:00
size_t len, size_t maxlen) {
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
int tries = 0;
int status;
DEBUG("STK500V2: stk500v2_command(");
Resolve signed/unsigned comparisons in stk500.c and stk500v2.c
2022-07-24 18:41:42 +00:00
for (size_t i=0; i<len; i++)
DEBUG("0x%02x ",buf[i]);
DEBUG(", %d)\n", (int) len);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
retry:
tries++;
// send the command to the programmer
stk500v2_send(pgm,buf,len);
// attempt to read the status back
status = stk500v2_recv(pgm,buf,maxlen);
Print meaningful ISP clock error message Currently, only some JTAG3 compatible programmers are able to detect that the connected target's main clock speed actually is too low for for the programmer. In this case, these will return 0xCC.
2022-11-27 21:52:45 +00:00
DEBUG("STK500V2: stk500v2_command() received content: [ ");
for (size_t i=0; i<len; i++)
DEBUG("0x%02x ",buf[i]);
DEBUG("], length %d\n", (int) len);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
// if we got a successful readback, return
if (status > 0) {
DEBUG(" = %d\n",status);
Submitted by John Voltz: add AVR053 oscillator calibration. * main.c: Add the -O option. * pgm.c: Add the hook for the perform_osccal() method. * pgm.h: (Ditto.) * stk500v2.c: Implement perform_osccal(). * avrdude.1: Document the -O option. * doc/avrdude.texi: (Ditto.) Partially closes bug #17487: AVR RC oscillator calibration routine not supported (feature request) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@671 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-09 14:34:24 +00:00
if (status < 2) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("short reply\n");
Submitted by John Voltz: add AVR053 oscillator calibration. * main.c: Add the -O option. * pgm.c: Add the hook for the perform_osccal() method. * pgm.h: (Ditto.) * stk500v2.c: Implement perform_osccal(). * avrdude.1: Document the -O option. * doc/avrdude.texi: (Ditto.) Partially closes bug #17487: AVR RC oscillator calibration routine not supported (feature request) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@671 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-09 14:34:24 +00:00
return -1;
}
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
if (buf[0] == CMD_XPROG_SETMODE || buf[0] == CMD_XPROG) {
/*
* Decode XPROG wrapper errors.
*/
const char *msg;
int i;
/*
* For CMD_XPROG_SETMODE, the status is returned in buf[1].
* For CMD_XPROG, buf[1] contains the XPRG_CMD_* command, and
* buf[2] contains the status.
*/
i = buf[0] == CMD_XPROG_SETMODE? 1: 2;
if (buf[i] != XPRG_ERR_OK) {
switch (buf[i]) {
case XPRG_ERR_FAILED: msg = "Failed"; break;
case XPRG_ERR_COLLISION: msg = "Collision"; break;
case XPRG_ERR_TIMEOUT: msg = "Timeout"; break;
default: msg = "Unknown"; break;
}
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("%s: %s\n",
buf[0]==CMD_XPROG_SETMODE? "CMD_XPROG_SETMODE": "CMD_XPROG", msg);
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
return -1;
}
return 0;
} else {
/*
* Decode STK500v2 errors.
*/
In stk500v2_command(), detect warning status codes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@951 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-12-14 17:17:33 +00:00
if (buf[1] >= STATUS_CMD_TOUT && buf[1] < 0xa0) {
const char *msg;
char msgbuf[30];
switch (buf[1]) {
case STATUS_CMD_TOUT:
msg = "Command timed out";
break;
case STATUS_RDY_BSY_TOUT:
msg = "Sampling of the RDY/nBSY pin timed out";
break;
case STATUS_SET_PARAM_MISSING:
msg = "The `Set Device Parameters' have not been "
"executed in advance of this command";
Read programmer serial number from libusb or hidusb (#1223) * Silence compiler warnings * Fix a fall-through error in switch * Slightly reformat programmer info printout Now the target voltage is printed out along with the HW version, FW version and serial number, and they all allign nicely. When in ISP mode, the "SCK period" value is printed _after_ the target voltage readout, and _before_ the additional clocks the programmer holds, such as JTAG, PDI and UPDI clocks. Co-authored-by: Stefan Rueger <stefan.rueger@urclocks.com>
2022-12-21 19:11:31 +00:00
break;
In stk500v2_command(), detect warning status codes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@951 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-12-14 17:17:33 +00:00
default:
sprintf(msgbuf, "unknown, code 0x%02x", buf[1]);
msg = msgbuf;
break;
}
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_warning("%s\n", msg);
In stk500v2_command(), treat warnings as errors rather than success. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1006 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-13 15:14:53 +00:00
} else if (buf[1] == STATUS_CMD_OK) {
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
return status;
In stk500v2_command(), treat warnings as errors rather than success. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1006 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-13 15:14:53 +00:00
} else if (buf[1] == STATUS_CMD_FAILED) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("command failed\n");
Print meaningful ISP clock error message Currently, only some JTAG3 compatible programmers are able to detect that the connected target's main clock speed actually is too low for for the programmer. In this case, these will return 0xCC.
2022-11-27 21:52:45 +00:00
} else if (buf[1] == STATUS_CLOCK_ERROR) {
pmsg_error("target clock speed error\n");
return -2;
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
} else if (buf[1] == STATUS_CMD_UNKNOWN) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unknown command\n");
In stk500v2_command(), treat warnings as errors rather than success. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1006 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-13 15:14:53 +00:00
} else {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unknown status 0x%02x\n", buf[1]);
In stk500v2_command(), treat warnings as errors rather than success. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1006 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-13 15:14:53 +00:00
}
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
return -1;
}
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
}
// otherwise try to sync up again
status = stk500v2_getsync(pgm);
if (status != 0) {
Submitted by Brett Hagman: Add support for the "Wiring" board/bootloader * wiring.c: New file. * wiring.h: (Ditto.) * Makefile.am: Add new files. * stk500v2_private.h: Reorganize so some functions and struct pdata are globally known. * stk500v2.c: (Ditto.) * stk500v2.h: (Ditto.) * lexer.l: Add new programmer keywords. * config_gram.y: (Ditto.) * avrdude.conf.in: Add "wiring" programmer entry. * avrdude.1: Document the new programmer. * doc/avrdude.texi: (Ditto.) * AUTHORS: Add Brett Hagman. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@987 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-08-26 20:22:09 +00:00
if (tries > RETRIES) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("failed to execute command 0x%02x\n", buf[0]);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -1;
} else
goto retry;
}
DEBUG(" = 0\n");
return 0;
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_cmd(const PROGRAMMER *pgm, const unsigned char *cmd,
bug #39794: warnings when building avrdude 6.0rc1 under CentOS 6.4 * buspirate.c: Turn the "cmd" argument of the various methods into a "const unsigned char *"; while doing this, declare all arrays being passed as arguments to be pointers rather than arrays, as the latter obfuscates the way arrays are being passed to a callee in C. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1196 81a1dc3b-b13d-400b-aceb-764788c761c2
2013-09-02 20:22:53 +00:00
unsigned char *res)
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
{
unsigned char buf[8];
int result;
DEBUG("STK500V2: stk500v2_cmd(%02x,%02x,%02x,%02x)\n",cmd[0],cmd[1],cmd[2],cmd[3]);
buf[0] = CMD_SPI_MULTI;
buf[1] = 4;
buf[2] = 4;
buf[3] = 0;
buf[4] = cmd[0];
buf[5] = cmd[1];
buf[6] = cmd[2];
buf[7] = cmd[3];
result = stk500v2_command(pgm, buf, 8, sizeof(buf));
Submitted by John Voltz: add AVR053 oscillator calibration. * main.c: Add the -O option. * pgm.c: Add the hook for the perform_osccal() method. * pgm.h: (Ditto.) * stk500v2.c: Implement perform_osccal(). * avrdude.1: Document the -O option. * doc/avrdude.texi: (Ditto.) Partially closes bug #17487: AVR RC oscillator calibration routine not supported (feature request) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@671 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-09 14:34:24 +00:00
if (result < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("send command failed\n");
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -1;
Submitted by John Voltz: add AVR053 oscillator calibration. * main.c: Add the -O option. * pgm.c: Add the hook for the perform_osccal() method. * pgm.h: (Ditto.) * stk500v2.c: Implement perform_osccal(). * avrdude.1: Document the -O option. * doc/avrdude.texi: (Ditto.) Partially closes bug #17487: AVR RC oscillator calibration routine not supported (feature request) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@671 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-09 14:34:24 +00:00
} else if (result < 6) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("short reply, len = %d\n", result);
Submitted by John Voltz: add AVR053 oscillator calibration. * main.c: Add the -O option. * pgm.c: Add the hook for the perform_osccal() method. * pgm.h: (Ditto.) * stk500v2.c: Implement perform_osccal(). * avrdude.1: Document the -O option. * doc/avrdude.texi: (Ditto.) Partially closes bug #17487: AVR RC oscillator calibration routine not supported (feature request) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@671 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-09 14:34:24 +00:00
return -1;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
}
res[0] = buf[2];
res[1] = buf[3];
res[2] = buf[4];
res[3] = buf[5];
return 0;
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_jtag3_cmd(const PROGRAMMER *pgm, const unsigned char *cmd,
bug #39794: warnings when building avrdude 6.0rc1 under CentOS 6.4 * buspirate.c: Turn the "cmd" argument of the various methods into a "const unsigned char *"; while doing this, declare all arrays being passed as arguments to be pointers rather than arrays, as the latter obfuscates the way arrays are being passed to a callee in C. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1196 81a1dc3b-b13d-400b-aceb-764788c761c2
2013-09-02 20:22:53 +00:00
unsigned char *res)
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
{
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("not available in JTAGICE3\n");
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
return -1;
}
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
/*
* issue the 'chip erase' command to the AVR device
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_chip_erase(const PROGRAMMER *pgm, const AVRPART *p) {
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
int result;
unsigned char buf[16];
if (p->op[AVR_OP_CHIP_ERASE] == NULL) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("chip erase instruction not defined for part %s\n", p->desc);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -1;
}
pgm->pgm_led(pgm, ON);
buf[0] = CMD_CHIP_ERASE_ISP;
buf[1] = p->chip_erase_delay / 1000;
buf[2] = 0; // use delay (?)
Initialise memory before avr_set_bits() calls in stk500.c and stk500v2.c
2022-07-24 17:20:35 +00:00
memset(buf+3, 0, 4);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
avr_set_bits(p->op[AVR_OP_CHIP_ERASE], buf+3);
result = stk500v2_command(pgm, buf, 7, sizeof(buf));
Fix JTAGICE mkII by removing the initialize call after CMD_CHIP_ERASE_ISP (#1172) - The extra call to initialize() sends CMD_ENTER_PROGMODE_ISP causing a flash verification mismatch
2022-11-22 01:17:19 +00:00
usleep(p->chip_erase_delay); // should not be needed
if (PDATA(pgm)->pgmtype != PGMTYPE_JTAGICE_MKII) { // skip for JTAGICE mkII (FW v7.39)
pgm->initialize(pgm, p); // should not be needed
}
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
pgm->pgm_led(pgm, OFF);
bug #22882: Erase Cycle Counter does not work for stk500v2 * stk500v2.c (stk500v2_chip_erase,stk500hv_chip_erase): Return the expected 0 for success rather than a protocol-dependant number. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@785 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-11-04 13:15:21 +00:00
return result >= 0? 0: -1;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
}
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
/*
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
* issue the 'chip erase' command to the AVR device, generic HV mode
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500hv_chip_erase(const PROGRAMMER *pgm, const AVRPART *p, enum hvmode mode) {
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
int result;
unsigned char buf[3];
pgm->pgm_led(pgm, ON);
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
if (mode == PPMODE) {
buf[0] = CMD_CHIP_ERASE_PP;
buf[1] = p->chiperasepulsewidth;
buf[2] = p->chiperasepolltimeout;
} else {
buf[0] = CMD_CHIP_ERASE_HVSP;
Unreverse the argument order for CMD_CHIP_ERASE_HVSP; Atmel says AVR068 is right, and stk500.exe is wrong. (This reverts rev 1.19) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@611 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-08-17 16:01:34 +00:00
buf[1] = p->chiperasepolltimeout;
buf[2] = p->chiperasetime;
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
}
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
result = stk500v2_command(pgm, buf, 3, sizeof(buf));
usleep(p->chip_erase_delay);
pgm->initialize(pgm, p);
pgm->pgm_led(pgm, OFF);
bug #22882: Erase Cycle Counter does not work for stk500v2 * stk500v2.c (stk500v2_chip_erase,stk500hv_chip_erase): Return the expected 0 for success rather than a protocol-dependant number. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@785 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-11-04 13:15:21 +00:00
return result >= 0? 0: -1;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
}
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
/*
* issue the 'chip erase' command to the AVR device, parallel mode
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500pp_chip_erase(const PROGRAMMER *pgm, const AVRPART *p) {
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
return stk500hv_chip_erase(pgm, p, PPMODE);
}
/*
* issue the 'chip erase' command to the AVR device, HVSP mode
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500hvsp_chip_erase(const PROGRAMMER *pgm, const AVRPART *p) {
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
return stk500hv_chip_erase(pgm, p, HVSPMODE);
}
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
static struct
{
unsigned int state;
const char *description;
} connection_status[] =
{
Deprecate mosi/miso in favour of sdo/sdi See https://www.oshwa.org/a-resolution-to-redefine-spi-signal-names/
2022-11-22 17:04:05 +00:00
{ STATUS_CONN_FAIL_SDO, "SDO fail" },
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
{ STATUS_CONN_FAIL_RST, "RST fail" },
{ STATUS_CONN_FAIL_SCK, "SCK fail" },
{ STATUS_TGT_NOT_DETECTED, "Target not detected" },
{ STATUS_TGT_REVERSE_INSERTED, "Target reverse inserted" },
};
bug #29636: AVRDude issues invalid CMD_CHECK_TARGET_CONNECTION on the AVRISP-MKII * stk500v2.c (stk500v2_program_enable): Rewrite the logic to explain ISP activation failures. * stk500v2_private.h: Fix the various STATUS_* constants; AVR069 and AVR079 disagreed in their values, even though they are apparently implementing the same logic behind. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@972 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-08-16 18:15:19 +00:00
/*
* Max length of returned message is the sum of all the description
* strings in the table above, plus 2 characters for separation.
* Currently, this is 76 chars.
*/
static void
stk500v2_translate_conn_status(unsigned char status, char *msg)
{
size_t i;
int need_comma;
*msg = 0;
need_comma = 0;
for (i = 0;
i < sizeof connection_status / sizeof connection_status[0];
i++)
{
if ((status & connection_status[i].state) != 0)
{
if (need_comma)
strcat(msg, ", ");
strcat(msg, connection_status[i].description);
need_comma = 1;
}
}
if (*msg == 0)
sprintf(msg, "Unknown status 0x%02x", status);
}
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
/*
* issue the 'program enable' command to the AVR device
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_program_enable(const PROGRAMMER *pgm, const AVRPART *p) {
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
unsigned char buf[16];
bug #29636: AVRDude issues invalid CMD_CHECK_TARGET_CONNECTION on the AVRISP-MKII * stk500v2.c (stk500v2_program_enable): Rewrite the logic to explain ISP activation failures. * stk500v2_private.h: Fix the various STATUS_* constants; AVR069 and AVR079 disagreed in their values, even though they are apparently implementing the same logic behind. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@972 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-08-16 18:15:19 +00:00
char msg[100]; /* see remarks above about size needed */
Submitted by "Malte" and John McCorquodale: patch #7876 JTAGICE mkII fails to connect to attiny if debugwire is enabled AND target has a very slow clock * jtagmkII.c (jtagmkII_getsync): When leaving debugWIRE mode temporarily, immediately retry with ISP, rather than leaving. * stk500v2 (stk500v2_program_enable): Implemented similar logic for the JTAGICE3. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1183 81a1dc3b-b13d-400b-aceb-764788c761c2
2013-05-17 16:23:55 +00:00
int rv, tries;
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
PDATA(pgm)->lastpart = p;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
if (p->op[AVR_OP_PGM_ENABLE] == NULL) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("program enable instruction not defined for part %s\n", p->desc);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -1;
}
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
if (PDATA(pgm)->pgmtype == PGMTYPE_STK500 ||
PDATA(pgm)->pgmtype == PGMTYPE_STK600)
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
/* Activate AVR-style (low active) RESET */
stk500v2_setparm_real(pgm, PARAM_RESET_POLARITY, 0x01);
Submitted by "Malte" and John McCorquodale: patch #7876 JTAGICE mkII fails to connect to attiny if debugwire is enabled AND target has a very slow clock * jtagmkII.c (jtagmkII_getsync): When leaving debugWIRE mode temporarily, immediately retry with ISP, rather than leaving. * stk500v2 (stk500v2_program_enable): Implemented similar logic for the JTAGICE3. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1183 81a1dc3b-b13d-400b-aceb-764788c761c2
2013-05-17 16:23:55 +00:00
tries = 0;
retry:
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
buf[0] = CMD_ENTER_PROGMODE_ISP;
buf[1] = p->timeout;
buf[2] = p->stabdelay;
buf[3] = p->cmdexedelay;
stk500v2_program_enable(): Fix a typo (synchloops vs. synchcycles). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@598 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-08-09 19:56:48 +00:00
buf[4] = p->synchloops;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
buf[5] = p->bytedelay;
buf[6] = p->pollvalue;
buf[7] = p->pollindex;
Initialise memory before avr_set_bits() calls in stk500.c and stk500v2.c
2022-07-24 17:20:35 +00:00
memset(buf+8, 0, 4);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
avr_set_bits(p->op[AVR_OP_PGM_ENABLE], buf+8);
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
rv = stk500v2_command(pgm, buf, 12, sizeof(buf));
if (rv < 0) {
bug #29636: AVRDude issues invalid CMD_CHECK_TARGET_CONNECTION on the AVRISP-MKII * stk500v2.c (stk500v2_program_enable): Rewrite the logic to explain ISP activation failures. * stk500v2_private.h: Fix the various STATUS_* constants; AVR069 and AVR079 disagreed in their values, even though they are apparently implementing the same logic behind. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@972 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-08-16 18:15:19 +00:00
switch (PDATA(pgm)->pgmtype)
{
case PGMTYPE_STK600:
case PGMTYPE_AVRISP_MKII:
if (stk500v2_getparm(pgm, PARAM_STATUS_TGT_CONN, &buf[0]) != 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("cannot get connection status\n");
bug #29636: AVRDude issues invalid CMD_CHECK_TARGET_CONNECTION on the AVRISP-MKII * stk500v2.c (stk500v2_program_enable): Rewrite the logic to explain ISP activation failures. * stk500v2_private.h: Fix the various STATUS_* constants; AVR069 and AVR079 disagreed in their values, even though they are apparently implementing the same logic behind. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@972 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-08-16 18:15:19 +00:00
} else {
stk500v2_translate_conn_status(buf[0], msg);
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("bad AVRISPmkII connection status: %s\n", msg);
bug #29636: AVRDude issues invalid CMD_CHECK_TARGET_CONNECTION on the AVRISP-MKII * stk500v2.c (stk500v2_program_enable): Rewrite the logic to explain ISP activation failures. * stk500v2_private.h: Fix the various STATUS_* constants; AVR069 and AVR079 disagreed in their values, even though they are apparently implementing the same logic behind. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@972 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-08-16 18:15:19 +00:00
}
break;
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
* jtag3.c: Make jtag3_command() public * jtag3.h: (Dito.) * jtag3_private.h: Add two new commands * stk500v2.c: Implement the "MonCon disable" hack that allows temporarily falling back to ISP when trying to talk to a part that has debugWIRE enabled git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1123 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 21:03:06 +00:00
case PGMTYPE_JTAGICE3:
Offload the programming interface info from part->flags to part->prog_modes flags now just hold parameters of the JTAG interface and some secondary serial, parallel, pseudo parallel info. This separation brings clarity. It used to be hard to augur whether a part has an ISP interface: (part->flags & (AVRPART_HAS_PDI | AVRPART_AVR32 | AVRPART_HAS_TPI | AVRPART_HAS_UPDI)) == 0 && (part->flags & AVRPART_SERIALOK) != 0 or had HVSP or HVPP capability, for that matter. Now it is just, eg, part->prog_modes & PM_ISP part->prog_modes & PM_HVPP
2022-08-30 15:33:42 +00:00
if (buf[1] == STATUS_CMD_FAILED && (p->prog_modes & PM_debugWIRE)) {
* jtag3.c: Make jtag3_command() public * jtag3.h: (Dito.) * jtag3_private.h: Add two new commands * stk500v2.c: Implement the "MonCon disable" hack that allows temporarily falling back to ISP when trying to talk to a part that has debugWIRE enabled git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1123 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 21:03:06 +00:00
unsigned char cmd[4], *resp;
/* Try debugWIRE, and MONCON_DISABLE */
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_notice2("no response in ISP mode, trying debugWIRE\n");
* jtag3.c: Make jtag3_command() public * jtag3.h: (Dito.) * jtag3_private.h: Add two new commands * stk500v2.c: Implement the "MonCon disable" hack that allows temporarily falling back to ISP when trying to talk to a part that has debugWIRE enabled git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1123 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 21:03:06 +00:00
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
PROGRAMMER *pgmcp = pgm_dup(pgm);
pgmcp->cookie = PDATA(pgm)->chained_pdata;
* jtag3.c: Make jtag3_command() public * jtag3.h: (Dito.) * jtag3_private.h: Add two new commands * stk500v2.c: Implement the "MonCon disable" hack that allows temporarily falling back to ISP when trying to talk to a part that has debugWIRE enabled git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1123 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 21:03:06 +00:00
cmd[0] = PARM3_CONN_DW;
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
if (jtag3_setparm(pgmcp, SCOPE_AVR, 1, PARM3_CONNECTION, cmd, 1) < 0) {
pgm_free(pgmcp);
* jtag3.c: Make jtag3_command() public * jtag3.h: (Dito.) * jtag3_private.h: Add two new commands * stk500v2.c: Implement the "MonCon disable" hack that allows temporarily falling back to ISP when trying to talk to a part that has debugWIRE enabled git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1123 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 21:03:06 +00:00
break;
}
cmd[0] = SCOPE_AVR;
cmd[1] = CMD3_SIGN_ON;
cmd[2] = cmd[3] = 0;
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
if (jtag3_command(pgmcp, cmd, 4, &resp, "AVR sign-on") >= 0) {
* jtag3.c: Make jtag3_command() public * jtag3.h: (Dito.) * jtag3_private.h: Add two new commands * stk500v2.c: Implement the "MonCon disable" hack that allows temporarily falling back to ISP when trying to talk to a part that has debugWIRE enabled git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1123 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 21:03:06 +00:00
free(resp);
cmd[1] = CMD3_START_DW_DEBUG;
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
if (jtag3_command(pgmcp, cmd, 4, &resp, "start DW debug") >= 0) {
* jtag3.c: Make jtag3_command() public * jtag3.h: (Dito.) * jtag3_private.h: Add two new commands * stk500v2.c: Implement the "MonCon disable" hack that allows temporarily falling back to ISP when trying to talk to a part that has debugWIRE enabled git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1123 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 21:03:06 +00:00
free(resp);
cmd[1] = CMD3_MONCON_DISABLE;
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
if (jtag3_command(pgmcp, cmd, 3, &resp, "MonCon disable") >= 0)
* jtag3.c: Make jtag3_command() public * jtag3.h: (Dito.) * jtag3_private.h: Add two new commands * stk500v2.c: Implement the "MonCon disable" hack that allows temporarily falling back to ISP when trying to talk to a part that has debugWIRE enabled git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1123 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 21:03:06 +00:00
free(resp);
}
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
pgm_free(pgmcp);
Submitted by "Malte" and John McCorquodale: patch #7876 JTAGICE mkII fails to connect to attiny if debugwire is enabled AND target has a very slow clock * jtagmkII.c (jtagmkII_getsync): When leaving debugWIRE mode temporarily, immediately retry with ISP, rather than leaving. * stk500v2 (stk500v2_program_enable): Implemented similar logic for the JTAGICE3. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1183 81a1dc3b-b13d-400b-aceb-764788c761c2
2013-05-17 16:23:55 +00:00
if (tries++ > 3) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unable to return from debugWIRE to ISP\n");
Submitted by "Malte" and John McCorquodale: patch #7876 JTAGICE mkII fails to connect to attiny if debugwire is enabled AND target has a very slow clock * jtagmkII.c (jtagmkII_getsync): When leaving debugWIRE mode temporarily, immediately retry with ISP, rather than leaving. * stk500v2 (stk500v2_program_enable): Implemented similar logic for the JTAGICE3. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1183 81a1dc3b-b13d-400b-aceb-764788c761c2
2013-05-17 16:23:55 +00:00
break;
}
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_warning("target prepared for ISP, signed off\n");
imsg_warning("now retrying without power-cycling the target\n");
Submitted by "Malte" and John McCorquodale: patch #7876 JTAGICE mkII fails to connect to attiny if debugwire is enabled AND target has a very slow clock * jtagmkII.c (jtagmkII_getsync): When leaving debugWIRE mode temporarily, immediately retry with ISP, rather than leaving. * stk500v2 (stk500v2_program_enable): Implemented similar logic for the JTAGICE3. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1183 81a1dc3b-b13d-400b-aceb-764788c761c2
2013-05-17 16:23:55 +00:00
goto retry;
* jtag3.c: Make jtag3_command() public * jtag3.h: (Dito.) * jtag3_private.h: Add two new commands * stk500v2.c: Implement the "MonCon disable" hack that allows temporarily falling back to ISP when trying to talk to a part that has debugWIRE enabled git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1123 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 21:03:06 +00:00
}
break;
bug #29636: AVRDude issues invalid CMD_CHECK_TARGET_CONNECTION on the AVRISP-MKII * stk500v2.c (stk500v2_program_enable): Rewrite the logic to explain ISP activation failures. * stk500v2_private.h: Fix the various STATUS_* constants; AVR069 and AVR079 disagreed in their values, even though they are apparently implementing the same logic behind. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@972 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-08-16 18:15:19 +00:00
default:
/* cannot report anything for other pgmtypes */
break;
}
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
}
return rv;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
}
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
/*
* issue the 'program enable' command to the AVR device, parallel mode
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500pp_program_enable(const PROGRAMMER *pgm, const AVRPART *p) {
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
unsigned char buf[16];
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
PDATA(pgm)->lastpart = p;
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
buf[0] = CMD_ENTER_PROGMODE_PP;
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
buf[1] = p->hventerstabdelay;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
buf[2] = p->progmodedelay;
buf[3] = p->latchcycles;
buf[4] = p->togglevtg;
buf[5] = p->poweroffdelay;
buf[6] = p->resetdelayms;
buf[7] = p->resetdelayus;
return stk500v2_command(pgm, buf, 8, sizeof(buf));
}
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
/*
* issue the 'program enable' command to the AVR device, HVSP mode
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500hvsp_program_enable(const PROGRAMMER *pgm, const AVRPART *p) {
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
unsigned char buf[16];
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
PDATA(pgm)->lastpart = p;
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
buf[0] = PDATA(pgm)->pgmtype == PGMTYPE_STK600?
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
CMD_ENTER_PROGMODE_HVSP_STK600:
CMD_ENTER_PROGMODE_HVSP;
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
buf[1] = p->hventerstabdelay;
buf[2] = p->hvspcmdexedelay;
buf[3] = p->synchcycles;
buf[4] = p->latchcycles;
buf[5] = p->togglevtg;
buf[6] = p->poweroffdelay;
buf[7] = p->resetdelayms;
buf[8] = p->resetdelayus;
return stk500v2_command(pgm, buf, 9, sizeof(buf));
}
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
/*
* initialize the AVR device and prepare it to accept commands
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_initialize(const PROGRAMMER *pgm, const AVRPART *p) {
* stk500v2.c: Use stk500isp_read_byte/stk500isp_write_byte for every byte-wide access (rather than JTAGICE3 only). This finally obsoletes the use of the prehistoric SPI_MULTI command where AVRDUDE used to assemble all the low-level ISP stuff by itself. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1336 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-10-06 22:15:29 +00:00
LNODEID ln;
AVRMEM * m;
Minor cosmetic changes: STK500 firmware version numbers are M.NN, so always display the minor number as two digits. Examine the response to the sign-on command to see which programmer hardware we are talking to, and then restrict the STK500 topcard display to devices detected as STK500. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@614 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-08-18 21:23:34 +00:00
Cleanup the open/close handling to avoid accessing unallocated memory (in the atexit handler) in case of bailing out. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@910 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-13 12:44:54 +00:00
if ((PDATA(pgm)->pgmtype == PGMTYPE_STK600 ||
PDATA(pgm)->pgmtype == PGMTYPE_AVRISP_MKII ||
PDATA(pgm)->pgmtype == PGMTYPE_JTAGICE_MKII) != 0
Offload the programming interface info from part->flags to part->prog_modes flags now just hold parameters of the JTAG interface and some secondary serial, parallel, pseudo parallel info. This separation brings clarity. It used to be hard to augur whether a part has an ISP interface: (part->flags & (AVRPART_HAS_PDI | AVRPART_AVR32 | AVRPART_HAS_TPI | AVRPART_HAS_UPDI)) == 0 && (part->flags & AVRPART_SERIALOK) != 0 or had HVSP or HVPP capability, for that matter. Now it is just, eg, part->prog_modes & PM_ISP part->prog_modes & PM_HVPP
2022-08-30 15:33:42 +00:00
&& (p->prog_modes & (PM_PDI | PM_TPI)) != 0) {
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
/*
* This is an ATxmega device, must use XPROG protocol for the
* remaining actions.
*/
Offload the programming interface info from part->flags to part->prog_modes flags now just hold parameters of the JTAG interface and some secondary serial, parallel, pseudo parallel info. This separation brings clarity. It used to be hard to augur whether a part has an ISP interface: (part->flags & (AVRPART_HAS_PDI | AVRPART_AVR32 | AVRPART_HAS_TPI | AVRPART_HAS_UPDI)) == 0 && (part->flags & AVRPART_SERIALOK) != 0 or had HVSP or HVPP capability, for that matter. Now it is just, eg, part->prog_modes & PM_ISP part->prog_modes & PM_HVPP
2022-08-30 15:33:42 +00:00
if (p->prog_modes & PM_PDI) {
// Find the border between application and boot area
* stk500v2_private.h (struct pdata): add boot_start * stk500v2.c: For the "flash" pseudo-memory of Xmega devices, distinguish addresses between "application" and "boot" area. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1081 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-04-19 13:32:34 +00:00
AVRMEM *bootmem = avr_locate_mem(p, "boot");
AVRMEM *flashmem = avr_locate_mem(p, "flash");
if (bootmem == NULL || flashmem == NULL) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("cannot locate flash or boot memories\n");
* stk500v2_private.h (struct pdata): add boot_start * stk500v2.c: For the "flash" pseudo-memory of Xmega devices, distinguish addresses between "application" and "boot" area. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1081 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-04-19 13:32:34 +00:00
} else {
PDATA(pgm)->boot_start = bootmem->offset - flashmem->offset;
}
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
// stk600_setup_xprog(pgm); [moved to pgm->enable()]
Cleanup the open/close handling to avoid accessing unallocated memory (in the atexit handler) in case of bailing out. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@910 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-13 12:44:54 +00:00
} else {
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
// stk600_setup_isp(pgm); [moved to pgm->enable()]
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
}
* stk500v2.c: Use stk500isp_read_byte/stk500isp_write_byte for every byte-wide access (rather than JTAGICE3 only). This finally obsoletes the use of the prehistoric SPI_MULTI command where AVRDUDE used to assemble all the low-level ISP stuff by itself. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1336 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-10-06 22:15:29 +00:00
/*
* Examine the avrpart's memory definitions, and initialize the page
* caches. For devices/memory that are not page oriented, treat
* them as page size 1 for EEPROM, and 2 for flash.
*/
PDATA(pgm)->flash_pagesize = 2;
PDATA(pgm)->eeprom_pagesize = 1;
for (ln = lfirst(p->mem); ln; ln = lnext(ln)) {
m = ldata(ln);
if (strcmp(m->desc, "flash") == 0) {
bug #60753: Patch #1436 breaks multiple programmer/device combinations on MacOS BigSur * avr.c: compare page_size > 1 instead of != 0 * stk500v2.c: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1452 81a1dc3b-b13d-400b-aceb-764788c761c2
2021-06-27 20:26:05 +00:00
if (m->page_size > 1) {
* stk500v2.c: Use stk500isp_read_byte/stk500isp_write_byte for every byte-wide access (rather than JTAGICE3 only). This finally obsoletes the use of the prehistoric SPI_MULTI command where AVRDUDE used to assemble all the low-level ISP stuff by itself. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1336 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-10-06 22:15:29 +00:00
if (m->page_size > 256)
PDATA(pgm)->flash_pagesize = 256;
else
PDATA(pgm)->flash_pagesize = m->page_size;
}
} else if (strcmp(m->desc, "eeprom") == 0) {
bug #60753: Patch #1436 breaks multiple programmer/device combinations on MacOS BigSur * avr.c: compare page_size > 1 instead of != 0 * stk500v2.c: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1452 81a1dc3b-b13d-400b-aceb-764788c761c2
2021-06-27 20:26:05 +00:00
if (m->page_size > 1)
* stk500v2.c: Use stk500isp_read_byte/stk500isp_write_byte for every byte-wide access (rather than JTAGICE3 only). This finally obsoletes the use of the prehistoric SPI_MULTI command where AVRDUDE used to assemble all the low-level ISP stuff by itself. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1336 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-10-06 22:15:29 +00:00
PDATA(pgm)->eeprom_pagesize = m->page_size;
}
}
free(PDATA(pgm)->flash_pagecache);
free(PDATA(pgm)->eeprom_pagecache);
if ((PDATA(pgm)->flash_pagecache = malloc(PDATA(pgm)->flash_pagesize)) == NULL) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("out of memory\n");
* stk500v2.c: Use stk500isp_read_byte/stk500isp_write_byte for every byte-wide access (rather than JTAGICE3 only). This finally obsoletes the use of the prehistoric SPI_MULTI command where AVRDUDE used to assemble all the low-level ISP stuff by itself. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1336 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-10-06 22:15:29 +00:00
return -1;
}
if ((PDATA(pgm)->eeprom_pagecache = malloc(PDATA(pgm)->eeprom_pagesize)) == NULL) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("out of memory\n");
* stk500v2.c: Use stk500isp_read_byte/stk500isp_write_byte for every byte-wide access (rather than JTAGICE3 only). This finally obsoletes the use of the prehistoric SPI_MULTI command where AVRDUDE used to assemble all the low-level ISP stuff by itself. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1336 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-10-06 22:15:29 +00:00
free(PDATA(pgm)->flash_pagecache);
return -1;
}
PDATA(pgm)->flash_pageaddr = PDATA(pgm)->eeprom_pageaddr = (unsigned long)-1L;
bug #34027: avrdude AT90S1200 Problem (part 2 - stk500v2 and relatives) * stk500v2.c (stk500v2_initialize): For the AT90S1200, release /RESET for a moment before reinitializing, as this is required by its programming protocol. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1004 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-08-30 07:19:21 +00:00
if (p->flags & AVRPART_IS_AT90S1200) {
/*
* AT90S1200 needs a positive reset pulse after a chip erase.
*/
pgm->disable(pgm);
usleep(10000);
}
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return pgm->program_enable(pgm, p);
}
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
/*
* initialize the AVR device and prepare it to accept commands
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_jtag3_initialize(const PROGRAMMER *pgm, const AVRPART *p) {
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
unsigned char parm[4], *resp;
LNODEID ln;
AVRMEM * m;
Distinguish different JTAG programming modes
2022-09-20 17:36:46 +00:00
// FIXME: condition below looks fishy, suspect the code wants !(p->prog_modes & (PM_debugWIRE | PM_JTAG | PM_JTAGmkI /* | PM_XMEGAJTAG | PM_AVR32JTAG */))
Offload the programming interface info from part->flags to part->prog_modes flags now just hold parameters of the JTAG interface and some secondary serial, parallel, pseudo parallel info. This separation brings clarity. It used to be hard to augur whether a part has an ISP interface: (part->flags & (AVRPART_HAS_PDI | AVRPART_AVR32 | AVRPART_HAS_TPI | AVRPART_HAS_UPDI)) == 0 && (part->flags & AVRPART_SERIALOK) != 0 or had HVSP or HVPP capability, for that matter. Now it is just, eg, part->prog_modes & PM_ISP part->prog_modes & PM_HVPP
2022-08-30 15:33:42 +00:00
if (p->prog_modes & (PM_PDI | PM_TPI)) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("part %s has no ISP interface\n", p->desc);
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
return -1;
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
PROGRAMMER *pgmcp = pgm_dup(pgm);
pgmcp->cookie = PDATA(pgm)->chained_pdata;
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
Offload the programming interface info from part->flags to part->prog_modes flags now just hold parameters of the JTAG interface and some secondary serial, parallel, pseudo parallel info. This separation brings clarity. It used to be hard to augur whether a part has an ISP interface: (part->flags & (AVRPART_HAS_PDI | AVRPART_AVR32 | AVRPART_HAS_TPI | AVRPART_HAS_UPDI)) == 0 && (part->flags & AVRPART_SERIALOK) != 0 or had HVSP or HVPP capability, for that matter. Now it is just, eg, part->prog_modes & PM_ISP part->prog_modes & PM_HVPP
2022-08-30 15:33:42 +00:00
if (p->prog_modes & PM_debugWIRE)
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
parm[0] = PARM3_ARCH_TINY;
else
parm[0] = PARM3_ARCH_MEGA;
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
if (jtag3_setparm(pgmcp, SCOPE_AVR, 0, PARM3_ARCH, parm, 1) < 0) {
pgm_free(pgmcp);
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
return -1;
}
parm[0] = PARM3_SESS_PROGRAMMING;
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
if (jtag3_setparm(pgmcp, SCOPE_AVR, 0, PARM3_SESS_PURPOSE, parm, 1) < 0) {
pgm_free(pgmcp);
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
return -1;
}
parm[0] = PARM3_CONN_ISP;
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
if (jtag3_setparm(pgmcp, SCOPE_AVR, 1, PARM3_CONNECTION, parm, 1) < 0) {
pgm_free(pgmcp);
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
return -1;
}
parm[0] = SCOPE_AVR_ISP;
parm[1] = 0x1e;
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
jtag3_send(pgmcp, parm, 2);
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
if (jtag3_recv(pgmcp, &resp) > 0)
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
free(resp);
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
free(pgmcp);
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
/*
* Examine the avrpart's memory definitions, and initialize the page
* caches. For devices/memory that are not page oriented, treat
* them as page size 1 for EEPROM, and 2 for flash.
*/
PDATA(pgm)->flash_pagesize = 2;
PDATA(pgm)->eeprom_pagesize = 1;
for (ln = lfirst(p->mem); ln; ln = lnext(ln)) {
m = ldata(ln);
if (strcmp(m->desc, "flash") == 0) {
bug #60753: Patch #1436 breaks multiple programmer/device combinations on MacOS BigSur * avr.c: compare page_size > 1 instead of != 0 * stk500v2.c: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1452 81a1dc3b-b13d-400b-aceb-764788c761c2
2021-06-27 20:26:05 +00:00
if (m->page_size > 1) {
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
if (m->page_size > 256)
PDATA(pgm)->flash_pagesize = 256;
else
PDATA(pgm)->flash_pagesize = m->page_size;
}
} else if (strcmp(m->desc, "eeprom") == 0) {
bug #60753: Patch #1436 breaks multiple programmer/device combinations on MacOS BigSur * avr.c: compare page_size > 1 instead of != 0 * stk500v2.c: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1452 81a1dc3b-b13d-400b-aceb-764788c761c2
2021-06-27 20:26:05 +00:00
if (m->page_size > 1)
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
PDATA(pgm)->eeprom_pagesize = m->page_size;
}
}
free(PDATA(pgm)->flash_pagecache);
free(PDATA(pgm)->eeprom_pagecache);
if ((PDATA(pgm)->flash_pagecache = malloc(PDATA(pgm)->flash_pagesize)) == NULL) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("out of memory\n");
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
return -1;
}
if ((PDATA(pgm)->eeprom_pagecache = malloc(PDATA(pgm)->eeprom_pagesize)) == NULL) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("out of memory\n");
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
free(PDATA(pgm)->flash_pagecache);
return -1;
}
PDATA(pgm)->flash_pageaddr = PDATA(pgm)->eeprom_pageaddr = (unsigned long)-1L;
return pgm->program_enable(pgm, p);
}
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
/*
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
* initialize the AVR device and prepare it to accept commands, generic HV mode
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500hv_initialize(const PROGRAMMER *pgm, const AVRPART *p, enum hvmode mode) {
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
unsigned char buf[CTL_STACK_SIZE + 1];
int result;
LNODEID ln;
AVRMEM * m;
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
if (p->ctl_stack_type != (mode == PPMODE? CTL_STACK_PP: CTL_STACK_HVSP)) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("%s programming control stack not defined for part %s\n",
mode == PPMODE? "parallel": "high-voltage serial", p->desc);
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
return -1;
}
buf[0] = CMD_SET_CONTROL_STACK;
memcpy(buf + 1, p->controlstack, CTL_STACK_SIZE);
result = stk500v2_command(pgm, buf, CTL_STACK_SIZE + 1, sizeof(buf));
Submitted by John Voltz: add AVR053 oscillator calibration. * main.c: Add the -O option. * pgm.c: Add the hook for the perform_osccal() method. * pgm.h: (Ditto.) * stk500v2.c: Implement perform_osccal(). * avrdude.1: Document the -O option. * doc/avrdude.texi: (Ditto.) Partially closes bug #17487: AVR RC oscillator calibration routine not supported (feature request) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@671 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-09 14:34:24 +00:00
if (result < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unable to set control stack\n");
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
return -1;
}
/*
* Examine the avrpart's memory definitions, and initialize the page
* caches. For devices/memory that are not page oriented, treat
* them as page size 1 for EEPROM, and 2 for flash.
*/
main.c, pgm.c, pgm.h: Add setup and teardown hooks to the programmer definition. If present, call the setup hook immediately after finding the respective programmer object, and schedule the teardown hook to be called upon exit. This allows the programmer implementation to dynamically allocate private programmer data. avr910.c, butterfly.c, jtagmkI.c, jtagmkII.c, stk500v2.c, usbasp.c, usbtiny.c: Convert static programmer data into dynamically allocated data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@764 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-11-07 20:36:12 +00:00
PDATA(pgm)->flash_pagesize = 2;
PDATA(pgm)->eeprom_pagesize = 1;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
for (ln = lfirst(p->mem); ln; ln = lnext(ln)) {
m = ldata(ln);
if (strcmp(m->desc, "flash") == 0) {
bug #60753: Patch #1436 breaks multiple programmer/device combinations on MacOS BigSur * avr.c: compare page_size > 1 instead of != 0 * stk500v2.c: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1452 81a1dc3b-b13d-400b-aceb-764788c761c2
2021-06-27 20:26:05 +00:00
if (m->page_size > 1) {
Xmega page erase implementation for JTAGICEmkII * jtagmkII.c: Handle flash pages sizes > 256 bytes, implement page_erase() method * avrdude.conf.in: Change flash pagesize for all Xmega devices to 512 bytes * avr.c: Implement auto_erase, using page_erase if available * avr.h: Remove unused parameters from avr_read(), replace unused parameter in avr_write)() by auto_erase * stk500v2.c: Handle flash page sizes > 256 bytes * update.c (do_op): Handle new updateflags parameter * main.c: Implement auto_erase as page_erase if possible * update.h (enum updateflags): New enum * pgm.h (struct programmer_t): Add page_erase method git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1089 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-05-04 10:02:30 +00:00
if (m->page_size > 256)
PDATA(pgm)->flash_pagesize = 256;
else
PDATA(pgm)->flash_pagesize = m->page_size;
}
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
} else if (strcmp(m->desc, "eeprom") == 0) {
bug #60753: Patch #1436 breaks multiple programmer/device combinations on MacOS BigSur * avr.c: compare page_size > 1 instead of != 0 * stk500v2.c: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1452 81a1dc3b-b13d-400b-aceb-764788c761c2
2021-06-27 20:26:05 +00:00
if (m->page_size > 1)
main.c, pgm.c, pgm.h: Add setup and teardown hooks to the programmer definition. If present, call the setup hook immediately after finding the respective programmer object, and schedule the teardown hook to be called upon exit. This allows the programmer implementation to dynamically allocate private programmer data. avr910.c, butterfly.c, jtagmkI.c, jtagmkII.c, stk500v2.c, usbasp.c, usbtiny.c: Convert static programmer data into dynamically allocated data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@764 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-11-07 20:36:12 +00:00
PDATA(pgm)->eeprom_pagesize = m->page_size;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
}
}
main.c, pgm.c, pgm.h: Add setup and teardown hooks to the programmer definition. If present, call the setup hook immediately after finding the respective programmer object, and schedule the teardown hook to be called upon exit. This allows the programmer implementation to dynamically allocate private programmer data. avr910.c, butterfly.c, jtagmkI.c, jtagmkII.c, stk500v2.c, usbasp.c, usbtiny.c: Convert static programmer data into dynamically allocated data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@764 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-11-07 20:36:12 +00:00
free(PDATA(pgm)->flash_pagecache);
free(PDATA(pgm)->eeprom_pagecache);
if ((PDATA(pgm)->flash_pagecache = malloc(PDATA(pgm)->flash_pagesize)) == NULL) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("out of memory\n");
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
return -1;
}
main.c, pgm.c, pgm.h: Add setup and teardown hooks to the programmer definition. If present, call the setup hook immediately after finding the respective programmer object, and schedule the teardown hook to be called upon exit. This allows the programmer implementation to dynamically allocate private programmer data. avr910.c, butterfly.c, jtagmkI.c, jtagmkII.c, stk500v2.c, usbasp.c, usbtiny.c: Convert static programmer data into dynamically allocated data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@764 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-11-07 20:36:12 +00:00
if ((PDATA(pgm)->eeprom_pagecache = malloc(PDATA(pgm)->eeprom_pagesize)) == NULL) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("out of memory\n");
main.c, pgm.c, pgm.h: Add setup and teardown hooks to the programmer definition. If present, call the setup hook immediately after finding the respective programmer object, and schedule the teardown hook to be called upon exit. This allows the programmer implementation to dynamically allocate private programmer data. avr910.c, butterfly.c, jtagmkI.c, jtagmkII.c, stk500v2.c, usbasp.c, usbtiny.c: Convert static programmer data into dynamically allocated data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@764 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-11-07 20:36:12 +00:00
free(PDATA(pgm)->flash_pagecache);
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
return -1;
}
main.c, pgm.c, pgm.h: Add setup and teardown hooks to the programmer definition. If present, call the setup hook immediately after finding the respective programmer object, and schedule the teardown hook to be called upon exit. This allows the programmer implementation to dynamically allocate private programmer data. avr910.c, butterfly.c, jtagmkI.c, jtagmkII.c, stk500v2.c, usbasp.c, usbtiny.c: Convert static programmer data into dynamically allocated data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@764 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-11-07 20:36:12 +00:00
PDATA(pgm)->flash_pageaddr = PDATA(pgm)->eeprom_pageaddr = (unsigned long)-1L;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
return pgm->program_enable(pgm, p);
}
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
/*
* initialize the AVR device and prepare it to accept commands, PP mode
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500pp_initialize(const PROGRAMMER *pgm, const AVRPART *p) {
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
return stk500hv_initialize(pgm, p, PPMODE);
}
/*
* initialize the AVR device and prepare it to accept commands, HVSP mode
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500hvsp_initialize(const PROGRAMMER *pgm, const AVRPART *p) {
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
return stk500hv_initialize(pgm, p, HVSPMODE);
}
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static void stk500v2_jtag3_disable(const PROGRAMMER *pgm) {
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
unsigned char buf[16];
int result;
free(PDATA(pgm)->flash_pagecache);
PDATA(pgm)->flash_pagecache = NULL;
free(PDATA(pgm)->eeprom_pagecache);
PDATA(pgm)->eeprom_pagecache = NULL;
buf[0] = CMD_LEAVE_PROGMODE_ISP;
buf[1] = 1; // preDelay;
buf[2] = 1; // postDelay;
result = stk500v2_command(pgm, buf, 3, sizeof(buf));
if (result < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unable to leave programming mode\n");
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
}
return;
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static void stk500v2_disable(const PROGRAMMER *pgm) {
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
unsigned char buf[16];
int result;
buf[0] = CMD_LEAVE_PROGMODE_ISP;
buf[1] = 1; // preDelay;
buf[2] = 1; // postDelay;
result = stk500v2_command(pgm, buf, 3, sizeof(buf));
Submitted by John Voltz: add AVR053 oscillator calibration. * main.c: Add the -O option. * pgm.c: Add the hook for the perform_osccal() method. * pgm.h: (Ditto.) * stk500v2.c: Implement perform_osccal(). * avrdude.1: Document the -O option. * doc/avrdude.texi: (Ditto.) Partially closes bug #17487: AVR RC oscillator calibration routine not supported (feature request) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@671 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-09 14:34:24 +00:00
if (result < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unable to leave programming mode\n");
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
}
return;
}
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
/*
* Leave programming mode, generic HV mode
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static void stk500hv_disable(const PROGRAMMER *pgm, enum hvmode mode) {
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
unsigned char buf[16];
int result;
main.c, pgm.c, pgm.h: Add setup and teardown hooks to the programmer definition. If present, call the setup hook immediately after finding the respective programmer object, and schedule the teardown hook to be called upon exit. This allows the programmer implementation to dynamically allocate private programmer data. avr910.c, butterfly.c, jtagmkI.c, jtagmkII.c, stk500v2.c, usbasp.c, usbtiny.c: Convert static programmer data into dynamically allocated data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@764 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-11-07 20:36:12 +00:00
free(PDATA(pgm)->flash_pagecache);
PDATA(pgm)->flash_pagecache = NULL;
free(PDATA(pgm)->eeprom_pagecache);
PDATA(pgm)->eeprom_pagecache = NULL;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
buf[0] = mode == PPMODE? CMD_LEAVE_PROGMODE_PP:
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
(PDATA(pgm)->pgmtype == PGMTYPE_STK600?
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
CMD_LEAVE_PROGMODE_HVSP_STK600:
CMD_LEAVE_PROGMODE_HVSP);
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
buf[1] = 15; // p->hvleavestabdelay;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
buf[2] = 15; // p->resetdelay;
result = stk500v2_command(pgm, buf, 3, sizeof(buf));
Submitted by John Voltz: add AVR053 oscillator calibration. * main.c: Add the -O option. * pgm.c: Add the hook for the perform_osccal() method. * pgm.h: (Ditto.) * stk500v2.c: Implement perform_osccal(). * avrdude.1: Document the -O option. * doc/avrdude.texi: (Ditto.) Partially closes bug #17487: AVR RC oscillator calibration routine not supported (feature request) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@671 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-09 14:34:24 +00:00
if (result < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unable to leave programming mode\n");
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
}
return;
}
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
/*
* Leave programming mode, PP mode
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static void stk500pp_disable(const PROGRAMMER *pgm) {
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
stk500hv_disable(pgm, PPMODE);
}
/*
* Leave programming mode, HVSP mode
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static void stk500hvsp_disable(const PROGRAMMER *pgm) {
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
stk500hv_disable(pgm, HVSPMODE);
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static void stk500v2_enable(PROGRAMMER *pgm, const AVRPART *p) {
// Previously stk500v2_initialize() set up pgm
if(pgm->initialize == stk500v2_initialize) {
if((PDATA(pgm)->pgmtype == PGMTYPE_STK600 ||
PDATA(pgm)->pgmtype == PGMTYPE_AVRISP_MKII ||
PDATA(pgm)->pgmtype == PGMTYPE_JTAGICE_MKII) != 0
Offload the programming interface info from part->flags to part->prog_modes flags now just hold parameters of the JTAG interface and some secondary serial, parallel, pseudo parallel info. This separation brings clarity. It used to be hard to augur whether a part has an ISP interface: (part->flags & (AVRPART_HAS_PDI | AVRPART_AVR32 | AVRPART_HAS_TPI | AVRPART_HAS_UPDI)) == 0 && (part->flags & AVRPART_SERIALOK) != 0 or had HVSP or HVPP capability, for that matter. Now it is just, eg, part->prog_modes & PM_ISP part->prog_modes & PM_HVPP
2022-08-30 15:33:42 +00:00
&& (p->prog_modes & (PM_PDI | PM_TPI)) != 0) {
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
stk600_setup_xprog(pgm);
} else {
stk600_setup_isp(pgm);
}
}
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return;
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_open(PROGRAMMER *pgm, const char *port) {
Implemented basic serial code refactoring for upcoming SerialUPDI implementation git-svn-id: svn://svn.savannah.nongnu.org/avrdude/branches/serial_refactoring@1511 81a1dc3b-b13d-400b-aceb-764788c761c2
2021-12-08 10:09:52 +00:00
union pinfo pinfo = { .serialinfo.baud = 115200, .serialinfo.cflags = SERIAL_8N1 };
Contributed by dcm@mit.edu: add support for the AVRISP mkII device. (Savannah patch #4789.) * serial.h: Declare usb_serdev_frame device descriptor. * stk500v2.c: Implementation of the AVRISP mkII handling. * usb_libusb.c: Add USB handling for short-frame delimited AVRISP mkII USB protocol; add distinction of different devices in usbdev_open(). * jtagmkII.c: Tell usbdev_open() to search for the JTAG ICE mkII. * usbdevs.h: (New file.) * Makefile.am: Add usbdevs.h, as well as some other forgotten files "make distcheck" complained about. * avrdude.conf.in: Add more aliases for the AVRISP mkII. * avrdude.1: Document how to use the AVRISP mkII. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@564 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-12 23:13:50 +00:00
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
DEBUG("STK500V2: stk500v2_open()\n");
if (pgm->baudrate)
Implemented basic serial code refactoring for upcoming SerialUPDI implementation git-svn-id: svn://svn.savannah.nongnu.org/avrdude/branches/serial_refactoring@1511 81a1dc3b-b13d-400b-aceb-764788c761c2
2021-12-08 10:09:52 +00:00
pinfo.serialinfo.baud = pgm->baudrate;
Contributed by dcm@mit.edu: add support for the AVRISP mkII device. (Savannah patch #4789.) * serial.h: Declare usb_serdev_frame device descriptor. * stk500v2.c: Implementation of the AVRISP mkII handling. * usb_libusb.c: Add USB handling for short-frame delimited AVRISP mkII USB protocol; add distinction of different devices in usbdev_open(). * jtagmkII.c: Tell usbdev_open() to search for the JTAG ICE mkII. * usbdevs.h: (New file.) * Makefile.am: Add usbdevs.h, as well as some other forgotten files "make distcheck" complained about. * avrdude.conf.in: Add more aliases for the AVRISP mkII. * avrdude.1: Document how to use the AVRISP mkII. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@564 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-12 23:13:50 +00:00
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
PDATA(pgm)->pgmtype = PGMTYPE_UNKNOWN;
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
Submitted by Christian Starkjohann: patch #5507: Support for AVR-Doper USB programmer in HID mode * configure.ac: Add hooks to detect the Win32 HID library, as well as the existence of <ddk/hidsdi.h>. * Makefile.am: Add new files. * my_ddk_hidsdi.h: (New file.) * ser_avrdoper.c: (New file.) * serial.h: Add declaration for avrdoper_serdev. * stk500v2.c: Add hook to divert to the AVR Doper code. * avrdude.1: Document the AVR Doper support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@704 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-12-20 23:43:34 +00:00
if(strcasecmp(port, "avrdoper") == 0){
Remove libhid support in ser_avrdoper.c in favor of libhidapi
2022-02-18 19:05:52 +00:00
#if defined(HAVE_LIBHIDAPI)
Submitted by Christian Starkjohann: patch #5507: Support for AVR-Doper USB programmer in HID mode * configure.ac: Add hooks to detect the Win32 HID library, as well as the existence of <ddk/hidsdi.h>. * Makefile.am: Add new files. * my_ddk_hidsdi.h: (New file.) * ser_avrdoper.c: (New file.) * serial.h: Add declaration for avrdoper_serdev. * stk500v2.c: Add hook to divert to the AVR Doper code. * avrdude.1: Document the AVR Doper support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@704 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-12-20 23:43:34 +00:00
serdev = &avrdoper_serdev;
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
PDATA(pgm)->pgmtype = PGMTYPE_STK500;
Submitted by Christian Starkjohann: patch #5507: Support for AVR-Doper USB programmer in HID mode * configure.ac: Add hooks to detect the Win32 HID library, as well as the existence of <ddk/hidsdi.h>. * Makefile.am: Add new files. * my_ddk_hidsdi.h: (New file.) * ser_avrdoper.c: (New file.) * serial.h: Add declaration for avrdoper_serdev. * stk500v2.c: Add hook to divert to the AVR Doper code. * avrdude.1: Document the AVR Doper support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@704 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-12-20 23:43:34 +00:00
#else
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("avrdoper requires avrdude with libhidapi support\n");
Submitted by Christian Starkjohann: patch #5507: Support for AVR-Doper USB programmer in HID mode * configure.ac: Add hooks to detect the Win32 HID library, as well as the existence of <ddk/hidsdi.h>. * Makefile.am: Add new files. * my_ddk_hidsdi.h: (New file.) * ser_avrdoper.c: (New file.) * serial.h: Add declaration for avrdoper_serdev. * stk500v2.c: Add hook to divert to the AVR Doper code. * avrdude.1: Document the AVR Doper support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@704 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-12-20 23:43:34 +00:00
return -1;
#endif
}
Contributed by dcm@mit.edu: add support for the AVRISP mkII device. (Savannah patch #4789.) * serial.h: Declare usb_serdev_frame device descriptor. * stk500v2.c: Implementation of the AVRISP mkII handling. * usb_libusb.c: Add USB handling for short-frame delimited AVRISP mkII USB protocol; add distinction of different devices in usbdev_open(). * jtagmkII.c: Tell usbdev_open() to search for the JTAG ICE mkII. * usbdevs.h: (New file.) * Makefile.am: Add usbdevs.h, as well as some other forgotten files "make distcheck" complained about. * avrdude.conf.in: Add more aliases for the AVRISP mkII. * avrdude.1: Document how to use the AVRISP mkII. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@564 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-12 23:13:50 +00:00
/*
* If the port name starts with "usb", divert the serial routines
* to the USB ones. The serial_open() function for USB overrides
* the meaning of the "baud" parameter to be the USB device ID to
* search for.
*/
if (strncmp(port, "usb", 3) == 0) {
Submitted by <eolson@mit.edu>: * stk500v2.c (stk500v2_open): patch #5273: Emit error message if user requests usb and no libusb support git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@608 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-08-14 21:30:04 +00:00
#if defined(HAVE_LIBUSB)
Contributed by dcm@mit.edu: add support for the AVRISP mkII device. (Savannah patch #4789.) * serial.h: Declare usb_serdev_frame device descriptor. * stk500v2.c: Implementation of the AVRISP mkII handling. * usb_libusb.c: Add USB handling for short-frame delimited AVRISP mkII USB protocol; add distinction of different devices in usbdev_open(). * jtagmkII.c: Tell usbdev_open() to search for the JTAG ICE mkII. * usbdevs.h: (New file.) * Makefile.am: Add usbdevs.h, as well as some other forgotten files "make distcheck" complained about. * avrdude.conf.in: Add more aliases for the AVRISP mkII. * avrdude.1: Document how to use the AVRISP mkII. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@564 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-12 23:13:50 +00:00
serdev = &usb_serdev_frame;
Change the second parameter of the ser_open method from "baud" into a "union pinfo", so the USB parameters can be passed without hacks. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1276 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-02-21 13:44:11 +00:00
pinfo.usbinfo.vid = USB_VENDOR_ATMEL;
pinfo.usbinfo.flags = 0;
pinfo.usbinfo.pid = USB_DEVICE_AVRISPMKII;
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
PDATA(pgm)->pgmtype = PGMTYPE_AVRISP_MKII;
Contributed by dcm@mit.edu: add support for the AVRISP mkII device. (Savannah patch #4789.) * serial.h: Declare usb_serdev_frame device descriptor. * stk500v2.c: Implementation of the AVRISP mkII handling. * usb_libusb.c: Add USB handling for short-frame delimited AVRISP mkII USB protocol; add distinction of different devices in usbdev_open(). * jtagmkII.c: Tell usbdev_open() to search for the JTAG ICE mkII. * usbdevs.h: (New file.) * Makefile.am: Add usbdevs.h, as well as some other forgotten files "make distcheck" complained about. * avrdude.conf.in: Add more aliases for the AVRISP mkII. * avrdude.1: Document how to use the AVRISP mkII. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@564 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-12 23:13:50 +00:00
pgm->set_sck_period = stk500v2_set_sck_period_mk2;
Change all the USB details (endpoint numbers, max transfer size etc.) to a per-programmer adjustable value. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1108 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-11-26 16:24:56 +00:00
pgm->fd.usb.max_xfer = USBDEV_MAX_XFER_MKII;
pgm->fd.usb.rep = USBDEV_BULK_EP_READ_MKII;
pgm->fd.usb.wep = USBDEV_BULK_EP_WRITE_MKII;
pgm->fd.usb.eep = 0; /* no seperate EP for events */
Submitted by <eolson@mit.edu>: * stk500v2.c (stk500v2_open): patch #5273: Emit error message if user requests usb and no libusb support git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@608 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-08-14 21:30:04 +00:00
#else
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("avrdude was compiled without usb support\n");
Submitted by <eolson@mit.edu>: * stk500v2.c (stk500v2_open): patch #5273: Emit error message if user requests usb and no libusb support git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@608 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-08-14 21:30:04 +00:00
return -1;
Contributed by dcm@mit.edu: add support for the AVRISP mkII device. (Savannah patch #4789.) * serial.h: Declare usb_serdev_frame device descriptor. * stk500v2.c: Implementation of the AVRISP mkII handling. * usb_libusb.c: Add USB handling for short-frame delimited AVRISP mkII USB protocol; add distinction of different devices in usbdev_open(). * jtagmkII.c: Tell usbdev_open() to search for the JTAG ICE mkII. * usbdevs.h: (New file.) * Makefile.am: Add usbdevs.h, as well as some other forgotten files "make distcheck" complained about. * avrdude.conf.in: Add more aliases for the AVRISP mkII. * avrdude.1: Document how to use the AVRISP mkII. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@564 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-12 23:13:50 +00:00
#endif
Submitted by <eolson@mit.edu>: * stk500v2.c (stk500v2_open): patch #5273: Emit error message if user requests usb and no libusb support git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@608 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-08-14 21:30:04 +00:00
}
Contributed by dcm@mit.edu: add support for the AVRISP mkII device. (Savannah patch #4789.) * serial.h: Declare usb_serdev_frame device descriptor. * stk500v2.c: Implementation of the AVRISP mkII handling. * usb_libusb.c: Add USB handling for short-frame delimited AVRISP mkII USB protocol; add distinction of different devices in usbdev_open(). * jtagmkII.c: Tell usbdev_open() to search for the JTAG ICE mkII. * usbdevs.h: (New file.) * Makefile.am: Add usbdevs.h, as well as some other forgotten files "make distcheck" complained about. * avrdude.conf.in: Add more aliases for the AVRISP mkII. * avrdude.1: Document how to use the AVRISP mkII. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@564 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-12 23:13:50 +00:00
strcpy(pgm->port, port);
Change the second parameter of the ser_open method from "baud" into a "union pinfo", so the USB parameters can be passed without hacks. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1276 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-02-21 13:44:11 +00:00
if (serial_open(port, pinfo, &pgm->fd)==-1) {
* serial_open() calls will now return -1 on error (no call to exit()) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@948 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-10-22 14:29:56 +00:00
return -1;
}
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
Extend the comment for the serial number function Hint that it is actually a function pointer rather than a serial number itself (in whatever form).
2023-01-01 21:36:50 +00:00
// Get USB serial number function if function pointer is present
Check whether serial_serno is not NULL before calling it. For devices connected through a real serial device rather than USB, there is no transport-layer serial number available. stk500v2.c (and jtag3.c) tried to always call this function. Closes #1254
2023-01-01 21:23:44 +00:00
if (serial_serno)
pgm->usbsn = serial_serno();
Read programmer serial number from libusb or hidusb (#1223) * Silence compiler warnings * Fix a fall-through error in switch * Slightly reformat programmer info printout Now the target voltage is printed out along with the HW version, FW version and serial number, and they all allign nicely. When in ISP mode, the "SCK period" value is printed _after_ the target voltage readout, and _before_ the additional clocks the programmer holds, such as JTAG, PDI and UPDI clocks. Co-authored-by: Stefan Rueger <stefan.rueger@urclocks.com>
2022-12-21 19:11:31 +00:00
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
/*
* drain any extraneous input
*/
stk500v2_drain(pgm, 0);
stk500v2_getsync(pgm);
stk500v2_drain(pgm, 0);
* main.c, stk500v2.c: Added patch 4804 from eolson@mit.edu Which stops sck from being writtend needlessly git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@568 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-17 21:11:39 +00:00
if (pgm->bitclock != 0.0) {
if (pgm->set_sck_period(pgm, pgm->bitclock) != 0)
return -1;
}
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return 0;
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk600_open(PROGRAMMER *pgm, const char *port) {
Implemented basic serial code refactoring for upcoming SerialUPDI implementation git-svn-id: svn://svn.savannah.nongnu.org/avrdude/branches/serial_refactoring@1511 81a1dc3b-b13d-400b-aceb-764788c761c2
2021-12-08 10:09:52 +00:00
union pinfo pinfo = { .serialinfo.baud = 115200, .serialinfo.cflags = SERIAL_8N1 };
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
DEBUG("STK500V2: stk600_open()\n");
if (pgm->baudrate)
Implemented basic serial code refactoring for upcoming SerialUPDI implementation git-svn-id: svn://svn.savannah.nongnu.org/avrdude/branches/serial_refactoring@1511 81a1dc3b-b13d-400b-aceb-764788c761c2
2021-12-08 10:09:52 +00:00
pinfo.serialinfo.baud = pgm->baudrate;
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
PDATA(pgm)->pgmtype = PGMTYPE_UNKNOWN;
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
/*
* If the port name starts with "usb", divert the serial routines
* to the USB ones. The serial_open() function for USB overrides
* the meaning of the "baud" parameter to be the USB device ID to
* search for.
*/
if (strncmp(port, "usb", 3) == 0) {
#if defined(HAVE_LIBUSB)
serdev = &usb_serdev_frame;
Change the second parameter of the ser_open method from "baud" into a "union pinfo", so the USB parameters can be passed without hacks. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1276 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-02-21 13:44:11 +00:00
pinfo.usbinfo.vid = USB_VENDOR_ATMEL;
pinfo.usbinfo.flags = 0;
pinfo.usbinfo.pid = USB_DEVICE_STK600;
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
PDATA(pgm)->pgmtype = PGMTYPE_STK600;
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
pgm->set_sck_period = stk600_set_sck_period;
Change all the USB details (endpoint numbers, max transfer size etc.) to a per-programmer adjustable value. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1108 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-11-26 16:24:56 +00:00
pgm->fd.usb.max_xfer = USBDEV_MAX_XFER_MKII;
* usbdefs.h (USBDEV_BULK_EP_WRITE_STK600) (USBDEV_BULK_EP_READ_STK600): new define values * stk500v2.c (stk600_open): use the STK600 EP values, as they are different from AVRISPmkII git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1126 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-18 10:45:23 +00:00
pgm->fd.usb.rep = USBDEV_BULK_EP_READ_STK600;
pgm->fd.usb.wep = USBDEV_BULK_EP_WRITE_STK600;
Change all the USB details (endpoint numbers, max transfer size etc.) to a per-programmer adjustable value. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1108 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-11-26 16:24:56 +00:00
pgm->fd.usb.eep = 0; /* no seperate EP for events */
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
#else
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("avrdude was compiled without usb support\n");
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
return -1;
#endif
}
strcpy(pgm->port, port);
Change the second parameter of the ser_open method from "baud" into a "union pinfo", so the USB parameters can be passed without hacks. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1276 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-02-21 13:44:11 +00:00
if (serial_open(port, pinfo, &pgm->fd)==-1) {
* serial_open() calls will now return -1 on error (no call to exit()) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@948 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-10-22 14:29:56 +00:00
return -1;
}
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
/*
* drain any extraneous input
*/
stk500v2_drain(pgm, 0);
stk500v2_getsync(pgm);
stk500v2_drain(pgm, 0);
if (pgm->bitclock != 0.0) {
if (pgm->set_sck_period(pgm, pgm->bitclock) != 0)
return -1;
}
return 0;
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static void stk500v2_close(PROGRAMMER *pgm) {
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
DEBUG("STK500V2: stk500v2_close()\n");
Make the code compile warning-free: - declare a dummy "struct timezone" for some Win32 systems (MinGW) - fix a few printf() argument types - get rid off the prevailing "all filedescriptors are of type int" attitude The last item required a large sweep across the code, in order to replace all "int fd"s by "struct filedescriptor *fd"s, and pass pointers (as we cannot pass a union directly). In return, the code is now supposed to be fully 64-bit safe, rather than relying on a 64-bit pointer being converted to a (32-bit) int and back to a pointer as we did previously. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@694 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-12-11 12:47:35 +00:00
serial_close(&pgm->fd);
pgm->fd.ifd = -1;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_loadaddr(const PROGRAMMER *pgm, unsigned int addr) {
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
unsigned char buf[16];
int result;
DEBUG("STK500V2: stk500v2_loadaddr(%d)\n",addr);
buf[0] = CMD_LOAD_ADDRESS;
buf[1] = (addr >> 24) & 0xff;
buf[2] = (addr >> 16) & 0xff;
buf[3] = (addr >> 8) & 0xff;
buf[4] = addr & 0xff;
result = stk500v2_command(pgm, buf, 5, sizeof(buf));
Submitted by John Voltz: add AVR053 oscillator calibration. * main.c: Add the -O option. * pgm.c: Add the hook for the perform_osccal() method. * pgm.h: (Ditto.) * stk500v2.c: Implement perform_osccal(). * avrdude.1: Document the -O option. * doc/avrdude.texi: (Ditto.) Partially closes bug #17487: AVR RC oscillator calibration routine not supported (feature request) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@671 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-09 14:34:24 +00:00
if (result < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unable to set load address\n");
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -1;
}
return 0;
}
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
/*
* Read a single byte, generic HV mode
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500hv_read_byte(const PROGRAMMER *pgm, const AVRPART *p, const AVRMEM *mem,
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
unsigned long addr, unsigned char * value,
enum hvmode mode)
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
{
int result, cmdlen = 2;
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
unsigned char buf[266];
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
unsigned long paddr = 0UL, *paddr_ptr = NULL;
unsigned int pagesize = 0, use_ext_addr = 0, addrshift = 0;
unsigned char *cache_ptr = NULL;
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_notice2("stk500hv_read_byte(.., %s, 0x%lx, ...)\n", mem->desc, addr);
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
if (strcmp(mem->desc, "flash") == 0) {
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
buf[0] = mode == PPMODE? CMD_READ_FLASH_PP: CMD_READ_FLASH_HVSP;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
cmdlen = 3;
Xmega page erase implementation for JTAGICEmkII * jtagmkII.c: Handle flash pages sizes > 256 bytes, implement page_erase() method * avrdude.conf.in: Change flash pagesize for all Xmega devices to 512 bytes * avr.c: Implement auto_erase, using page_erase if available * avr.h: Remove unused parameters from avr_read(), replace unused parameter in avr_write)() by auto_erase * stk500v2.c: Handle flash page sizes > 256 bytes * update.c (do_op): Handle new updateflags parameter * main.c: Implement auto_erase as page_erase if possible * update.h (enum updateflags): New enum * pgm.h (struct programmer_t): Add page_erase method git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1089 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-05-04 10:02:30 +00:00
pagesize = PDATA(pgm)->flash_pagesize;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
paddr = addr & ~(pagesize - 1);
main.c, pgm.c, pgm.h: Add setup and teardown hooks to the programmer definition. If present, call the setup hook immediately after finding the respective programmer object, and schedule the teardown hook to be called upon exit. This allows the programmer implementation to dynamically allocate private programmer data. avr910.c, butterfly.c, jtagmkI.c, jtagmkII.c, stk500v2.c, usbasp.c, usbtiny.c: Convert static programmer data into dynamically allocated data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@764 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-11-07 20:36:12 +00:00
paddr_ptr = &PDATA(pgm)->flash_pageaddr;
cache_ptr = PDATA(pgm)->flash_pagecache;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
addrshift = 1;
/*
* If bit 31 is set, this indicates that the following read/write
* operation will be performed on a memory that is larger than
* 64KBytes. This is an indication to STK500 that a load extended
* address must be executed.
*/
if (mem->op[AVR_OP_LOAD_EXT_ADDR] != NULL) {
use_ext_addr = (1U << 31);
}
} else if (strcmp(mem->desc, "eeprom") == 0) {
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
buf[0] = mode == PPMODE? CMD_READ_EEPROM_PP: CMD_READ_EEPROM_HVSP;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
cmdlen = 3;
pagesize = mem->page_size;
if (pagesize == 0)
pagesize = 1;
paddr = addr & ~(pagesize - 1);
main.c, pgm.c, pgm.h: Add setup and teardown hooks to the programmer definition. If present, call the setup hook immediately after finding the respective programmer object, and schedule the teardown hook to be called upon exit. This allows the programmer implementation to dynamically allocate private programmer data. avr910.c, butterfly.c, jtagmkI.c, jtagmkII.c, stk500v2.c, usbasp.c, usbtiny.c: Convert static programmer data into dynamically allocated data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@764 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-11-07 20:36:12 +00:00
paddr_ptr = &PDATA(pgm)->eeprom_pageaddr;
cache_ptr = PDATA(pgm)->eeprom_pagecache;
Add more parameters for PP mode. Fix the non-paged write operations for old AVRs. In avrdude.conf.in, use the new PP mode parameters; add PP mode definitions for AT90S8515. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@591 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-19 21:25:08 +00:00
} else if (strcmp(mem->desc, "lfuse") == 0 ||
strcmp(mem->desc, "fuse") == 0) {
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
buf[0] = mode == PPMODE? CMD_READ_FUSE_PP: CMD_READ_FUSE_HVSP;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
addr = 0;
} else if (strcmp(mem->desc, "hfuse") == 0) {
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
buf[0] = mode == PPMODE? CMD_READ_FUSE_PP: CMD_READ_FUSE_HVSP;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
addr = 1;
} else if (strcmp(mem->desc, "efuse") == 0) {
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
buf[0] = mode == PPMODE? CMD_READ_FUSE_PP: CMD_READ_FUSE_HVSP;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
addr = 2;
} else if (strcmp(mem->desc, "lock") == 0) {
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
buf[0] = mode == PPMODE? CMD_READ_LOCK_PP: CMD_READ_LOCK_HVSP;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
} else if (strcmp(mem->desc, "calibration") == 0) {
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
buf[0] = mode == PPMODE? CMD_READ_OSCCAL_PP: CMD_READ_OSCCAL_HVSP;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
} else if (strcmp(mem->desc, "signature") == 0) {
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
buf[0] = mode == PPMODE? CMD_READ_SIGNATURE_PP: CMD_READ_SIGNATURE_HVSP;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
}
/*
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
* In HV mode, we have to use paged reads for flash and
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
* EEPROM, and cache the results in a page cache.
*
* Page cache validation is based on "{flash,eeprom}_pageaddr"
* (holding the base address of the most recent cache fill
* operation). This variable is set to (unsigned long)-1L when the
* cache needs to be invalidated.
*/
if (pagesize && paddr == *paddr_ptr) {
*value = cache_ptr[addr & (pagesize - 1)];
return 0;
}
if (cmdlen == 3) {
/* long command, fill in # of bytes */
Fix all bugs in the stk500pp implementation. Eliminate pagebuf, and use stack-allocated buffers instead. The pagesize of all current AVRs is at most 256 only anyway, and this is unlikely to change with the STK500v2 protocol. The previous pagebuf implementation suffered from some possible buffer overrun. In stk500pp_write_page(), do always write full pages, rather than attempting to write a partial last page which did not get written at all. Fill the remaining bytes with 0xff. For (paged) write operations, correctly synthesize the mode byte. This mode byte is very different from the ISP mode byte (sigh). In stk500pp_read_byte(), when performing read operations on paged memory, start reading at the previous page boundary rather than the current address. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@589 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-17 21:25:59 +00:00
buf[1] = (pagesize >> 8) & 0xff;
buf[2] = pagesize & 0xff;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
/* flash and EEPROM reads require the load address command */
Various fixes for Xmega devices. * avrdude.conf.in: Correctly declare EEPROM page sizes for all Xmega devices (0x20 instead of 0x100). * avr.c: If a memory region has a page size declared, try using the paged IO routines regardless of the target memory name. Xmega EEPROM requires to be written in paged mode. Correctly use a long (rather than unsigned long) variable to evaluate the success status of the paged mode write attempt. * stk500v2.c: Don't apply TIF space offsets twice (bug #27995: AVRDUDE 5.8svn fails to program and read XMEGA); use stk500v2_loadaddr() prior to paged mode (EEPROM and flash) writes, otherwise programming of flash areas will fail; while being there, check the return value of stk500v2_loadaddr() everywhere; use the correct write/erase mode bits (same as AVR Studio does). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@907 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-12 15:42:40 +00:00
if (stk500v2_loadaddr(pgm, use_ext_addr | (paddr >> addrshift)) < 0)
return -1;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
} else {
Fix all bugs in the stk500pp implementation. Eliminate pagebuf, and use stack-allocated buffers instead. The pagesize of all current AVRs is at most 256 only anyway, and this is unlikely to change with the STK500v2 protocol. The previous pagebuf implementation suffered from some possible buffer overrun. In stk500pp_write_page(), do always write full pages, rather than attempting to write a partial last page which did not get written at all. Fill the remaining bytes with 0xff. For (paged) write operations, correctly synthesize the mode byte. This mode byte is very different from the ISP mode byte (sigh). In stk500pp_read_byte(), when performing read operations on paged memory, start reading at the previous page boundary rather than the current address. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@589 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-17 21:25:59 +00:00
buf[1] = addr;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
}
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_notice2("stk500hv_read_byte(): sending read memory command: ");
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
Fix all bugs in the stk500pp implementation. Eliminate pagebuf, and use stack-allocated buffers instead. The pagesize of all current AVRs is at most 256 only anyway, and this is unlikely to change with the STK500v2 protocol. The previous pagebuf implementation suffered from some possible buffer overrun. In stk500pp_write_page(), do always write full pages, rather than attempting to write a partial last page which did not get written at all. Fill the remaining bytes with 0xff. For (paged) write operations, correctly synthesize the mode byte. This mode byte is very different from the ISP mode byte (sigh). In stk500pp_read_byte(), when performing read operations on paged memory, start reading at the previous page boundary rather than the current address. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@589 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-17 21:25:59 +00:00
result = stk500v2_command(pgm, buf, cmdlen, sizeof(buf));
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
Submitted by John Voltz: add AVR053 oscillator calibration. * main.c: Add the -O option. * pgm.c: Add the hook for the perform_osccal() method. * pgm.h: (Ditto.) * stk500v2.c: Implement perform_osccal(). * avrdude.1: Document the -O option. * doc/avrdude.texi: (Ditto.) Partially closes bug #17487: AVR RC oscillator calibration routine not supported (feature request) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@671 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-09 14:34:24 +00:00
if (result < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("timeout/error communicating with programmer\n");
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
return -1;
}
if (pagesize) {
*paddr_ptr = paddr;
Fix all bugs in the stk500pp implementation. Eliminate pagebuf, and use stack-allocated buffers instead. The pagesize of all current AVRs is at most 256 only anyway, and this is unlikely to change with the STK500v2 protocol. The previous pagebuf implementation suffered from some possible buffer overrun. In stk500pp_write_page(), do always write full pages, rather than attempting to write a partial last page which did not get written at all. Fill the remaining bytes with 0xff. For (paged) write operations, correctly synthesize the mode byte. This mode byte is very different from the ISP mode byte (sigh). In stk500pp_read_byte(), when performing read operations on paged memory, start reading at the previous page boundary rather than the current address. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@589 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-17 21:25:59 +00:00
memcpy(cache_ptr, buf + 2, pagesize);
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
*value = cache_ptr[addr & (pagesize - 1)];
} else {
Fix all bugs in the stk500pp implementation. Eliminate pagebuf, and use stack-allocated buffers instead. The pagesize of all current AVRs is at most 256 only anyway, and this is unlikely to change with the STK500v2 protocol. The previous pagebuf implementation suffered from some possible buffer overrun. In stk500pp_write_page(), do always write full pages, rather than attempting to write a partial last page which did not get written at all. Fill the remaining bytes with 0xff. For (paged) write operations, correctly synthesize the mode byte. This mode byte is very different from the ISP mode byte (sigh). In stk500pp_read_byte(), when performing read operations on paged memory, start reading at the previous page boundary rather than the current address. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@589 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-17 21:25:59 +00:00
*value = buf[2];
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
}
return 0;
}
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
/*
* Read a single byte, PP mode
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500pp_read_byte(const PROGRAMMER *pgm, const AVRPART *p, const AVRMEM *mem,
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
unsigned long addr, unsigned char * value)
{
return stk500hv_read_byte(pgm, p, mem, addr, value, PPMODE);
}
/*
* Read a single byte, HVSP mode
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500hvsp_read_byte(const PROGRAMMER *pgm, const AVRPART *p, const AVRMEM *mem,
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
unsigned long addr, unsigned char * value)
{
return stk500hv_read_byte(pgm, p, mem, addr, value, HVSPMODE);
}
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
/*
* Read a single byte, ISP mode
*
* By now, only used on the JTAGICE3 which does not implement the
* CMD_SPI_MULTI SPI passthrough command.
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500isp_read_byte(const PROGRAMMER *pgm, const AVRPART *p, const AVRMEM *mem,
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
unsigned long addr, unsigned char * value)
{
int result, pollidx;
unsigned char buf[6];
unsigned long paddr = 0UL, *paddr_ptr = NULL;
unsigned int pagesize = 0;
unsigned char *cache_ptr = NULL;
OPCODE *op;
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_notice2("stk500isp_read_byte(.., %s, 0x%lx, ...)\n", mem->desc, addr);
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
if (strcmp(mem->desc, "flash") == 0 ||
strcmp(mem->desc, "eeprom") == 0) {
// use paged access, and cache result
if (strcmp(mem->desc, "flash") == 0) {
pagesize = PDATA(pgm)->flash_pagesize;
paddr = addr & ~(pagesize - 1);
paddr_ptr = &PDATA(pgm)->flash_pageaddr;
cache_ptr = PDATA(pgm)->flash_pagecache;
} else {
pagesize = mem->page_size;
if (pagesize == 0)
pagesize = 1;
paddr = addr & ~(pagesize - 1);
paddr_ptr = &PDATA(pgm)->eeprom_pageaddr;
cache_ptr = PDATA(pgm)->eeprom_pagecache;
}
if (paddr == *paddr_ptr) {
*value = cache_ptr[addr & (pagesize - 1)];
return 0;
}
if (stk500v2_paged_load(pgm, p, mem, pagesize, paddr, pagesize) < 0)
return -1;
*paddr_ptr = paddr;
memcpy(cache_ptr, &mem->buf[paddr], pagesize);
*value = cache_ptr[addr & (pagesize - 1)];
return 0;
}
if (strcmp(mem->desc, "lfuse") == 0 ||
strcmp(mem->desc, "fuse") == 0) {
buf[0] = CMD_READ_FUSE_ISP;
addr = 0;
} else if (strcmp(mem->desc, "hfuse") == 0) {
buf[0] = CMD_READ_FUSE_ISP;
addr = 1;
} else if (strcmp(mem->desc, "efuse") == 0) {
buf[0] = CMD_READ_FUSE_ISP;
addr = 2;
} else if (strcmp(mem->desc, "lock") == 0) {
buf[0] = CMD_READ_LOCK_ISP;
} else if (strcmp(mem->desc, "calibration") == 0) {
buf[0] = CMD_READ_OSCCAL_ISP;
} else if (strcmp(mem->desc, "signature") == 0) {
buf[0] = CMD_READ_SIGNATURE_ISP;
}
if ((op = mem->op[AVR_OP_READ]) == NULL) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("invalid operation AVR_OP_READ on %s memory\n", mem->desc);
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
return -1;
}
Initialise memory before avr_set_bits() calls in stk500.c and stk500v2.c
2022-07-24 17:20:35 +00:00
memset(buf+2, 0, 4);
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
avr_set_bits(op, buf + 2);
if ((pollidx = avr_get_output_index(op)) == -1) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_warning("cannot determine pollidx to read %s memory\n", mem->desc);
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
pollidx = 3;
}
buf[1] = pollidx + 1;
avr_set_addr(op, buf + 2, addr);
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_notice2("stk500isp_read_byte(): sending read memory command: ");
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
result = stk500v2_command(pgm, buf, 6, sizeof(buf));
if (result < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("timeout/error communicating with programmer\n");
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
return -1;
}
*value = buf[2];
return 0;
}
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
/*
* Write one byte, generic HV mode
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500hv_write_byte(const PROGRAMMER *pgm, const AVRPART *p, const AVRMEM *mem,
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
unsigned long addr, unsigned char data,
enum hvmode mode)
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
{
Add more parameters for PP mode. Fix the non-paged write operations for old AVRs. In avrdude.conf.in, use the new PP mode parameters; add PP mode definitions for AT90S8515. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@591 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-19 21:25:08 +00:00
int result, cmdlen, timeout = 0, pulsewidth = 0;
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
unsigned char buf[266];
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
unsigned long paddr = 0UL, *paddr_ptr = NULL;
unsigned int pagesize = 0, use_ext_addr = 0, addrshift = 0;
unsigned char *cache_ptr = NULL;
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_notice2("stk500hv_write_byte(.., %s, 0x%lx, ...)\n", mem->desc, addr);
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
if (strcmp(mem->desc, "flash") == 0) {
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
buf[0] = mode == PPMODE? CMD_PROGRAM_FLASH_PP: CMD_PROGRAM_FLASH_HVSP;
Xmega page erase implementation for JTAGICEmkII * jtagmkII.c: Handle flash pages sizes > 256 bytes, implement page_erase() method * avrdude.conf.in: Change flash pagesize for all Xmega devices to 512 bytes * avr.c: Implement auto_erase, using page_erase if available * avr.h: Remove unused parameters from avr_read(), replace unused parameter in avr_write)() by auto_erase * stk500v2.c: Handle flash page sizes > 256 bytes * update.c (do_op): Handle new updateflags parameter * main.c: Implement auto_erase as page_erase if possible * update.h (enum updateflags): New enum * pgm.h (struct programmer_t): Add page_erase method git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1089 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-05-04 10:02:30 +00:00
pagesize = PDATA(pgm)->flash_pagesize;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
paddr = addr & ~(pagesize - 1);
main.c, pgm.c, pgm.h: Add setup and teardown hooks to the programmer definition. If present, call the setup hook immediately after finding the respective programmer object, and schedule the teardown hook to be called upon exit. This allows the programmer implementation to dynamically allocate private programmer data. avr910.c, butterfly.c, jtagmkI.c, jtagmkII.c, stk500v2.c, usbasp.c, usbtiny.c: Convert static programmer data into dynamically allocated data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@764 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-11-07 20:36:12 +00:00
paddr_ptr = &PDATA(pgm)->flash_pageaddr;
cache_ptr = PDATA(pgm)->flash_pagecache;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
addrshift = 1;
/*
* If bit 31 is set, this indicates that the following read/write
* operation will be performed on a memory that is larger than
* 64KBytes. This is an indication to STK500 that a load extended
* address must be executed.
*/
if (mem->op[AVR_OP_LOAD_EXT_ADDR] != NULL) {
use_ext_addr = (1U << 31);
}
} else if (strcmp(mem->desc, "eeprom") == 0) {
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
buf[0] = mode == PPMODE? CMD_PROGRAM_EEPROM_PP: CMD_PROGRAM_EEPROM_HVSP;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
pagesize = mem->page_size;
if (pagesize == 0)
pagesize = 1;
paddr = addr & ~(pagesize - 1);
main.c, pgm.c, pgm.h: Add setup and teardown hooks to the programmer definition. If present, call the setup hook immediately after finding the respective programmer object, and schedule the teardown hook to be called upon exit. This allows the programmer implementation to dynamically allocate private programmer data. avr910.c, butterfly.c, jtagmkI.c, jtagmkII.c, stk500v2.c, usbasp.c, usbtiny.c: Convert static programmer data into dynamically allocated data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@764 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-11-07 20:36:12 +00:00
paddr_ptr = &PDATA(pgm)->eeprom_pageaddr;
cache_ptr = PDATA(pgm)->eeprom_pagecache;
Add more parameters for PP mode. Fix the non-paged write operations for old AVRs. In avrdude.conf.in, use the new PP mode parameters; add PP mode definitions for AT90S8515. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@591 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-19 21:25:08 +00:00
} else if (strcmp(mem->desc, "lfuse") == 0 ||
strcmp(mem->desc, "fuse") == 0) {
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
buf[0] = mode == PPMODE? CMD_PROGRAM_FUSE_PP: CMD_PROGRAM_FUSE_HVSP;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
addr = 0;
Add more parameters for PP mode. Fix the non-paged write operations for old AVRs. In avrdude.conf.in, use the new PP mode parameters; add PP mode definitions for AT90S8515. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@591 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-19 21:25:08 +00:00
pulsewidth = p->programfusepulsewidth;
timeout = p->programfusepolltimeout;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
} else if (strcmp(mem->desc, "hfuse") == 0) {
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
buf[0] = mode == PPMODE? CMD_PROGRAM_FUSE_PP: CMD_PROGRAM_FUSE_HVSP;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
addr = 1;
Add more parameters for PP mode. Fix the non-paged write operations for old AVRs. In avrdude.conf.in, use the new PP mode parameters; add PP mode definitions for AT90S8515. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@591 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-19 21:25:08 +00:00
pulsewidth = p->programfusepulsewidth;
timeout = p->programfusepolltimeout;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
} else if (strcmp(mem->desc, "efuse") == 0) {
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
buf[0] = mode == PPMODE? CMD_PROGRAM_FUSE_PP: CMD_PROGRAM_FUSE_HVSP;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
addr = 2;
Add more parameters for PP mode. Fix the non-paged write operations for old AVRs. In avrdude.conf.in, use the new PP mode parameters; add PP mode definitions for AT90S8515. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@591 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-19 21:25:08 +00:00
pulsewidth = p->programfusepulsewidth;
timeout = p->programfusepolltimeout;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
} else if (strcmp(mem->desc, "lock") == 0) {
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
buf[0] = mode == PPMODE? CMD_PROGRAM_LOCK_PP: CMD_PROGRAM_LOCK_HVSP;
Add more parameters for PP mode. Fix the non-paged write operations for old AVRs. In avrdude.conf.in, use the new PP mode parameters; add PP mode definitions for AT90S8515. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@591 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-19 21:25:08 +00:00
pulsewidth = p->programlockpulsewidth;
timeout = p->programlockpolltimeout;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
} else {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unsupported memory type %s\n", mem->desc);
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
return -1;
}
cmdlen = 5 + pagesize;
/*
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
* In HV mode, we have to use paged writes for flash and
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
* EEPROM. As both, flash and EEPROM cells can only be programmed
* from `1' to `0' bits (even EEPROM does not support auto-erase in
* parallel mode), we just pre-fill the page cache with 0xff, so all
* those cells that are outside our current address will remain
* unaffected.
*/
if (pagesize) {
memset(cache_ptr, 0xff, pagesize);
cache_ptr[addr & (pagesize - 1)] = data;
/* long command, fill in # of bytes */
Fix all bugs in the stk500pp implementation. Eliminate pagebuf, and use stack-allocated buffers instead. The pagesize of all current AVRs is at most 256 only anyway, and this is unlikely to change with the STK500v2 protocol. The previous pagebuf implementation suffered from some possible buffer overrun. In stk500pp_write_page(), do always write full pages, rather than attempting to write a partial last page which did not get written at all. Fill the remaining bytes with 0xff. For (paged) write operations, correctly synthesize the mode byte. This mode byte is very different from the ISP mode byte (sigh). In stk500pp_read_byte(), when performing read operations on paged memory, start reading at the previous page boundary rather than the current address. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@589 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-17 21:25:59 +00:00
buf[1] = (pagesize >> 8) & 0xff;
buf[2] = pagesize & 0xff;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
Fix all bugs in the stk500pp implementation. Eliminate pagebuf, and use stack-allocated buffers instead. The pagesize of all current AVRs is at most 256 only anyway, and this is unlikely to change with the STK500v2 protocol. The previous pagebuf implementation suffered from some possible buffer overrun. In stk500pp_write_page(), do always write full pages, rather than attempting to write a partial last page which did not get written at all. Fill the remaining bytes with 0xff. For (paged) write operations, correctly synthesize the mode byte. This mode byte is very different from the ISP mode byte (sigh). In stk500pp_read_byte(), when performing read operations on paged memory, start reading at the previous page boundary rather than the current address. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@589 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-17 21:25:59 +00:00
/*
* Synthesize the mode byte. This is simpler than adding yet
* another parameter to the avrdude.conf file. We calculate the
* bits corresponding to the page size, as explained in AVR068.
* We set bit 7, to indicate this is to actually write the page to
* the target device. We set bit 6 to indicate this is the very
* last page to be programmed, whatever this means -- we just
* pretend we don't know any better. ;-) Bit 0 is set if this is
* a paged memory, which means it has a page size of more than 2.
*/
Add more parameters for PP mode. Fix the non-paged write operations for old AVRs. In avrdude.conf.in, use the new PP mode parameters; add PP mode definitions for AT90S8515. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@591 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-19 21:25:08 +00:00
buf[3] = 0x80 | 0x40;
if (pagesize > 2) {
Replace all occurences of exit() in potential library code by appropriate return values. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1301 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-05-16 15:52:25 +00:00
unsigned int rv = stk500v2_mode_for_pagesize(pagesize);
if (rv == 0)
return -1;
buf[3] |= rv;
Fix all bugs in the stk500pp implementation. Eliminate pagebuf, and use stack-allocated buffers instead. The pagesize of all current AVRs is at most 256 only anyway, and this is unlikely to change with the STK500v2 protocol. The previous pagebuf implementation suffered from some possible buffer overrun. In stk500pp_write_page(), do always write full pages, rather than attempting to write a partial last page which did not get written at all. Fill the remaining bytes with 0xff. For (paged) write operations, correctly synthesize the mode byte. This mode byte is very different from the ISP mode byte (sigh). In stk500pp_read_byte(), when performing read operations on paged memory, start reading at the previous page boundary rather than the current address. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@589 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-17 21:25:59 +00:00
buf[3] |= 0x01;
Add more parameters for PP mode. Fix the non-paged write operations for old AVRs. In avrdude.conf.in, use the new PP mode parameters; add PP mode definitions for AT90S8515. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@591 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-19 21:25:08 +00:00
}
Fix all bugs in the stk500pp implementation. Eliminate pagebuf, and use stack-allocated buffers instead. The pagesize of all current AVRs is at most 256 only anyway, and this is unlikely to change with the STK500v2 protocol. The previous pagebuf implementation suffered from some possible buffer overrun. In stk500pp_write_page(), do always write full pages, rather than attempting to write a partial last page which did not get written at all. Fill the remaining bytes with 0xff. For (paged) write operations, correctly synthesize the mode byte. This mode byte is very different from the ISP mode byte (sigh). In stk500pp_read_byte(), when performing read operations on paged memory, start reading at the previous page boundary rather than the current address. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@589 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-17 21:25:59 +00:00
buf[4] = mem->delay;
memcpy(buf + 5, cache_ptr, pagesize);
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
/* flash and EEPROM reads require the load address command */
Various fixes for Xmega devices. * avrdude.conf.in: Correctly declare EEPROM page sizes for all Xmega devices (0x20 instead of 0x100). * avr.c: If a memory region has a page size declared, try using the paged IO routines regardless of the target memory name. Xmega EEPROM requires to be written in paged mode. Correctly use a long (rather than unsigned long) variable to evaluate the success status of the paged mode write attempt. * stk500v2.c: Don't apply TIF space offsets twice (bug #27995: AVRDUDE 5.8svn fails to program and read XMEGA); use stk500v2_loadaddr() prior to paged mode (EEPROM and flash) writes, otherwise programming of flash areas will fail; while being there, check the return value of stk500v2_loadaddr() everywhere; use the correct write/erase mode bits (same as AVR Studio does). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@907 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-12 15:42:40 +00:00
if (stk500v2_loadaddr(pgm, use_ext_addr | (paddr >> addrshift)) < 0)
return -1;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
} else {
Fix all bugs in the stk500pp implementation. Eliminate pagebuf, and use stack-allocated buffers instead. The pagesize of all current AVRs is at most 256 only anyway, and this is unlikely to change with the STK500v2 protocol. The previous pagebuf implementation suffered from some possible buffer overrun. In stk500pp_write_page(), do always write full pages, rather than attempting to write a partial last page which did not get written at all. Fill the remaining bytes with 0xff. For (paged) write operations, correctly synthesize the mode byte. This mode byte is very different from the ISP mode byte (sigh). In stk500pp_read_byte(), when performing read operations on paged memory, start reading at the previous page boundary rather than the current address. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@589 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-17 21:25:59 +00:00
buf[1] = addr;
buf[2] = data;
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
if (mode == PPMODE) {
buf[3] = pulsewidth;
buf[4] = timeout;
} else {
buf[3] = timeout;
cmdlen--;
}
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
}
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_notice2("stk500hv_write_byte(): sending write memory command: ");
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
Fix all bugs in the stk500pp implementation. Eliminate pagebuf, and use stack-allocated buffers instead. The pagesize of all current AVRs is at most 256 only anyway, and this is unlikely to change with the STK500v2 protocol. The previous pagebuf implementation suffered from some possible buffer overrun. In stk500pp_write_page(), do always write full pages, rather than attempting to write a partial last page which did not get written at all. Fill the remaining bytes with 0xff. For (paged) write operations, correctly synthesize the mode byte. This mode byte is very different from the ISP mode byte (sigh). In stk500pp_read_byte(), when performing read operations on paged memory, start reading at the previous page boundary rather than the current address. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@589 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-17 21:25:59 +00:00
result = stk500v2_command(pgm, buf, cmdlen, sizeof(buf));
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
Submitted by John Voltz: add AVR053 oscillator calibration. * main.c: Add the -O option. * pgm.c: Add the hook for the perform_osccal() method. * pgm.h: (Ditto.) * stk500v2.c: Implement perform_osccal(). * avrdude.1: Document the -O option. * doc/avrdude.texi: (Ditto.) Partially closes bug #17487: AVR RC oscillator calibration routine not supported (feature request) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@671 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-09 14:34:24 +00:00
if (result < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("timeout/error communicating with programmer\n");
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
return -1;
}
if (pagesize) {
/* Invalidate the page cache. */
*paddr_ptr = (unsigned long)-1L;
}
return 0;
}
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
/*
* Write one byte, PP mode
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500pp_write_byte(const PROGRAMMER *pgm, const AVRPART *p, const AVRMEM *mem,
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
unsigned long addr, unsigned char data)
{
return stk500hv_write_byte(pgm, p, mem, addr, data, PPMODE);
}
/*
* Write one byte, HVSP mode
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500hvsp_write_byte(const PROGRAMMER *pgm, const AVRPART *p, const AVRMEM *mem,
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
unsigned long addr, unsigned char data)
{
return stk500hv_write_byte(pgm, p, mem, addr, data, HVSPMODE);
}
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
/*
* Write one byte, ISP mode
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500isp_write_byte(const PROGRAMMER *pgm, const AVRPART *p, const AVRMEM *mem,
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
unsigned long addr, unsigned char data)
{
int result;
unsigned char buf[5];
unsigned long paddr = 0UL, *paddr_ptr = NULL;
unsigned int pagesize = 0;
unsigned char *cache_ptr = NULL;
OPCODE *op;
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_notice2("stk500isp_write_byte(.., %s, 0x%lx, ...)\n", mem->desc, addr);
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
if (strcmp(mem->desc, "flash") == 0 ||
strcmp(mem->desc, "eeprom") == 0) {
if (strcmp(mem->desc, "flash") == 0) {
pagesize = PDATA(pgm)->flash_pagesize;
paddr = addr & ~(pagesize - 1);
paddr_ptr = &PDATA(pgm)->flash_pageaddr;
cache_ptr = PDATA(pgm)->flash_pagecache;
Annual changelog rotation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1393 81a1dc3b-b13d-400b-aceb-764788c761c2
2017-03-05 18:21:08 +00:00
if ((mem->mode & 1) == 0)
/* old, unpaged device, really write single bytes */
pagesize = 1;
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
} else {
pagesize = mem->page_size;
if (pagesize == 0)
pagesize = 1;
paddr = addr & ~(pagesize - 1);
paddr_ptr = &PDATA(pgm)->eeprom_pageaddr;
cache_ptr = PDATA(pgm)->eeprom_pagecache;
}
/*
* stk500v2.c: Use stk500isp_read_byte/stk500isp_write_byte for every byte-wide access (rather than JTAGICE3 only). This finally obsoletes the use of the prehistoric SPI_MULTI command where AVRDUDE used to assemble all the low-level ISP stuff by itself. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1336 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-10-06 22:15:29 +00:00
* We use paged writes for flash and EEPROM, reading back the
* current page first, modify the byte to write, and write out the
* entire page.
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
*/
* stk500v2.c: Use stk500isp_read_byte/stk500isp_write_byte for every byte-wide access (rather than JTAGICE3 only). This finally obsoletes the use of the prehistoric SPI_MULTI command where AVRDUDE used to assemble all the low-level ISP stuff by itself. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1336 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-10-06 22:15:29 +00:00
if (stk500v2_paged_load(pgm, p, mem, pagesize, paddr, pagesize) < 0)
return -1;
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
* stk500v2.c: Use stk500isp_read_byte/stk500isp_write_byte for every byte-wide access (rather than JTAGICE3 only). This finally obsoletes the use of the prehistoric SPI_MULTI command where AVRDUDE used to assemble all the low-level ISP stuff by itself. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1336 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-10-06 22:15:29 +00:00
memcpy(cache_ptr, mem->buf + paddr, pagesize);
*paddr_ptr = paddr;
cache_ptr[addr & (pagesize - 1)] = data;
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
memcpy(mem->buf + paddr, cache_ptr, pagesize);
* stk500v2.c: Use stk500isp_read_byte/stk500isp_write_byte for every byte-wide access (rather than JTAGICE3 only). This finally obsoletes the use of the prehistoric SPI_MULTI command where AVRDUDE used to assemble all the low-level ISP stuff by itself. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1336 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-10-06 22:15:29 +00:00
stk500v2_paged_write(pgm, p, mem, pagesize, paddr, pagesize);
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
return 0;
}
memset(buf, 0, sizeof buf);
if (strcmp(mem->desc, "lfuse") == 0 ||
strcmp(mem->desc, "fuse") == 0) {
buf[0] = CMD_PROGRAM_FUSE_ISP;
addr = 0;
} else if (strcmp(mem->desc, "hfuse") == 0) {
buf[0] = CMD_PROGRAM_FUSE_ISP;
addr = 1;
} else if (strcmp(mem->desc, "efuse") == 0) {
buf[0] = CMD_PROGRAM_FUSE_ISP;
addr = 2;
} else if (strcmp(mem->desc, "lock") == 0) {
buf[0] = CMD_PROGRAM_LOCK_ISP;
} else {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unsupported memory type: %s\n", mem->desc);
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
return -1;
}
if ((op = mem->op[AVR_OP_WRITE]) == NULL) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("no AVR_OP_WRITE for %s memory\n", mem->desc);
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
return -1;
}
avr_set_bits(op, buf + 1);
avr_set_addr(op, buf + 1, addr);
avr_set_input(op, buf + 1, data);
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_notice2("stk500isp_write_byte(): sending write memory command: ");
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
result = stk500v2_command(pgm, buf, 5, sizeof(buf));
if (result < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("timeout/error communicating with programmer\n");
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
return -1;
}
bug #42267: jtag3isp fails to read lock and fuse bytes directly after changing lock byte * stk500v2.c (stk500isp_write_byte): As a workaround for broken tool firmware, add 10 ms of delay before returning from any single-byte write operation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1337 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-10-07 15:21:14 +00:00
/*
* Prevent verification readback to be too fast, see
* https://savannah.nongnu.org/bugs/index.php?42267
*
* After all, this is just an ugly hack working around some
* brokeness in the Atmel firmware starting with the AVRISPmkII (the
* old JTAGICEmkII isn't affected). Let's hope 10 ms of additional
* delay are good enough for everyone.
*/
usleep(10000);
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
return 0;
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_paged_write(const PROGRAMMER *pgm, const AVRPART *p, const AVRMEM *m,
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
unsigned int page_size,
unsigned int addr, unsigned int n_bytes)
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
{
Fix all "unused variable" warnings that were encountered. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1200 81a1dc3b-b13d-400b-aceb-764788c761c2
2013-09-03 11:05:13 +00:00
unsigned int block_size, last_addr, addrshift, use_ext_addr;
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
unsigned int maxaddr = addr + n_bytes;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
unsigned char commandbuf[10];
unsigned char buf[266];
unsigned char cmds[4];
int result;
I thought I had already committed this but I don't see it in CVS. This fixes EEPROM access using the STK500V2 programmer, partially undoing part of a previous general fixup commit. Choose the correct read/write operations with the stk500v2 program function - the correct one depends on the memory type. EEPROM is byte addressable so uses read/write. FLASH is word addressable and so uses read_lo/write_lo. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@501 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-09-16 21:23:14 +00:00
OPCODE * rop, * wop;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
DEBUG("STK500V2: stk500v2_paged_write(..,%s,%u,%u,%u)\n",
m->desc, page_size, addr, n_bytes);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
if (page_size == 0) page_size = 256;
Implement extended addressing needed for the ATmega256x devices. Document ATmega256x support. Also document Solaris port defaults in avrdude.texi. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@584 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-05-25 14:38:08 +00:00
addrshift = 0;
use_ext_addr = 0;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
// determine which command is to be used
if (strcmp(m->desc, "flash") == 0) {
Implement extended addressing needed for the ATmega256x devices. Document ATmega256x support. Also document Solaris port defaults in avrdude.texi. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@584 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-05-25 14:38:08 +00:00
addrshift = 1;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
commandbuf[0] = CMD_PROGRAM_FLASH_ISP;
Implement extended addressing needed for the ATmega256x devices. Document ATmega256x support. Also document Solaris port defaults in avrdude.texi. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@584 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-05-25 14:38:08 +00:00
/*
* If bit 31 is set, this indicates that the following read/write
* operation will be performed on a memory that is larger than
* 64KBytes. This is an indication to STK500 that a load extended
* address must be executed.
*/
if (m->op[AVR_OP_LOAD_EXT_ADDR] != NULL) {
use_ext_addr = (1U << 31);
}
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
} else if (strcmp(m->desc, "eeprom") == 0) {
commandbuf[0] = CMD_PROGRAM_EEPROM_ISP;
}
commandbuf[4] = m->delay;
Implement extended addressing needed for the ATmega256x devices. Document ATmega256x support. Also document Solaris port defaults in avrdude.texi. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@584 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-05-25 14:38:08 +00:00
if (addrshift == 0) {
I thought I had already committed this but I don't see it in CVS. This fixes EEPROM access using the STK500V2 programmer, partially undoing part of a previous general fixup commit. Choose the correct read/write operations with the stk500v2 program function - the correct one depends on the memory type. EEPROM is byte addressable so uses read/write. FLASH is word addressable and so uses read_lo/write_lo. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@501 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-09-16 21:23:14 +00:00
wop = m->op[AVR_OP_WRITE];
rop = m->op[AVR_OP_READ];
}
else {
wop = m->op[AVR_OP_WRITE_LO];
rop = m->op[AVR_OP_READ_LO];
}
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
// if the memory is paged, load the appropriate commands into the buffer
if (m->mode & 0x01) {
I thought I had already committed this but I don't see it in CVS. This fixes EEPROM access using the STK500V2 programmer, partially undoing part of a previous general fixup commit. Choose the correct read/write operations with the stk500v2 program function - the correct one depends on the memory type. EEPROM is byte addressable so uses read/write. FLASH is word addressable and so uses read_lo/write_lo. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@501 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-09-16 21:23:14 +00:00
commandbuf[3] = m->mode | 0x80; // yes, write the page to flash
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
if (m->op[AVR_OP_LOADPAGE_LO] == NULL) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("loadpage instruction not defined for part %s\n", p->desc);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -1;
}
Initialise memory before avr_set_bits() calls in stk500.c and stk500v2.c
2022-07-24 17:20:35 +00:00
memset(cmds, 0, sizeof cmds);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
avr_set_bits(m->op[AVR_OP_LOADPAGE_LO], cmds);
commandbuf[5] = cmds[0];
if (m->op[AVR_OP_WRITEPAGE] == NULL) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("write page instruction not defined for part %s\n", p->desc);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -1;
}
Initialise memory before avr_set_bits() calls in stk500.c and stk500v2.c
2022-07-24 17:20:35 +00:00
memset(cmds, 0, sizeof cmds);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
avr_set_bits(m->op[AVR_OP_WRITEPAGE], cmds);
commandbuf[6] = cmds[0];
// otherwise, we need to load different commands in
I thought I had already committed this but I don't see it in CVS. This fixes EEPROM access using the STK500V2 programmer, partially undoing part of a previous general fixup commit. Choose the correct read/write operations with the stk500v2 program function - the correct one depends on the memory type. EEPROM is byte addressable so uses read/write. FLASH is word addressable and so uses read_lo/write_lo. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@501 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-09-16 21:23:14 +00:00
}
else {
commandbuf[3] = m->mode | 0x80; // yes, write the words to flash
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
I thought I had already committed this but I don't see it in CVS. This fixes EEPROM access using the STK500V2 programmer, partially undoing part of a previous general fixup commit. Choose the correct read/write operations with the stk500v2 program function - the correct one depends on the memory type. EEPROM is byte addressable so uses read/write. FLASH is word addressable and so uses read_lo/write_lo. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@501 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-09-16 21:23:14 +00:00
if (wop == NULL) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("write instruction not defined for part %s\n", p->desc);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -1;
}
Initialise memory before avr_set_bits() calls in stk500.c and stk500v2.c
2022-07-24 17:20:35 +00:00
memset(cmds, 0, sizeof cmds);
I thought I had already committed this but I don't see it in CVS. This fixes EEPROM access using the STK500V2 programmer, partially undoing part of a previous general fixup commit. Choose the correct read/write operations with the stk500v2 program function - the correct one depends on the memory type. EEPROM is byte addressable so uses read/write. FLASH is word addressable and so uses read_lo/write_lo. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@501 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-09-16 21:23:14 +00:00
avr_set_bits(wop, cmds);
This is patch # 4338, obsoletes patch #4327, provides fixes for bugs #13693, #13871, and #14212. This provides bug fixes to the STK500V2 programmer type. From the patch information: - incorrect token used from avrdude.conf.in - wrong command sent to programmer, hence no write to eeprom. - programmer was said to start writing at 0x0000 and continue page by page and was not repositionned when a gap was found in the hex file, or when the hex file start address was not 0x0000. Hence the verify procedure was correct, not the write procedure. - speed up of flash write to skip empty pages (full of 0xFF) by re-enabling a dedicated function for that task. - stk500v2_paged_load() was not returning the number of byte read, so empty hex files were generated when reading memory. Submitted by: Bernard Fouch <bernard.fouche@kuantic.com> git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@490 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-08-28 22:23:35 +00:00
commandbuf[5] = cmds[0];
commandbuf[6] = 0;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
}
// the read command is common to both methods
I thought I had already committed this but I don't see it in CVS. This fixes EEPROM access using the STK500V2 programmer, partially undoing part of a previous general fixup commit. Choose the correct read/write operations with the stk500v2 program function - the correct one depends on the memory type. EEPROM is byte addressable so uses read/write. FLASH is word addressable and so uses read_lo/write_lo. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@501 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-09-16 21:23:14 +00:00
if (rop == NULL) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("read instruction not defined for part %s\n", p->desc);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -1;
}
Initialise memory before avr_set_bits() calls in stk500.c and stk500v2.c
2022-07-24 17:20:35 +00:00
memset(cmds, 0, sizeof cmds);
I thought I had already committed this but I don't see it in CVS. This fixes EEPROM access using the STK500V2 programmer, partially undoing part of a previous general fixup commit. Choose the correct read/write operations with the stk500v2 program function - the correct one depends on the memory type. EEPROM is byte addressable so uses read/write. FLASH is word addressable and so uses read_lo/write_lo. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@501 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-09-16 21:23:14 +00:00
avr_set_bits(rop, cmds);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
commandbuf[7] = cmds[0];
commandbuf[8] = m->readback[0];
commandbuf[9] = m->readback[1];
Implement extended addressing needed for the ATmega256x devices. Document ATmega256x support. Also document Solaris port defaults in avrdude.texi. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@584 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-05-25 14:38:08 +00:00
last_addr=UINT_MAX; /* impossible address */
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
for (; addr < maxaddr; addr += page_size) {
if ((maxaddr - addr) < page_size)
block_size = maxaddr - addr;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
else
block_size = page_size;
This is patch # 4338, obsoletes patch #4327, provides fixes for bugs #13693, #13871, and #14212. This provides bug fixes to the STK500V2 programmer type. From the patch information: - incorrect token used from avrdude.conf.in - wrong command sent to programmer, hence no write to eeprom. - programmer was said to start writing at 0x0000 and continue page by page and was not repositionned when a gap was found in the hex file, or when the hex file start address was not 0x0000. Hence the verify procedure was correct, not the write procedure. - speed up of flash write to skip empty pages (full of 0xFF) by re-enabling a dedicated function for that task. - stk500v2_paged_load() was not returning the number of byte read, so empty hex files were generated when reading memory. Submitted by: Bernard Fouch <bernard.fouche@kuantic.com> git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@490 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-08-28 22:23:35 +00:00
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
DEBUG("block_size at addr %d is %d\n",addr,block_size);
memcpy(buf,commandbuf,sizeof(commandbuf));
buf[1] = block_size >> 8;
buf[2] = block_size & 0xff;
Implement extended addressing needed for the ATmega256x devices. Document ATmega256x support. Also document Solaris port defaults in avrdude.texi. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@584 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-05-25 14:38:08 +00:00
if((last_addr==UINT_MAX)||(last_addr+block_size != addr)){
Various fixes for Xmega devices. * avrdude.conf.in: Correctly declare EEPROM page sizes for all Xmega devices (0x20 instead of 0x100). * avr.c: If a memory region has a page size declared, try using the paged IO routines regardless of the target memory name. Xmega EEPROM requires to be written in paged mode. Correctly use a long (rather than unsigned long) variable to evaluate the success status of the paged mode write attempt. * stk500v2.c: Don't apply TIF space offsets twice (bug #27995: AVRDUDE 5.8svn fails to program and read XMEGA); use stk500v2_loadaddr() prior to paged mode (EEPROM and flash) writes, otherwise programming of flash areas will fail; while being there, check the return value of stk500v2_loadaddr() everywhere; use the correct write/erase mode bits (same as AVR Studio does). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@907 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-12 15:42:40 +00:00
if (stk500v2_loadaddr(pgm, use_ext_addr | (addr >> addrshift)) < 0)
return -1;
This is patch # 4338, obsoletes patch #4327, provides fixes for bugs #13693, #13871, and #14212. This provides bug fixes to the STK500V2 programmer type. From the patch information: - incorrect token used from avrdude.conf.in - wrong command sent to programmer, hence no write to eeprom. - programmer was said to start writing at 0x0000 and continue page by page and was not repositionned when a gap was found in the hex file, or when the hex file start address was not 0x0000. Hence the verify procedure was correct, not the write procedure. - speed up of flash write to skip empty pages (full of 0xFF) by re-enabling a dedicated function for that task. - stk500v2_paged_load() was not returning the number of byte read, so empty hex files were generated when reading memory. Submitted by: Bernard Fouch <bernard.fouche@kuantic.com> git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@490 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-08-28 22:23:35 +00:00
}
last_addr=addr;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
memcpy(buf+10,m->buf+addr, block_size);
result = stk500v2_command(pgm,buf,block_size+10, sizeof(buf));
Submitted by John Voltz: add AVR053 oscillator calibration. * main.c: Add the -O option. * pgm.c: Add the hook for the perform_osccal() method. * pgm.h: (Ditto.) * stk500v2.c: Implement perform_osccal(). * avrdude.1: Document the -O option. * doc/avrdude.texi: (Ditto.) Partially closes bug #17487: AVR RC oscillator calibration routine not supported (feature request) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@671 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-09 14:34:24 +00:00
if (result < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("write command failed\n");
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -1;
}
}
return n_bytes;
}
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
/*
* Write pages of flash/EEPROM, generic HV mode
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500hv_paged_write(const PROGRAMMER *pgm, const AVRPART *p, const AVRMEM *m,
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
unsigned int page_size,
unsigned int addr, unsigned int n_bytes,
enum hvmode mode)
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
{
Fix all "unused variable" warnings that were encountered. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1200 81a1dc3b-b13d-400b-aceb-764788c761c2
2013-09-03 11:05:13 +00:00
unsigned int block_size, last_addr, addrshift, use_ext_addr;
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
unsigned int maxaddr = addr + n_bytes;
Fix all bugs in the stk500pp implementation. Eliminate pagebuf, and use stack-allocated buffers instead. The pagesize of all current AVRs is at most 256 only anyway, and this is unlikely to change with the STK500v2 protocol. The previous pagebuf implementation suffered from some possible buffer overrun. In stk500pp_write_page(), do always write full pages, rather than attempting to write a partial last page which did not get written at all. Fill the remaining bytes with 0xff. For (paged) write operations, correctly synthesize the mode byte. This mode byte is very different from the ISP mode byte (sigh). In stk500pp_read_byte(), when performing read operations on paged memory, start reading at the previous page boundary rather than the current address. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@589 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-17 21:25:59 +00:00
unsigned char commandbuf[5], buf[266];
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
int result;
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
DEBUG("STK500V2: stk500hv_paged_write(..,%s,%u,%u,%u)\n",
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
m->desc, page_size, addr, n_bytes);
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
addrshift = 0;
use_ext_addr = 0;
// determine which command is to be used
if (strcmp(m->desc, "flash") == 0) {
addrshift = 1;
main.c, pgm.c, pgm.h: Add setup and teardown hooks to the programmer definition. If present, call the setup hook immediately after finding the respective programmer object, and schedule the teardown hook to be called upon exit. This allows the programmer implementation to dynamically allocate private programmer data. avr910.c, butterfly.c, jtagmkI.c, jtagmkII.c, stk500v2.c, usbasp.c, usbtiny.c: Convert static programmer data into dynamically allocated data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@764 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-11-07 20:36:12 +00:00
PDATA(pgm)->flash_pageaddr = (unsigned long)-1L;
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
commandbuf[0] = mode == PPMODE? CMD_PROGRAM_FLASH_PP: CMD_PROGRAM_FLASH_HVSP;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
/*
* If bit 31 is set, this indicates that the following read/write
* operation will be performed on a memory that is larger than
* 64KBytes. This is an indication to STK500 that a load extended
* address must be executed.
*/
if (m->op[AVR_OP_LOAD_EXT_ADDR] != NULL) {
use_ext_addr = (1U << 31);
}
} else if (strcmp(m->desc, "eeprom") == 0) {
main.c, pgm.c, pgm.h: Add setup and teardown hooks to the programmer definition. If present, call the setup hook immediately after finding the respective programmer object, and schedule the teardown hook to be called upon exit. This allows the programmer implementation to dynamically allocate private programmer data. avr910.c, butterfly.c, jtagmkI.c, jtagmkII.c, stk500v2.c, usbasp.c, usbtiny.c: Convert static programmer data into dynamically allocated data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@764 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-11-07 20:36:12 +00:00
PDATA(pgm)->eeprom_pageaddr = (unsigned long)-1L;
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
commandbuf[0] = mode == PPMODE? CMD_PROGRAM_EEPROM_PP: CMD_PROGRAM_EEPROM_HVSP;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
}
Fix all bugs in the stk500pp implementation. Eliminate pagebuf, and use stack-allocated buffers instead. The pagesize of all current AVRs is at most 256 only anyway, and this is unlikely to change with the STK500v2 protocol. The previous pagebuf implementation suffered from some possible buffer overrun. In stk500pp_write_page(), do always write full pages, rather than attempting to write a partial last page which did not get written at all. Fill the remaining bytes with 0xff. For (paged) write operations, correctly synthesize the mode byte. This mode byte is very different from the ISP mode byte (sigh). In stk500pp_read_byte(), when performing read operations on paged memory, start reading at the previous page boundary rather than the current address. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@589 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-17 21:25:59 +00:00
/*
* Synthesize the mode byte. This is simpler than adding yet
* another parameter to the avrdude.conf file. We calculate the
* bits corresponding to the page size, as explained in AVR068. We
* set bit 7, to indicate this is to actually write the page to the
* target device. We set bit 6 to indicate this is the very last
* page to be programmed, whatever this means -- we just pretend we
* don't know any better. ;-) Finally, we set bit 0 to say this is
* a paged memory, after all, that's why we got here at all.
*/
Add more parameters for PP mode. Fix the non-paged write operations for old AVRs. In avrdude.conf.in, use the new PP mode parameters; add PP mode definitions for AT90S8515. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@591 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-19 21:25:08 +00:00
commandbuf[3] = 0x80 | 0x40;
if (page_size > 2) {
Replace all occurences of exit() in potential library code by appropriate return values. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1301 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-05-16 15:52:25 +00:00
unsigned int rv = stk500v2_mode_for_pagesize(page_size);
if (rv == 0)
return -1;
commandbuf[3] |= rv;
Add more parameters for PP mode. Fix the non-paged write operations for old AVRs. In avrdude.conf.in, use the new PP mode parameters; add PP mode definitions for AT90S8515. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@591 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-19 21:25:08 +00:00
commandbuf[3] |= 0x01;
}
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
commandbuf[4] = m->delay;
Add more parameters for PP mode. Fix the non-paged write operations for old AVRs. In avrdude.conf.in, use the new PP mode parameters; add PP mode definitions for AT90S8515. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@591 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-19 21:25:08 +00:00
if (page_size == 0) page_size = 256;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
last_addr = UINT_MAX; /* impossible address */
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
for (; addr < maxaddr; addr += page_size) {
if ((maxaddr - addr) < page_size)
block_size = maxaddr - addr;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
else
block_size = page_size;
DEBUG("block_size at addr %d is %d\n",addr,block_size);
Fix all bugs in the stk500pp implementation. Eliminate pagebuf, and use stack-allocated buffers instead. The pagesize of all current AVRs is at most 256 only anyway, and this is unlikely to change with the STK500v2 protocol. The previous pagebuf implementation suffered from some possible buffer overrun. In stk500pp_write_page(), do always write full pages, rather than attempting to write a partial last page which did not get written at all. Fill the remaining bytes with 0xff. For (paged) write operations, correctly synthesize the mode byte. This mode byte is very different from the ISP mode byte (sigh). In stk500pp_read_byte(), when performing read operations on paged memory, start reading at the previous page boundary rather than the current address. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@589 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-17 21:25:59 +00:00
memcpy(buf, commandbuf, sizeof(commandbuf));
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
Fix all bugs in the stk500pp implementation. Eliminate pagebuf, and use stack-allocated buffers instead. The pagesize of all current AVRs is at most 256 only anyway, and this is unlikely to change with the STK500v2 protocol. The previous pagebuf implementation suffered from some possible buffer overrun. In stk500pp_write_page(), do always write full pages, rather than attempting to write a partial last page which did not get written at all. Fill the remaining bytes with 0xff. For (paged) write operations, correctly synthesize the mode byte. This mode byte is very different from the ISP mode byte (sigh). In stk500pp_read_byte(), when performing read operations on paged memory, start reading at the previous page boundary rather than the current address. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@589 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-17 21:25:59 +00:00
buf[1] = page_size >> 8;
buf[2] = page_size & 0xff;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
if ((last_addr == UINT_MAX) || (last_addr + block_size != addr)) {
Various fixes for Xmega devices. * avrdude.conf.in: Correctly declare EEPROM page sizes for all Xmega devices (0x20 instead of 0x100). * avr.c: If a memory region has a page size declared, try using the paged IO routines regardless of the target memory name. Xmega EEPROM requires to be written in paged mode. Correctly use a long (rather than unsigned long) variable to evaluate the success status of the paged mode write attempt. * stk500v2.c: Don't apply TIF space offsets twice (bug #27995: AVRDUDE 5.8svn fails to program and read XMEGA); use stk500v2_loadaddr() prior to paged mode (EEPROM and flash) writes, otherwise programming of flash areas will fail; while being there, check the return value of stk500v2_loadaddr() everywhere; use the correct write/erase mode bits (same as AVR Studio does). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@907 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-12 15:42:40 +00:00
if (stk500v2_loadaddr(pgm, use_ext_addr | (addr >> addrshift)) < 0)
return -1;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
}
last_addr=addr;
Fix all bugs in the stk500pp implementation. Eliminate pagebuf, and use stack-allocated buffers instead. The pagesize of all current AVRs is at most 256 only anyway, and this is unlikely to change with the STK500v2 protocol. The previous pagebuf implementation suffered from some possible buffer overrun. In stk500pp_write_page(), do always write full pages, rather than attempting to write a partial last page which did not get written at all. Fill the remaining bytes with 0xff. For (paged) write operations, correctly synthesize the mode byte. This mode byte is very different from the ISP mode byte (sigh). In stk500pp_read_byte(), when performing read operations on paged memory, start reading at the previous page boundary rather than the current address. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@589 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-17 21:25:59 +00:00
memcpy(buf + 5, m->buf + addr, block_size);
if (block_size != page_size)
memset(buf + 5 + block_size, 0xff, page_size - block_size);
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
Fix all bugs in the stk500pp implementation. Eliminate pagebuf, and use stack-allocated buffers instead. The pagesize of all current AVRs is at most 256 only anyway, and this is unlikely to change with the STK500v2 protocol. The previous pagebuf implementation suffered from some possible buffer overrun. In stk500pp_write_page(), do always write full pages, rather than attempting to write a partial last page which did not get written at all. Fill the remaining bytes with 0xff. For (paged) write operations, correctly synthesize the mode byte. This mode byte is very different from the ISP mode byte (sigh). In stk500pp_read_byte(), when performing read operations on paged memory, start reading at the previous page boundary rather than the current address. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@589 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-17 21:25:59 +00:00
result = stk500v2_command(pgm, buf, page_size + 5, sizeof(buf));
Submitted by John Voltz: add AVR053 oscillator calibration. * main.c: Add the -O option. * pgm.c: Add the hook for the perform_osccal() method. * pgm.h: (Ditto.) * stk500v2.c: Implement perform_osccal(). * avrdude.1: Document the -O option. * doc/avrdude.texi: (Ditto.) Partially closes bug #17487: AVR RC oscillator calibration routine not supported (feature request) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@671 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-09 14:34:24 +00:00
if (result < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("write command failed\n");
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
return -1;
}
}
return n_bytes;
}
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
/*
* Write pages of flash/EEPROM, PP mode
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500pp_paged_write(const PROGRAMMER *pgm, const AVRPART *p, const AVRMEM *m,
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
unsigned int page_size,
unsigned int addr, unsigned int n_bytes)
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
{
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
return stk500hv_paged_write(pgm, p, m, page_size, addr, n_bytes, PPMODE);
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
}
/*
* Write pages of flash/EEPROM, HVSP mode
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500hvsp_paged_write(const PROGRAMMER *pgm, const AVRPART *p, const AVRMEM *m,
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
unsigned int page_size,
unsigned int addr, unsigned int n_bytes)
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
{
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
return stk500hv_paged_write(pgm, p, m, page_size, addr, n_bytes, HVSPMODE);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_paged_load(const PROGRAMMER *pgm, const AVRPART *p, const AVRMEM *m,
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
unsigned int page_size,
unsigned int addr, unsigned int n_bytes)
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
{
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
unsigned int block_size, hiaddr, addrshift, use_ext_addr;
unsigned int maxaddr = addr + n_bytes;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
unsigned char commandbuf[4];
unsigned char buf[275]; // max buffer size for stk500v2 at this point
unsigned char cmds[4];
This is patch # 4338, obsoletes patch #4327, provides fixes for bugs #13693, #13871, and #14212. This provides bug fixes to the STK500V2 programmer type. From the patch information: - incorrect token used from avrdude.conf.in - wrong command sent to programmer, hence no write to eeprom. - programmer was said to start writing at 0x0000 and continue page by page and was not repositionned when a gap was found in the hex file, or when the hex file start address was not 0x0000. Hence the verify procedure was correct, not the write procedure. - speed up of flash write to skip empty pages (full of 0xFF) by re-enabling a dedicated function for that task. - stk500v2_paged_load() was not returning the number of byte read, so empty hex files were generated when reading memory. Submitted by: Bernard Fouch <bernard.fouche@kuantic.com> git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@490 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-08-28 22:23:35 +00:00
int result;
I thought I had already committed this but I don't see it in CVS. This fixes EEPROM access using the STK500V2 programmer, partially undoing part of a previous general fixup commit. Choose the correct read/write operations with the stk500v2 program function - the correct one depends on the memory type. EEPROM is byte addressable so uses read/write. FLASH is word addressable and so uses read_lo/write_lo. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@501 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-09-16 21:23:14 +00:00
OPCODE * rop;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
DEBUG("STK500V2: stk500v2_paged_load(..,%s,%u,%u,%u)\n",
m->desc, page_size, addr, n_bytes);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
page_size = m->readsize;
I thought I had already committed this but I don't see it in CVS. This fixes EEPROM access using the STK500V2 programmer, partially undoing part of a previous general fixup commit. Choose the correct read/write operations with the stk500v2 program function - the correct one depends on the memory type. EEPROM is byte addressable so uses read/write. FLASH is word addressable and so uses read_lo/write_lo. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@501 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-09-16 21:23:14 +00:00
rop = m->op[AVR_OP_READ];
Implement extended addressing needed for the ATmega256x devices. Document ATmega256x support. Also document Solaris port defaults in avrdude.texi. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@584 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-05-25 14:38:08 +00:00
hiaddr = UINT_MAX;
addrshift = 0;
use_ext_addr = 0;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
// determine which command is to be used
if (strcmp(m->desc, "flash") == 0) {
commandbuf[0] = CMD_READ_FLASH_ISP;
I thought I had already committed this but I don't see it in CVS. This fixes EEPROM access using the STK500V2 programmer, partially undoing part of a previous general fixup commit. Choose the correct read/write operations with the stk500v2 program function - the correct one depends on the memory type. EEPROM is byte addressable so uses read/write. FLASH is word addressable and so uses read_lo/write_lo. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@501 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-09-16 21:23:14 +00:00
rop = m->op[AVR_OP_READ_LO];
Implement extended addressing needed for the ATmega256x devices. Document ATmega256x support. Also document Solaris port defaults in avrdude.texi. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@584 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-05-25 14:38:08 +00:00
addrshift = 1;
/*
* If bit 31 is set, this indicates that the following read/write
* operation will be performed on a memory that is larger than
* 64KBytes. This is an indication to STK500 that a load extended
* address must be executed.
*/
if (m->op[AVR_OP_LOAD_EXT_ADDR] != NULL) {
use_ext_addr = (1U << 31);
}
}
I thought I had already committed this but I don't see it in CVS. This fixes EEPROM access using the STK500V2 programmer, partially undoing part of a previous general fixup commit. Choose the correct read/write operations with the stk500v2 program function - the correct one depends on the memory type. EEPROM is byte addressable so uses read/write. FLASH is word addressable and so uses read_lo/write_lo. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@501 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-09-16 21:23:14 +00:00
else if (strcmp(m->desc, "eeprom") == 0) {
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
commandbuf[0] = CMD_READ_EEPROM_ISP;
}
// the read command is common to both methods
I thought I had already committed this but I don't see it in CVS. This fixes EEPROM access using the STK500V2 programmer, partially undoing part of a previous general fixup commit. Choose the correct read/write operations with the stk500v2 program function - the correct one depends on the memory type. EEPROM is byte addressable so uses read/write. FLASH is word addressable and so uses read_lo/write_lo. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@501 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-09-16 21:23:14 +00:00
if (rop == NULL) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("read instruction not defined for part %s\n", p->desc);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -1;
}
Initialise memory before avr_set_bits() calls in stk500.c and stk500v2.c
2022-07-24 17:20:35 +00:00
memset(cmds, 0, sizeof cmds);
I thought I had already committed this but I don't see it in CVS. This fixes EEPROM access using the STK500V2 programmer, partially undoing part of a previous general fixup commit. Choose the correct read/write operations with the stk500v2 program function - the correct one depends on the memory type. EEPROM is byte addressable so uses read/write. FLASH is word addressable and so uses read_lo/write_lo. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@501 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-09-16 21:23:14 +00:00
avr_set_bits(rop, cmds);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
commandbuf[3] = cmds[0];
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
for (; addr < maxaddr; addr += page_size) {
if ((maxaddr - addr) < page_size)
block_size = maxaddr - addr;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
else
block_size = page_size;
DEBUG("block_size at addr %d is %d\n",addr,block_size);
memcpy(buf,commandbuf,sizeof(commandbuf));
buf[1] = block_size >> 8;
buf[2] = block_size & 0xff;
Implement extended addressing needed for the ATmega256x devices. Document ATmega256x support. Also document Solaris port defaults in avrdude.texi. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@584 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-05-25 14:38:08 +00:00
// Ensure a new "load extended address" will be issued
// when crossing a 64 KB boundary in flash.
if (hiaddr != (addr & ~0xFFFF)) {
hiaddr = addr & ~0xFFFF;
Various fixes for Xmega devices. * avrdude.conf.in: Correctly declare EEPROM page sizes for all Xmega devices (0x20 instead of 0x100). * avr.c: If a memory region has a page size declared, try using the paged IO routines regardless of the target memory name. Xmega EEPROM requires to be written in paged mode. Correctly use a long (rather than unsigned long) variable to evaluate the success status of the paged mode write attempt. * stk500v2.c: Don't apply TIF space offsets twice (bug #27995: AVRDUDE 5.8svn fails to program and read XMEGA); use stk500v2_loadaddr() prior to paged mode (EEPROM and flash) writes, otherwise programming of flash areas will fail; while being there, check the return value of stk500v2_loadaddr() everywhere; use the correct write/erase mode bits (same as AVR Studio does). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@907 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-12 15:42:40 +00:00
if (stk500v2_loadaddr(pgm, use_ext_addr | (addr >> addrshift)) < 0)
return -1;
Implement extended addressing needed for the ATmega256x devices. Document ATmega256x support. Also document Solaris port defaults in avrdude.texi. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@584 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-05-25 14:38:08 +00:00
}
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
result = stk500v2_command(pgm,buf,4,sizeof(buf));
Submitted by John Voltz: add AVR053 oscillator calibration. * main.c: Add the -O option. * pgm.c: Add the hook for the perform_osccal() method. * pgm.h: (Ditto.) * stk500v2.c: Implement perform_osccal(). * avrdude.1: Document the -O option. * doc/avrdude.texi: (Ditto.) Partially closes bug #17487: AVR RC oscillator calibration routine not supported (feature request) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@671 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-09 14:34:24 +00:00
if (result < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("read command failed\n");
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -1;
}
#if 0
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
for (i=0; i<page_size; i++) {
msg_info("%02X", buf[2+i]);
if (i%16 == 15)
msg_info("\n");
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
}
#endif
memcpy(&m->buf[addr], &buf[2], block_size);
}
This is patch # 4338, obsoletes patch #4327, provides fixes for bugs #13693, #13871, and #14212. This provides bug fixes to the STK500V2 programmer type. From the patch information: - incorrect token used from avrdude.conf.in - wrong command sent to programmer, hence no write to eeprom. - programmer was said to start writing at 0x0000 and continue page by page and was not repositionned when a gap was found in the hex file, or when the hex file start address was not 0x0000. Hence the verify procedure was correct, not the write procedure. - speed up of flash write to skip empty pages (full of 0xFF) by re-enabling a dedicated function for that task. - stk500v2_paged_load() was not returning the number of byte read, so empty hex files were generated when reading memory. Submitted by: Bernard Fouch <bernard.fouche@kuantic.com> git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@490 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-08-28 22:23:35 +00:00
return n_bytes;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
}
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
/*
* Read pages of flash/EEPROM, generic HV mode
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500hv_paged_load(const PROGRAMMER *pgm, const AVRPART *p, const AVRMEM *m,
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
unsigned int page_size,
unsigned int addr, unsigned int n_bytes,
enum hvmode mode)
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
{
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
unsigned int block_size, hiaddr, addrshift, use_ext_addr;
unsigned int maxaddr = addr + n_bytes;
Fix all bugs in the stk500pp implementation. Eliminate pagebuf, and use stack-allocated buffers instead. The pagesize of all current AVRs is at most 256 only anyway, and this is unlikely to change with the STK500v2 protocol. The previous pagebuf implementation suffered from some possible buffer overrun. In stk500pp_write_page(), do always write full pages, rather than attempting to write a partial last page which did not get written at all. Fill the remaining bytes with 0xff. For (paged) write operations, correctly synthesize the mode byte. This mode byte is very different from the ISP mode byte (sigh). In stk500pp_read_byte(), when performing read operations on paged memory, start reading at the previous page boundary rather than the current address. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@589 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-17 21:25:59 +00:00
unsigned char commandbuf[3], buf[266];
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
int result;
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
DEBUG("STK500V2: stk500hv_paged_load(..,%s,%u,%u,%u)\n",
m->desc, page_size, addr, n_bytes);
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
page_size = m->readsize;
hiaddr = UINT_MAX;
addrshift = 0;
use_ext_addr = 0;
// determine which command is to be used
if (strcmp(m->desc, "flash") == 0) {
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
commandbuf[0] = mode == PPMODE? CMD_READ_FLASH_PP: CMD_READ_FLASH_HVSP;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
addrshift = 1;
/*
* If bit 31 is set, this indicates that the following read/write
* operation will be performed on a memory that is larger than
* 64KBytes. This is an indication to STK500 that a load extended
* address must be executed.
*/
if (m->op[AVR_OP_LOAD_EXT_ADDR] != NULL) {
use_ext_addr = (1U << 31);
}
}
else if (strcmp(m->desc, "eeprom") == 0) {
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
commandbuf[0] = mode == PPMODE? CMD_READ_EEPROM_PP: CMD_READ_EEPROM_HVSP;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
}
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
for (; addr < maxaddr; addr += page_size) {
if ((maxaddr - addr) < page_size)
block_size = maxaddr - addr;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
else
block_size = page_size;
DEBUG("block_size at addr %d is %d\n",addr,block_size);
Fix all bugs in the stk500pp implementation. Eliminate pagebuf, and use stack-allocated buffers instead. The pagesize of all current AVRs is at most 256 only anyway, and this is unlikely to change with the STK500v2 protocol. The previous pagebuf implementation suffered from some possible buffer overrun. In stk500pp_write_page(), do always write full pages, rather than attempting to write a partial last page which did not get written at all. Fill the remaining bytes with 0xff. For (paged) write operations, correctly synthesize the mode byte. This mode byte is very different from the ISP mode byte (sigh). In stk500pp_read_byte(), when performing read operations on paged memory, start reading at the previous page boundary rather than the current address. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@589 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-17 21:25:59 +00:00
memcpy(buf, commandbuf, sizeof(commandbuf));
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
Fix all bugs in the stk500pp implementation. Eliminate pagebuf, and use stack-allocated buffers instead. The pagesize of all current AVRs is at most 256 only anyway, and this is unlikely to change with the STK500v2 protocol. The previous pagebuf implementation suffered from some possible buffer overrun. In stk500pp_write_page(), do always write full pages, rather than attempting to write a partial last page which did not get written at all. Fill the remaining bytes with 0xff. For (paged) write operations, correctly synthesize the mode byte. This mode byte is very different from the ISP mode byte (sigh). In stk500pp_read_byte(), when performing read operations on paged memory, start reading at the previous page boundary rather than the current address. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@589 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-17 21:25:59 +00:00
buf[1] = block_size >> 8;
buf[2] = block_size & 0xff;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
// Ensure a new "load extended address" will be issued
// when crossing a 64 KB boundary in flash.
if (hiaddr != (addr & ~0xFFFF)) {
hiaddr = addr & ~0xFFFF;
Various fixes for Xmega devices. * avrdude.conf.in: Correctly declare EEPROM page sizes for all Xmega devices (0x20 instead of 0x100). * avr.c: If a memory region has a page size declared, try using the paged IO routines regardless of the target memory name. Xmega EEPROM requires to be written in paged mode. Correctly use a long (rather than unsigned long) variable to evaluate the success status of the paged mode write attempt. * stk500v2.c: Don't apply TIF space offsets twice (bug #27995: AVRDUDE 5.8svn fails to program and read XMEGA); use stk500v2_loadaddr() prior to paged mode (EEPROM and flash) writes, otherwise programming of flash areas will fail; while being there, check the return value of stk500v2_loadaddr() everywhere; use the correct write/erase mode bits (same as AVR Studio does). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@907 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-12 15:42:40 +00:00
if (stk500v2_loadaddr(pgm, use_ext_addr | (addr >> addrshift)) < 0)
return -1;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
}
Fix all bugs in the stk500pp implementation. Eliminate pagebuf, and use stack-allocated buffers instead. The pagesize of all current AVRs is at most 256 only anyway, and this is unlikely to change with the STK500v2 protocol. The previous pagebuf implementation suffered from some possible buffer overrun. In stk500pp_write_page(), do always write full pages, rather than attempting to write a partial last page which did not get written at all. Fill the remaining bytes with 0xff. For (paged) write operations, correctly synthesize the mode byte. This mode byte is very different from the ISP mode byte (sigh). In stk500pp_read_byte(), when performing read operations on paged memory, start reading at the previous page boundary rather than the current address. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@589 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-17 21:25:59 +00:00
result = stk500v2_command(pgm, buf, 3, sizeof(buf));
Submitted by John Voltz: add AVR053 oscillator calibration. * main.c: Add the -O option. * pgm.c: Add the hook for the perform_osccal() method. * pgm.h: (Ditto.) * stk500v2.c: Implement perform_osccal(). * avrdude.1: Document the -O option. * doc/avrdude.texi: (Ditto.) Partially closes bug #17487: AVR RC oscillator calibration routine not supported (feature request) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@671 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-09 14:34:24 +00:00
if (result < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("read command failed\n");
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
return -1;
}
#if 0
for (i = 0; i < page_size; i++) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
msg_info("%02X", buf[2 + i]);
if (i % 16 == 15)
msg_info("\n");
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
}
#endif
Fix all bugs in the stk500pp implementation. Eliminate pagebuf, and use stack-allocated buffers instead. The pagesize of all current AVRs is at most 256 only anyway, and this is unlikely to change with the STK500v2 protocol. The previous pagebuf implementation suffered from some possible buffer overrun. In stk500pp_write_page(), do always write full pages, rather than attempting to write a partial last page which did not get written at all. Fill the remaining bytes with 0xff. For (paged) write operations, correctly synthesize the mode byte. This mode byte is very different from the ISP mode byte (sigh). In stk500pp_read_byte(), when performing read operations on paged memory, start reading at the previous page boundary rather than the current address. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@589 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-17 21:25:59 +00:00
memcpy(&m->buf[addr], &buf[2], block_size);
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
}
return n_bytes;
}
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
/*
* Read pages of flash/EEPROM, PP mode
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500pp_paged_load(const PROGRAMMER *pgm, const AVRPART *p, const AVRMEM *m,
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
unsigned int page_size,
unsigned int addr, unsigned int n_bytes)
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
{
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
return stk500hv_paged_load(pgm, p, m, page_size, addr, n_bytes, PPMODE);
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
}
/*
* Read pages of flash/EEPROM, HVSP mode
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500hvsp_paged_load(const PROGRAMMER *pgm, const AVRPART *p, const AVRMEM *m,
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
unsigned int page_size,
unsigned int addr, unsigned int n_bytes)
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
{
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
return stk500hv_paged_load(pgm, p, m, page_size, addr, n_bytes, HVSPMODE);
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
}
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_page_erase(const PROGRAMMER *pgm, const AVRPART *p, const AVRMEM *m,
Xmega page erase implementation for XPROG (AVRISPmkII, STK600) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1090 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-05-04 15:47:26 +00:00
unsigned int addr)
{
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("this function must never be called\n");
Xmega page erase implementation for XPROG (AVRISPmkII, STK600) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1090 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-05-04 15:47:26 +00:00
return -1;
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_set_vtarget(const PROGRAMMER *pgm, double v) {
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
unsigned char uaref, utarg;
utarg = (unsigned)((v + 0.049) * 10);
if (stk500v2_getparm(pgm, PARAM_VADJUST, &uaref) != 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("cannot obtain V[aref]\n");
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -1;
}
if (uaref > utarg) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_warning("reducing V[aref] from %.1f to %.1f\n", uaref/10.0, v);
if (stk500v2_setparm(pgm, PARAM_VADJUST, utarg) != 0)
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -1;
}
return stk500v2_setparm(pgm, PARAM_VTARGET, utarg);
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_set_varef(const PROGRAMMER *pgm, unsigned int chan /* unused */,
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
double v)
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
{
unsigned char uaref, utarg;
uaref = (unsigned)((v + 0.049) * 10);
if (stk500v2_getparm(pgm, PARAM_VTARGET, &utarg) != 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("cannot obtain V[target]\n");
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -1;
}
if (uaref > utarg) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("V[aref] must not be greater than "
"V[target] = %.1f\n", utarg/10.0);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -1;
}
return stk500v2_setparm(pgm, PARAM_VADJUST, uaref);
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_set_fosc(const PROGRAMMER *pgm, double v) {
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
int fosc;
unsigned char prescale, cmatch;
static unsigned ps[] = {
1, 8, 32, 64, 128, 256, 1024
};
Resolve signed/unsigned comparisons in stk500.c and stk500v2.c
2022-07-24 18:41:42 +00:00
size_t idx;
int rc;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
prescale = cmatch = 0;
if (v > 0.0) {
if (v > STK500V2_XTAL / 2) {
const char *unit;
if (v > 1e6) {
v /= 1e6;
unit = "MHz";
} else if (v > 1e3) {
v /= 1e3;
unit = "kHz";
} else
unit = "Hz";
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_warning("f = %.3f %s too high, using %.3f MHz\n",
v, unit, STK500V2_XTAL / 2e6);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
fosc = STK500V2_XTAL / 2;
} else
fosc = (unsigned)v;
for (idx = 0; idx < sizeof(ps) / sizeof(ps[0]); idx++) {
Resolve signed/unsigned comparisons in stk500.c and stk500v2.c
2022-07-24 18:41:42 +00:00
if ((unsigned) fosc >= STK500V2_XTAL / (256 * ps[idx] * 2)) {
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
/* this prescaler value can handle our frequency */
prescale = idx + 1;
cmatch = (unsigned)(STK500V2_XTAL / (2 * fosc * ps[idx])) - 1;
break;
}
}
if (idx == sizeof(ps) / sizeof(ps[0])) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_warning("f = %u Hz too low, %u Hz min\n",
fosc, STK500V2_XTAL / (256 * 1024 * 2));
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -1;
}
}
if ((rc = stk500v2_setparm(pgm, PARAM_OSC_PSCALE, prescale)) != 0
|| (rc = stk500v2_setparm(pgm, PARAM_OSC_CMATCH, cmatch)) != 0)
return rc;
return 0;
}
Contributed by dcm@mit.edu: add support for the AVRISP mkII device. (Savannah patch #4789.) * serial.h: Declare usb_serdev_frame device descriptor. * stk500v2.c: Implementation of the AVRISP mkII handling. * usb_libusb.c: Add USB handling for short-frame delimited AVRISP mkII USB protocol; add distinction of different devices in usbdev_open(). * jtagmkII.c: Tell usbdev_open() to search for the JTAG ICE mkII. * usbdevs.h: (New file.) * Makefile.am: Add usbdevs.h, as well as some other forgotten files "make distcheck" complained about. * avrdude.conf.in: Add more aliases for the AVRISP mkII. * avrdude.1: Document how to use the AVRISP mkII. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@564 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-12 23:13:50 +00:00
/* The list of SCK frequencies supported by the AVRISP mkII, as listed
* in AVR069 */
Major code cleanup. - Make all internal functions "static". - Make sure each module's header and implementation file match. - Remove all library-like functionality from main.c, so only the actual frontend remains in main.c. - Add C++ brackets to all header files. That effectively leaves the various module C files as something like an "avrdude library", with main.c being the currently only frontend program for that library. In theory, it should be possible to write different frontends using the same library backend functions though. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@722 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-01-24 22:43:46 +00:00
static double avrispmkIIfreqs[] = {
Contributed by dcm@mit.edu: add support for the AVRISP mkII device. (Savannah patch #4789.) * serial.h: Declare usb_serdev_frame device descriptor. * stk500v2.c: Implementation of the AVRISP mkII handling. * usb_libusb.c: Add USB handling for short-frame delimited AVRISP mkII USB protocol; add distinction of different devices in usbdev_open(). * jtagmkII.c: Tell usbdev_open() to search for the JTAG ICE mkII. * usbdevs.h: (New file.) * Makefile.am: Add usbdevs.h, as well as some other forgotten files "make distcheck" complained about. * avrdude.conf.in: Add more aliases for the AVRISP mkII. * avrdude.1: Document how to use the AVRISP mkII. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@564 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-12 23:13:50 +00:00
8000000, 4000000, 2000000, 1000000, 500000, 250000, 125000,
96386, 89888, 84211, 79208, 74767, 70797, 67227, 64000,
61069, 58395, 55945, 51613, 49690, 47905, 46243, 43244,
41885, 39409, 38278, 36200, 34335, 32654, 31129, 29740,
28470, 27304, 25724, 24768, 23461, 22285, 21221, 20254,
19371, 18562, 17583, 16914, 16097, 15356, 14520, 13914,
13224, 12599, 12031, 11511, 10944, 10431, 9963, 9468,
9081, 8612, 8239, 7851, 7498, 7137, 6809, 6478, 6178,
5879, 5607, 5359, 5093, 4870, 4633, 4418, 4209, 4019,
3823, 3645, 3474, 3310, 3161, 3011, 2869, 2734, 2611,
2484, 2369, 2257, 2152, 2052, 1956, 1866, 1779, 1695,
1615, 1539, 1468, 1398, 1333, 1271, 1212, 1155, 1101,
1049, 1000, 953, 909, 866, 826, 787, 750, 715, 682,
650, 619, 590, 563, 536, 511, 487, 465, 443, 422,
402, 384, 366, 349, 332, 317, 302, 288, 274, 261,
249, 238, 226, 216, 206, 196, 187, 178, 170, 162,
154, 147, 140, 134, 128, 122, 116, 111, 105, 100,
95.4, 90.9, 86.6, 82.6, 78.7, 75.0, 71.5, 68.2,
65.0, 61.9, 59.0, 56.3, 53.6, 51.1
};
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_set_sck_period_mk2(const PROGRAMMER *pgm, double v) {
Resolve signed/unsigned comparisons in stk500.c and stk500v2.c
2022-07-24 18:41:42 +00:00
size_t i;
Contributed by dcm@mit.edu: add support for the AVRISP mkII device. (Savannah patch #4789.) * serial.h: Declare usb_serdev_frame device descriptor. * stk500v2.c: Implementation of the AVRISP mkII handling. * usb_libusb.c: Add USB handling for short-frame delimited AVRISP mkII USB protocol; add distinction of different devices in usbdev_open(). * jtagmkII.c: Tell usbdev_open() to search for the JTAG ICE mkII. * usbdevs.h: (New file.) * Makefile.am: Add usbdevs.h, as well as some other forgotten files "make distcheck" complained about. * avrdude.conf.in: Add more aliases for the AVRISP mkII. * avrdude.1: Document how to use the AVRISP mkII. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@564 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-12 23:13:50 +00:00
Fix handling of SCK parameter table in AVRISP2 Submitted by Adrian Klieber patch #9820: Fix some out-of-bounds/uninitialized issues git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1440 81a1dc3b-b13d-400b-aceb-764788c761c2
2020-09-16 21:15:30 +00:00
for (i = 0; i < sizeof(avrispmkIIfreqs) / sizeof(avrispmkIIfreqs[0]); i++) {
Contributed by dcm@mit.edu: add support for the AVRISP mkII device. (Savannah patch #4789.) * serial.h: Declare usb_serdev_frame device descriptor. * stk500v2.c: Implementation of the AVRISP mkII handling. * usb_libusb.c: Add USB handling for short-frame delimited AVRISP mkII USB protocol; add distinction of different devices in usbdev_open(). * jtagmkII.c: Tell usbdev_open() to search for the JTAG ICE mkII. * usbdevs.h: (New file.) * Makefile.am: Add usbdevs.h, as well as some other forgotten files "make distcheck" complained about. * avrdude.conf.in: Add more aliases for the AVRISP mkII. * avrdude.1: Document how to use the AVRISP mkII. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@564 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-12 23:13:50 +00:00
if (1 / avrispmkIIfreqs[i] >= v)
break;
}
Fix handling of SCK parameter table in AVRISP2 Submitted by Adrian Klieber patch #9820: Fix some out-of-bounds/uninitialized issues git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1440 81a1dc3b-b13d-400b-aceb-764788c761c2
2020-09-16 21:15:30 +00:00
if (i >= sizeof(avrispmkIIfreqs) / sizeof(avrispmkIIfreqs[0])) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("invalid SCK period: %g\n", v);
Fix handling of SCK parameter table in AVRISP2 Submitted by Adrian Klieber patch #9820: Fix some out-of-bounds/uninitialized issues git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1440 81a1dc3b-b13d-400b-aceb-764788c761c2
2020-09-16 21:15:30 +00:00
return -1;
}
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
msg_notice2("Using p = %.2f us for SCK (param = %d)\n",
1000000 / avrispmkIIfreqs[i], (int) i);
Contributed by dcm@mit.edu: add support for the AVRISP mkII device. (Savannah patch #4789.) * serial.h: Declare usb_serdev_frame device descriptor. * stk500v2.c: Implementation of the AVRISP mkII handling. * usb_libusb.c: Add USB handling for short-frame delimited AVRISP mkII USB protocol; add distinction of different devices in usbdev_open(). * jtagmkII.c: Tell usbdev_open() to search for the JTAG ICE mkII. * usbdevs.h: (New file.) * Makefile.am: Add usbdevs.h, as well as some other forgotten files "make distcheck" complained about. * avrdude.conf.in: Add more aliases for the AVRISP mkII. * avrdude.1: Document how to use the AVRISP mkII. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@564 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-12 23:13:50 +00:00
return stk500v2_setparm(pgm, PARAM_SCK_DURATION, i);
}
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
Fix all bugs in the stk500pp implementation. Eliminate pagebuf, and use stack-allocated buffers instead. The pagesize of all current AVRs is at most 256 only anyway, and this is unlikely to change with the STK500v2 protocol. The previous pagebuf implementation suffered from some possible buffer overrun. In stk500pp_write_page(), do always write full pages, rather than attempting to write a partial last page which did not get written at all. Fill the remaining bytes with 0xff. For (paged) write operations, correctly synthesize the mode byte. This mode byte is very different from the ISP mode byte (sigh). In stk500pp_read_byte(), when performing read operations on paged memory, start reading at the previous page boundary rather than the current address. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@589 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-17 21:25:59 +00:00
/*
* Return the "mode" value for the parallel and HVSP modes that
* corresponds to the pagesize.
*/
static unsigned int stk500v2_mode_for_pagesize(unsigned int pagesize)
{
switch (pagesize)
{
case 256: return 0u << 1;
case 2: return 1u << 1;
case 4: return 2u << 1;
case 8: return 3u << 1;
case 16: return 4u << 1;
case 32: return 5u << 1;
case 64: return 6u << 1;
case 128: return 7u << 1;
}
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("invalid pagesize: %u\n", pagesize);
Replace all occurences of exit() in potential library code by appropriate return values. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1301 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-05-16 15:52:25 +00:00
return 0;
Fix all bugs in the stk500pp implementation. Eliminate pagebuf, and use stack-allocated buffers instead. The pagesize of all current AVRs is at most 256 only anyway, and this is unlikely to change with the STK500v2 protocol. The previous pagebuf implementation suffered from some possible buffer overrun. In stk500pp_write_page(), do always write full pages, rather than attempting to write a partial last page which did not get written at all. Fill the remaining bytes with 0xff. For (paged) write operations, correctly synthesize the mode byte. This mode byte is very different from the ISP mode byte (sigh). In stk500pp_read_byte(), when performing read operations on paged memory, start reading at the previous page boundary rather than the current address. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@589 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-17 21:25:59 +00:00
}
* stk500v2.c (stk500v2_set_sck_period): Revamp this to match the description/pseudo-code in appnote AVR068. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1229 81a1dc3b-b13d-400b-aceb-764788c761c2
2013-09-13 21:32:00 +00:00
/*
* See pseudo-code in AVR068
*
* This algorithm only fits for the STK500 itself. For the (old)
* AVRISP, the resulting ISP clock is only half. While this would be
* easy to fix in the algorithm, we'd need to add another
* configuration flag for this to the config file. Given the old
* AVRISP devices are virtually no longer around (and the AVRISPmkII
* uses a different algorithm below), it's probably not worth the
* hassle.
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_set_sck_period(const PROGRAMMER *pgm, double v) {
* stk500v2.c (stk500v2_set_sck_period): Revamp this to match the description/pseudo-code in appnote AVR068. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1229 81a1dc3b-b13d-400b-aceb-764788c761c2
2013-09-13 21:32:00 +00:00
unsigned int d;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
unsigned char dur;
* stk500v2.c (stk500v2_set_sck_period): Revamp this to match the description/pseudo-code in appnote AVR068. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1229 81a1dc3b-b13d-400b-aceb-764788c761c2
2013-09-13 21:32:00 +00:00
double f = 1 / v;
if (f >= 1.8432E6)
d = 0;
else if (f > 460.8E3)
d = 1;
else if (f > 115.2E3)
d = 2;
else if (f > 57.6E3)
d = 3;
else
d = (unsigned int)ceil(1 / (24 * f / (double)STK500V2_XTAL) - 10.0 / 12.0);
if (d >= 255)
d = 254;
dur = d;
return stk500v2_setparm(pgm, PARAM_SCK_DURATION, dur);
}
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static double stk500v2_sck_to_us(const PROGRAMMER *pgm, unsigned char dur) {
* stk500v2.c (stk500v2_set_sck_period): Revamp this to match the description/pseudo-code in appnote AVR068. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1229 81a1dc3b-b13d-400b-aceb-764788c761c2
2013-09-13 21:32:00 +00:00
double x;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
* stk500v2.c (stk500v2_set_sck_period): Revamp this to match the description/pseudo-code in appnote AVR068. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1229 81a1dc3b-b13d-400b-aceb-764788c761c2
2013-09-13 21:32:00 +00:00
if (dur == 0)
return 0.5425;
if (dur == 1)
return 2.17;
if (dur == 2)
return 8.68;
if (dur == 3)
return 17.36;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
* stk500v2.c (stk500v2_set_sck_period): Revamp this to match the description/pseudo-code in appnote AVR068. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1229 81a1dc3b-b13d-400b-aceb-764788c761c2
2013-09-13 21:32:00 +00:00
x = (double)dur + 10.0 / 12.0;
x = 1.0 / x;
x /= 24.0;
x *= (double)STK500V2_XTAL;
return 1E6 / x;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk600_set_vtarget(const PROGRAMMER *pgm, double v) {
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
unsigned char utarg;
unsigned int uaref;
int rv;
utarg = (unsigned)((v + 0.049) * 10);
if (stk500v2_getparm2(pgm, PARAM2_AREF0, &uaref) != 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("cannot obtain V[aref][0]\n");
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
return -1;
}
if (uaref > (unsigned)utarg * 10) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_warning("reducing V[aref][0] from %.2f to %.1f\n", uaref/100.0, v);
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
uaref = 10 * (unsigned)utarg;
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
if (stk500v2_setparm2(pgm, PARAM2_AREF0, uaref) != 0)
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
return -1;
}
if (stk500v2_getparm2(pgm, PARAM2_AREF1, &uaref) != 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("cannot obtain V[aref][1]\n");
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
return -1;
}
if (uaref > (unsigned)utarg * 10) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_warning("reducing V[aref][1] from %.2f to %.1f\n", uaref/100.0, v);
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
uaref = 10 * (unsigned)utarg;
if (stk500v2_setparm2(pgm, PARAM2_AREF1, uaref)
!= 0)
return -1;
}
/*
* Vtarget on the STK600 can only be adjusted while not being in
* programming mode.
*/
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
if (PDATA(pgm)->lastpart)
pgm->disable(pgm);
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
rv = stk500v2_setparm(pgm, PARAM_VTARGET, utarg);
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
if (PDATA(pgm)->lastpart)
pgm->program_enable(pgm, PDATA(pgm)->lastpart);
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
return rv;
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk600_set_varef(const PROGRAMMER *pgm, unsigned int chan, double v) {
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
unsigned char utarg;
unsigned int uaref;
uaref = (unsigned)((v + 0.0049) * 100);
if (stk500v2_getparm(pgm, PARAM_VTARGET, &utarg) != 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("cannot obtain V[target]\n");
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
return -1;
}
if (uaref > (unsigned)utarg * 10) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("V[aref] must not be greater than V[target] = %.1f\n", utarg/10.0);
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
return -1;
}
switch (chan)
{
case 0:
return stk500v2_setparm2(pgm, PARAM2_AREF0, uaref);
case 1:
return stk500v2_setparm2(pgm, PARAM2_AREF1, uaref);
default:
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("invalid channel %d\n", chan);
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
return -1;
}
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk600_set_fosc(const PROGRAMMER *pgm, double v) {
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
unsigned int oct, dac;
oct = 1.443 * log(v / 1039.0);
dac = 2048 - (2078.0 * pow(2, (double)(10 + oct))) / v;
return stk500v2_setparm2(pgm, PARAM2_CLOCK_CONF, (oct << 12) | (dac << 2));
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk600_set_sck_period(const PROGRAMMER *pgm, double v) {
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
unsigned int sck;
sck = ceil((16e6 / (2 * 1.0 / v)) - 1);
if (sck >= 4096)
sck = 4095;
return stk500v2_setparm2(pgm, PARAM2_SCK_DURATION, sck);
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_jtag3_set_sck_period(const PROGRAMMER *pgm, double v) {
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
unsigned char value[3];
unsigned int sck;
if (v < 1E-6)
sck = 0x400;
else if (v > 1E-3)
sck = 1;
else
sck = 1.0 / (1000.0 * v);
value[0] = CMD_SET_SCK;
value[1] = sck & 0xff;
value[2] = (sck >> 8) & 0xff;
if (stk500v2_jtag3_send(pgm, value, 3) < 0)
return -1;
if (stk500v2_jtag3_recv(pgm, value, 3) < 0)
return -1;
return 0;
}
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_getparm(const PROGRAMMER *pgm, unsigned char parm, unsigned char *value) {
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
unsigned char buf[32];
buf[0] = CMD_GET_PARAMETER;
buf[1] = parm;
if (stk500v2_command(pgm, buf, 2, sizeof(buf)) < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unable to get parameter 0x%02x\n", parm);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -1;
}
*value = buf[2];
return 0;
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_setparm_real(const PROGRAMMER *pgm, unsigned char parm, unsigned char value) {
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
unsigned char buf[32];
buf[0] = CMD_SET_PARAMETER;
buf[1] = parm;
buf[2] = value;
if (stk500v2_command(pgm, buf, 3, sizeof(buf)) < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("\n%s: stk500v2_setparm(): unable to set parameter 0x%02x\n",
progname, parm);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return -1;
}
return 0;
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_setparm(const PROGRAMMER *pgm, unsigned char parm, unsigned char value) {
Address compiler warnings in 4 source files
2022-08-12 13:58:21 +00:00
unsigned char current_value = value;
* main.c, stk500v2.c: Added patch 4804 from eolson@mit.edu Which stops sck from being writtend needlessly git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@568 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-17 21:11:39 +00:00
int res;
res = stk500v2_getparm(pgm, parm, &current_value);
Address compiler warnings in 4 source files
2022-08-12 13:58:21 +00:00
if (res < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unable to get parameter 0x%02x\n", parm);
Address compiler warnings in 4 source files
2022-08-12 13:58:21 +00:00
return -1;
}
* main.c, stk500v2.c: Added patch 4804 from eolson@mit.edu Which stops sck from being writtend needlessly git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@568 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-17 21:11:39 +00:00
// don't issue a write if the correct value is already set.
added verbose level in avrdude_message() git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1321 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-06-13 20:07:40 +00:00
if (value == current_value) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_notice2("skipping parameter write; parameter value already set\n");
* main.c, stk500v2.c: Added patch 4804 from eolson@mit.edu Which stops sck from being writtend needlessly git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@568 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-01-17 21:11:39 +00:00
return 0;
}
return stk500v2_setparm_real(pgm, parm, value);
}
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_getparm2(const PROGRAMMER *pgm, unsigned char parm, unsigned int *value) {
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
unsigned char buf[32];
buf[0] = CMD_GET_PARAMETER;
buf[1] = parm;
if (stk500v2_command(pgm, buf, 2, sizeof(buf)) < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unable to get parameter 0x%02x\n", parm);
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
return -1;
}
*value = ((unsigned)buf[2] << 8) | buf[3];
return 0;
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_setparm2(const PROGRAMMER *pgm, unsigned char parm, unsigned int value) {
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
unsigned char buf[32];
buf[0] = CMD_SET_PARAMETER;
buf[1] = parm;
buf[2] = value >> 8;
buf[3] = value;
if (stk500v2_command(pgm, buf, 4, sizeof(buf)) < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("\n%s: stk500v2_setparm2(): unable to set parameter 0x%02x\n",
progname, parm);
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
return -1;
}
return 0;
}
static const char *stk600_get_cardname(const struct carddata *table,
size_t nele, int id)
{
const struct carddata *cdp;
if (id == 0xFF)
/* 0xFF means this card is not present at all. */
return "Not present";
for (cdp = table; nele > 0; cdp++, nele--)
if (cdp->id == id)
return cdp->name;
return "Unknown";
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static void stk500v2_display(const PROGRAMMER *pgm, const char *p) {
Fix some possibly uninitialized variables in stk500v2_display() Submitted by Adrian Klieber patch #9820: Fix some out-of-bounds/uninitialized issues git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1441 81a1dc3b-b13d-400b-aceb-764788c761c2
2020-09-16 21:19:50 +00:00
unsigned char maj = 0, min = 0, hdw = 0, topcard = 0,
maj_s1 = 0, min_s1 = 0, maj_s2 = 0, min_s2 = 0;
unsigned int rev = 0;
Some more STK500 vs. AVRISP cleanup: don't announce STK500-only parameters on AVRISPs. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@615 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-08-19 21:02:45 +00:00
const char *topcard_name, *pgmname;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
switch (PDATA(pgm)->pgmtype) {
Some more STK500 vs. AVRISP cleanup: don't announce STK500-only parameters on AVRISPs. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@615 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-08-19 21:02:45 +00:00
case PGMTYPE_UNKNOWN: pgmname = "Unknown"; break;
case PGMTYPE_STK500: pgmname = "STK500"; break;
case PGMTYPE_AVRISP: pgmname = "AVRISP"; break;
case PGMTYPE_AVRISP_MKII: pgmname = "AVRISP mkII"; break;
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
case PGMTYPE_STK600: pgmname = "STK600"; break;
Some more STK500 vs. AVRISP cleanup: don't announce STK500-only parameters on AVRISPs. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@615 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-08-19 21:02:45 +00:00
default: pgmname = "None";
}
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
if (PDATA(pgm)->pgmtype != PGMTYPE_JTAGICE_MKII &&
PDATA(pgm)->pgmtype != PGMTYPE_JTAGICE3) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
msg_info("%sProgrammer Model: %s\n", p, pgmname);
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
stk500v2_getparm(pgm, PARAM_HW_VER, &hdw);
stk500v2_getparm(pgm, PARAM_SW_MAJOR, &maj);
stk500v2_getparm(pgm, PARAM_SW_MINOR, &min);
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
msg_info("%sHardware Version: %d\n", p, hdw);
Read programmer serial number from libusb or hidusb (#1223) * Silence compiler warnings * Fix a fall-through error in switch * Slightly reformat programmer info printout Now the target voltage is printed out along with the HW version, FW version and serial number, and they all allign nicely. When in ISP mode, the "SCK period" value is printed _after_ the target voltage readout, and _before_ the additional clocks the programmer holds, such as JTAG, PDI and UPDI clocks. Co-authored-by: Stefan Rueger <stefan.rueger@urclocks.com>
2022-12-21 19:11:31 +00:00
if (pgm->usbsn && *pgm->usbsn)
msg_info("%sSerial number : %s\n", p, pgm->usbsn);
Replace gratuitous references to slave and master with technical terms
2022-11-22 17:11:33 +00:00
msg_info("%sFirmware Version Controller : %d.%02d\n", p, maj, min);
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
if (PDATA(pgm)->pgmtype == PGMTYPE_STK600) {
Replace gratuitous references to slave and master with technical terms
2022-11-22 17:11:33 +00:00
stk500v2_getparm(pgm, PARAM_SW_MAJOR_PERIPHERY1, &maj_s1);
stk500v2_getparm(pgm, PARAM_SW_MINOR_PERIPHERY1, &min_s1);
stk500v2_getparm(pgm, PARAM_SW_MAJOR_PERIPHERY2, &maj_s2);
stk500v2_getparm(pgm, PARAM_SW_MINOR_PERIPHERY2, &min_s2);
msg_info("%sFirmware Version Periphery 1: %d.%02d\n", p, maj_s1, min_s1);
msg_info("%sFirmware Version Periphery 2: %d.%02d\n", p, maj_s2, min_s2);
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
}
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
}
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
if (PDATA(pgm)->pgmtype == PGMTYPE_STK500) {
Minor cosmetic changes: STK500 firmware version numbers are M.NN, so always display the minor number as two digits. Examine the response to the sign-on command to see which programmer hardware we are talking to, and then restrict the STK500 topcard display to devices detected as STK500. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@614 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-08-18 21:23:34 +00:00
stk500v2_getparm(pgm, PARAM_TOPCARD_DETECT, &topcard);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
switch (topcard) {
case 0xAA: topcard_name = "STK501"; break;
case 0x55: topcard_name = "STK502"; break;
case 0xFA: topcard_name = "STK503"; break;
case 0xEE: topcard_name = "STK504"; break;
case 0xE4: topcard_name = "STK505"; break;
case 0xDD: topcard_name = "STK520"; break;
default: topcard_name = "Unknown"; break;
}
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
msg_info("%sTopcard : %s\n", p, topcard_name);
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
} else if (PDATA(pgm)->pgmtype == PGMTYPE_STK600) {
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
stk500v2_getparm(pgm, PARAM_ROUTINGCARD_ID, &topcard);
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
msg_info("%sRouting card : %s\n", p,
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
stk600_get_cardname(routing_cards,
sizeof routing_cards / sizeof routing_cards[0],
topcard));
stk500v2_getparm(pgm, PARAM_SOCKETCARD_ID, &topcard);
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
msg_info("%sSocket card : %s\n", p,
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
stk600_get_cardname(socket_cards,
sizeof socket_cards / sizeof socket_cards[0],
topcard));
stk500v2_getparm2(pgm, PARAM2_RC_ID_TABLE_REV, &rev);
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
msg_info("%sRC_ID table rev : %d\n", p, rev);
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
stk500v2_getparm2(pgm, PARAM2_EC_ID_TABLE_REV, &rev);
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
msg_info("%sEC_ID table rev : %d\n", p, rev);
Read programmer serial number from libusb or hidusb (#1223) * Silence compiler warnings * Fix a fall-through error in switch * Slightly reformat programmer info printout Now the target voltage is printed out along with the HW version, FW version and serial number, and they all allign nicely. When in ISP mode, the "SCK period" value is printed _after_ the target voltage readout, and _before_ the additional clocks the programmer holds, such as JTAG, PDI and UPDI clocks. Co-authored-by: Stefan Rueger <stefan.rueger@urclocks.com>
2022-12-21 19:11:31 +00:00
} else if (PDATA(pgm)->pgmtype == PGMTYPE_JTAGICE3) {
PROGRAMMER *pgmcp = pgm_dup(pgm);
pgmcp->cookie = PDATA(pgm)->chained_pdata;
jtag3_display(pgmcp, p);
msg_info("\n");
pgm_free(pgmcp);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
}
Revamp terminal output: progress bar, callback and stdout/stderr (#1132) * Print parms output to stdout * Flush terminal writes and other minor changes * Prepare terminal for periodic calls to programmer to reset bootloader WDT * Only show progress reports for memories > 32 bytes or on -vv * Freeze progress bar on serious error * Allow cached r/w byte routines to be used in pgm->read_byte and pgm->write_byte
2022-10-23 20:56:45 +00:00
stk500v2_print_parms1(pgm, p, stderr);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return;
}
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
static double
f_to_kHz_MHz(double f, const char **unit)
{
if (f > 1e6) {
f /= 1e6;
*unit = "MHz";
} else if (f > 1e3) {
f /= 1000;
*unit = "kHz";
} else
*unit = "Hz";
return f;
}
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
Revamp terminal output: progress bar, callback and stdout/stderr (#1132) * Print parms output to stdout * Flush terminal writes and other minor changes * Prepare terminal for periodic calls to programmer to reset bootloader WDT * Only show progress reports for memories > 32 bytes or on -vv * Freeze progress bar on serious error * Allow cached r/w byte routines to be used in pgm->read_byte and pgm->write_byte
2022-10-23 20:56:45 +00:00
static void stk500v2_print_parms1(const PROGRAMMER *pgm, const char *p, FILE *fp) {
Address compiler warnings in 4 source files
2022-08-12 13:58:21 +00:00
unsigned char vtarget = 0, vadjust = 0, osc_pscale = 0, osc_cmatch = 0, sck_duration =0; //XXX 0 is not correct, check caller
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
unsigned int sck_stk600, clock_conf, dac, oct, varef;
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
unsigned char vtarget_jtag[4];
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
int prescale;
double f;
const char *unit;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
Address compiler warnings in 4 source files
2022-08-12 13:58:21 +00:00
memset(vtarget_jtag, 0, sizeof vtarget_jtag);
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
if (PDATA(pgm)->pgmtype == PGMTYPE_JTAGICE_MKII) {
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
PROGRAMMER *pgmcp = pgm_dup(pgm);
pgmcp->cookie = PDATA(pgm)->chained_pdata;
jtagmkII_getparm(pgmcp, PAR_OCD_VTARGET, vtarget_jtag);
pgm_free(pgmcp);
Revamp terminal output: progress bar, callback and stdout/stderr (#1132) * Print parms output to stdout * Flush terminal writes and other minor changes * Prepare terminal for periodic calls to programmer to reset bootloader WDT * Only show progress reports for memories > 32 bytes or on -vv * Freeze progress bar on serious error * Allow cached r/w byte routines to be used in pgm->read_byte and pgm->write_byte
2022-10-23 20:56:45 +00:00
fmsg_out(fp, "%sVtarget : %.1f V\n", p, b2_to_u16(vtarget_jtag) / 1000.0);
Read programmer serial number from libusb or hidusb (#1223) * Silence compiler warnings * Fix a fall-through error in switch * Slightly reformat programmer info printout Now the target voltage is printed out along with the HW version, FW version and serial number, and they all allign nicely. When in ISP mode, the "SCK period" value is printed _after_ the target voltage readout, and _before_ the additional clocks the programmer holds, such as JTAG, PDI and UPDI clocks. Co-authored-by: Stefan Rueger <stefan.rueger@urclocks.com>
2022-12-21 19:11:31 +00:00
} else if (PDATA(pgm)->pgmtype != PGMTYPE_JTAGICE3) {
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
stk500v2_getparm(pgm, PARAM_VTARGET, &vtarget);
Revamp terminal output: progress bar, callback and stdout/stderr (#1132) * Print parms output to stdout * Flush terminal writes and other minor changes * Prepare terminal for periodic calls to programmer to reset bootloader WDT * Only show progress reports for memories > 32 bytes or on -vv * Freeze progress bar on serious error * Allow cached r/w byte routines to be used in pgm->read_byte and pgm->write_byte
2022-10-23 20:56:45 +00:00
fmsg_out(fp, "%sVtarget : %.1f V\n", p, vtarget / 10.0);
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
}
Some more STK500 vs. AVRISP cleanup: don't announce STK500-only parameters on AVRISPs. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@615 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-08-19 21:02:45 +00:00
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
switch (PDATA(pgm)->pgmtype) {
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
case PGMTYPE_STK500:
stk500v2_getparm(pgm, PARAM_SCK_DURATION, &sck_duration);
Some more STK500 vs. AVRISP cleanup: don't announce STK500-only parameters on AVRISPs. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@615 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-08-19 21:02:45 +00:00
stk500v2_getparm(pgm, PARAM_VADJUST, &vadjust);
stk500v2_getparm(pgm, PARAM_OSC_PSCALE, &osc_pscale);
stk500v2_getparm(pgm, PARAM_OSC_CMATCH, &osc_cmatch);
Revamp terminal output: progress bar, callback and stdout/stderr (#1132) * Print parms output to stdout * Flush terminal writes and other minor changes * Prepare terminal for periodic calls to programmer to reset bootloader WDT * Only show progress reports for memories > 32 bytes or on -vv * Freeze progress bar on serious error * Allow cached r/w byte routines to be used in pgm->read_byte and pgm->write_byte
2022-10-23 20:56:45 +00:00
fmsg_out(fp, "%sSCK period : %.1f us\n", p,
* stk500v2.c (stk500v2_set_sck_period): Revamp this to match the description/pseudo-code in appnote AVR068. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1229 81a1dc3b-b13d-400b-aceb-764788c761c2
2013-09-13 21:32:00 +00:00
stk500v2_sck_to_us(pgm, sck_duration));
Revamp terminal output: progress bar, callback and stdout/stderr (#1132) * Print parms output to stdout * Flush terminal writes and other minor changes * Prepare terminal for periodic calls to programmer to reset bootloader WDT * Only show progress reports for memories > 32 bytes or on -vv * Freeze progress bar on serious error * Allow cached r/w byte routines to be used in pgm->read_byte and pgm->write_byte
2022-10-23 20:56:45 +00:00
fmsg_out(fp, "%sVaref : %.1f V\n", p, vadjust / 10.0);
fmsg_out(fp, "%sOscillator : ", p);
Some more STK500 vs. AVRISP cleanup: don't announce STK500-only parameters on AVRISPs. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@615 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-08-19 21:02:45 +00:00
if (osc_pscale == 0)
Revamp terminal output: progress bar, callback and stdout/stderr (#1132) * Print parms output to stdout * Flush terminal writes and other minor changes * Prepare terminal for periodic calls to programmer to reset bootloader WDT * Only show progress reports for memories > 32 bytes or on -vv * Freeze progress bar on serious error * Allow cached r/w byte routines to be used in pgm->read_byte and pgm->write_byte
2022-10-23 20:56:45 +00:00
fmsg_out(fp, "Off\n");
Some more STK500 vs. AVRISP cleanup: don't announce STK500-only parameters on AVRISPs. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@615 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-08-19 21:02:45 +00:00
else {
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
prescale = 1;
f = STK500V2_XTAL / 2;
Some more STK500 vs. AVRISP cleanup: don't announce STK500-only parameters on AVRISPs. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@615 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-08-19 21:02:45 +00:00
switch (osc_pscale) {
case 2: prescale = 8; break;
case 3: prescale = 32; break;
case 4: prescale = 64; break;
case 5: prescale = 128; break;
case 6: prescale = 256; break;
case 7: prescale = 1024; break;
}
f /= prescale;
f /= (osc_cmatch + 1);
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
f = f_to_kHz_MHz(f, &unit);
Revamp terminal output: progress bar, callback and stdout/stderr (#1132) * Print parms output to stdout * Flush terminal writes and other minor changes * Prepare terminal for periodic calls to programmer to reset bootloader WDT * Only show progress reports for memories > 32 bytes or on -vv * Freeze progress bar on serious error * Allow cached r/w byte routines to be used in pgm->read_byte and pgm->write_byte
2022-10-23 20:56:45 +00:00
fmsg_out(fp, "%.3f %s\n", f, unit);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
}
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
break;
case PGMTYPE_AVRISP_MKII:
case PGMTYPE_JTAGICE_MKII:
stk500v2_getparm(pgm, PARAM_SCK_DURATION, &sck_duration);
Revamp terminal output: progress bar, callback and stdout/stderr (#1132) * Print parms output to stdout * Flush terminal writes and other minor changes * Prepare terminal for periodic calls to programmer to reset bootloader WDT * Only show progress reports for memories > 32 bytes or on -vv * Freeze progress bar on serious error * Allow cached r/w byte routines to be used in pgm->read_byte and pgm->write_byte
2022-10-23 20:56:45 +00:00
fmsg_out(fp, "%sSCK period : %.2f us\n", p,
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
(float) 1000000 / avrispmkIIfreqs[sck_duration]);
break;
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
case PGMTYPE_JTAGICE3:
{
unsigned char cmd[4];
cmd[0] = CMD_GET_SCK;
Read programmer serial number from libusb or hidusb (#1223) * Silence compiler warnings * Fix a fall-through error in switch * Slightly reformat programmer info printout Now the target voltage is printed out along with the HW version, FW version and serial number, and they all allign nicely. When in ISP mode, the "SCK period" value is printed _after_ the target voltage readout, and _before_ the additional clocks the programmer holds, such as JTAG, PDI and UPDI clocks. Co-authored-by: Stefan Rueger <stefan.rueger@urclocks.com>
2022-12-21 19:11:31 +00:00
if (stk500v2_jtag3_send(pgm, cmd, 1) >= 0 && stk500v2_jtag3_recv(pgm, cmd, 4) >= 2) {
unsigned int sck = cmd[1] | (cmd[2] << 8);
fmsg_out(fp, "%sSCK period : %.2f us\n", p, (1E6 / (1000.0 * sck)));
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
}
Read programmer serial number from libusb or hidusb (#1223) * Silence compiler warnings * Fix a fall-through error in switch * Slightly reformat programmer info printout Now the target voltage is printed out along with the HW version, FW version and serial number, and they all allign nicely. When in ISP mode, the "SCK period" value is printed _after_ the target voltage readout, and _before_ the additional clocks the programmer holds, such as JTAG, PDI and UPDI clocks. Co-authored-by: Stefan Rueger <stefan.rueger@urclocks.com>
2022-12-21 19:11:31 +00:00
PROGRAMMER *pgmcp = pgm_dup(pgm);
pgmcp->cookie = PDATA(pgm)->chained_pdata;
pgmcp->id = lcreat(NULL, 0);
// Copy pgm->id contents over to pgmcp->id
for(LNODEID ln=lfirst(pgm->id); ln; ln=lnext(ln))
ladd(pgmcp->id, cfg_strdup("stk500v2_print_parms1()", ldata(ln)));
jtag3_print_parms1(pgmcp, p, fp);
pgm_free(pgmcp);
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
}
break;
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
case PGMTYPE_STK600:
stk500v2_getparm2(pgm, PARAM2_AREF0, &varef);
Revamp terminal output: progress bar, callback and stdout/stderr (#1132) * Print parms output to stdout * Flush terminal writes and other minor changes * Prepare terminal for periodic calls to programmer to reset bootloader WDT * Only show progress reports for memories > 32 bytes or on -vv * Freeze progress bar on serious error * Allow cached r/w byte routines to be used in pgm->read_byte and pgm->write_byte
2022-10-23 20:56:45 +00:00
fmsg_out(fp, "%sVaref 0 : %.2f V\n", p, varef / 100.0);
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
stk500v2_getparm2(pgm, PARAM2_AREF1, &varef);
Revamp terminal output: progress bar, callback and stdout/stderr (#1132) * Print parms output to stdout * Flush terminal writes and other minor changes * Prepare terminal for periodic calls to programmer to reset bootloader WDT * Only show progress reports for memories > 32 bytes or on -vv * Freeze progress bar on serious error * Allow cached r/w byte routines to be used in pgm->read_byte and pgm->write_byte
2022-10-23 20:56:45 +00:00
fmsg_out(fp, "%sVaref 1 : %.2f V\n", p, varef / 100.0);
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
stk500v2_getparm2(pgm, PARAM2_SCK_DURATION, &sck_stk600);
Revamp terminal output: progress bar, callback and stdout/stderr (#1132) * Print parms output to stdout * Flush terminal writes and other minor changes * Prepare terminal for periodic calls to programmer to reset bootloader WDT * Only show progress reports for memories > 32 bytes or on -vv * Freeze progress bar on serious error * Allow cached r/w byte routines to be used in pgm->read_byte and pgm->write_byte
2022-10-23 20:56:45 +00:00
fmsg_out(fp, "%sSCK period : %.2f us\n", p,
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
(float) (sck_stk600 + 1) / 8.0);
stk500v2_getparm2(pgm, PARAM2_CLOCK_CONF, &clock_conf);
oct = (clock_conf & 0xf000) >> 12u;
dac = (clock_conf & 0x0ffc) >> 2u;
f = pow(2, (double)oct) * 2078.0 / (2 - (double)dac / 1024.0);
f = f_to_kHz_MHz(f, &unit);
Revamp terminal output: progress bar, callback and stdout/stderr (#1132) * Print parms output to stdout * Flush terminal writes and other minor changes * Prepare terminal for periodic calls to programmer to reset bootloader WDT * Only show progress reports for memories > 32 bytes or on -vv * Freeze progress bar on serious error * Allow cached r/w byte routines to be used in pgm->read_byte and pgm->write_byte
2022-10-23 20:56:45 +00:00
fmsg_out(fp, "%sOscillator : %.3f %s\n",
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
p, f, unit);
break;
default:
Revamp terminal output: progress bar, callback and stdout/stderr (#1132) * Print parms output to stdout * Flush terminal writes and other minor changes * Prepare terminal for periodic calls to programmer to reset bootloader WDT * Only show progress reports for memories > 32 bytes or on -vv * Freeze progress bar on serious error * Allow cached r/w byte routines to be used in pgm->read_byte and pgm->write_byte
2022-10-23 20:56:45 +00:00
fmsg_out(fp, "%sSCK period : %.1f us\n", p,
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
sck_duration * 8.0e6 / STK500V2_XTAL + 0.05);
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
break;
}
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
return;
}
Revamp terminal output: progress bar, callback and stdout/stderr (#1132) * Print parms output to stdout * Flush terminal writes and other minor changes * Prepare terminal for periodic calls to programmer to reset bootloader WDT * Only show progress reports for memories > 32 bytes or on -vv * Freeze progress bar on serious error * Allow cached r/w byte routines to be used in pgm->read_byte and pgm->write_byte
2022-10-23 20:56:45 +00:00
static void stk500v2_print_parms(const PROGRAMMER *pgm, FILE *fp) {
stk500v2_print_parms1(pgm, "", fp);
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_perform_osccal(const PROGRAMMER *pgm) {
Submitted by John Voltz: add AVR053 oscillator calibration. * main.c: Add the -O option. * pgm.c: Add the hook for the perform_osccal() method. * pgm.h: (Ditto.) * stk500v2.c: Implement perform_osccal(). * avrdude.1: Document the -O option. * doc/avrdude.texi: (Ditto.) Partially closes bug #17487: AVR RC oscillator calibration routine not supported (feature request) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@671 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-09 14:34:24 +00:00
unsigned char buf[32];
int rv;
buf[0] = CMD_OSCCAL;
rv = stk500v2_command(pgm, buf, 1, sizeof(buf));
if (rv < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unable to perform oscillator calibaration\n");
Submitted by John Voltz: add AVR053 oscillator calibration. * main.c: Add the -O option. * pgm.c: Add the hook for the perform_osccal() method. * pgm.h: (Ditto.) * stk500v2.c: Implement perform_osccal(). * avrdude.1: Document the -O option. * doc/avrdude.texi: (Ditto.) Partially closes bug #17487: AVR RC oscillator calibration routine not supported (feature request) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@671 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-09 14:34:24 +00:00
return -1;
}
return 0;
}
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
/*
* Wrapper functions for the JTAG ICE mkII in ISP mode. This mode
* uses the normal JTAG ICE mkII packet stream to communicate with the
* ICE, but then encapsulates AVRISP mkII commands using
* CMND_ISP_PACKET.
*/
/*
* Open a JTAG ICE mkII in ISP mode.
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_jtagmkII_open(PROGRAMMER *pgm, const char *port) {
Change the second parameter of the ser_open method from "baud" into a "union pinfo", so the USB parameters can be passed without hacks. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1276 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-02-21 13:44:11 +00:00
union pinfo pinfo;
Fix a serious memory corruption problem resulting out of the chaining of both, the stk500v2 and the jtagmkII programmers for some programming hardware (JTAG ICE mkII and AVR Dragon running in ISP, HVSP or PP mode), where both programmers have to maintain their private programmer data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@834 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-07-06 22:10:20 +00:00
void *mycookie;
bug #33114: Segfault after setting the DWEN fuse with Dragon * jtagII.c (jtagmkII_getsync): Instead of exit()ing from deep within the tree when detecting the "need debugWIRE" situation, properly pass this up as a return code. * jtagII_private.h (JTAGII_GETSYNC_FAIL_GRACEFUL): New constant. * stk500v2.c (stk500v2_jtagmkII_open): Don't tell anything anymore when receiving a JTAGII_GETSYNC_FAIL_GRACEFUL from jtagmkII_getsync(); silently give up (all necessary has been said already). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@983 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-08-26 10:05:09 +00:00
int rv;
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_notice2("stk500v2_jtagmkII_open()\n");
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
/*
* The JTAG ICE mkII always starts with a baud rate of 19200 Bd upon
* attaching. If the config file or command-line parameters specify
* a higher baud rate, we switch to it later on, after establishing
* the connection with the ICE.
*/
Implemented basic serial code refactoring for upcoming SerialUPDI implementation git-svn-id: svn://svn.savannah.nongnu.org/avrdude/branches/serial_refactoring@1511 81a1dc3b-b13d-400b-aceb-764788c761c2
2021-12-08 10:09:52 +00:00
pinfo.serialinfo.baud = 19200;
pinfo.serialinfo.cflags = SERIAL_8N1;
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
/*
* If the port name starts with "usb", divert the serial routines
* to the USB ones. The serial_open() function for USB overrides
* the meaning of the "baud" parameter to be the USB device ID to
* search for.
*/
if (strncmp(port, "usb", 3) == 0) {
#if defined(HAVE_LIBUSB)
serdev = &usb_serdev;
Change the second parameter of the ser_open method from "baud" into a "union pinfo", so the USB parameters can be passed without hacks. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1276 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-02-21 13:44:11 +00:00
pinfo.usbinfo.vid = USB_VENDOR_ATMEL;
pinfo.usbinfo.flags = 0;
pinfo.usbinfo.pid = USB_DEVICE_JTAGICEMKII;
Change all the USB details (endpoint numbers, max transfer size etc.) to a per-programmer adjustable value. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1108 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-11-26 16:24:56 +00:00
pgm->fd.usb.max_xfer = USBDEV_MAX_XFER_MKII;
pgm->fd.usb.rep = USBDEV_BULK_EP_READ_MKII;
pgm->fd.usb.wep = USBDEV_BULK_EP_WRITE_MKII;
pgm->fd.usb.eep = 0; /* no seperate EP for events */
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
#else
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("avrdude was compiled without usb support\n");
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
return -1;
#endif
}
strcpy(pgm->port, port);
Change the second parameter of the ser_open method from "baud" into a "union pinfo", so the USB parameters can be passed without hacks. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1276 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-02-21 13:44:11 +00:00
if (serial_open(port, pinfo, &pgm->fd)==-1) {
* serial_open() calls will now return -1 on error (no call to exit()) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@948 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-10-22 14:29:56 +00:00
return -1;
}
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
/*
* drain any extraneous input
*/
stk500v2_drain(pgm, 0);
Fix a serious memory corruption problem resulting out of the chaining of both, the stk500v2 and the jtagmkII programmers for some programming hardware (JTAG ICE mkII and AVR Dragon running in ISP, HVSP or PP mode), where both programmers have to maintain their private programmer data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@834 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-07-06 22:10:20 +00:00
mycookie = pgm->cookie;
pgm->cookie = PDATA(pgm)->chained_pdata;
bug #33114: Segfault after setting the DWEN fuse with Dragon * jtagII.c (jtagmkII_getsync): Instead of exit()ing from deep within the tree when detecting the "need debugWIRE" situation, properly pass this up as a return code. * jtagII_private.h (JTAGII_GETSYNC_FAIL_GRACEFUL): New constant. * stk500v2.c (stk500v2_jtagmkII_open): Don't tell anything anymore when receiving a JTAGII_GETSYNC_FAIL_GRACEFUL from jtagmkII_getsync(); silently give up (all necessary has been said already). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@983 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-08-26 10:05:09 +00:00
if ((rv = jtagmkII_getsync(pgm, EMULATOR_MODE_SPI)) != 0) {
if (rv != JTAGII_GETSYNC_FAIL_GRACEFUL)
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unable to sync with the JTAG ICE mkII in ISP mode\n");
Fix a serious memory corruption problem resulting out of the chaining of both, the stk500v2 and the jtagmkII programmers for some programming hardware (JTAG ICE mkII and AVR Dragon running in ISP, HVSP or PP mode), where both programmers have to maintain their private programmer data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@834 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-07-06 22:10:20 +00:00
pgm->cookie = mycookie;
Cleanup the open/close handling to avoid accessing unallocated memory (in the atexit handler) in case of bailing out. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@910 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-13 12:44:54 +00:00
return -1;
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
}
Fix a serious memory corruption problem resulting out of the chaining of both, the stk500v2 and the jtagmkII programmers for some programming hardware (JTAG ICE mkII and AVR Dragon running in ISP, HVSP or PP mode), where both programmers have to maintain their private programmer data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@834 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-07-06 22:10:20 +00:00
pgm->cookie = mycookie;
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
PDATA(pgm)->pgmtype = PGMTYPE_JTAGICE_MKII;
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
if (pgm->bitclock != 0.0) {
if (pgm->set_sck_period(pgm, pgm->bitclock) != 0)
return -1;
}
return 0;
}
bug #28516: Linux/Dragon: Error message on exit * stk500v2.c: Fix the "bad response to GO command: RSP_ILLEGAL_EMULATOR_MODE" message. jtagmkII_close() has been called with the wrong pgm->cookie. Wrap it inside stk500v2_jtagmkII_close(), adjusting the cookie data appropriately. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@895 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-08 16:01:37 +00:00
/*
* Close an AVR Dragon or JTAG ICE mkII in ISP/HVSP/PP mode.
*/
static void stk500v2_jtagmkII_close(PROGRAMMER * pgm)
{
void *mycookie;
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_notice2("stk500v2_jtagmkII_close()\n");
bug #28516: Linux/Dragon: Error message on exit * stk500v2.c: Fix the "bad response to GO command: RSP_ILLEGAL_EMULATOR_MODE" message. jtagmkII_close() has been called with the wrong pgm->cookie. Wrap it inside stk500v2_jtagmkII_close(), adjusting the cookie data appropriately. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@895 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-08 16:01:37 +00:00
mycookie = pgm->cookie;
pgm->cookie = PDATA(pgm)->chained_pdata;
jtagmkII_close(pgm);
pgm->cookie = mycookie;
}
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
/*
* Close JTAGICE3.
*/
static void stk500v2_jtag3_close(PROGRAMMER * pgm)
{
void *mycookie;
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_notice2("stk500v2_jtag3_close()\n");
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
mycookie = pgm->cookie;
pgm->cookie = PDATA(pgm)->chained_pdata;
jtag3_close(pgm);
pgm->cookie = mycookie;
}
Add support for the AVR Dragon (JTAG and ISP mode). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@675 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-26 21:14:10 +00:00
/*
* Wrapper functions for the AVR Dragon in ISP mode. This mode
* uses the normal JTAG ICE mkII packet stream to communicate with the
* ICE, but then encapsulates AVRISP mkII commands using
* CMND_ISP_PACKET.
*/
/*
* Open an AVR Dragon in ISP mode.
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_dragon_isp_open(PROGRAMMER *pgm, const char *port) {
Change the second parameter of the ser_open method from "baud" into a "union pinfo", so the USB parameters can be passed without hacks. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1276 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-02-21 13:44:11 +00:00
union pinfo pinfo;
Fix a serious memory corruption problem resulting out of the chaining of both, the stk500v2 and the jtagmkII programmers for some programming hardware (JTAG ICE mkII and AVR Dragon running in ISP, HVSP or PP mode), where both programmers have to maintain their private programmer data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@834 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-07-06 22:10:20 +00:00
void *mycookie;
Add support for the AVR Dragon (JTAG and ISP mode). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@675 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-26 21:14:10 +00:00
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_notice2("stk500v2_dragon_isp_open()\n");
Add support for the AVR Dragon (JTAG and ISP mode). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@675 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-26 21:14:10 +00:00
/*
* The JTAG ICE mkII always starts with a baud rate of 19200 Bd upon
* attaching. If the config file or command-line parameters specify
* a higher baud rate, we switch to it later on, after establishing
* the connection with the ICE.
*/
Implemented basic serial code refactoring for upcoming SerialUPDI implementation git-svn-id: svn://svn.savannah.nongnu.org/avrdude/branches/serial_refactoring@1511 81a1dc3b-b13d-400b-aceb-764788c761c2
2021-12-08 10:09:52 +00:00
pinfo.serialinfo.baud = 19200;
pinfo.serialinfo.cflags = SERIAL_8N1;
Add support for the AVR Dragon (JTAG and ISP mode). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@675 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-26 21:14:10 +00:00
/*
* If the port name starts with "usb", divert the serial routines
* to the USB ones. The serial_open() function for USB overrides
* the meaning of the "baud" parameter to be the USB device ID to
* search for.
*/
if (strncmp(port, "usb", 3) == 0) {
#if defined(HAVE_LIBUSB)
serdev = &usb_serdev;
Change the second parameter of the ser_open method from "baud" into a "union pinfo", so the USB parameters can be passed without hacks. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1276 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-02-21 13:44:11 +00:00
pinfo.usbinfo.vid = USB_VENDOR_ATMEL;
pinfo.usbinfo.flags = 0;
pinfo.usbinfo.pid = USB_DEVICE_AVRDRAGON;
Change all the USB details (endpoint numbers, max transfer size etc.) to a per-programmer adjustable value. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1108 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-11-26 16:24:56 +00:00
pgm->fd.usb.max_xfer = USBDEV_MAX_XFER_MKII;
pgm->fd.usb.rep = USBDEV_BULK_EP_READ_MKII;
pgm->fd.usb.wep = USBDEV_BULK_EP_WRITE_MKII;
pgm->fd.usb.eep = 0; /* no seperate EP for events */
Add support for the AVR Dragon (JTAG and ISP mode). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@675 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-26 21:14:10 +00:00
#else
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("avrdude was compiled without usb support\n");
Add support for the AVR Dragon (JTAG and ISP mode). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@675 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-26 21:14:10 +00:00
return -1;
#endif
}
strcpy(pgm->port, port);
Change the second parameter of the ser_open method from "baud" into a "union pinfo", so the USB parameters can be passed without hacks. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1276 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-02-21 13:44:11 +00:00
if (serial_open(port, pinfo, &pgm->fd)==-1) {
* serial_open() calls will now return -1 on error (no call to exit()) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@948 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-10-22 14:29:56 +00:00
return -1;
}
Add support for the AVR Dragon (JTAG and ISP mode). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@675 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-26 21:14:10 +00:00
/*
* drain any extraneous input
*/
stk500v2_drain(pgm, 0);
Fix a serious memory corruption problem resulting out of the chaining of both, the stk500v2 and the jtagmkII programmers for some programming hardware (JTAG ICE mkII and AVR Dragon running in ISP, HVSP or PP mode), where both programmers have to maintain their private programmer data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@834 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-07-06 22:10:20 +00:00
mycookie = pgm->cookie;
pgm->cookie = PDATA(pgm)->chained_pdata;
Add support for the AVR Dragon (JTAG and ISP mode). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@675 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-26 21:14:10 +00:00
if (jtagmkII_getsync(pgm, EMULATOR_MODE_SPI) != 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unable to sync with the AVR Dragon in ISP mode\n");
Fix a serious memory corruption problem resulting out of the chaining of both, the stk500v2 and the jtagmkII programmers for some programming hardware (JTAG ICE mkII and AVR Dragon running in ISP, HVSP or PP mode), where both programmers have to maintain their private programmer data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@834 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-07-06 22:10:20 +00:00
pgm->cookie = mycookie;
Cleanup the open/close handling to avoid accessing unallocated memory (in the atexit handler) in case of bailing out. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@910 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-13 12:44:54 +00:00
return -1;
Add support for the AVR Dragon (JTAG and ISP mode). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@675 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-26 21:14:10 +00:00
}
Fix a serious memory corruption problem resulting out of the chaining of both, the stk500v2 and the jtagmkII programmers for some programming hardware (JTAG ICE mkII and AVR Dragon running in ISP, HVSP or PP mode), where both programmers have to maintain their private programmer data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@834 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-07-06 22:10:20 +00:00
pgm->cookie = mycookie;
Add support for the AVR Dragon (JTAG and ISP mode). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@675 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-26 21:14:10 +00:00
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
PDATA(pgm)->pgmtype = PGMTYPE_JTAGICE_MKII;
Add support for the AVR Dragon (JTAG and ISP mode). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@675 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-26 21:14:10 +00:00
if (pgm->bitclock != 0.0) {
if (pgm->set_sck_period(pgm, pgm->bitclock) != 0)
return -1;
}
return 0;
}
Implement and document HVSP and PP modes for the AVR Dragon. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@677 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-11-01 21:47:25 +00:00
/*
* Wrapper functions for the AVR Dragon in HV mode. This mode
* uses the normal JTAG ICE mkII packet stream to communicate with the
* ICE, but then encapsulates AVRISP mkII commands using
* CMND_ISP_PACKET.
*/
/*
* Open an AVR Dragon in HV mode (HVSP or parallel).
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_dragon_hv_open(PROGRAMMER *pgm, const char *port) {
Change the second parameter of the ser_open method from "baud" into a "union pinfo", so the USB parameters can be passed without hacks. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1276 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-02-21 13:44:11 +00:00
union pinfo pinfo;
Implement and document HVSP and PP modes for the AVR Dragon. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@677 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-11-01 21:47:25 +00:00
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_notice2("stk500v2_dragon_hv_open()\n");
Implement and document HVSP and PP modes for the AVR Dragon. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@677 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-11-01 21:47:25 +00:00
/*
* The JTAG ICE mkII always starts with a baud rate of 19200 Bd upon
* attaching. If the config file or command-line parameters specify
* a higher baud rate, we switch to it later on, after establishing
* the connection with the ICE.
*/
Implemented basic serial code refactoring for upcoming SerialUPDI implementation git-svn-id: svn://svn.savannah.nongnu.org/avrdude/branches/serial_refactoring@1511 81a1dc3b-b13d-400b-aceb-764788c761c2
2021-12-08 10:09:52 +00:00
pinfo.serialinfo.baud = 19200;
pinfo.serialinfo.cflags = SERIAL_8N1;
Implement and document HVSP and PP modes for the AVR Dragon. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@677 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-11-01 21:47:25 +00:00
/*
* If the port name starts with "usb", divert the serial routines
* to the USB ones. The serial_open() function for USB overrides
* the meaning of the "baud" parameter to be the USB device ID to
* search for.
*/
if (strncmp(port, "usb", 3) == 0) {
#if defined(HAVE_LIBUSB)
serdev = &usb_serdev;
Change the second parameter of the ser_open method from "baud" into a "union pinfo", so the USB parameters can be passed without hacks. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1276 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-02-21 13:44:11 +00:00
pinfo.usbinfo.vid = USB_VENDOR_ATMEL;
pinfo.usbinfo.flags = 0;
pinfo.usbinfo.pid = USB_DEVICE_AVRDRAGON;
Change all the USB details (endpoint numbers, max transfer size etc.) to a per-programmer adjustable value. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1108 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-11-26 16:24:56 +00:00
pgm->fd.usb.max_xfer = USBDEV_MAX_XFER_MKII;
pgm->fd.usb.rep = USBDEV_BULK_EP_READ_MKII;
pgm->fd.usb.wep = USBDEV_BULK_EP_WRITE_MKII;
pgm->fd.usb.eep = 0; /* no seperate EP for events */
Implement and document HVSP and PP modes for the AVR Dragon. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@677 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-11-01 21:47:25 +00:00
#else
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("avrdude was compiled without usb support\n");
Implement and document HVSP and PP modes for the AVR Dragon. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@677 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-11-01 21:47:25 +00:00
return -1;
#endif
}
strcpy(pgm->port, port);
Change the second parameter of the ser_open method from "baud" into a "union pinfo", so the USB parameters can be passed without hacks. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1276 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-02-21 13:44:11 +00:00
if (serial_open(port, pinfo, &pgm->fd)==-1) {
* serial_open() calls will now return -1 on error (no call to exit()) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@948 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-10-22 14:29:56 +00:00
return -1;
}
Implement and document HVSP and PP modes for the AVR Dragon. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@677 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-11-01 21:47:25 +00:00
/*
* drain any extraneous input
*/
stk500v2_drain(pgm, 0);
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
PROGRAMMER *pgmcp = pgm_dup(pgm);
pgmcp->cookie = PDATA(pgm)->chained_pdata;
if (jtagmkII_getsync(pgmcp, EMULATOR_MODE_HV) != 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unable to sync with the AVR Dragon in HV mode\n");
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
pgm_free(pgmcp);
Cleanup the open/close handling to avoid accessing unallocated memory (in the atexit handler) in case of bailing out. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@910 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-13 12:44:54 +00:00
return -1;
Implement and document HVSP and PP modes for the AVR Dragon. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@677 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-11-01 21:47:25 +00:00
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
pgm_free(pgmcp);
Turn all non-const static data into instance data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@807 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-02-25 09:20:38 +00:00
PDATA(pgm)->pgmtype = PGMTYPE_JTAGICE_MKII;
Implement and document HVSP and PP modes for the AVR Dragon. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@677 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-11-01 21:47:25 +00:00
if (pgm->bitclock != 0.0) {
if (pgm->set_sck_period(pgm, pgm->bitclock) != 0)
return -1;
}
return 0;
}
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
/*
* Wrapper functions for the JTAGICE3 in ISP mode. This mode
* uses the normal JTAGICE3 packet stream to communicate with the
* ICE, but then encapsulates AVRISP mkII commands using
* scope AVRISP.
*/
/*
* Open a JTAGICE3 in ISP mode.
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk500v2_jtag3_open(PROGRAMMER *pgm, const char *port) {
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
void *mycookie;
int rv;
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_notice2("stk500v2_jtag3_open()\n");
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
Turn the usbpid parameter of the programmer into a list of PIDs. Make the JTAGICE3 programmer handle a list of PIDs, by trying each of them in sequence. Use a single, central jtag3_open_common() function to handle the common code of all jtag3_open_* functions. Centralize all USB VID/PID definitions in usbdevs.h. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1283 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-02-27 13:06:03 +00:00
if (jtag3_open_common(pgm, port) < 0)
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
return -1;
mycookie = pgm->cookie;
pgm->cookie = PDATA(pgm)->chained_pdata;
if ((rv = jtag3_getsync(pgm, 42)) != 0) {
if (rv != JTAGII_GETSYNC_FAIL_GRACEFUL)
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unable to sync with the JTAGICE3 in ISP mode\n");
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
pgm->cookie = mycookie;
return -1;
}
pgm->cookie = mycookie;
PDATA(pgm)->pgmtype = PGMTYPE_JTAGICE3;
if (pgm->bitclock != 0.0) {
if (pgm->set_sck_period(pgm, pgm->bitclock) != 0)
return -1;
}
return 0;
}
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
/*
* XPROG wrapper
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk600_xprog_command(const PROGRAMMER *pgm, unsigned char *b,
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
unsigned int cmdsize, unsigned int responsesize)
{
unsigned char *newb;
unsigned int s;
int rv;
if (cmdsize < responsesize)
s = responsesize;
else
s = cmdsize;
if ((newb = malloc(s + 1)) == 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("out of memory\n");
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
return -1;
}
newb[0] = CMD_XPROG;
memcpy(newb + 1, b, cmdsize);
rv = stk500v2_command(pgm, newb, cmdsize + 1, responsesize + 1);
if (rv == 0) {
memcpy(b, newb + 1, responsesize);
}
free(newb);
return rv;
}
/*
* issue the 'program enable' command to the AVR device, XPROG version
*/
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk600_xprog_program_enable(const PROGRAMMER *pgm, const AVRPART *p) {
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
unsigned char buf[16];
unsigned int eepagesize = 42;
unsigned int nvm_base;
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
AVRMEM *mem = NULL;
int use_tpi;
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
Offload the programming interface info from part->flags to part->prog_modes flags now just hold parameters of the JTAG interface and some secondary serial, parallel, pseudo parallel info. This separation brings clarity. It used to be hard to augur whether a part has an ISP interface: (part->flags & (AVRPART_HAS_PDI | AVRPART_AVR32 | AVRPART_HAS_TPI | AVRPART_HAS_UPDI)) == 0 && (part->flags & AVRPART_SERIALOK) != 0 or had HVSP or HVPP capability, for that matter. Now it is just, eg, part->prog_modes & PM_ISP part->prog_modes & PM_HVPP
2022-08-30 15:33:42 +00:00
use_tpi = (p->prog_modes & PM_TPI) != 0;
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
if (!use_tpi) {
if (p->nvm_base == 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("no nvm_base parameter for PDI device\n");
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
return -1;
}
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
if ((mem = avr_locate_mem(p, "eeprom")) != NULL) {
bug #60753: Patch #1436 breaks multiple programmer/device combinations on MacOS BigSur * avr.c: compare page_size > 1 instead of != 0 * stk500v2.c: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1452 81a1dc3b-b13d-400b-aceb-764788c761c2
2021-06-27 20:26:05 +00:00
if (mem->page_size <= 1) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("no EEPROM page_size parameter for PDI device\n");
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
return -1;
}
eepagesize = mem->page_size;
}
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
}
buf[0] = CMD_XPROG_SETMODE;
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
buf[1] = use_tpi? XPRG_MODE_TPI: XPRG_MODE_PDI;
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
if (stk500v2_command(pgm, buf, 2, sizeof(buf)) < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("CMD_XPROG_SETMODE(XPRG_MODE_%s) failed\n", use_tpi? "TPI": "PDI");
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
return -1;
}
buf[0] = XPRG_CMD_ENTER_PROGMODE;
if (stk600_xprog_command(pgm, buf, 1, 2) < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("XPRG_CMD_ENTER_PROGMODE failed\n");
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
return -1;
}
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
if (use_tpi) {
/*
* Whatever all that might mean, it matches what AVR Studio
* does.
*/
if (stk500v2_setparm_real(pgm, PARAM_DISCHARGEDELAY, 232) < 0)
return -1;
buf[0] = XPRG_CMD_SET_PARAM;
Initial commit for jtag3 TPI support. Can read signature on ATtiny104 XPlained mini, but nothing else yet.
2022-11-21 20:18:13 +00:00
buf[1] = XPRG_PARAM_NVMCMD_ADDR;
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
buf[2] = 51;
if (stk600_xprog_command(pgm, buf, 3, 2) < 0) {
Initial commit for jtag3 TPI support. Can read signature on ATtiny104 XPlained mini, but nothing else yet.
2022-11-21 20:18:13 +00:00
pmsg_error("XPRG_CMD_SET_PARAM(XPRG_PARAM_NVMCMD_ADDR) failed\n");
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
return -1;
}
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
buf[0] = XPRG_CMD_SET_PARAM;
Initial commit for jtag3 TPI support. Can read signature on ATtiny104 XPlained mini, but nothing else yet.
2022-11-21 20:18:13 +00:00
buf[1] = XPRG_PARAM_NVMCSR_ADDR;
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
buf[2] = 50;
if (stk600_xprog_command(pgm, buf, 3, 2) < 0) {
Initial commit for jtag3 TPI support. Can read signature on ATtiny104 XPlained mini, but nothing else yet.
2022-11-21 20:18:13 +00:00
pmsg_error("XPRG_CMD_SET_PARAM(XPRG_PARAM_NVMCSR_ADDR) failed\n");
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
return -1;
}
} else {
buf[0] = XPRG_CMD_SET_PARAM;
buf[1] = XPRG_PARAM_NVMBASE;
nvm_base = p->nvm_base;
/*
* The 0x01000000 appears to be an indication to the programmer
* that the respective address is located in IO (i.e., SRAM)
* memory address space rather than flash. This is not documented
* anywhere in AVR079 but matches what AVR Studio does.
*/
nvm_base |= 0x01000000;
buf[2] = nvm_base >> 24;
buf[3] = nvm_base >> 16;
buf[4] = nvm_base >> 8;
buf[5] = nvm_base;
if (stk600_xprog_command(pgm, buf, 6, 2) < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("XPRG_CMD_SET_PARAM(XPRG_PARAM_NVMBASE) failed\n");
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
return -1;
}
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
if (mem != NULL) {
buf[0] = XPRG_CMD_SET_PARAM;
buf[1] = XPRG_PARAM_EEPPAGESIZE;
buf[2] = eepagesize >> 8;
buf[3] = eepagesize;
if (stk600_xprog_command(pgm, buf, 4, 2) < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("XPRG_CMD_SET_PARAM(XPRG_PARAM_EEPPAGESIZE) failed\n");
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
return -1;
}
}
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
}
return 0;
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static unsigned char stk600_xprog_memtype(const PROGRAMMER *pgm, unsigned long addr) {
* stk500v2_private.h (struct pdata): add boot_start * stk500v2.c: For the "flash" pseudo-memory of Xmega devices, distinguish addresses between "application" and "boot" area. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1081 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-04-19 13:32:34 +00:00
if (addr >= PDATA(pgm)->boot_start)
return XPRG_MEM_TYPE_BOOT;
else
return XPRG_MEM_TYPE_APPL;
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static void stk600_xprog_disable(const PROGRAMMER *pgm) {
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
unsigned char buf[2];
buf[0] = XPRG_CMD_LEAVE_PROGMODE;
if (stk600_xprog_command(pgm, buf, 1, 2) < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("XPRG_CMD_LEAVE_PROGMODE failed\n");
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
}
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk600_xprog_write_byte(const PROGRAMMER *pgm, const AVRPART *p, const AVRMEM *mem,
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
unsigned long addr, unsigned char data)
{
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
unsigned char b[9 + 256];
int need_erase = 0;
unsigned char write_size = 1;
unsigned char memcode;
memset(b, 0, sizeof(b));
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
* stk500v2_private.h (struct pdata): add boot_start * stk500v2.c: For the "flash" pseudo-memory of Xmega devices, distinguish addresses between "application" and "boot" area. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1081 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-04-19 13:32:34 +00:00
if (strcmp(mem->desc, "flash") == 0) {
memcode = stk600_xprog_memtype(pgm, addr);
} else if (strcmp(mem->desc, "application") == 0 ||
strcmp(mem->desc, "apptable") == 0) {
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
memcode = XPRG_MEM_TYPE_APPL;
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
} else if (strcmp(mem->desc, "boot") == 0) {
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
memcode = XPRG_MEM_TYPE_BOOT;
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
} else if (strcmp(mem->desc, "eeprom") == 0) {
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
memcode = XPRG_MEM_TYPE_EEPROM;
bug #30451: Accessing some Xmega memory sections gives not supported error * stk500v2.c: Handle all Xmega memory sections (except "prodsig" which is not documented in AVR079) * fileio.c: Treat the "boot", "application", and "apptable" regions (which are actually subregions of "flash") all as being flash, i.e. suppress trailing 0xFF bytes when reading them * avr.c: (Dito.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1075 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-03-29 14:32:18 +00:00
} else if (strncmp(mem->desc, "lock", strlen("lock")) == 0) {
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
memcode = XPRG_MEM_TYPE_LOCKBITS;
Handle writing fuse bytes (on the Xmega). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@820 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-06-24 21:18:36 +00:00
} else if (strncmp(mem->desc, "fuse", strlen("fuse")) == 0) {
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
memcode = XPRG_MEM_TYPE_FUSE;
Offload the programming interface info from part->flags to part->prog_modes flags now just hold parameters of the JTAG interface and some secondary serial, parallel, pseudo parallel info. This separation brings clarity. It used to be hard to augur whether a part has an ISP interface: (part->flags & (AVRPART_HAS_PDI | AVRPART_AVR32 | AVRPART_HAS_TPI | AVRPART_HAS_UPDI)) == 0 && (part->flags & AVRPART_SERIALOK) != 0 or had HVSP or HVPP capability, for that matter. Now it is just, eg, part->prog_modes & PM_ISP part->prog_modes & PM_HVPP
2022-08-30 15:33:42 +00:00
if (p->prog_modes & PM_TPI)
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
/*
* TPI devices need a mystic erase prior to writing their
* fuses.
*/
need_erase = 1;
Support both userrow and usersig names
2022-02-28 18:46:47 +00:00
} else if (strcmp(mem->desc, "usersig") == 0 ||
strcmp(mem->desc, "userrow") == 0) {
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
memcode = XPRG_MEM_TYPE_USERSIG;
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
} else {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unknown memory %s\n", mem->desc);
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
return -1;
}
addr += mem->offset;
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
if (need_erase) {
b[0] = XPRG_CMD_ERASE;
b[1] = XPRG_ERASE_CONFIG;
b[2] = mem->offset >> 24;
b[3] = mem->offset >> 16;
b[4] = mem->offset >> 8;
b[5] = mem->offset + 1;
if (stk600_xprog_command(pgm, b, 6, 2) < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("XPRG_CMD_ERASE(XPRG_ERASE_CONFIG) failed\n");
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
return -1;
}
}
Offload the programming interface info from part->flags to part->prog_modes flags now just hold parameters of the JTAG interface and some secondary serial, parallel, pseudo parallel info. This separation brings clarity. It used to be hard to augur whether a part has an ISP interface: (part->flags & (AVRPART_HAS_PDI | AVRPART_AVR32 | AVRPART_HAS_TPI | AVRPART_HAS_UPDI)) == 0 && (part->flags & AVRPART_SERIALOK) != 0 or had HVSP or HVPP capability, for that matter. Now it is just, eg, part->prog_modes & PM_ISP part->prog_modes & PM_HVPP
2022-08-30 15:33:42 +00:00
if (p->prog_modes & PM_TPI) {
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
/*
* Some TPI memories (configuration aka. fuse) require a
* larger write block size. We record that as a blocksize in
* avrdude.conf.
*/
if (mem->blocksize != 0)
write_size = mem->blocksize;
}
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
b[0] = XPRG_CMD_WRITE_MEM;
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
b[1] = memcode;
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
b[2] = 0; /* pagemode: non-paged write */
b[3] = addr >> 24;
b[4] = addr >> 16;
b[5] = addr >> 8;
b[6] = addr;
b[7] = 0;
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
b[8] = write_size;
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
b[9] = data;
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
if (stk600_xprog_command(pgm, b, 9 + write_size, 2) < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("XPRG_CMD_WRITE_MEM failed\n");
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
return -1;
}
return 0;
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk600_xprog_read_byte(const PROGRAMMER *pgm, const AVRPART *p, const AVRMEM *mem,
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
unsigned long addr, unsigned char * value)
{
unsigned char b[8];
* stk500v2_private.h (struct pdata): add boot_start * stk500v2.c: For the "flash" pseudo-memory of Xmega devices, distinguish addresses between "application" and "boot" area. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1081 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-04-19 13:32:34 +00:00
if (strcmp(mem->desc, "flash") == 0) {
b[1] = stk600_xprog_memtype(pgm, addr);
} else if (strcmp(mem->desc, "application") == 0 ||
strcmp(mem->desc, "apptable") == 0) {
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
b[1] = XPRG_MEM_TYPE_APPL;
} else if (strcmp(mem->desc, "boot") == 0) {
b[1] = XPRG_MEM_TYPE_BOOT;
} else if (strcmp(mem->desc, "eeprom") == 0) {
b[1] = XPRG_MEM_TYPE_EEPROM;
} else if (strcmp(mem->desc, "signature") == 0) {
b[1] = XPRG_MEM_TYPE_APPL;
} else if (strncmp(mem->desc, "fuse", strlen("fuse")) == 0) {
b[1] = XPRG_MEM_TYPE_FUSE;
bug #30451: Accessing some Xmega memory sections gives not supported error * stk500v2.c: Handle all Xmega memory sections (except "prodsig" which is not documented in AVR079) * fileio.c: Treat the "boot", "application", and "apptable" regions (which are actually subregions of "flash") all as being flash, i.e. suppress trailing 0xFF bytes when reading them * avr.c: (Dito.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1075 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-03-29 14:32:18 +00:00
} else if (strncmp(mem->desc, "lock", strlen("lock")) == 0) {
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
b[1] = XPRG_MEM_TYPE_LOCKBITS;
Submitted by Joseph Coffland: patch #8957: Allow reading prodsig memory from stk500v2 on xmega devices * stk500v2.c (stk600_xprog_read_byte, stk600_xprog_paged_load): treat "prodsig" memory space as equivalent to "calibration" git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1469 81a1dc3b-b13d-400b-aceb-764788c761c2
2021-11-11 20:59:25 +00:00
} else if (strcmp(mem->desc, "calibration") == 0 ||
strcmp(mem->desc, "prodsig") == 0) {
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
b[1] = XPRG_MEM_TYPE_FACTORY_CALIBRATION;
Support both userrow and usersig names
2022-02-28 18:46:47 +00:00
} else if (strcmp(mem->desc, "usersig") == 0 ||
strcmp(mem->desc, "userrow") == 0) {
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
b[1] = XPRG_MEM_TYPE_USERSIG;
} else {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unknown memory %s\n", mem->desc);
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
return -1;
}
addr += mem->offset;
b[0] = XPRG_CMD_READ_MEM;
b[2] = addr >> 24;
b[3] = addr >> 16;
b[4] = addr >> 8;
b[5] = addr;
b[6] = 0;
b[7] = 1;
if (stk600_xprog_command(pgm, b, 8, 3) < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("XPRG_CMD_READ_MEM failed\n");
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
return -1;
}
*value = b[2];
return 0;
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk600_xprog_paged_load(const PROGRAMMER *pgm, const AVRPART *p, const AVRMEM *mem,
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
unsigned int page_size,
unsigned int addr, unsigned int n_bytes)
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
{
unsigned char *b;
unsigned int offset;
unsigned char memtype;
* stk500v2_private.h (struct pdata): add boot_start * stk500v2.c: For the "flash" pseudo-memory of Xmega devices, distinguish addresses between "application" and "boot" area. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1081 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-04-19 13:32:34 +00:00
int n_bytes_orig = n_bytes, dynamic_memtype = 0;
Various fixes for Xmega devices. * avrdude.conf.in: Correctly declare EEPROM page sizes for all Xmega devices (0x20 instead of 0x100). * avr.c: If a memory region has a page size declared, try using the paged IO routines regardless of the target memory name. Xmega EEPROM requires to be written in paged mode. Correctly use a long (rather than unsigned long) variable to evaluate the success status of the paged mode write attempt. * stk500v2.c: Don't apply TIF space offsets twice (bug #27995: AVRDUDE 5.8svn fails to program and read XMEGA); use stk500v2_loadaddr() prior to paged mode (EEPROM and flash) writes, otherwise programming of flash areas will fail; while being there, check the return value of stk500v2_loadaddr() everywhere; use the correct write/erase mode bits (same as AVR Studio does). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@907 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-12 15:42:40 +00:00
unsigned long use_ext_addr = 0;
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
/*
* The XPROG read command supports at most 256 bytes in one
* transfer.
*/
if (page_size > 256)
page_size = 256; /* not really a page size anymore */
/*
* Fancy offsets everywhere.
* This is probably what AVR079 means when writing about the
* "TIF address space".
*/
* stk500v2_private.h (struct pdata): add boot_start * stk500v2.c: For the "flash" pseudo-memory of Xmega devices, distinguish addresses between "application" and "boot" area. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1081 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-04-19 13:32:34 +00:00
if (strcmp(mem->desc, "flash") == 0) {
memtype = 0;
dynamic_memtype = 1;
if (mem->size > 64 * 1024)
use_ext_addr = (1UL << 31);
} else if (strcmp(mem->desc, "application") == 0 ||
strcmp(mem->desc, "apptable") == 0) {
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
memtype = XPRG_MEM_TYPE_APPL;
Various fixes for Xmega devices. * avrdude.conf.in: Correctly declare EEPROM page sizes for all Xmega devices (0x20 instead of 0x100). * avr.c: If a memory region has a page size declared, try using the paged IO routines regardless of the target memory name. Xmega EEPROM requires to be written in paged mode. Correctly use a long (rather than unsigned long) variable to evaluate the success status of the paged mode write attempt. * stk500v2.c: Don't apply TIF space offsets twice (bug #27995: AVRDUDE 5.8svn fails to program and read XMEGA); use stk500v2_loadaddr() prior to paged mode (EEPROM and flash) writes, otherwise programming of flash areas will fail; while being there, check the return value of stk500v2_loadaddr() everywhere; use the correct write/erase mode bits (same as AVR Studio does). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@907 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-12 15:42:40 +00:00
if (mem->size > 64 * 1024)
use_ext_addr = (1UL << 31);
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
} else if (strcmp(mem->desc, "boot") == 0) {
memtype = XPRG_MEM_TYPE_BOOT;
Various fixes for Xmega devices. * avrdude.conf.in: Correctly declare EEPROM page sizes for all Xmega devices (0x20 instead of 0x100). * avr.c: If a memory region has a page size declared, try using the paged IO routines regardless of the target memory name. Xmega EEPROM requires to be written in paged mode. Correctly use a long (rather than unsigned long) variable to evaluate the success status of the paged mode write attempt. * stk500v2.c: Don't apply TIF space offsets twice (bug #27995: AVRDUDE 5.8svn fails to program and read XMEGA); use stk500v2_loadaddr() prior to paged mode (EEPROM and flash) writes, otherwise programming of flash areas will fail; while being there, check the return value of stk500v2_loadaddr() everywhere; use the correct write/erase mode bits (same as AVR Studio does). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@907 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-12 15:42:40 +00:00
// Do we have to consider the total amount of flash
// instead to decide whether to use extended addressing?
if (mem->size > 64 * 1024)
use_ext_addr = (1UL << 31);
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
} else if (strcmp(mem->desc, "eeprom") == 0) {
memtype = XPRG_MEM_TYPE_EEPROM;
* stk500v2.c (stk600_xprog_paged_load, stk600_xprog_paged_write): Fix regression in the AVRISPmkII/STK600 TPI handling introduced by the USBasp's TPI implementation which added a pagesize even for the minor memory regions of TPI devices. Also fix wrong offset introduced by the memory tagging patch. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1009 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-15 13:04:36 +00:00
} else if (strcmp(mem->desc, "signature") == 0) {
memtype = XPRG_MEM_TYPE_APPL;
} else if (strncmp(mem->desc, "fuse", strlen("fuse")) == 0) {
memtype = XPRG_MEM_TYPE_FUSE;
bug #30451: Accessing some Xmega memory sections gives not supported error * stk500v2.c: Handle all Xmega memory sections (except "prodsig" which is not documented in AVR079) * fileio.c: Treat the "boot", "application", and "apptable" regions (which are actually subregions of "flash") all as being flash, i.e. suppress trailing 0xFF bytes when reading them * avr.c: (Dito.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1075 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-03-29 14:32:18 +00:00
} else if (strncmp(mem->desc, "lock", strlen("lock")) == 0) {
* stk500v2.c (stk600_xprog_paged_load, stk600_xprog_paged_write): Fix regression in the AVRISPmkII/STK600 TPI handling introduced by the USBasp's TPI implementation which added a pagesize even for the minor memory regions of TPI devices. Also fix wrong offset introduced by the memory tagging patch. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1009 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-15 13:04:36 +00:00
memtype = XPRG_MEM_TYPE_LOCKBITS;
Submitted by Joseph Coffland: patch #8957: Allow reading prodsig memory from stk500v2 on xmega devices * stk500v2.c (stk600_xprog_read_byte, stk600_xprog_paged_load): treat "prodsig" memory space as equivalent to "calibration" git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1469 81a1dc3b-b13d-400b-aceb-764788c761c2
2021-11-11 20:59:25 +00:00
} else if (strcmp(mem->desc, "calibration") == 0 ||
strcmp(mem->desc, "prodsig") == 0) {
* stk500v2.c (stk600_xprog_paged_load, stk600_xprog_paged_write): Fix regression in the AVRISPmkII/STK600 TPI handling introduced by the USBasp's TPI implementation which added a pagesize even for the minor memory regions of TPI devices. Also fix wrong offset introduced by the memory tagging patch. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1009 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-15 13:04:36 +00:00
memtype = XPRG_MEM_TYPE_FACTORY_CALIBRATION;
Support both userrow and usersig names
2022-02-28 18:46:47 +00:00
} else if (strcmp(mem->desc, "usersig") == 0 ||
strcmp(mem->desc, "userrow") == 0) {
* stk500v2.c (stk600_xprog_paged_load, stk600_xprog_paged_write): Fix regression in the AVRISPmkII/STK600 TPI handling introduced by the USBasp's TPI implementation which added a pagesize even for the minor memory regions of TPI devices. Also fix wrong offset introduced by the memory tagging patch. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1009 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-15 13:04:36 +00:00
memtype = XPRG_MEM_TYPE_USERSIG;
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
} else {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unknown paged memory %s\n", mem->desc);
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
return -1;
}
* stk500v2.c (stk600_xprog_paged_load, stk600_xprog_paged_write): Fix regression in the AVRISPmkII/STK600 TPI handling introduced by the USBasp's TPI implementation which added a pagesize even for the minor memory regions of TPI devices. Also fix wrong offset introduced by the memory tagging patch. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1009 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-15 13:04:36 +00:00
offset = addr;
addr += mem->offset;
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
if ((b = malloc(page_size + 2)) == NULL) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("out of memory\n");
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
return -1;
}
Fixed following findings reported by cppcheck * avr910.c:625 (error) Possible null pointer dereference: cmd - otherwise it is redundant to check if cmd is null at line 624 * avr910.c:626 (error) Possible null pointer dereference: cmd - otherwise it is redundant to check if cmd is null at line 624 * avr910.c:168 (information) The scope of the variable 'devtype_1st' can be reduced * avr910.c:169 (information) The scope of the variable 'dev_supported' can be reduced * avrftdi.c:647 (error) Using sizeof for array given as function argument returns the size of pointer. * stk500v2.c:3347 (error) Memory leak: b * stk500v2.c:3452 (error) Memory leak: b * usbasp.c:554 (error) Using sizeof for array given as function argument returns the size of pointer. * usbasp.c:485 (information) The scope of the variable 'dly' can be reduced git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1035 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-01-08 17:09:40 +00:00
if (stk500v2_loadaddr(pgm, use_ext_addr) < 0) {
free(b);
Various fixes for Xmega devices. * avrdude.conf.in: Correctly declare EEPROM page sizes for all Xmega devices (0x20 instead of 0x100). * avr.c: If a memory region has a page size declared, try using the paged IO routines regardless of the target memory name. Xmega EEPROM requires to be written in paged mode. Correctly use a long (rather than unsigned long) variable to evaluate the success status of the paged mode write attempt. * stk500v2.c: Don't apply TIF space offsets twice (bug #27995: AVRDUDE 5.8svn fails to program and read XMEGA); use stk500v2_loadaddr() prior to paged mode (EEPROM and flash) writes, otherwise programming of flash areas will fail; while being there, check the return value of stk500v2_loadaddr() everywhere; use the correct write/erase mode bits (same as AVR Studio does). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@907 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-12 15:42:40 +00:00
return -1;
Fixed following findings reported by cppcheck * avr910.c:625 (error) Possible null pointer dereference: cmd - otherwise it is redundant to check if cmd is null at line 624 * avr910.c:626 (error) Possible null pointer dereference: cmd - otherwise it is redundant to check if cmd is null at line 624 * avr910.c:168 (information) The scope of the variable 'devtype_1st' can be reduced * avr910.c:169 (information) The scope of the variable 'dev_supported' can be reduced * avrftdi.c:647 (error) Using sizeof for array given as function argument returns the size of pointer. * stk500v2.c:3347 (error) Memory leak: b * stk500v2.c:3452 (error) Memory leak: b * usbasp.c:554 (error) Using sizeof for array given as function argument returns the size of pointer. * usbasp.c:485 (information) The scope of the variable 'dly' can be reduced git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1035 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-01-08 17:09:40 +00:00
}
Various fixes for Xmega devices. * avrdude.conf.in: Correctly declare EEPROM page sizes for all Xmega devices (0x20 instead of 0x100). * avr.c: If a memory region has a page size declared, try using the paged IO routines regardless of the target memory name. Xmega EEPROM requires to be written in paged mode. Correctly use a long (rather than unsigned long) variable to evaluate the success status of the paged mode write attempt. * stk500v2.c: Don't apply TIF space offsets twice (bug #27995: AVRDUDE 5.8svn fails to program and read XMEGA); use stk500v2_loadaddr() prior to paged mode (EEPROM and flash) writes, otherwise programming of flash areas will fail; while being there, check the return value of stk500v2_loadaddr() everywhere; use the correct write/erase mode bits (same as AVR Studio does). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@907 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-12 15:42:40 +00:00
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
while (n_bytes != 0) {
* stk500v2_private.h (struct pdata): add boot_start * stk500v2.c: For the "flash" pseudo-memory of Xmega devices, distinguish addresses between "application" and "boot" area. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1081 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-04-19 13:32:34 +00:00
if (dynamic_memtype)
memtype = stk600_xprog_memtype(pgm, addr - mem->offset);
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
b[0] = XPRG_CMD_READ_MEM;
b[1] = memtype;
b[2] = addr >> 24;
b[3] = addr >> 16;
b[4] = addr >> 8;
b[5] = addr;
b[6] = page_size >> 8;
b[7] = page_size;
if (stk600_xprog_command(pgm, b, 8, page_size + 2) < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("XPRG_CMD_READ_MEM failed\n");
Fixed following findings reported by cppcheck * avr910.c:625 (error) Possible null pointer dereference: cmd - otherwise it is redundant to check if cmd is null at line 624 * avr910.c:626 (error) Possible null pointer dereference: cmd - otherwise it is redundant to check if cmd is null at line 624 * avr910.c:168 (information) The scope of the variable 'devtype_1st' can be reduced * avr910.c:169 (information) The scope of the variable 'dev_supported' can be reduced * avrftdi.c:647 (error) Using sizeof for array given as function argument returns the size of pointer. * stk500v2.c:3347 (error) Memory leak: b * stk500v2.c:3452 (error) Memory leak: b * usbasp.c:554 (error) Using sizeof for array given as function argument returns the size of pointer. * usbasp.c:485 (information) The scope of the variable 'dly' can be reduced git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1035 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-01-08 17:09:40 +00:00
free(b);
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
return -1;
}
memcpy(mem->buf + offset, b + 2, page_size);
if (n_bytes < page_size) {
n_bytes = page_size;
}
offset += page_size;
addr += page_size;
n_bytes -= page_size;
}
free(b);
return n_bytes_orig;
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk600_xprog_paged_write(const PROGRAMMER *pgm, const AVRPART *p, const AVRMEM *mem,
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
unsigned int page_size,
unsigned int addr, unsigned int n_bytes)
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
{
unsigned char *b;
unsigned int offset;
unsigned char memtype;
* stk500v2_private.h (struct pdata): add boot_start * stk500v2.c: For the "flash" pseudo-memory of Xmega devices, distinguish addresses between "application" and "boot" area. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1081 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-04-19 13:32:34 +00:00
int n_bytes_orig = n_bytes, dynamic_memtype = 0;
* stk500v2.c (stk600_xprog_paged_write): Fix a fatal miscalculation of the number of bytes to be written which caused a malloc chunk corruption. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@778 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-29 08:39:15 +00:00
size_t writesize;
Various fixes for Xmega devices. * avrdude.conf.in: Correctly declare EEPROM page sizes for all Xmega devices (0x20 instead of 0x100). * avr.c: If a memory region has a page size declared, try using the paged IO routines regardless of the target memory name. Xmega EEPROM requires to be written in paged mode. Correctly use a long (rather than unsigned long) variable to evaluate the success status of the paged mode write attempt. * stk500v2.c: Don't apply TIF space offsets twice (bug #27995: AVRDUDE 5.8svn fails to program and read XMEGA); use stk500v2_loadaddr() prior to paged mode (EEPROM and flash) writes, otherwise programming of flash areas will fail; while being there, check the return value of stk500v2_loadaddr() everywhere; use the correct write/erase mode bits (same as AVR Studio does). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@907 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-12 15:42:40 +00:00
unsigned long use_ext_addr = 0;
unsigned char writemode;
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
/*
* The XPROG read command supports at most 256 bytes in one
* transfer.
*/
if (page_size > 512) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("cannot handle page size > 512\n");
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
return -1;
}
/*
* Fancy offsets everywhere.
* This is probably what AVR079 means when writing about the
* "TIF address space".
*/
* stk500v2_private.h (struct pdata): add boot_start * stk500v2.c: For the "flash" pseudo-memory of Xmega devices, distinguish addresses between "application" and "boot" area. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1081 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-04-19 13:32:34 +00:00
if (strcmp(mem->desc, "flash") == 0) {
memtype = 0;
dynamic_memtype = 1;
writemode = (1 << XPRG_MEM_WRITE_WRITE);
if (mem->size > 64 * 1024)
use_ext_addr = (1UL << 31);
} else if (strcmp(mem->desc, "application") == 0 ||
strcmp(mem->desc, "apptable") == 0) {
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
memtype = XPRG_MEM_TYPE_APPL;
Various fixes for Xmega devices. * avrdude.conf.in: Correctly declare EEPROM page sizes for all Xmega devices (0x20 instead of 0x100). * avr.c: If a memory region has a page size declared, try using the paged IO routines regardless of the target memory name. Xmega EEPROM requires to be written in paged mode. Correctly use a long (rather than unsigned long) variable to evaluate the success status of the paged mode write attempt. * stk500v2.c: Don't apply TIF space offsets twice (bug #27995: AVRDUDE 5.8svn fails to program and read XMEGA); use stk500v2_loadaddr() prior to paged mode (EEPROM and flash) writes, otherwise programming of flash areas will fail; while being there, check the return value of stk500v2_loadaddr() everywhere; use the correct write/erase mode bits (same as AVR Studio does). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@907 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-12 15:42:40 +00:00
writemode = (1 << XPRG_MEM_WRITE_WRITE);
if (mem->size > 64 * 1024)
use_ext_addr = (1UL << 31);
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
} else if (strcmp(mem->desc, "boot") == 0) {
memtype = XPRG_MEM_TYPE_BOOT;
Various fixes for Xmega devices. * avrdude.conf.in: Correctly declare EEPROM page sizes for all Xmega devices (0x20 instead of 0x100). * avr.c: If a memory region has a page size declared, try using the paged IO routines regardless of the target memory name. Xmega EEPROM requires to be written in paged mode. Correctly use a long (rather than unsigned long) variable to evaluate the success status of the paged mode write attempt. * stk500v2.c: Don't apply TIF space offsets twice (bug #27995: AVRDUDE 5.8svn fails to program and read XMEGA); use stk500v2_loadaddr() prior to paged mode (EEPROM and flash) writes, otherwise programming of flash areas will fail; while being there, check the return value of stk500v2_loadaddr() everywhere; use the correct write/erase mode bits (same as AVR Studio does). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@907 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-12 15:42:40 +00:00
writemode = (1 << XPRG_MEM_WRITE_WRITE);
// Do we have to consider the total amount of flash
// instead to decide whether to use extended addressing?
if (mem->size > 64 * 1024)
use_ext_addr = (1UL << 31);
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
} else if (strcmp(mem->desc, "eeprom") == 0) {
memtype = XPRG_MEM_TYPE_EEPROM;
Various fixes for Xmega devices. * avrdude.conf.in: Correctly declare EEPROM page sizes for all Xmega devices (0x20 instead of 0x100). * avr.c: If a memory region has a page size declared, try using the paged IO routines regardless of the target memory name. Xmega EEPROM requires to be written in paged mode. Correctly use a long (rather than unsigned long) variable to evaluate the success status of the paged mode write attempt. * stk500v2.c: Don't apply TIF space offsets twice (bug #27995: AVRDUDE 5.8svn fails to program and read XMEGA); use stk500v2_loadaddr() prior to paged mode (EEPROM and flash) writes, otherwise programming of flash areas will fail; while being there, check the return value of stk500v2_loadaddr() everywhere; use the correct write/erase mode bits (same as AVR Studio does). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@907 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-12 15:42:40 +00:00
writemode = (1 << XPRG_MEM_WRITE_WRITE) | (1 << XPRG_MEM_WRITE_ERASE);
* stk500v2.c (stk600_xprog_paged_load, stk600_xprog_paged_write): Fix regression in the AVRISPmkII/STK600 TPI handling introduced by the USBasp's TPI implementation which added a pagesize even for the minor memory regions of TPI devices. Also fix wrong offset introduced by the memory tagging patch. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1009 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-15 13:04:36 +00:00
} else if (strcmp(mem->desc, "signature") == 0) {
memtype = XPRG_MEM_TYPE_APPL;
writemode = (1 << XPRG_MEM_WRITE_WRITE);
} else if (strncmp(mem->desc, "fuse", strlen("fuse")) == 0) {
memtype = XPRG_MEM_TYPE_FUSE;
writemode = (1 << XPRG_MEM_WRITE_WRITE);
bug #30451: Accessing some Xmega memory sections gives not supported error * stk500v2.c: Handle all Xmega memory sections (except "prodsig" which is not documented in AVR079) * fileio.c: Treat the "boot", "application", and "apptable" regions (which are actually subregions of "flash") all as being flash, i.e. suppress trailing 0xFF bytes when reading them * avr.c: (Dito.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1075 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-03-29 14:32:18 +00:00
} else if (strncmp(mem->desc, "lock", strlen("lock")) == 0) {
* stk500v2.c (stk600_xprog_paged_load, stk600_xprog_paged_write): Fix regression in the AVRISPmkII/STK600 TPI handling introduced by the USBasp's TPI implementation which added a pagesize even for the minor memory regions of TPI devices. Also fix wrong offset introduced by the memory tagging patch. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1009 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-15 13:04:36 +00:00
memtype = XPRG_MEM_TYPE_LOCKBITS;
writemode = (1 << XPRG_MEM_WRITE_WRITE);
} else if (strcmp(mem->desc, "calibration") == 0) {
memtype = XPRG_MEM_TYPE_FACTORY_CALIBRATION;
writemode = (1 << XPRG_MEM_WRITE_WRITE);
Support both userrow and usersig names
2022-02-28 18:46:47 +00:00
} else if (strcmp(mem->desc, "usersig") == 0 ||
strcmp(mem->desc, "userrow") == 0) {
* stk500v2.c (stk600_xprog_paged_load, stk600_xprog_paged_write): Fix regression in the AVRISPmkII/STK600 TPI handling introduced by the USBasp's TPI implementation which added a pagesize even for the minor memory regions of TPI devices. Also fix wrong offset introduced by the memory tagging patch. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1009 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-15 13:04:36 +00:00
memtype = XPRG_MEM_TYPE_USERSIG;
writemode = (1 << XPRG_MEM_WRITE_WRITE);
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
} else {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unknown paged memory %s\n", mem->desc);
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
return -1;
}
* stk500v2.c (stk600_xprog_paged_load, stk600_xprog_paged_write): Fix regression in the AVRISPmkII/STK600 TPI handling introduced by the USBasp's TPI implementation which added a pagesize even for the minor memory regions of TPI devices. Also fix wrong offset introduced by the memory tagging patch. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1009 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-15 13:04:36 +00:00
offset = addr;
Mega-commit to bring in memory tagging. Each memory image byte is now tagged as it's being read from a file. Only bytes read from a file will be written or verified (modulo page granularity requirements). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1007 81a1dc3b-b13d-400b-aceb-764788c761c2
2011-09-14 21:49:42 +00:00
addr += mem->offset;
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
if ((b = malloc(page_size + 9)) == NULL) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("out of memory\n");
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
return -1;
}
Fixed following findings reported by cppcheck * avr910.c:625 (error) Possible null pointer dereference: cmd - otherwise it is redundant to check if cmd is null at line 624 * avr910.c:626 (error) Possible null pointer dereference: cmd - otherwise it is redundant to check if cmd is null at line 624 * avr910.c:168 (information) The scope of the variable 'devtype_1st' can be reduced * avr910.c:169 (information) The scope of the variable 'dev_supported' can be reduced * avrftdi.c:647 (error) Using sizeof for array given as function argument returns the size of pointer. * stk500v2.c:3347 (error) Memory leak: b * stk500v2.c:3452 (error) Memory leak: b * usbasp.c:554 (error) Using sizeof for array given as function argument returns the size of pointer. * usbasp.c:485 (information) The scope of the variable 'dly' can be reduced git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1035 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-01-08 17:09:40 +00:00
if (stk500v2_loadaddr(pgm, use_ext_addr) < 0) {
free(b);
Various fixes for Xmega devices. * avrdude.conf.in: Correctly declare EEPROM page sizes for all Xmega devices (0x20 instead of 0x100). * avr.c: If a memory region has a page size declared, try using the paged IO routines regardless of the target memory name. Xmega EEPROM requires to be written in paged mode. Correctly use a long (rather than unsigned long) variable to evaluate the success status of the paged mode write attempt. * stk500v2.c: Don't apply TIF space offsets twice (bug #27995: AVRDUDE 5.8svn fails to program and read XMEGA); use stk500v2_loadaddr() prior to paged mode (EEPROM and flash) writes, otherwise programming of flash areas will fail; while being there, check the return value of stk500v2_loadaddr() everywhere; use the correct write/erase mode bits (same as AVR Studio does). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@907 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-12 15:42:40 +00:00
return -1;
Fixed following findings reported by cppcheck * avr910.c:625 (error) Possible null pointer dereference: cmd - otherwise it is redundant to check if cmd is null at line 624 * avr910.c:626 (error) Possible null pointer dereference: cmd - otherwise it is redundant to check if cmd is null at line 624 * avr910.c:168 (information) The scope of the variable 'devtype_1st' can be reduced * avr910.c:169 (information) The scope of the variable 'dev_supported' can be reduced * avrftdi.c:647 (error) Using sizeof for array given as function argument returns the size of pointer. * stk500v2.c:3347 (error) Memory leak: b * stk500v2.c:3452 (error) Memory leak: b * usbasp.c:554 (error) Using sizeof for array given as function argument returns the size of pointer. * usbasp.c:485 (information) The scope of the variable 'dly' can be reduced git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1035 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-01-08 17:09:40 +00:00
}
Various fixes for Xmega devices. * avrdude.conf.in: Correctly declare EEPROM page sizes for all Xmega devices (0x20 instead of 0x100). * avr.c: If a memory region has a page size declared, try using the paged IO routines regardless of the target memory name. Xmega EEPROM requires to be written in paged mode. Correctly use a long (rather than unsigned long) variable to evaluate the success status of the paged mode write attempt. * stk500v2.c: Don't apply TIF space offsets twice (bug #27995: AVRDUDE 5.8svn fails to program and read XMEGA); use stk500v2_loadaddr() prior to paged mode (EEPROM and flash) writes, otherwise programming of flash areas will fail; while being there, check the return value of stk500v2_loadaddr() everywhere; use the correct write/erase mode bits (same as AVR Studio does). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@907 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-12 15:42:40 +00:00
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
while (n_bytes != 0) {
* stk500v2_private.h (struct pdata): add boot_start * stk500v2.c: For the "flash" pseudo-memory of Xmega devices, distinguish addresses between "application" and "boot" area. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1081 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-04-19 13:32:34 +00:00
if (dynamic_memtype)
memtype = stk600_xprog_memtype(pgm, addr - mem->offset);
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
if (page_size > 256) {
/*
* AVR079 is not quite clear. While it suggests that
* downloading up to 512 bytes (256 words) were OK, it
* obviously isn't -- 512-byte pages on the ATxmega128A1
* are getting corrupted when written as a single piece.
* It writes random junk somewhere beyond byte 256.
* Splitting it into 256 byte chunks, and only setting the
* erase page / write page bits in the final chunk helps.
*/
if (page_size % 256 != 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("page size not multiple of 256\n");
Fixed following findings reported by cppcheck * avr910.c:625 (error) Possible null pointer dereference: cmd - otherwise it is redundant to check if cmd is null at line 624 * avr910.c:626 (error) Possible null pointer dereference: cmd - otherwise it is redundant to check if cmd is null at line 624 * avr910.c:168 (information) The scope of the variable 'devtype_1st' can be reduced * avr910.c:169 (information) The scope of the variable 'dev_supported' can be reduced * avrftdi.c:647 (error) Using sizeof for array given as function argument returns the size of pointer. * stk500v2.c:3347 (error) Memory leak: b * stk500v2.c:3452 (error) Memory leak: b * usbasp.c:554 (error) Using sizeof for array given as function argument returns the size of pointer. * usbasp.c:485 (information) The scope of the variable 'dly' can be reduced git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1035 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-01-08 17:09:40 +00:00
free(b);
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
return -1;
}
unsigned int chunk;
for (chunk = 0; chunk < page_size; chunk += 256) {
* stk500v2.c (stk600_xprog_paged_write): Fix a fatal miscalculation of the number of bytes to be written which caused a malloc chunk corruption. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@778 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-29 08:39:15 +00:00
if (n_bytes < 256) {
memset(b + 9 + n_bytes, 0xff, 256 - n_bytes);
writesize = n_bytes;
} else {
writesize = 256;
}
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
b[0] = XPRG_CMD_WRITE_MEM;
b[1] = memtype;
Xmega page erase implementation for XPROG (AVRISPmkII, STK600) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1090 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-05-04 15:47:26 +00:00
b[2] = writemode;
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
b[3] = addr >> 24;
b[4] = addr >> 16;
b[5] = addr >> 8;
b[6] = addr;
b[7] = 1;
b[8] = 0;
* stk500v2.c (stk600_xprog_paged_write): Fix a fatal miscalculation of the number of bytes to be written which caused a malloc chunk corruption. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@778 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-29 08:39:15 +00:00
memcpy(b + 9, mem->buf + offset, writesize);
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
if (stk600_xprog_command(pgm, b, 256 + 9, 2) < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("XPRG_CMD_WRITE_MEM failed\n");
Fixed following findings reported by cppcheck * avr910.c:625 (error) Possible null pointer dereference: cmd - otherwise it is redundant to check if cmd is null at line 624 * avr910.c:626 (error) Possible null pointer dereference: cmd - otherwise it is redundant to check if cmd is null at line 624 * avr910.c:168 (information) The scope of the variable 'devtype_1st' can be reduced * avr910.c:169 (information) The scope of the variable 'dev_supported' can be reduced * avrftdi.c:647 (error) Using sizeof for array given as function argument returns the size of pointer. * stk500v2.c:3347 (error) Memory leak: b * stk500v2.c:3452 (error) Memory leak: b * usbasp.c:554 (error) Using sizeof for array given as function argument returns the size of pointer. * usbasp.c:485 (information) The scope of the variable 'dly' can be reduced git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1035 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-01-08 17:09:40 +00:00
free(b);
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
return -1;
}
if (n_bytes < 256)
n_bytes = 256;
offset += 256;
addr += 256;
n_bytes -= 256;
}
} else {
* stk500v2.c (stk600_xprog_paged_write): Fix a fatal miscalculation of the number of bytes to be written which caused a malloc chunk corruption. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@778 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-29 08:39:15 +00:00
if (n_bytes < page_size) {
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
/*
* This can easily happen if the input file was not a
* multiple of the page size.
*/
memset(b + 9 + n_bytes, 0xff, page_size - n_bytes);
* stk500v2.c (stk600_xprog_paged_write): Fix a fatal miscalculation of the number of bytes to be written which caused a malloc chunk corruption. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@778 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-29 08:39:15 +00:00
writesize = n_bytes;
} else {
writesize = page_size;
}
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
b[0] = XPRG_CMD_WRITE_MEM;
b[1] = memtype;
Various fixes for Xmega devices. * avrdude.conf.in: Correctly declare EEPROM page sizes for all Xmega devices (0x20 instead of 0x100). * avr.c: If a memory region has a page size declared, try using the paged IO routines regardless of the target memory name. Xmega EEPROM requires to be written in paged mode. Correctly use a long (rather than unsigned long) variable to evaluate the success status of the paged mode write attempt. * stk500v2.c: Don't apply TIF space offsets twice (bug #27995: AVRDUDE 5.8svn fails to program and read XMEGA); use stk500v2_loadaddr() prior to paged mode (EEPROM and flash) writes, otherwise programming of flash areas will fail; while being there, check the return value of stk500v2_loadaddr() everywhere; use the correct write/erase mode bits (same as AVR Studio does). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@907 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-12 15:42:40 +00:00
b[2] = writemode;
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
b[3] = addr >> 24;
b[4] = addr >> 16;
b[5] = addr >> 8;
b[6] = addr;
b[7] = page_size >> 8;
b[8] = page_size;
* stk500v2.c (stk600_xprog_paged_write): Fix a fatal miscalculation of the number of bytes to be written which caused a malloc chunk corruption. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@778 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-29 08:39:15 +00:00
memcpy(b + 9, mem->buf + offset, writesize);
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
if (stk600_xprog_command(pgm, b, page_size + 9, 2) < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("XPRG_CMD_WRITE_MEM failed\n");
Fixed following findings reported by cppcheck * avr910.c:625 (error) Possible null pointer dereference: cmd - otherwise it is redundant to check if cmd is null at line 624 * avr910.c:626 (error) Possible null pointer dereference: cmd - otherwise it is redundant to check if cmd is null at line 624 * avr910.c:168 (information) The scope of the variable 'devtype_1st' can be reduced * avr910.c:169 (information) The scope of the variable 'dev_supported' can be reduced * avrftdi.c:647 (error) Using sizeof for array given as function argument returns the size of pointer. * stk500v2.c:3347 (error) Memory leak: b * stk500v2.c:3452 (error) Memory leak: b * usbasp.c:554 (error) Using sizeof for array given as function argument returns the size of pointer. * usbasp.c:485 (information) The scope of the variable 'dly' can be reduced git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1035 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-01-08 17:09:40 +00:00
free(b);
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
return -1;
}
if (n_bytes < page_size)
n_bytes = page_size;
offset += page_size;
addr += page_size;
n_bytes -= page_size;
}
}
free(b);
return n_bytes_orig;
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk600_xprog_chip_erase(const PROGRAMMER *pgm, const AVRPART *p) {
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
unsigned char b[6];
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
AVRMEM *mem;
unsigned int addr = 0;
Offload the programming interface info from part->flags to part->prog_modes flags now just hold parameters of the JTAG interface and some secondary serial, parallel, pseudo parallel info. This separation brings clarity. It used to be hard to augur whether a part has an ISP interface: (part->flags & (AVRPART_HAS_PDI | AVRPART_AVR32 | AVRPART_HAS_TPI | AVRPART_HAS_UPDI)) == 0 && (part->flags & AVRPART_SERIALOK) != 0 or had HVSP or HVPP capability, for that matter. Now it is just, eg, part->prog_modes & PM_ISP part->prog_modes & PM_HVPP
2022-08-30 15:33:42 +00:00
if (p->prog_modes & PM_TPI) {
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
if ((mem = avr_locate_mem(p, "flash")) == NULL) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("no FLASH definition found for TPI device\n");
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
return -1;
}
addr = mem->offset + 1;
}
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
b[0] = XPRG_CMD_ERASE;
b[1] = XPRG_ERASE_CHIP;
Implement TPI mode for AVRISPmkII/STK600. Add ATtiny4/5/9/10 to avrdude.conf.in. Document TPI and new device support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@916 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-15 16:36:13 +00:00
b[2] = addr >> 24;
b[3] = addr >> 16;
b[4] = addr >> 8;
b[5] = addr;
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
if (stk600_xprog_command(pgm, b, 6, 2) < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("XPRG_CMD_ERASE(XPRG_ERASE_CHIP) failed\n");
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
return -1;
}
return 0;
}
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
static int stk600_xprog_page_erase(const PROGRAMMER *pgm, const AVRPART *p, const AVRMEM *m,
Xmega page erase implementation for XPROG (AVRISPmkII, STK600) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1090 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-05-04 15:47:26 +00:00
unsigned int addr)
{
unsigned char b[6];
if (strcmp(m->desc, "flash") == 0) {
bug #39893: Verification failure with AVRISPmkII and Xmega * stk500v2.c (stk600_xprog_page_erase): Fix argument that is passed to stk600_xprog_memtype() git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1188 81a1dc3b-b13d-400b-aceb-764788c761c2
2013-08-30 06:49:40 +00:00
b[1] = stk600_xprog_memtype(pgm, addr) == XPRG_MEM_TYPE_APPL?
Xmega page erase implementation for XPROG (AVRISPmkII, STK600) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1090 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-05-04 15:47:26 +00:00
XPRG_ERASE_APP_PAGE: XPRG_ERASE_BOOT_PAGE;
} else if (strcmp(m->desc, "application") == 0 ||
strcmp(m->desc, "apptable") == 0) {
b[1] = XPRG_ERASE_APP_PAGE;
} else if (strcmp(m->desc, "boot") == 0) {
b[1] = XPRG_ERASE_BOOT_PAGE;
} else if (strcmp(m->desc, "eeprom") == 0) {
b[1] = XPRG_ERASE_EEPROM_PAGE;
Support both userrow and usersig names
2022-02-28 18:46:47 +00:00
} else if (strcmp(m->desc, "usersig") == 0 ||
strcmp(m->desc, "userrow") == 0) {
Submitted by Travis Griggs: bug #38307: Can't write usersig of an xmega256a3 * stk500v2.c (stk600_xprog_page_erase): allow erasing the usersig space. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1199 81a1dc3b-b13d-400b-aceb-764788c761c2
2013-09-02 21:10:44 +00:00
b[1] = XPRG_ERASE_USERSIG;
Xmega page erase implementation for XPROG (AVRISPmkII, STK600) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1090 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-05-04 15:47:26 +00:00
} else {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("unknown paged memory %s\n", m->desc);
Xmega page erase implementation for XPROG (AVRISPmkII, STK600) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1090 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-05-04 15:47:26 +00:00
return -1;
}
addr += m->offset;
b[0] = XPRG_CMD_ERASE;
b[2] = addr >> 24;
b[3] = addr >> 16;
b[4] = addr >> 8;
b[5] = addr;
if (stk600_xprog_command(pgm, b, 6, 2) < 0) {
Review and overhaul AVRDUDE's messaging system (#1126) * Change avrdude_message(MSG_XYZ, ...) to msg_xyz(...) * Define and use pmsg_xyz(...) instead of msg_xyz("%s: ...", progname, ...) * Review and change avrdude_message() levels - Introduce new levels warning, error and ext_error - Distribute info level to info, warning, error, ext_error - Assign levels (more) consistently - Unify grammar, punctuation and style of messages * Use imsg_xyz() to print indented messages * Show function name in errors and warnings on -v * Reduce effective verbosity level by number of -q above one
2022-10-17 14:44:55 +00:00
pmsg_error("XPRG_CMD_ERASE(%d) failed\n", b[1]);
Xmega page erase implementation for XPROG (AVRISPmkII, STK600) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1090 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-05-04 15:47:26 +00:00
return -1;
}
return 0;
}
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
/*
* Modify pgm's methods for XPROG operation.
*/
static void stk600_setup_xprog(PROGRAMMER * pgm)
{
pgm->program_enable = stk600_xprog_program_enable;
pgm->disable = stk600_xprog_disable;
pgm->read_byte = stk600_xprog_read_byte;
pgm->write_byte = stk600_xprog_write_byte;
pgm->paged_load = stk600_xprog_paged_load;
pgm->paged_write = stk600_xprog_paged_write;
Xmega page erase implementation for XPROG (AVRISPmkII, STK600) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1090 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-05-04 15:47:26 +00:00
pgm->page_erase = stk600_xprog_page_erase;
First implementation of ATxmega support. By now, only the PDI mode of the STK600 is supported. Single-byte EEPROM (and flash) updates do not work yet. * avr.c: "boot" memory is a candidate memory region for paged operations, besides "flash" and "eeprom". * avrdude.conf.in: add ATxmega128A1 and ATxmega128A1revD * avrpart.h: add the AVRPART_HAS_PDI flag (used to distinguish ATxmega parts from classic AVRs), the nvm_base part field, and the offset field for a memory region. * config_gram.y: add "has_pdi", "nvm_base", and "offset" * lexer.l: (Ditto.) * main.c: disable auto_erase for ATxmega parts * stk500v2.c: implement the XPROG functionality, and divert to this for ATxmega parts * avrdude.1: Document the changes. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@777 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-07-26 22:53:40 +00:00
pgm->chip_erase = stk600_xprog_chip_erase;
}
Implement and document HVSP and PP modes for the AVR Dragon. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@677 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-11-01 21:47:25 +00:00
Cleanup the open/close handling to avoid accessing unallocated memory (in the atexit handler) in case of bailing out. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@910 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-13 12:44:54 +00:00
/*
* Modify pgm's methods for ISP operation.
*/
static void stk600_setup_isp(PROGRAMMER * pgm)
{
pgm->program_enable = stk500v2_program_enable;
pgm->disable = stk500v2_disable;
* stk500v2.c: Use stk500isp_read_byte/stk500isp_write_byte for every byte-wide access (rather than JTAGICE3 only). This finally obsoletes the use of the prehistoric SPI_MULTI command where AVRDUDE used to assemble all the low-level ISP stuff by itself. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1336 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-10-06 22:15:29 +00:00
pgm->read_byte = stk500isp_read_byte;
pgm->write_byte = stk500isp_write_byte;
Cleanup the open/close handling to avoid accessing unallocated memory (in the atexit handler) in case of bailing out. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@910 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-13 12:44:54 +00:00
pgm->paged_load = stk500v2_paged_load;
pgm->paged_write = stk500v2_paged_write;
Xmega page erase implementation for XPROG (AVRISPmkII, STK600) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1090 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-05-04 15:47:26 +00:00
pgm->page_erase = stk500v2_page_erase;
Cleanup the open/close handling to avoid accessing unallocated memory (in the atexit handler) in case of bailing out. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@910 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-13 12:44:54 +00:00
pgm->chip_erase = stk500v2_chip_erase;
}
Parser does not need to know all programmer types now, new programmers will update only the table in pgm_type.c. * config_gram.y, lexer.l: removed programmer type keywords, use now locate_programmer_type() function * pgm_type.[ch]: added new files for table of programmer types * main.c: allow list of programmer types by -c ?type * avrdude.conf.in: changed all type keywords to quoted strings * doc/avrdude.texi: changed description of type definition, list of valid types is now included from generated file * doc/Makefile.am: generate list of programmer types for doc * all programmers [hc]: add xxx_desc string for description of programmer git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1051 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-01-31 17:03:43 +00:00
const char stk500v2_desc[] = "Atmel STK500 Version 2.x firmware";
Cleanup the open/close handling to avoid accessing unallocated memory (in the atexit handler) in case of bailing out. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@910 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-13 12:44:54 +00:00
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
void stk500v2_initpgm(PROGRAMMER *pgm) {
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
strcpy(pgm->type, "STK500V2");
/*
* mandatory functions
*/
pgm->initialize = stk500v2_initialize;
pgm->display = stk500v2_display;
pgm->enable = stk500v2_enable;
pgm->disable = stk500v2_disable;
pgm->program_enable = stk500v2_program_enable;
pgm->chip_erase = stk500v2_chip_erase;
pgm->cmd = stk500v2_cmd;
pgm->open = stk500v2_open;
pgm->close = stk500v2_close;
* stk500v2.c: Use stk500isp_read_byte/stk500isp_write_byte for every byte-wide access (rather than JTAGICE3 only). This finally obsoletes the use of the prehistoric SPI_MULTI command where AVRDUDE used to assemble all the low-level ISP stuff by itself. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1336 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-10-06 22:15:29 +00:00
pgm->read_byte = stk500isp_read_byte;
pgm->write_byte = stk500isp_write_byte;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
/*
* optional functions
*/
pgm->paged_write = stk500v2_paged_write;
pgm->paged_load = stk500v2_paged_load;
Xmega page erase implementation for XPROG (AVRISPmkII, STK600) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1090 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-05-04 15:47:26 +00:00
pgm->page_erase = stk500v2_page_erase;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
pgm->print_parms = stk500v2_print_parms;
pgm->set_vtarget = stk500v2_set_vtarget;
pgm->set_varef = stk500v2_set_varef;
pgm->set_fosc = stk500v2_set_fosc;
pgm->set_sck_period = stk500v2_set_sck_period;
Submitted by John Voltz: add AVR053 oscillator calibration. * main.c: Add the -O option. * pgm.c: Add the hook for the perform_osccal() method. * pgm.h: (Ditto.) * stk500v2.c: Implement perform_osccal(). * avrdude.1: Document the -O option. * doc/avrdude.texi: (Ditto.) Partially closes bug #17487: AVR RC oscillator calibration routine not supported (feature request) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@671 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-09 14:34:24 +00:00
pgm->perform_osccal = stk500v2_perform_osccal;
main.c, pgm.c, pgm.h: Add setup and teardown hooks to the programmer definition. If present, call the setup hook immediately after finding the respective programmer object, and schedule the teardown hook to be called upon exit. This allows the programmer implementation to dynamically allocate private programmer data. avr910.c, butterfly.c, jtagmkI.c, jtagmkII.c, stk500v2.c, usbasp.c, usbtiny.c: Convert static programmer data into dynamically allocated data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@764 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-11-07 20:36:12 +00:00
pgm->setup = stk500v2_setup;
pgm->teardown = stk500v2_teardown;
Mega-commit to bring in both, the STK500v2 support from Erik Walthinsen, as well as JTAG ICE mkII support (by me). Erik's submission has been cleaned up a little bit, mostly to add his name and the current year to the copyright of the new file, remove trailing white space before importing the files, and fix the minor syntax errors in his avrdude.conf.in additions (missing semicolons). The JTAG ICE mkII support should be considered alpha to beta quality at this point. Few things are still to be done, like defering the hfuse (OCDEN) tweaks until they are really required. Also, for reasons not yet known, the target MCU doesn't start to run after signing off from the ICE, it needs a power-cycle first (at least on my STK500). Note that for the JTAG ICE, I did change a few things in the internal API. Notably I made the serial receive timeout configurable by the backends via an exported variable (done in both the Posix and the Win32 implementation), and I made the serial_recv() function return a -1 instead of bailing out with exit(1) upon encountering a receive timeout (currently only done in the Posix implementation). Both measures together allow me to receive a datastreem from the ICE at 115 kbps on a somewhat lossy PCI multi-UART card that occasionally drops a character. The JTAG ICE mkII protocol has enough of safety layers to allow recovering from these events, but the previous code wasn't prepared for any kind of recovery. The Win32 change for this still has to be done, and the traditional drivers need to be converted to exit(1) upon encountering a timeout (as they're now getting a -1 returned they didn't see before in that case). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@451 81a1dc3b-b13d-400b-aceb-764788c761c2
2005-05-10 19:17:12 +00:00
pgm->page_size = 256;
}
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
Parser does not need to know all programmer types now, new programmers will update only the table in pgm_type.c. * config_gram.y, lexer.l: removed programmer type keywords, use now locate_programmer_type() function * pgm_type.[ch]: added new files for table of programmer types * main.c: allow list of programmer types by -c ?type * avrdude.conf.in: changed all type keywords to quoted strings * doc/avrdude.texi: changed description of type definition, list of valid types is now included from generated file * doc/Makefile.am: generate list of programmer types for doc * all programmers [hc]: add xxx_desc string for description of programmer git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1051 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-01-31 17:03:43 +00:00
const char stk500pp_desc[] = "Atmel STK500 V2 in parallel programming mode";
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
void stk500pp_initpgm(PROGRAMMER *pgm) {
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
strcpy(pgm->type, "STK500PP");
/*
* mandatory functions
*/
pgm->initialize = stk500pp_initialize;
pgm->display = stk500v2_display;
pgm->enable = stk500v2_enable;
pgm->disable = stk500pp_disable;
pgm->program_enable = stk500pp_program_enable;
pgm->chip_erase = stk500pp_chip_erase;
pgm->open = stk500v2_open;
pgm->close = stk500v2_close;
Replace the fallback of avr_read_byte() and avr_write_byte() to avr_read_byte_default() and avr_write_byte_default (resp.) by directly calling the latter functions from within all programmers that don't implement their own read_byte()/write_byte() methods. In turn, make the read_byte() and write_byte() methods mandatory, and the cmd() method (direct ISP command) optional instead (it's effectively mandatory for any programmer using avr_read_byte_default()/avr_write_byte_default() though). Remove all the pointless cmd() method stubs from those programmers that don't need it. Eliminate avr_read_byte() as it was now completely identical to pgm->read_byte(). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@684 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-11-20 15:04:09 +00:00
pgm->read_byte = stk500pp_read_byte;
pgm->write_byte = stk500pp_write_byte;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
/*
* optional functions
*/
pgm->paged_write = stk500pp_paged_write;
pgm->paged_load = stk500pp_paged_load;
pgm->print_parms = stk500v2_print_parms;
pgm->set_vtarget = stk500v2_set_vtarget;
pgm->set_varef = stk500v2_set_varef;
pgm->set_fosc = stk500v2_set_fosc;
pgm->set_sck_period = stk500v2_set_sck_period;
main.c, pgm.c, pgm.h: Add setup and teardown hooks to the programmer definition. If present, call the setup hook immediately after finding the respective programmer object, and schedule the teardown hook to be called upon exit. This allows the programmer implementation to dynamically allocate private programmer data. avr910.c, butterfly.c, jtagmkI.c, jtagmkII.c, stk500v2.c, usbasp.c, usbtiny.c: Convert static programmer data into dynamically allocated data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@764 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-11-07 20:36:12 +00:00
pgm->setup = stk500v2_setup;
pgm->teardown = stk500v2_teardown;
First stab at an implementation of the STK500 parallel programming feature (v2 firmware only), named "stk500pp". Still not yet complete: EEPROM writes not working, documentation missing, only ATmega16 parameters available in avrdude.conf.in, some parameters not yet implemented. * avrdude.conf.in: Add sample parameters for PP mode to ATmega16. * avrpart.h: Add the parallel programming control parameters. * avrpart.c: (Ditto.) * config_gram.y: Add stk500pp configuration grammar. * lexer.l: Add stk500pp token recognition. * stk500v2.h: Add declaration for stk500pp_initpgm(). * stk500v2.c: Add stk500pp implementation. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@587 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-16 21:30:14 +00:00
pgm->page_size = 256;
}
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
Parser does not need to know all programmer types now, new programmers will update only the table in pgm_type.c. * config_gram.y, lexer.l: removed programmer type keywords, use now locate_programmer_type() function * pgm_type.[ch]: added new files for table of programmer types * main.c: allow list of programmer types by -c ?type * avrdude.conf.in: changed all type keywords to quoted strings * doc/avrdude.texi: changed description of type definition, list of valid types is now included from generated file * doc/Makefile.am: generate list of programmer types for doc * all programmers [hc]: add xxx_desc string for description of programmer git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1051 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-01-31 17:03:43 +00:00
const char stk500hvsp_desc[] = "Atmel STK500 V2 in high-voltage serial programming mode";
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
void stk500hvsp_initpgm(PROGRAMMER *pgm) {
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
strcpy(pgm->type, "STK500HVSP");
/*
* mandatory functions
*/
pgm->initialize = stk500hvsp_initialize;
pgm->display = stk500v2_display;
pgm->enable = stk500v2_enable;
pgm->disable = stk500hvsp_disable;
pgm->program_enable = stk500hvsp_program_enable;
pgm->chip_erase = stk500hvsp_chip_erase;
pgm->open = stk500v2_open;
pgm->close = stk500v2_close;
Replace the fallback of avr_read_byte() and avr_write_byte() to avr_read_byte_default() and avr_write_byte_default (resp.) by directly calling the latter functions from within all programmers that don't implement their own read_byte()/write_byte() methods. In turn, make the read_byte() and write_byte() methods mandatory, and the cmd() method (direct ISP command) optional instead (it's effectively mandatory for any programmer using avr_read_byte_default()/avr_write_byte_default() though). Remove all the pointless cmd() method stubs from those programmers that don't need it. Eliminate avr_read_byte() as it was now completely identical to pgm->read_byte(). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@684 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-11-20 15:04:09 +00:00
pgm->read_byte = stk500hvsp_read_byte;
pgm->write_byte = stk500hvsp_write_byte;
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
/*
* optional functions
*/
pgm->paged_write = stk500hvsp_paged_write;
pgm->paged_load = stk500hvsp_paged_load;
pgm->print_parms = stk500v2_print_parms;
pgm->set_vtarget = stk500v2_set_vtarget;
pgm->set_varef = stk500v2_set_varef;
pgm->set_fosc = stk500v2_set_fosc;
pgm->set_sck_period = stk500v2_set_sck_period;
main.c, pgm.c, pgm.h: Add setup and teardown hooks to the programmer definition. If present, call the setup hook immediately after finding the respective programmer object, and schedule the teardown hook to be called upon exit. This allows the programmer implementation to dynamically allocate private programmer data. avr910.c, butterfly.c, jtagmkI.c, jtagmkII.c, stk500v2.c, usbasp.c, usbtiny.c: Convert static programmer data into dynamically allocated data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@764 81a1dc3b-b13d-400b-aceb-764788c761c2
2007-11-07 20:36:12 +00:00
pgm->setup = stk500v2_setup;
pgm->teardown = stk500v2_teardown;
Implement STK500 (v2) HVSP mode. * stk500v2.c: Add new functions for HVSP support. * stk500v2.h: Add prototype for the stk500hvsp programmer. * avrpart.h: Add fields to struct avrpart for new features. * config_gram.y: Extend the configuration syntax for new features required for HVSP support. * lexer.l: (Ditto.) * avrdude.conf.in: Add HVSP support for ATtiny13 and ATtiny45 as an example. * avrdude.1: Document stk500hvsp. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@595 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-07-21 21:53:49 +00:00
pgm->page_size = 256;
}
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
Parser does not need to know all programmer types now, new programmers will update only the table in pgm_type.c. * config_gram.y, lexer.l: removed programmer type keywords, use now locate_programmer_type() function * pgm_type.[ch]: added new files for table of programmer types * main.c: allow list of programmer types by -c ?type * avrdude.conf.in: changed all type keywords to quoted strings * doc/avrdude.texi: changed description of type definition, list of valid types is now included from generated file * doc/Makefile.am: generate list of programmer types for doc * all programmers [hc]: add xxx_desc string for description of programmer git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1051 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-01-31 17:03:43 +00:00
const char stk500v2_jtagmkII_desc[] = "Atmel JTAG ICE mkII in ISP mode";
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
void stk500v2_jtagmkII_initpgm(PROGRAMMER *pgm) {
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
strcpy(pgm->type, "JTAGMKII_ISP");
/*
* mandatory functions
*/
pgm->initialize = stk500v2_initialize;
pgm->display = stk500v2_display;
pgm->enable = stk500v2_enable;
pgm->disable = stk500v2_disable;
pgm->program_enable = stk500v2_program_enable;
pgm->chip_erase = stk500v2_chip_erase;
pgm->cmd = stk500v2_cmd;
pgm->open = stk500v2_jtagmkII_open;
bug #28516: Linux/Dragon: Error message on exit * stk500v2.c: Fix the "bad response to GO command: RSP_ILLEGAL_EMULATOR_MODE" message. jtagmkII_close() has been called with the wrong pgm->cookie. Wrap it inside stk500v2_jtagmkII_close(), adjusting the cookie data appropriately. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@895 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-08 16:01:37 +00:00
pgm->close = stk500v2_jtagmkII_close;
* stk500v2.c: Use stk500isp_read_byte/stk500isp_write_byte for every byte-wide access (rather than JTAGICE3 only). This finally obsoletes the use of the prehistoric SPI_MULTI command where AVRDUDE used to assemble all the low-level ISP stuff by itself. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1336 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-10-06 22:15:29 +00:00
pgm->read_byte = stk500isp_read_byte;
pgm->write_byte = stk500isp_write_byte;
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
/*
* optional functions
*/
pgm->paged_write = stk500v2_paged_write;
pgm->paged_load = stk500v2_paged_load;
Xmega page erase implementation for XPROG (AVRISPmkII, STK600) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1090 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-05-04 15:47:26 +00:00
pgm->page_erase = stk500v2_page_erase;
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
pgm->print_parms = stk500v2_print_parms;
pgm->set_sck_period = stk500v2_set_sck_period_mk2;
Submitted by John Voltz: add AVR053 oscillator calibration. * main.c: Add the -O option. * pgm.c: Add the hook for the perform_osccal() method. * pgm.h: (Ditto.) * stk500v2.c: Implement perform_osccal(). * avrdude.1: Document the -O option. * doc/avrdude.texi: (Ditto.) Partially closes bug #17487: AVR RC oscillator calibration routine not supported (feature request) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@671 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-09 14:34:24 +00:00
pgm->perform_osccal = stk500v2_perform_osccal;
Fix a serious memory corruption problem resulting out of the chaining of both, the stk500v2 and the jtagmkII programmers for some programming hardware (JTAG ICE mkII and AVR Dragon running in ISP, HVSP or PP mode), where both programmers have to maintain their private programmer data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@834 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-07-06 22:10:20 +00:00
pgm->setup = stk500v2_jtagmkII_setup;
pgm->teardown = stk500v2_jtagmkII_teardown;
Add support for the JTAG ICE mkII in ISP mode. * avrdude.conf.in (jtag2isp): New programmer entry. * config_gram.y: Add K_JTAG_MKII_ISP. * jtagmkII.c: Restructure and export some more functions. * jtagmkII.h: Declare exported functions. * jtagmkII_private.h: Prepare file to be included in stk500v2.c. * lexer.l: Add jtagmkii_isp token. * stk500v2.c: Implement glue to jtagmkII.c. * stk500v2.h: Declare stk500v2_jtagmkII_initpgm(). * avrdude.1: Document the new programmer support. * doc/avrdude.texi: (Ditto.) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@643 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-09-06 20:06:07 +00:00
pgm->page_size = 256;
}
Add support for the AVR Dragon (JTAG and ISP mode). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@675 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-26 21:14:10 +00:00
Parser does not need to know all programmer types now, new programmers will update only the table in pgm_type.c. * config_gram.y, lexer.l: removed programmer type keywords, use now locate_programmer_type() function * pgm_type.[ch]: added new files for table of programmer types * main.c: allow list of programmer types by -c ?type * avrdude.conf.in: changed all type keywords to quoted strings * doc/avrdude.texi: changed description of type definition, list of valid types is now included from generated file * doc/Makefile.am: generate list of programmer types for doc * all programmers [hc]: add xxx_desc string for description of programmer git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1051 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-01-31 17:03:43 +00:00
const char stk500v2_dragon_isp_desc[] = "Atmel AVR Dragon in ISP mode";
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
void stk500v2_dragon_isp_initpgm(PROGRAMMER *pgm) {
Add support for the AVR Dragon (JTAG and ISP mode). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@675 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-26 21:14:10 +00:00
strcpy(pgm->type, "DRAGON_ISP");
/*
* mandatory functions
*/
pgm->initialize = stk500v2_initialize;
pgm->display = stk500v2_display;
pgm->enable = stk500v2_enable;
pgm->disable = stk500v2_disable;
pgm->program_enable = stk500v2_program_enable;
pgm->chip_erase = stk500v2_chip_erase;
pgm->cmd = stk500v2_cmd;
pgm->open = stk500v2_dragon_isp_open;
bug #28516: Linux/Dragon: Error message on exit * stk500v2.c: Fix the "bad response to GO command: RSP_ILLEGAL_EMULATOR_MODE" message. jtagmkII_close() has been called with the wrong pgm->cookie. Wrap it inside stk500v2_jtagmkII_close(), adjusting the cookie data appropriately. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@895 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-08 16:01:37 +00:00
pgm->close = stk500v2_jtagmkII_close;
* stk500v2.c: Use stk500isp_read_byte/stk500isp_write_byte for every byte-wide access (rather than JTAGICE3 only). This finally obsoletes the use of the prehistoric SPI_MULTI command where AVRDUDE used to assemble all the low-level ISP stuff by itself. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1336 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-10-06 22:15:29 +00:00
pgm->read_byte = stk500isp_read_byte;
pgm->write_byte = stk500isp_write_byte;
Add support for the AVR Dragon (JTAG and ISP mode). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@675 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-26 21:14:10 +00:00
/*
* optional functions
*/
pgm->paged_write = stk500v2_paged_write;
pgm->paged_load = stk500v2_paged_load;
Xmega page erase implementation for XPROG (AVRISPmkII, STK600) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1090 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-05-04 15:47:26 +00:00
pgm->page_erase = stk500v2_page_erase;
Add support for the AVR Dragon (JTAG and ISP mode). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@675 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-26 21:14:10 +00:00
pgm->print_parms = stk500v2_print_parms;
pgm->set_sck_period = stk500v2_set_sck_period_mk2;
bug #27507: SIGSEGV when using avrdragon (avrdude 5.8) * stk500v2.c (stk500v2_dragon_isp_initpgm): Use stk500v2_jtagmkII_setup/stk500v2_jtagmkII_rather than their jtagII counterparts, to get the private data properly initialized. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@887 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-07 15:53:50 +00:00
pgm->setup = stk500v2_jtagmkII_setup;
pgm->teardown = stk500v2_jtagmkII_teardown;
Add support for the AVR Dragon (JTAG and ISP mode). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@675 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-10-26 21:14:10 +00:00
pgm->page_size = 256;
}
Implement and document HVSP and PP modes for the AVR Dragon. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@677 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-11-01 21:47:25 +00:00
Parser does not need to know all programmer types now, new programmers will update only the table in pgm_type.c. * config_gram.y, lexer.l: removed programmer type keywords, use now locate_programmer_type() function * pgm_type.[ch]: added new files for table of programmer types * main.c: allow list of programmer types by -c ?type * avrdude.conf.in: changed all type keywords to quoted strings * doc/avrdude.texi: changed description of type definition, list of valid types is now included from generated file * doc/Makefile.am: generate list of programmer types for doc * all programmers [hc]: add xxx_desc string for description of programmer git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1051 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-01-31 17:03:43 +00:00
const char stk500v2_dragon_pp_desc[] = "Atmel AVR Dragon in PP mode";
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
void stk500v2_dragon_pp_initpgm(PROGRAMMER *pgm) {
Implement and document HVSP and PP modes for the AVR Dragon. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@677 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-11-01 21:47:25 +00:00
strcpy(pgm->type, "DRAGON_PP");
/*
* mandatory functions
*/
pgm->initialize = stk500pp_initialize;
pgm->display = stk500v2_display;
pgm->enable = stk500v2_enable;
pgm->disable = stk500pp_disable;
pgm->program_enable = stk500pp_program_enable;
pgm->chip_erase = stk500pp_chip_erase;
pgm->open = stk500v2_dragon_hv_open;
bug #28516: Linux/Dragon: Error message on exit * stk500v2.c: Fix the "bad response to GO command: RSP_ILLEGAL_EMULATOR_MODE" message. jtagmkII_close() has been called with the wrong pgm->cookie. Wrap it inside stk500v2_jtagmkII_close(), adjusting the cookie data appropriately. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@895 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-08 16:01:37 +00:00
pgm->close = stk500v2_jtagmkII_close;
Replace the fallback of avr_read_byte() and avr_write_byte() to avr_read_byte_default() and avr_write_byte_default (resp.) by directly calling the latter functions from within all programmers that don't implement their own read_byte()/write_byte() methods. In turn, make the read_byte() and write_byte() methods mandatory, and the cmd() method (direct ISP command) optional instead (it's effectively mandatory for any programmer using avr_read_byte_default()/avr_write_byte_default() though). Remove all the pointless cmd() method stubs from those programmers that don't need it. Eliminate avr_read_byte() as it was now completely identical to pgm->read_byte(). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@684 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-11-20 15:04:09 +00:00
pgm->read_byte = stk500pp_read_byte;
pgm->write_byte = stk500pp_write_byte;
Implement and document HVSP and PP modes for the AVR Dragon. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@677 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-11-01 21:47:25 +00:00
/*
* optional functions
*/
pgm->paged_write = stk500pp_paged_write;
pgm->paged_load = stk500pp_paged_load;
pgm->print_parms = stk500v2_print_parms;
pgm->set_vtarget = stk500v2_set_vtarget;
pgm->set_varef = stk500v2_set_varef;
pgm->set_fosc = stk500v2_set_fosc;
pgm->set_sck_period = stk500v2_set_sck_period_mk2;
Fix a serious memory corruption problem resulting out of the chaining of both, the stk500v2 and the jtagmkII programmers for some programming hardware (JTAG ICE mkII and AVR Dragon running in ISP, HVSP or PP mode), where both programmers have to maintain their private programmer data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@834 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-07-06 22:10:20 +00:00
pgm->setup = stk500v2_jtagmkII_setup;
pgm->teardown = stk500v2_jtagmkII_teardown;
Implement and document HVSP and PP modes for the AVR Dragon. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@677 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-11-01 21:47:25 +00:00
pgm->page_size = 256;
}
Parser does not need to know all programmer types now, new programmers will update only the table in pgm_type.c. * config_gram.y, lexer.l: removed programmer type keywords, use now locate_programmer_type() function * pgm_type.[ch]: added new files for table of programmer types * main.c: allow list of programmer types by -c ?type * avrdude.conf.in: changed all type keywords to quoted strings * doc/avrdude.texi: changed description of type definition, list of valid types is now included from generated file * doc/Makefile.am: generate list of programmer types for doc * all programmers [hc]: add xxx_desc string for description of programmer git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1051 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-01-31 17:03:43 +00:00
const char stk500v2_dragon_hvsp_desc[] = "Atmel AVR Dragon in HVSP mode";
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
void stk500v2_dragon_hvsp_initpgm(PROGRAMMER *pgm) {
Implement and document HVSP and PP modes for the AVR Dragon. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@677 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-11-01 21:47:25 +00:00
strcpy(pgm->type, "DRAGON_HVSP");
/*
* mandatory functions
*/
pgm->initialize = stk500hvsp_initialize;
pgm->display = stk500v2_display;
pgm->enable = stk500v2_enable;
pgm->disable = stk500hvsp_disable;
pgm->program_enable = stk500hvsp_program_enable;
pgm->chip_erase = stk500hvsp_chip_erase;
pgm->open = stk500v2_dragon_hv_open;
bug #28516: Linux/Dragon: Error message on exit * stk500v2.c: Fix the "bad response to GO command: RSP_ILLEGAL_EMULATOR_MODE" message. jtagmkII_close() has been called with the wrong pgm->cookie. Wrap it inside stk500v2_jtagmkII_close(), adjusting the cookie data appropriately. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@895 81a1dc3b-b13d-400b-aceb-764788c761c2
2010-01-08 16:01:37 +00:00
pgm->close = stk500v2_jtagmkII_close;
Replace the fallback of avr_read_byte() and avr_write_byte() to avr_read_byte_default() and avr_write_byte_default (resp.) by directly calling the latter functions from within all programmers that don't implement their own read_byte()/write_byte() methods. In turn, make the read_byte() and write_byte() methods mandatory, and the cmd() method (direct ISP command) optional instead (it's effectively mandatory for any programmer using avr_read_byte_default()/avr_write_byte_default() though). Remove all the pointless cmd() method stubs from those programmers that don't need it. Eliminate avr_read_byte() as it was now completely identical to pgm->read_byte(). git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@684 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-11-20 15:04:09 +00:00
pgm->read_byte = stk500hvsp_read_byte;
pgm->write_byte = stk500hvsp_write_byte;
Implement and document HVSP and PP modes for the AVR Dragon. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@677 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-11-01 21:47:25 +00:00
/*
* optional functions
*/
pgm->paged_write = stk500hvsp_paged_write;
pgm->paged_load = stk500hvsp_paged_load;
pgm->print_parms = stk500v2_print_parms;
pgm->set_vtarget = stk500v2_set_vtarget;
pgm->set_varef = stk500v2_set_varef;
pgm->set_fosc = stk500v2_set_fosc;
pgm->set_sck_period = stk500v2_set_sck_period_mk2;
Fix a serious memory corruption problem resulting out of the chaining of both, the stk500v2 and the jtagmkII programmers for some programming hardware (JTAG ICE mkII and AVR Dragon running in ISP, HVSP or PP mode), where both programmers have to maintain their private programmer data. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@834 81a1dc3b-b13d-400b-aceb-764788c761c2
2009-07-06 22:10:20 +00:00
pgm->setup = stk500v2_jtagmkII_setup;
pgm->teardown = stk500v2_jtagmkII_teardown;
Implement and document HVSP and PP modes for the AVR Dragon. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@677 81a1dc3b-b13d-400b-aceb-764788c761c2
2006-11-01 21:47:25 +00:00
pgm->page_size = 256;
}
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
Parser does not need to know all programmer types now, new programmers will update only the table in pgm_type.c. * config_gram.y, lexer.l: removed programmer type keywords, use now locate_programmer_type() function * pgm_type.[ch]: added new files for table of programmer types * main.c: allow list of programmer types by -c ?type * avrdude.conf.in: changed all type keywords to quoted strings * doc/avrdude.texi: changed description of type definition, list of valid types is now included from generated file * doc/Makefile.am: generate list of programmer types for doc * all programmers [hc]: add xxx_desc string for description of programmer git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1051 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-01-31 17:03:43 +00:00
const char stk600_desc[] = "Atmel STK600";
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
void stk600_initpgm(PROGRAMMER *pgm) {
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
strcpy(pgm->type, "STK600");
/*
* mandatory functions
*/
pgm->initialize = stk500v2_initialize;
pgm->display = stk500v2_display;
pgm->enable = stk500v2_enable;
pgm->disable = stk500v2_disable;
pgm->program_enable = stk500v2_program_enable;
pgm->chip_erase = stk500v2_chip_erase;
pgm->cmd = stk500v2_cmd;
pgm->open = stk600_open;
pgm->close = stk500v2_close;
* stk500v2.c: Use stk500isp_read_byte/stk500isp_write_byte for every byte-wide access (rather than JTAGICE3 only). This finally obsoletes the use of the prehistoric SPI_MULTI command where AVRDUDE used to assemble all the low-level ISP stuff by itself. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1336 81a1dc3b-b13d-400b-aceb-764788c761c2
2014-10-06 22:15:29 +00:00
pgm->read_byte = stk500isp_read_byte;
pgm->write_byte = stk500isp_write_byte;
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
/*
* optional functions
*/
pgm->paged_write = stk500v2_paged_write;
pgm->paged_load = stk500v2_paged_load;
Xmega page erase implementation for XPROG (AVRISPmkII, STK600) git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1090 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-05-04 15:47:26 +00:00
pgm->page_erase = stk500v2_page_erase;
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
pgm->print_parms = stk500v2_print_parms;
pgm->set_vtarget = stk600_set_vtarget;
pgm->set_varef = stk600_set_varef;
pgm->set_fosc = stk600_set_fosc;
pgm->set_sck_period = stk600_set_sck_period;
pgm->perform_osccal = stk500v2_perform_osccal;
pgm->setup = stk500v2_setup;
pgm->teardown = stk500v2_teardown;
pgm->page_size = 256;
}
Parser does not need to know all programmer types now, new programmers will update only the table in pgm_type.c. * config_gram.y, lexer.l: removed programmer type keywords, use now locate_programmer_type() function * pgm_type.[ch]: added new files for table of programmer types * main.c: allow list of programmer types by -c ?type * avrdude.conf.in: changed all type keywords to quoted strings * doc/avrdude.texi: changed description of type definition, list of valid types is now included from generated file * doc/Makefile.am: generate list of programmer types for doc * all programmers [hc]: add xxx_desc string for description of programmer git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1051 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-01-31 17:03:43 +00:00
const char stk600pp_desc[] = "Atmel STK600 in parallel programming mode";
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
void stk600pp_initpgm(PROGRAMMER *pgm) {
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
strcpy(pgm->type, "STK600PP");
/*
* mandatory functions
*/
pgm->initialize = stk500pp_initialize;
pgm->display = stk500v2_display;
pgm->enable = stk500v2_enable;
pgm->disable = stk500pp_disable;
pgm->program_enable = stk500pp_program_enable;
pgm->chip_erase = stk500pp_chip_erase;
pgm->open = stk600_open;
pgm->close = stk500v2_close;
pgm->read_byte = stk500pp_read_byte;
pgm->write_byte = stk500pp_write_byte;
/*
* optional functions
*/
pgm->paged_write = stk500pp_paged_write;
pgm->paged_load = stk500pp_paged_load;
pgm->print_parms = stk500v2_print_parms;
pgm->set_vtarget = stk600_set_vtarget;
pgm->set_varef = stk600_set_varef;
pgm->set_fosc = stk600_set_fosc;
pgm->set_sck_period = stk600_set_sck_period;
pgm->setup = stk500v2_setup;
pgm->teardown = stk500v2_teardown;
pgm->page_size = 256;
}
Parser does not need to know all programmer types now, new programmers will update only the table in pgm_type.c. * config_gram.y, lexer.l: removed programmer type keywords, use now locate_programmer_type() function * pgm_type.[ch]: added new files for table of programmer types * main.c: allow list of programmer types by -c ?type * avrdude.conf.in: changed all type keywords to quoted strings * doc/avrdude.texi: changed description of type definition, list of valid types is now included from generated file * doc/Makefile.am: generate list of programmer types for doc * all programmers [hc]: add xxx_desc string for description of programmer git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1051 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-01-31 17:03:43 +00:00
const char stk600hvsp_desc[] = "Atmel STK600 in high-voltage serial programming mode";
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
void stk600hvsp_initpgm(PROGRAMMER *pgm) {
Add initial support for the Atmel STK600, for "classic" AVRs (AT90, ATtiny, ATmega) in both, ISP and high-voltage programming modes. * Makefile.am: Add -lm. * avrdude.conf.in: Add stk600, stk600pp, and stk600hvsp. * config_gram.y: Add support for the stk600* keywords. * lexer.l: (Ditto.) * pgm.h: Add the "chan" parameter to set_varef(). * stk500.c: (Ditto.) * serial.h: Add USB endpoint support to struct filedescriptor. * stk500v2.c: Implement the meat of the STK600 support. * stk500v2.h: Add new prototypes for stk600*() programmers. * stk500v2_private.h: Add new constants used in the STK600. * term.c: Add AREF channel support. * usb_libusb.c: Automatically determine the correct write endpoint ID, the STK600 uses 0x83 while all other tools use 0x82. Propagate the EP to use through struct filedescriptor. * usbdevs.h: Add the STK600 USB product ID. * tools/get-stk600-cards.xsl: XSL transformation for targetboards.xml to obtain the list of socket and routing card IDs, to be used in stk500v2.c (for displaying the names). * tools/get-stk600-devices.xsl: XSL transformation for targetboards.xml to obtain the table of socket/routing cards and their respective AVR device support for doc/avrdude.texi. * avrdude.1: Document all the STK600 stuff. * doc/avrdude.texi: Ditto. Added a new chapter for Programmer Specific Information. Thanks to Eirik Rasmussen from Atmel Norway for his support in getting this code running within that short amount of time! git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@768 81a1dc3b-b13d-400b-aceb-764788c761c2
2008-03-14 13:00:08 +00:00
strcpy(pgm->type, "STK600HVSP");
/*
* mandatory functions
*/
pgm->initialize = stk500hvsp_initialize;
pgm->display = stk500v2_display;
pgm->enable = stk500v2_enable;
pgm->disable = stk500hvsp_disable;
pgm->program_enable = stk500hvsp_program_enable;
pgm->chip_erase = stk500hvsp_chip_erase;
pgm->open = stk600_open;
pgm->close = stk500v2_close;
pgm->read_byte = stk500hvsp_read_byte;
pgm->write_byte = stk500hvsp_write_byte;
/*
* optional functions
*/
pgm->paged_write = stk500hvsp_paged_write;
pgm->paged_load = stk500hvsp_paged_load;
pgm->print_parms = stk500v2_print_parms;
pgm->set_vtarget = stk600_set_vtarget;
pgm->set_varef = stk600_set_varef;
pgm->set_fosc = stk600_set_fosc;
pgm->set_sck_period = stk600_set_sck_period;
pgm->setup = stk500v2_setup;
pgm->teardown = stk500v2_teardown;
pgm->page_size = 256;
}
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
const char stk500v2_jtag3_desc[] = "Atmel JTAGICE3 in ISP mode";
Use const in PROGRAMMER function arguments where appropriate In order to get meaningful const properties for the PROGRAMMER, AVRPART and AVRMEM arguments, some code needed to be moved around, otherwise a network of "tainted" assignments risked rendering nothing const: - Change void (*enable)(PROGRAMMER *pgm) to void (*enable)(PROGRAMMER *pgm, const AVRPART *p); this allows changes in the PROGRAMMER structure after the part is known. For example, use TPI, UPDI, PDI functions in that programmer appropriate to the part. This used to be done later in the process, eg, in the initialize() function, which "taints" all other programmer functions wrt const and sometimes requires other finessing with flags etc. Much clearer with the modified enable() interface. - Move TPI initpgm-type code from initialize() to enable() --- note that initpgm() does not have the info at the time when it is called whether or not TPI is required - buspirate.c: move pgm->flag to PDATA(pgm)->flag (so legitimate modification of the flag does not change PROGRAMMER structure) - Move AVRPART_INIT_SMC and AVRPART_WRITE bits from the flags field in AVRPART to jtagmkII.c's private data flags32 fiels as FLAGS32_INIT_SMC and FLAGS32_WRITE bits - Move the xbeeResetPin component to private data in stk500.c as this is needed by xbee when it saddles on the stk500 code (previously, the flags component of the part was re-dedicated to this) - Change the way the "chained" private data are used in jtag3.c whilst keeping the PROGRAMMER structure read-only otherwise - In stk500v2.c move the STK600 pgm update from stk500v2_initialize() to stk500v2_enable() so the former keeps the PROGRAMMER structure read-only (for const assertion). - In usbasp change the code from changing PROGRAMMER functions late to dispatching to TPI or regular SPI protocol functions at runtime; reason being the decision whether to use TPI protocol is done at run-time depending on the capability of the attached programmer Also fixes Issue #1071, the treatment of default eecr value.
2022-08-17 15:05:28 +00:00
void stk500v2_jtag3_initpgm(PROGRAMMER *pgm) {
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
strcpy(pgm->type, "JTAG3_ISP");
/*
* mandatory functions
*/
pgm->initialize = stk500v2_jtag3_initialize;
pgm->display = stk500v2_display;
pgm->enable = stk500v2_enable;
pgm->disable = stk500v2_jtag3_disable;
pgm->program_enable = stk500v2_program_enable;
pgm->chip_erase = stk500v2_chip_erase;
pgm->cmd = stk500v2_jtag3_cmd;
pgm->open = stk500v2_jtag3_open;
pgm->close = stk500v2_jtag3_close;
pgm->read_byte = stk500isp_read_byte;
pgm->write_byte = stk500isp_write_byte;
/*
* optional functions
*/
pgm->paged_write = stk500v2_paged_write;
pgm->paged_load = stk500v2_paged_load;
pgm->page_erase = stk500v2_page_erase;
pgm->print_parms = stk500v2_print_parms;
pgm->set_sck_period = stk500v2_jtag3_set_sck_period;
pgm->perform_osccal = stk500v2_perform_osccal;
pgm->setup = stk500v2_jtag3_setup;
pgm->teardown = stk500v2_jtag3_teardown;
pgm->page_size = 256;
Initial support for Power Debugger analog reading Voltage and current though channel A and B
2022-09-28 19:59:50 +00:00
if (strcmp(ldata(lfirst(pgm->id)), "powerdebugger_isp") == 0)
pgm->set_vtarget = jtag3_set_vtarget;
* pgm_type.c: Add "jtagice3_isp" programmer hook * avrdude.conf.in: Add "jtag3isp" programmer * jtag3.c: jtag3_setparm() is now public * jtag3.h: (Dito) * stk500v2_private.h: Command 0x1D is CMD_SPI_MULTI only for STK500v2, AVRISPmkII, and JTAGICEmkII; for JTAGICE3, it's CMD_SET_SCK now; also add CMD_GET_SCK * avrpart.c (avr_get_output_index): New function * avrpart.h: (Dito) * stk500v2.c: Implement the pasthrough programmer glue logic for JTAGICE3 in ISP mode * stk500v2.h: (Dito) * avrdude.1: Document the JTAGICE3 support. git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1119 81a1dc3b-b13d-400b-aceb-764788c761c2
2012-12-03 15:52:38 +00:00
}