avrdude/stk500v2.c

1100 lines
28 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>
* Copyright (C) 2006 David Moore
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
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*/
/* $Id$ */
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
*
* 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"
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <errno.h>
#include <unistd.h>
#include <sys/time.h>
#include <time.h>
#include "avr.h"
#include "pgm.h"
#include "stk500_private.h" // temp until all code converted
#include "stk500v2_private.h"
#include "serial.h"
#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
#define STK500V2_XTAL 7372800U
#if 0
#define DEBUG(format,args...) fprintf(stderr, format, ## args)
#else
#define DEBUG(format,args...)
#endif
#if 0
#define DEBUGRECV(format,args...) fprintf(stderr, format, ## args)
#else
#define DEBUGRECV(format,args...)
#endif
extern int verbose;
extern char * progname;
extern int do_cycles;
static unsigned char command_sequence = 1;
static int is_mk2; /* Is the device an AVRISP mkII? */
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
static int stk500v2_getparm(PROGRAMMER * pgm, unsigned char parm, unsigned char * value);
static int stk500v2_setparm(PROGRAMMER * pgm, unsigned char parm, unsigned char value);
static void stk500v2_print_parms1(PROGRAMMER * pgm, char * p);
static int stk500v2_is_page_empty(unsigned int address, int page_size,
const unsigned char *buf);
static int stk500v2_set_sck_period_mk2(PROGRAMMER * pgm, double v);
static int stk500v2_send_mk2(PROGRAMMER * pgm, unsigned char * data, size_t len)
{
if (serial_send(pgm->fd, data, len) != 0) {
fprintf(stderr,"%s: stk500_send_mk2(): failed to send command to serial port\n",progname);
exit(1);
}
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
static int stk500v2_send(PROGRAMMER * pgm, unsigned char * data, size_t len)
{
unsigned char buf[275 + 6]; // max MESSAGE_BODY of 275 bytes, 6 bytes overhead
int i;
if (is_mk2)
return stk500v2_send_mk2(pgm, data, 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
buf[0] = MESSAGE_START;
buf[1] = command_sequence;
buf[2] = len / 256;
buf[3] = len % 256;
buf[4] = TOKEN;
memcpy(buf+5, data, len);
// calculate the XOR checksum
buf[5+len] = 0;
for (i=0;i<5+len;i++)
buf[5+len] ^= buf[i];
DEBUG("STK500V2: stk500v2_send(");
for (i=0;i<len+6;i++) DEBUG("0x%02x ",buf[i]);
DEBUG(", %d)\n",len+6);
if (serial_send(pgm->fd, buf, len+6) != 0) {
fprintf(stderr,"%s: stk500_send(): failed to send command to serial port\n",progname);
exit(1);
}
return 0;
}
static int stk500v2_drain(PROGRAMMER * pgm, int display)
{
return serial_drain(pgm->fd, display);
}
static int stk500v2_recv_mk2(PROGRAMMER * pgm, unsigned char msg[],
size_t maxsize)
{
int rv;
rv = serial_recv(pgm->fd, msg, maxsize);
if (rv < 0) {
fprintf(stderr, "%s: stk500v2_recv_mk2: error in USB receive\n", progname);
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
static int stk500v2_recv(PROGRAMMER * pgm, unsigned char msg[], size_t maxsize) {
enum states { sINIT, sSTART, sSEQNUM, sSIZE1, sSIZE2, sTOKEN, sDATA, sCSUM, sDONE } state = sSTART;
int msglen = 0;
int curlen = 0;
int timeout = 0;
unsigned char c, checksum = 0;
long timeoutval = 5; // seconds
struct timeval tv;
double tstart, tnow;
if (is_mk2)
return stk500v2_recv_mk2(pgm, msg, 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
DEBUG("STK500V2: stk500v2_recv(): ");
gettimeofday(&tv, NULL);
tstart = tv.tv_sec;
while ( (state != sDONE ) && (!timeout) ) {
if (serial_recv(pgm->fd, &c, 1) < 0)
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:
DEBUGRECV("hoping for start token...");
if (c == MESSAGE_START) {
DEBUGRECV("got it\n");
checksum = MESSAGE_START;
state = sSEQNUM;
} else
DEBUGRECV("sorry\n");
break;
case sSEQNUM:
DEBUGRECV("hoping for sequence...\n");
if (c == command_sequence) {
DEBUGRECV("got it, incrementing\n");
state = sSIZE1;
command_sequence++;
} else {
DEBUGRECV("sorry\n");
state = sSTART;
}
break;
case sSIZE1:
DEBUGRECV("hoping for size LSB\n");
msglen = c*256;
state = sSIZE2;
break;
case sSIZE2:
DEBUGRECV("hoping for size MSB...");
msglen += c;
DEBUG(" msg is %d bytes\n",msglen);
state = sTOKEN;
break;
case sTOKEN:
if (c == TOKEN) state = sDATA;
else state = sSTART;
break;
case sDATA:
if (curlen < maxsize) {
msg[curlen] = c;
} else {
fprintf(stderr, "%s: stk500v2_recv(): buffer too small, received %d byte into %zd byte buffer\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
progname,curlen,maxsize);
return -2;
}
if ((curlen == 0) && (msg[0] == ANSWER_CKSUM_ERROR)) {
fprintf(stderr, "%s: stk500v2_recv(): previous packet sent with wrong checksum\n",
progname);
return -3;
}
curlen++;
if (curlen == msglen) state = sCSUM;
break;
case sCSUM:
if (checksum == 0) {
state = sDONE;
} else {
state = sSTART;
fprintf(stderr, "%s: stk500v2_recv(): checksum error\n",
progname);
return -4;
}
break;
default:
fprintf(stderr, "%s: stk500v2_recv(): unknown state\n",
progname);
return -5;
} /* switch */
gettimeofday(&tv, NULL);
tnow = tv.tv_sec;
if (tnow-tstart > timeoutval) { // wuff - signed/unsigned/overflow
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
fprintf(stderr, "%s: stk500_2_ReceiveMessage(): timeout\n",
progname);
return -1;
}
} /* while */
DEBUG("\n");
return msglen+6;
}
static int stk500v2_getsync(PROGRAMMER * pgm) {
int tries = 0;
unsigned char buf[1], resp[32];
int status;
DEBUG("STK500V2: stk500v2_getsync()\n");
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) {
if ((resp[0] == CMD_SIGN_ON) && (resp[1] == STATUS_CMD_OK)) {
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!
return 0;
} else {
if (tries > 33) {
fprintf(stderr,
"%s: stk500v2_getsync(): can't communicate with device: resp=0x%02x\n",
progname, resp[0]);
return -6;
} else
goto retry;
}
// or if we got a timeout
} else if (status == -1) {
if (tries > 33) {
fprintf(stderr,"%s: stk500v2_getsync(): timeout communicating with programmer\n",
progname);
return -1;
} else
goto retry;
// or any other error
} else {
if (tries > 33) {
fprintf(stderr,"%s: stk500v2_getsync(): error communicating with programmer: (%d)\n",
progname,status);
} else
goto retry;
}
return 0;
}
static int stk500v2_command(PROGRAMMER * pgm, unsigned char * buf, 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 i;
int tries = 0;
int status;
DEBUG("STK500V2: stk500v2_command(");
for (i=0;i<len;i++) DEBUG("0x%02hhx ",buf[i]);
DEBUG(", %d)\n",len);
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);
// if we got a successful readback, return
if (status > 0) {
DEBUG(" = %d\n",status);
return status;
}
// otherwise try to sync up again
status = stk500v2_getsync(pgm);
if (status != 0) {
if (tries > 33) {
fprintf(stderr,"%s: stk500v2_command(): failed miserably to execute command 0x%02x\n",
progname,buf[0]);
return -1;
} else
goto retry;
}
DEBUG(" = 0\n");
return 0;
}
static int stk500v2_cmd(PROGRAMMER * pgm, unsigned char cmd[4],
unsigned char res[4])
{
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));
if (buf[1] != STATUS_CMD_OK) {
fprintf(stderr, "%s: stk500v2_cmd(): failed to send command\n",
progname);
return -1;
}
res[0] = buf[2];
res[1] = buf[3];
res[2] = buf[4];
res[3] = buf[5];
return 0;
}
/*
* issue the 'chip erase' command to the AVR device
*/
static int stk500v2_chip_erase(PROGRAMMER * pgm, AVRPART * p)
{
int result;
unsigned char buf[16];
if (p->op[AVR_OP_CHIP_ERASE] == NULL) {
fprintf(stderr, "%s: stk500v2_chip_erase: chip erase instruction not defined for part \"%s\"\n",
progname, p->desc);
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 (?)
avr_set_bits(p->op[AVR_OP_CHIP_ERASE], buf+3);
result = stk500v2_command(pgm, buf, 7, sizeof(buf));
usleep(p->chip_erase_delay);
pgm->initialize(pgm, p);
pgm->pgm_led(pgm, OFF);
return result;
}
/*
* issue the 'program enable' command to the AVR device
*/
static int stk500v2_program_enable(PROGRAMMER * pgm, AVRPART * p)
{
unsigned char buf[16];
if (p->op[AVR_OP_PGM_ENABLE] == NULL) {
fprintf(stderr, "%s: stk500v2_program_enable(): program enable instruction not defined for part \"%s\"\n",
progname, p->desc);
return -1;
}
buf[0] = CMD_ENTER_PROGMODE_ISP;
buf[1] = p->timeout;
buf[2] = p->stabdelay;
buf[3] = p->cmdexedelay;
buf[4] = p->synchloops;
buf[5] = p->bytedelay;
buf[6] = p->pollvalue;
buf[7] = p->pollindex;
avr_set_bits(p->op[AVR_OP_PGM_ENABLE], buf+8);
return stk500v2_command(pgm, buf, 12, sizeof(buf));
}
/*
* initialize the AVR device and prepare it to accept commands
*/
static int stk500v2_initialize(PROGRAMMER * pgm, AVRPART * p)
{
return pgm->program_enable(pgm, p);
}
static void stk500v2_disable(PROGRAMMER * pgm)
{
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));
if (buf[1] != STATUS_CMD_OK) {
fprintf(stderr, "%s: stk500v2_disable(): failed to leave programming mode, got 0x%02x\n",
progname,buf[1]);
exit(1);
}
return;
}
static void stk500v2_enable(PROGRAMMER * pgm)
{
return;
}
static int stk500v2_open(PROGRAMMER * pgm, char * port)
{
long baud = 115200;
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)
baud = pgm->baudrate;
#if defined(HAVE_LIBUSB)
/*
* 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) {
serdev = &usb_serdev_frame;
baud = USB_DEVICE_AVRISPMKII;
is_mk2 = 1;
pgm->set_sck_period = stk500v2_set_sck_period_mk2;
}
#endif
strcpy(pgm->port, port);
pgm->fd = serial_open(port, baud);
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);
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;
}
static void stk500v2_close(PROGRAMMER * pgm)
{
DEBUG("STK500V2: stk500v2_close()\n");
serial_close(pgm->fd);
pgm->fd = -1;
}
static int stk500v2_loadaddr(PROGRAMMER * pgm, unsigned int addr)
{
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));
if (buf[1] != STATUS_CMD_OK) {
fprintf(stderr, "%s: stk500v2_loadaddr(): failed to set load address, got 0x%02x\n",
progname,buf[1]);
return -1;
}
return 0;
}
static int stk500v2_paged_write(PROGRAMMER * pgm, AVRPART * p, AVRMEM * m,
int page_size, int n_bytes)
{
int addr, block_size, last_addr;
int a_div=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
unsigned char commandbuf[10];
unsigned char buf[266];
unsigned char cmds[4];
int result;
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
DEBUG("STK500V2: stk500v2_paged_write(..,%s,%d,%d)\n",m->desc,page_size,n_bytes);
if (page_size == 0) page_size = 256;
// determine which command is to be used
if (strcmp(m->desc, "flash") == 0) {
a_div=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
commandbuf[0] = CMD_PROGRAM_FLASH_ISP;
} else if (strcmp(m->desc, "eeprom") == 0) {
commandbuf[0] = CMD_PROGRAM_EEPROM_ISP;
}
commandbuf[4] = m->delay;
if (a_div == 1) {
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) {
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) {
fprintf(stderr, "%s: stk500v2_paged_write: loadpage instruction not defined for part \"%s\"\n",
progname, p->desc);
return -1;
}
avr_set_bits(m->op[AVR_OP_LOADPAGE_LO], cmds);
commandbuf[5] = cmds[0];
if (m->op[AVR_OP_WRITEPAGE] == NULL) {
fprintf(stderr, "%s: stk500v2_paged_write: write page instruction not defined for part \"%s\"\n",
progname, p->desc);
return -1;
}
avr_set_bits(m->op[AVR_OP_WRITEPAGE], cmds);
commandbuf[6] = cmds[0];
// otherwise, we need to load different commands in
}
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
if (wop == NULL) {
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
fprintf(stderr, "%s: stk500v2_paged_write: write instruction not defined for part \"%s\"\n",
progname, p->desc);
return -1;
}
avr_set_bits(wop, cmds);
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
if (rop == NULL) {
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
fprintf(stderr, "%s: stk500v2_paged_write: read instruction not defined for part \"%s\"\n",
progname, p->desc);
return -1;
}
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];
last_addr=-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
for (addr=0; addr < n_bytes; addr += page_size) {
report_progress(addr,n_bytes,NULL);
if ((n_bytes-addr) < page_size)
block_size = n_bytes - addr;
else
block_size = page_size;
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);
if(commandbuf[0] == CMD_PROGRAM_FLASH_ISP){
if (stk500v2_is_page_empty(addr, block_size, m->buf)) {
continue;
}
}
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,commandbuf,sizeof(commandbuf));
buf[1] = block_size >> 8;
buf[2] = block_size & 0xff;
if((last_addr==-1)||(last_addr+block_size != addr)){
stk500v2_loadaddr(pgm, addr/a_div);
}
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));
if (buf[1] != STATUS_CMD_OK) {
fprintf(stderr,"%s: stk500v2_paged_write: write command failed with %d\n",
progname,buf[1]);
return -1;
}
}
return n_bytes;
}
static int stk500v2_is_page_empty(unsigned int address, int page_size,
const unsigned char *buf)
{
int i;
for(i = 0; i < page_size; i++) {
if(buf[address + i] != 0xFF) {
/* Page is not empty. */
return(0);
}
}
/* Page is empty. */
return(1);
}
static int stk500v2_paged_load(PROGRAMMER * pgm, AVRPART * p, AVRMEM * m,
int page_size, int n_bytes)
{
int addr, block_size;
unsigned char commandbuf[4];
unsigned char buf[275]; // max buffer size for stk500v2 at this point
unsigned char cmds[4];
int result;
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
DEBUG("STK500V2: stk500v2_paged_load(..,%s,%d,%d)\n",m->desc,page_size,n_bytes);
page_size = m->readsize;
rop = m->op[AVR_OP_READ];
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;
rop = m->op[AVR_OP_READ_LO];
}
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
if (rop == NULL) {
fprintf(stderr, "%s: stk500v2_paged_load: read instruction not defined for part \"%s\"\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
progname, p->desc);
return -1;
}
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];
stk500v2_loadaddr(pgm, 0);
for (addr=0; addr < n_bytes; addr += page_size) {
report_progress(addr, n_bytes,NULL);
if ((n_bytes-addr) < page_size)
block_size = n_bytes - addr;
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;
result = stk500v2_command(pgm,buf,4,sizeof(buf));
if (buf[1] != STATUS_CMD_OK) {
fprintf(stderr,"%s: stk500v2_paged_load: read command failed with %d\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
progname,buf[1]);
return -1;
}
#if 0
for (i=0;i<page_size;i++) {
fprintf(stderr,"%02X",buf[2+i]);
if (i%16 == 15) fprintf(stderr,"\n");
}
#endif
memcpy(&m->buf[addr], &buf[2], block_size);
}
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
}
static int stk500v2_set_vtarget(PROGRAMMER * pgm, double v)
{
unsigned char uaref, utarg;
utarg = (unsigned)((v + 0.049) * 10);
if (stk500v2_getparm(pgm, PARAM_VADJUST, &uaref) != 0) {
fprintf(stderr,
"%s: stk500v2_set_vtarget(): cannot obtain V[aref]\n",
progname);
return -1;
}
if (uaref > utarg) {
fprintf(stderr,
"%s: stk500v2_set_vtarget(): reducing V[aref] from %.1f to %.1f\n",
progname, uaref / 10.0, v);
if (stk500v2_setparm(pgm, PARAM_VADJUST, utarg)
!= 0)
return -1;
}
return stk500v2_setparm(pgm, PARAM_VTARGET, utarg);
}
static int stk500v2_set_varef(PROGRAMMER * pgm, double v)
{
unsigned char uaref, utarg;
uaref = (unsigned)((v + 0.049) * 10);
if (stk500v2_getparm(pgm, PARAM_VTARGET, &utarg) != 0) {
fprintf(stderr,
"%s: stk500v2_set_varef(): cannot obtain V[target]\n",
progname);
return -1;
}
if (uaref > utarg) {
fprintf(stderr,
"%s: stk500v2_set_varef(): V[aref] must not be greater than "
"V[target] = %.1f\n",
progname, utarg / 10.0);
return -1;
}
return stk500v2_setparm(pgm, PARAM_VADJUST, uaref);
}
static int stk500v2_set_fosc(PROGRAMMER * pgm, double v)
{
int fosc;
unsigned char prescale, cmatch;
static unsigned ps[] = {
1, 8, 32, 64, 128, 256, 1024
};
int idx, rc;
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";
fprintf(stderr,
"%s: stk500v2_set_fosc(): f = %.3f %s too high, using %.3f MHz\n",
progname, v, unit, STK500V2_XTAL / 2e6);
fosc = STK500V2_XTAL / 2;
} else
fosc = (unsigned)v;
for (idx = 0; idx < sizeof(ps) / sizeof(ps[0]); idx++) {
if (fosc >= STK500V2_XTAL / (256 * ps[idx] * 2)) {
/* 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])) {
fprintf(stderr, "%s: stk500v2_set_fosc(): f = %u Hz too low, %u Hz min\n",
progname, fosc, STK500V2_XTAL / (256 * 1024 * 2));
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;
}
/* The list of SCK frequencies supported by the AVRISP mkII, as listed
* in AVR069 */
double avrispmkIIfreqs[] = {
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
};
static int stk500v2_set_sck_period_mk2(PROGRAMMER * pgm, double v)
{
int i;
for (i = 0; i < sizeof(avrispmkIIfreqs); i++) {
if (1 / avrispmkIIfreqs[i] >= v)
break;
}
fprintf(stderr, "Using p = %.2f us for SCK (param = %d)\n",
1000000 / avrispmkIIfreqs[i], i);
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
/* This code assumes that each count of the SCK duration parameter
represents 8/f, where f is the clock frequency of the STK500V2 master
processors (not the target). This number comes from Atmel
application note AVR061. It appears that the STK500V2 bit bangs SCK.
For small duration values, the actual SCK width is larger than
expected. As the duration value increases, the SCK width error
diminishes. */
static int stk500v2_set_sck_period(PROGRAMMER * pgm, double v)
{
unsigned char dur;
double min, max;
min = 8.0 / STK500V2_XTAL;
max = 255 * min;
dur = v / min + 0.5;
if (v < min) {
dur = 1;
fprintf(stderr,
"%s: stk500v2_set_sck_period(): p = %.1f us too small, using %.1f us\n",
progname, v / 1e-6, dur * min / 1e-6);
} else if (v > max) {
dur = 255;
fprintf(stderr,
"%s: stk500v2_set_sck_period(): p = %.1f us too large, using %.1f us\n",
progname, v / 1e-6, dur * min / 1e-6);
}
return stk500v2_setparm(pgm, PARAM_SCK_DURATION, dur);
}
static int stk500v2_getparm(PROGRAMMER * pgm, unsigned char parm, unsigned char * value)
{
unsigned char buf[32];
buf[0] = CMD_GET_PARAMETER;
buf[1] = parm;
if (stk500v2_command(pgm, buf, 2, sizeof(buf)) < 0) {
fprintf(stderr,"%s: stk500v2_getparm(): failed to get parameter 0x%02x\n",
progname, parm);
return -1;
}
*value = buf[2];
return 0;
}
static int stk500v2_setparm_real(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) {
fprintf(stderr, "\n%s: stk500v2_setparm(): failed to set parameter 0x%02x\n",
progname, parm);
return -1;
}
return 0;
}
static int stk500v2_setparm(PROGRAMMER * pgm, unsigned char parm, unsigned char value)
{
unsigned char current_value;
int res;
res = stk500v2_getparm(pgm, parm, &current_value);
if (res < 0)
fprintf(stderr, "%s: Unable to get parameter 0x%02x\n", progname, parm);
// don't issue a write if the correct value is already set.
if (value == current_value) {
fprintf(stderr, "%s: Skipping paramter write; parameter value already set.\n", progname);
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
static void stk500v2_display(PROGRAMMER * pgm, char * p)
{
unsigned char maj, min, hdw, topcard;
const char *topcard_name;
stk500v2_getparm(pgm, PARAM_HW_VER, &hdw);
stk500v2_getparm(pgm, PARAM_SW_MAJOR, &maj);
stk500v2_getparm(pgm, PARAM_SW_MINOR, &min);
stk500v2_getparm(pgm, PARAM_TOPCARD_DETECT, &topcard);
fprintf(stderr, "%sHardware Version: %d\n", p, hdw);
fprintf(stderr, "%sFirmware Version: %d.%d\n", p, maj, min);
if (1) { // should check to see if it's a stk500 first
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;
}
fprintf(stderr, "%sTopcard : %s\n", p, topcard_name);
}
stk500v2_print_parms1(pgm, p);
return;
}
static void stk500v2_print_parms1(PROGRAMMER * pgm, char * p)
{
unsigned char vtarget, vadjust, osc_pscale, osc_cmatch, sck_duration;
stk500v2_getparm(pgm, PARAM_VTARGET, &vtarget);
stk500v2_getparm(pgm, PARAM_VADJUST, &vadjust);
stk500v2_getparm(pgm, PARAM_OSC_PSCALE, &osc_pscale);
stk500v2_getparm(pgm, PARAM_OSC_CMATCH, &osc_cmatch);
stk500v2_getparm(pgm, PARAM_SCK_DURATION, &sck_duration);
fprintf(stderr, "%sVtarget : %.1f V\n", p, vtarget / 10.0);
fprintf(stderr, "%sVaref : %.1f V\n", p, vadjust / 10.0);
fprintf(stderr, "%sOscillator : ", p);
if (osc_pscale == 0)
fprintf(stderr, "Off\n");
else {
int prescale = 1;
double f = STK500V2_XTAL / 2;
const char *unit;
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);
if (f > 1e6) {
f /= 1e6;
unit = "MHz";
} else if (f > 1e3) {
f /= 1000;
unit = "kHz";
} else
unit = "Hz";
fprintf(stderr, "%.3f %s\n", f, unit);
}
if (is_mk2)
fprintf(stderr, "%sSCK period : %.2f us\n", p,
(float) 1000000 / avrispmkIIfreqs[sck_duration]);
else
fprintf(stderr, "%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);
return;
}
static void stk500v2_print_parms(PROGRAMMER * pgm)
{
stk500v2_print_parms1(pgm, "");
}
void stk500v2_initpgm(PROGRAMMER * pgm)
{
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;
/*
* optional functions
*/
pgm->paged_write = stk500v2_paged_write;
pgm->paged_load = stk500v2_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;
pgm->page_size = 256;
}