bug #41561: AVRDUDE 6.0.1/USBasp doesn't write first bytes of

flash page
* usbasp.c (usbasp_spi_paged_write): Remove USBASP_BLOCKFLAG_LAST.
It is no longer needed, as we always write full pages now in paged
write mode.

This prevents each flash page from being written twice, where the
second attempt on older devices (ATmega64/128) signalled an immediate
"ready" condition when polling (likely a silicon bug) even though they
were still in progress.  This caused the USBasp firmware to proceed,
and fill the page buffer anew which eventually caused the data
corruption.



git-svn-id: svn://svn.savannah.nongnu.org/avrdude/trunk/avrdude@1343 81a1dc3b-b13d-400b-aceb-764788c761c2

ChangeLog

@@ -1,3 +1,11 @@
+2014-11-20  Joerg Wunsch <j.gnu@uriah.heep.sax.de>
+
+	bug #41561: AVRDUDE 6.0.1/USBasp doesn't write first bytes of
+	flash page
+	* usbasp.c (usbasp_spi_paged_write): Remove USBASP_BLOCKFLAG_LAST.
+	It is no longer needed, as we always write full pages now in paged
+	write mode.
+
 2014-11-19  Joerg Wunsch <j.gnu@uriah.heep.sax.de>
 
 	bug #43626: Inconsistent timeouts in stk500v2

NEWS

@@ -24,6 +24,8 @@ Current:
     - bug #22248: Read efuse error (partial fix)
     - bug #42267: jtag3isp fails to read lock and fuse bytes directly
                   after changing lock byte
+    - bug #41561: AVRDUDE 6.0.1/USBasp doesn't write first bytes of
+                  flash page
 
   * Internals:
     - ...

usbasp.c

@@ -795,7 +795,6 @@ static int usbasp_spi_paged_write(PROGRAMMER * pgm, AVRPART * p, AVRMEM * m,
 
     if (wbytes <= blocksize) {
       blocksize = wbytes;
-      blockflags |= USBASP_BLOCKFLAG_LAST;
     }
     wbytes -= blocksize;