Re: [avr-chat] Can avrdude-5.3.1 handle stk500v1 and ATtiny15?

[Erik Christiansen]

On Wed, Mar 07, 2007 at 07:51:32AM +0100, Joerg Wunsch wrote:
> Of course, I'm particularly interested whether the segfault still
> happens.

Yes, it's still there. I'll give you a look over my shoulder, with the
gdb run at the end, after confirming the build. Working back from the
outcome:

$ avrdude -c stk500v1 -p t15 -e -U flash:w:obj/flash.srec -F

avrdude: please define PAGEL and BS2 signals in the configuration file for part 
ATtiny15
avrdude: AVR device initialized and ready to accept instructions

Reading | ################################################## | 100% 0.01s

avrdude: Device signature = 0x1e9006
avrdude: erasing chip
avrdude: please define PAGEL and BS2 signals in the configuration file for part 
ATtiny15
avrdude: reading input file "obj/flash.srec"
avrdude: input file obj/flash.srec auto detected as Motorola S-Record
avrdude: writing flash (126 bytes):

Writing |                                                    | 0% 
0.00sSegmentation fault

That's not what's anticipated. Let's check that we actually have a new avrdude:

$ which avrdude
avrdude is hashed (/usr/local/bin/avrdude)
$ ll /usr/local/bin/avrdude
-rwxr-xr-x  1 root root 480608 2007-03-07 21:13 /usr/local/bin/avrdude*

For the sake of thoroughness, /usr/bin/install did that, just a moment ago:

$ sudo make install
Password:
make[1]: Entering directory `/usr/local/src/avr/avrdude-5.3.1'
make[2]: Entering directory `/usr/local/src/avr/avrdude-5.3.1'
test -z "/usr/local/bin" || mkdir -p -- "/usr/local/bin"
  /usr/bin/install -c 'avrdude' '/usr/local/bin/avrdude'
Backing up avrdude.conf in /usr/local/etc
test -z "/usr/local/etc" || mkdir -p -- "/usr/local/etc"
 /usr/bin/install -c -m 644 'avrdude.conf' '/usr/local/etc/avrdude.conf'
test -z "/usr/local/man/man1" || mkdir -p -- "/usr/local/man/man1"
 /usr/bin/install -c -m 644 './avrdude.1' '/usr/local/man/man1/avrdude.1'
make[2]: Leaving directory `/usr/local/src/avr/avrdude-5.3.1'
make[1]: Leaving directory `/usr/local/src/avr/avrdude-5.3.1'

And before that, a new avrdude-stk500.o was compiled and linked:

$ make
make  all-recursive
make[1]: Entering directory `/usr/local/src/avr/avrdude-5.3.1'
make[2]: Entering directory `/usr/local/src/avr/avrdude-5.3.1'
if gcc -DHAVE_CONFIG_H -I. -I. -I.  -DCONFIG_DIR=\"/usr/local/etc\"  -Wall   -g 
-O2 -MT avrdude-stk500.o -MD -MP -MF ".deps/avrdude-stk500.Tpo" -c -o 
avrdude-stk500.o `test -f 'stk500.c' || echo './'`stk500.c; \
then mv -f ".deps/avrdude-stk500.Tpo" ".deps/avrdude-stk500.Po"; else rm -f 
".deps/avrdude-stk500.Tpo"; exit 1; fi
gcc  -g -O2   -o avrdude  avrdude-config_gram.o avrdude-lexer.o avrdude-avr.o 
avrdude-avr910.o avrdude-avrpart.o avrdude-bitbang.o avrdude-butterfly.o 
avrdude-config.o avrdude-confwin.o avrdude-crc16.o avrdude-fileio.o 
avrdude-jtagmkI.o avrdude-jtagmkII.o avrdude-lists.o avrdude-main.o 
avrdude-par.o avrdude-pgm.o avrdude-ppi.o avrdude-ppiwin.o avrdude-safemode.o 
avrdude-serbb_posix.o avrdude-serbb_win32.o avrdude-ser_avrdoper.o 
avrdude-ser_posix.o avrdude-ser_win32.o avrdude-stk500.o avrdude-stk500v2.o 
avrdude-stk500generic.o avrdude-term.o avrdude-usbasp.o avrdude-usb_libusb.o  
make[2]: Leaving directory `/usr/local/src/avr/avrdude-5.3.1'
make[1]: Leaving directory `/usr/local/src/avr/avrdude-5.3.1'

Here it is:

-rw-r--r--  1 erik erik 38240 2007-03-07 21:13 
/usr/local/src/avr/avrdude-5.3.1/avrdude-stk500.o

Ah, but did the patch take?:

$ diff -w stk500.c stk500.c.orig
497,499d496
<       n_extparms = 1;         /* EEPROM only */
<     }
< 
510d506
<     if (n_extparms > 1) {
520d515
<     }
527a523
>   }

OK, time to follow your suggestion, and run avrdude under gdb:

(gdb) run  -c stk500v1 -p t15 -e -U flash:w:obj/flash.srec
Starting program: /usr/local/bin/avrdude -c stk500v1 -p t15 -e -U 
flash:w:obj/flash.srec

avrdude: please define PAGEL and BS2 signals in the configuration file for part 
ATtiny15
avrdude: AVR device initialized and ready to accept instructions

Reading | ################################################## | 100% 0.01s

avrdude: Device signature = 0x1e9006
avrdude: erasing chip
avrdude: please define PAGEL and BS2 signals in the configuration file for part 
ATtiny15
avrdude: reading input file "obj/flash.srec"
avrdude: input file obj/flash.srec auto detected as Motorola S-Record
avrdude: writing flash (126 bytes):

Writing |                                                    | 0% 0.00s
Program received signal SIGSEGV, Segmentation fault.
0x08060cf9 in stk500_paged_write (pgm=0x8089b18, p=0x81bb2c0, m=0x81bc638, 
page_size=128, n_bytes=Variable "n_bytes" is not available.
) at stk500.c:749
749         buf[i++] = Sync_CRC_EOP;

The code around line 749 doesn't instantly hit me in the face with a
plausible cause. Gaining familiarity would take longer that I'm afforded
tonight. Hopefully this much is a useful start.

Thanks for the v2 firmware. I'll hold off upgrading while we still need
a v1 testbed. (Or it's concluded that upgrading is the fix. ;-)

Many thanks.
Erik