Ok, I'll check it later...
looks like something odd in bfd...
~d

On Friday 11 October 2002 19:21, Galen Seitz wrote:
> > This is cause gdb puts software breakpoints as follows:
> > 1.Fill up the memory with some predefined pattern (0x0000) od word length
> > 2. Advance a counter by word length to jump over the filled bytes and
> > remembers a counter value.
> > 3. Save replaced code.
> >
> > When it hits a zero word , the simulator issues a signal - "interrupted
> > with a reason breakpoint". When this signal issued, the real counter (PC)
> > will be pointing the end of zeros. Then if finds a real code at saved
> > address and restores memory. Stop execution and revert a counter backward
> > by 2.
>
> That is how I would expect it to work.  However, my problem is that
> the breakpoint is never hit, apparently due to the breakpoint being
> set on the second word of a two word instruction.
>
> Here is another example.  I set breakpoints on 'main' and 'dummy', but
> when I run, it hits the 'dummy' breakpoint.
>
>       .global main
> main:
>       mov     #0x1234,r4
>       add     #1,r4
> dummy:
>       add     #2,r4
>       jmp     main
>
>
> bash$ msp430-gdb bkpt_test
> GNU gdb 5.1.1
> Copyright 2002 Free Software Foundation, Inc.
> GDB is free software, covered by the GNU General Public License, and you
> are welcome to change it and/or distribute copies of it under certain
> conditions. Type "show copying" to see the conditions.
> There is absolutely no warranty for GDB.  Type "show warranty" for details.
> This GDB was configured as "--host=i686-pc-linux-gnu --target=msp430"...
> (gdb) tar sim
> Connected to the simulator.
> (gdb) load bkpt_test
> (gdb) b main
> Breakpoint 1 at 0xfc42: file bkpt_test.S, line 4.
> (gdb) b dummy
> Breakpoint 2 at 0xfc46: file bkpt_test.S, line 7.
> (gdb) r
> Starting program: /export/home/galens/jobs/headon/sw/gdb_test/bkpt_test
>
> Breakpoint 2, dummy () at bkpt_test.S:7
> 7               add     #2,r4
> Current language:  auto; currently asm
> (gdb)
>
>
> thanks,
> galen
>
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Mspgcc-users mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/mspgcc-users



Reply via email to