Bug#569571: [armel] elf_machine_rel(): misaligned memory access
reassign 569571 gcc-4.4
retitle 569571 [armel] gcc-4.4 generates non-aligned relocations
thanks
On Fri, Feb 12, 2010 at 06:10:31PM +0100, Sascha Silbe wrote:
> On Fri, Feb 12, 2010 at 04:05:50PM +0100, Aurelien Jarno wrote:
>
> [gdb python -m 'import xapian']
>>> Program received signal SIGBUS, Bus error.
>>> elf_machine_rel (scope=<value optimized out>, reloc_mode=<value
>>> optimized out>, consider_profiling=<value optimized out>) at
>>> ../ports/sysdeps/arm/dl-machine.h:429
>>> 429 *reloc_addr += value;
>>
>> The problem is most probably a non word aligned relocation in the
>> program you are trying to run or one of its library. You can list them
>> with objdump -R file.
>
> This one perhaps?
>
> (sugar-jhbuild)sascha.silbe@flatty:~$ objdump -R
> /usr/lib/python2.5/site-packages/_xapian.so |grep '^[0-9a-f]*[13579bdf]
> '
> 000a0377 R_ARM_ABS32 __gxx_personality_v0
> (sugar-jhbuild)sascha.silbe@flatty:~$
>
> I guess it belongs to gcc as there's no match for "personality" inside
> the xapian-bindings sources, but a lot of shared libraries on the system
> contain exactly the same symbol.
> As _xapian.so is the only shared library I could find containing a
> not-word-aligned symbol (except for a few instances of "R_ARM_RELATIVE
> *ABS*") something might have gone wrong during building
> xapian-bindings?
>
I have try to do a rebuild of the package, but the issue is still there.
It most probably belongs to gcc or binutils, I am therefore reassigning
the bug.
--
Aurelien Jarno GPG: 1024D/F1BCDB73
aurelien@aurel32.net http://www.aurel32.net
Reply to: