This is the mail archive of the libc-ports@sources.redhat.com mailing list for the libc-ports project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: backtrace() from a pthread on ARM


On Tue, May 12, 2009 at 11:09 AM, Andreas Schwab <schwab@linux-m68k.org> wrote:
> Matt Fischer <mattfischer84@gmail.com> writes:
>
>> On a lot of the architectures (including m68k), it looks like it does.
>>  Problem is that since ARM has actively moved away from requiring a
>> frame pointer in their ABI, as Daniel mentioned, you can't be sure at
>> the time of the clone() call whether you should be modifying r11 at
>> all.
>
> Since r11 is not used for argument passing and the child part of the
> clone call never returns modifying the register shouldn't be a problem,
> is it?
>
> Andreas.
>
> --
> Andreas Schwab, schwab@linux-m68k.org
> GPG Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
> "And now for something completely different."
>

I thought of that after I sent the email above.  It does seem like
that wouldn't have any negative side effects, at least in my limited
understanding of how all of this machinery works.

At the very least, I'm going to try making that change locally, which
ought to fix the problem we're seeing.  Would mainline be interested
in a patch, or is this just labeled as undefined behavior that
shouldn't be messed with?

--Matt


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]