This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [patch] Fix disp-step-syscall.exp on some i386 targets
On 02/28/2012 04:40 PM, Jan Kratochvil wrote:
> I do not mind much but it makes some assumption if PC did not change it was by
> a syscall without checking it really was a syscall at all. There could be for
> example some "jmp *%ebx" with %ebx == _start and it would be falsely relocated
> by your patch back to its code location, ignoring its intended jump. The
> patch of mine would not relocate it as %orig_eax remained 0.
>
> But any code messing with the entry point address may confuse this
> autodetection anyway so these countercases are more hypothetical.
>
> What do you think about the %orig_eax verification?
It looks reasonable to me then, better than my approach. It would be
better if we can add some comments to explain this fix is a hack and why
we have to do in this way.
--
Yao (éå)