Re: [PATCH] gdb fix for catch-syscall.exp

[Simon Marchi <simark@simark.ca>]

On 2021-11-23 8:15 p.m., Carl Love wrote:
> Simon:
>
> On Tue, 2021-11-23 at 15:34 -0500, Simon Marchi wrote:
>> If you all agree that it's a bug, I would suggest reverting this
>> patch
>> and making a patch that kfails the test when on powerpc.  And
>> ideally,
>> someone should dig to understand why we don't see the return on
>> powerpc
>> (and fix it), but I'm not here to tell what other people should work
>> on :).
>
> OK, I think for powerpc the test should be a xfail.  Looking at the
> README, kfail is for gdb known issues, xfail is for issues in the
> environment including the OS.

If we can already determine that it's the kernel fault, then xfail is
appropriate.  But I haven't actually dug to find out who is at fault.

> I see xfail takes two arguments, the first one is the gdb bug number.
> So, I will need to file a bug before I finish this patch.

I just filed it here: https://sourceware.org/bugzilla/show_bug.cgi?id=28623

> I am hoping someone has a suggestion on how to improve my patch to add
> the xfail.  The issue is since Powerpc doesn't print the "Catchpoint 1
> (returned from syscall execve)" but rather stops at main, the test
> should not issue the continue command after the xfail.  In the case of
> the xfail, I need to pass 0 backup so the if statement in proc
> test_catch_syscall_execve can decide if the continue command should be
> issued or not.  The change is a bit messy having to pass the return
> value up.  Wondering if anyone has a better idea how to add the xfail?
>
> Note, so far I have only tested this on Powerpc.  Patch is not ready
> for committing.

I try to look at this later.  For now I will push a revert patch, since
we agree on the fact that this is a bug (and the test should not expect
the buggy behavior).

Simon