[PATCH 0/2] Fix powl on x86_64 for bz#33393

[Siddhesh Poyarekar]

On 2025-10-21 08:38, Paul Zimmermann wrote:
>         Hi Siddhesh,
> 
> looks good to me.
> 
> Tested on x86-64.
> 
> I confirm the test case from #33411 no longer fails with testrun.sh ./a.out.
> 
> Reviewed by: Paul Zimmermann <Paul.Zimmermann@inria.fr>

Thank you!

> Paul
> 
> PS: is that enough, or should I answer to the 2 patches?

This is sufficient for me to commit, but please respond to the 
individual patches so that there's a record of that on patchwork.  I 
don't think Patchwork actually tracks conversations (in a way that's 
easily visible anyway) on the cover letter.

Thanks,
Sid

>> From: Siddhesh Poyarekar <siddhesh@sourceware.org>
>> Cc: josmyers@redhat.com,
>> 	Paul.Zimmermann@inria.fr
>> Date: Mon, 20 Oct 2025 23:02:12 -0400
>>
>> Take two of the patch, which is now a patchset (albeit just two
>> patches).  This time there are no regressions in the testsuite and
>> there's also a test added to the existing test-powl after porting it to
>> libsupport.
>>
>> Siddhesh Poyarekar (2):
>>    test-powl: Port to libsupport
>>    Simplify powl computation for small integral y [BZ #33411]
>>
>>   math/test-powl.c            | 44 ++++++++++++++++++++------
>>   sysdeps/x86_64/fpu/e_powl.S | 61 +++++++++++++++++++++----------------
>>   2 files changed, 68 insertions(+), 37 deletions(-)
>>
>> -- 
>> 2.51.0
>>
>>
>