[PATCH v2 11/12] math: Consolidate erf/erfc definitions

[Adhemerval Zanella Netto]

On 29/10/25 17:49, H.J. Lu wrote:
> On Sat, Oct 25, 2025 at 2:27 AM DJ Delorie <dj@redhat.com> wrote:
>>
>>
>> Adhemerval Zanella Netto <adhemerval.zanella@linaro.org> writes:
>>> I did, sorry about that. And it makes sense, update patch below:
>>
>> LGTM
>> Reviewed-by: DJ Delorie <dj@redhat.com>
>>
> 
> I got
> 
> FAIL: math/test-double-erf
> FAIL: math/test-double-erfc
> FAIL: math/test-float32x-erf
> FAIL: math/test-float32x-erfc
> FAIL: math/test-float64-erf
> FAIL: math/test-float64-erfc
> FAIL: math/test-float64x-pow
> FAIL: math/test-ldouble-pow
> 
> with "-O2 -g -march=i686" on i686?
> 

I also noted it after adjusted the code to move some definition to
s_erf_common.c. It is fixed by 5ee722d3ac0da7f7e389af946317272957c7568b [1].

[1] https://sourceware.org/git/?p=glibc.git;a=commit;h=5ee722d3ac0da7f7e389af946317272957c7568b