Re: V2 [PATCH] benchtests: Restore the clock_gettime option

[Adhemerval Zanella <adhemerval.zanella@linaro.org>]

On 20/05/2020 15:14, H.J. Lu via Libc-alpha wrote:
> On Wed, May 20, 2020 at 11:05 AM Florian Weimer <fweimer@redhat.com> wrote:
>>
>> * Alexander Monakov via Libc-alpha:
>>
>>> I am well aware. Again: rdtsc does not count CPU cycles on recent
>>> Intel CPUs.
>>
>> H.J. probably has a different view on what those “recent Intel CPUs”
>> are. 8-) I have not reviewed the mechanics of the patch, but if we need
>> this for some CPUs, we should make the change.
>>
> 
> Here the patch with updated commit message:
> 
> commit 7621e38bf3c58b2d0359545f1f2898017fd89d05
> Author: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
> Date:   Tue Jan 29 17:43:45 2019 +0000
> 
>     Add generic hp-timing support
> 
> removed the clock_gettime option.  Restore the clock_gettime option for
> some x86 CPUs on which value from RDTSC may not be incremented at a fixed
> rate.
> 
> OK for master?

What kind of result discrepancies are you seeing using hp-timing.h on x86?
The clock_gettime help in what exactly here (it was not clear from
discussion, neither from patch submission)?  

I am asking because we rely on hp-timing.h to get the loader profiling,
so if this does provide accurate information in some cases it might be
the case to disable on ld.so/libc.so as well.