[PATCH v2] x86_64: Optimize ffsll function code size.

[Sunil Pandey]

On Wed, Jan 10, 2024 at 11:19 AM Carlos O'Donell <carlos@redhat.com> wrote:

> On 7/31/23 14:35, Sunil K Pandey via Libc-alpha wrote:
> > Ffsll function size is 17 byte, this patch optimizes size to 16 byte.
> > Currently ffsll function randomly regress by ~20%, depending on how
> > code get aligned.
> >
> > This patch fixes ffsll function random performance regression.
>
> Here is my suggestion.
>
> Commit this to glibc 2.39 as an incremental improvement that we can
> backport
> to any active branch.
>
> I have already approved the removal for glibc 2.40 via Adhemerval's patch.
>
> Reviewed-by: Carlos O'Donell <carlos@redhat.com>
>
> > Changes from v1:
> > - Further reduce size ffsll function size to 12 bytes.
> > ---
> >  sysdeps/x86_64/ffsll.c | 10 +++++-----
> >  1 file changed, 5 insertions(+), 5 deletions(-)
> >
> > diff --git a/sysdeps/x86_64/ffsll.c b/sysdeps/x86_64/ffsll.c
> > index a1c13d4906..6a5803c7c1 100644
> > --- a/sysdeps/x86_64/ffsll.c
> > +++ b/sysdeps/x86_64/ffsll.c
> > @@ -26,13 +26,13 @@ int
> >  ffsll (long long int x)
> >  {
> >    long long int cnt;
> > -  long long int tmp;
> >
> > -  asm ("bsfq %2,%0\n"                /* Count low bits in X and store
> in %1.  */
> > -       "cmoveq %1,%0\n"              /* If number was zero, use -1 as
> result.  */
> > -       : "=&r" (cnt), "=r" (tmp) : "rm" (x), "1" (-1));
> > +  asm ("mov $-1,%k0\n"       /* Intialize CNT to -1.  */
> > +       "bsf %1,%0\n" /* Count low bits in X and store in CNT.  */
> > +       "inc %k0\n"   /* Increment CNT by 1.  */
> > +       : "=&r" (cnt) : "r" (x));
> >
> > -  return cnt + 1;
> > +  return cnt;
> >  }
> >
> >  #ifndef __ILP32__
>
> --
> Cheers,
> Carlos.
>

I would like to backport this patch to release branches from 2.28 to 2.38.
Any comments or objections?

--Sunil
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://sourceware.org/pipermail/libc-alpha/attachments/20240124/bb348a21/attachment.htm>