This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH] Faster strlen


On Tue, Oct 09, 2012 at 08:02:33AM -0700, H.J. Lu wrote:
> On Sun, Oct 7, 2012 at 10:27 AM, OndÅej BÃlka <neleai@seznam.cz> wrote:
> > Hello, I investigated strlen bit more and improved pminub variant.
> >
> > I got upto 10% speedup by unrolling main loop. I did not measured
> > difference when I unrolled loop more.
> >
> > I also benchmarked atom and added variant which is identical to
> > strlen-sse2-pminub except bsf is replaced by table lookup.
> >
> > Last addition is attempt to generate VEX encoded strlen. I need only to
> > pass -mavx flag when compiling strlen_avx.S but do not know how.
> >
> 
> Please rename strlen_atom.S to strlen-no-bsf.S since it
> depends on bit_Slow_BSF, not Atom.

Perhaps we could add bit_Atom or something similar. 
I plan to use other atom specific optimizations like instruction
pairing. Strlen is bad example for this, all that I could get are 
perhaps 2 cycles when zeroing xmm registers.

> 
> Thanks.
> 
> -- 
> H.J.

-- 

The electricity substation in the car park blew up.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]