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]

PATCH: optimized libm single precision routines: erfcf, erff, expffor x86_64.

From: Dmitrieva Liubov <liubov dot dmitrieva at gmail dot com>
To: libc-alpha at sourceware dot org
Cc: marius dot cornea at intel dot com, Ulrich Drepper <drepper at gmail dot com>
Date: Mon, 6 Feb 2012 13:28:37 +0400
Subject: PATCH: optimized libm single precision routines: erfcf, erff, expffor x86_64.

Hello All,

Here is a patch with the following optimized libm single precision
routines: erfcf, erff, expf for x86_64.

Our performance tests show (in the table here is a ratio: time of test
with libm from current master branch / time of test with libm from
current master branch after applying the patch.)

 	on Istanbul	on    on Atom   on Neha-lem     on AVX
erfcf  1.53	                1.26        1.81                  1.06
erff    1.52	                1.32        2.02                  1.08
expf   2.33                 1.40        2.57                  6.55



ChangeLog:

2012-02-06  Liubov Dmitrieva  <liubov.dmitrieva@gmail.com>

	* sysdeps/x86_64/fpu/s_erff.S: New file.
	* sysdeps/x86_64/fpu/s_erfcf.S: New file.
	* sysdeps/x86_64/fpu/e_expf.S: New file.


--
Liubov Dmitrieva
Software Engineer
Intel Corporation

Attachment: single_precision_some_optimized_libm_routines.patch
Description: Binary data

Follow-Ups:
- Re: PATCH: optimized libm single precision routines: erfcf, erff,expf for x86_64.
  - From: Ulrich Drepper

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