From 5c74e47cd628a6fcd2e4eba021301e275f8a95a3 Mon Sep 17 00:00:00 2001 From: Ling Ma Date: Thu, 19 Jun 2014 13:00:56 -0700 Subject: Add x86_64 memset optimized for AVX2 In this patch we take advantage of HSW memory bandwidth, manage to reduce miss branch prediction by avoiding using branch instructions and force destination to be aligned with avx & avx2 instruction. The CPU2006 403.gcc benchmark indicates this patch improves performance from 26% to 59%. * sysdeps/x86_64/multiarch/Makefile: Add memset-avx2. * sysdeps/x86_64/multiarch/memset-avx2.S: New file. * sysdeps/x86_64/multiarch/memset.S: Likewise. * sysdeps/x86_64/multiarch/memset_chk.S: Likewise. * sysdeps/x86_64/multiarch/rtld-memset.S: Likewise. --- ChangeLog | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'ChangeLog') diff --git a/ChangeLog b/ChangeLog index bc8bc31a22..841d55e340 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,12 @@ +2014-06-19 Ling Ma + H.J. Lu + + * sysdeps/x86_64/multiarch/Makefile: Add memset-avx2. + * sysdeps/x86_64/multiarch/memset-avx2.S: New file. + * sysdeps/x86_64/multiarch/memset.S: Likewise. + * sysdeps/x86_64/multiarch/memset_chk.S: Likewise. + * sysdeps/x86_64/multiarch/rtld-memset.S: Likewise. + 2014-06-19 Andreas Schwab [BZ #17069] -- cgit v1.2.3