diff options
author | Noah Goldstein <goldstein.w.n@gmail.com> | 2022-07-13 16:32:59 -0700 |
---|---|---|
committer | Noah Goldstein <goldstein.w.n@gmail.com> | 2022-07-16 03:07:59 -0700 |
commit | ceabdcd130ca7043b0fcf2676183d79431d10493 (patch) | |
tree | f6ced435023b66ac1f3b519e5b26a60ef8f479dd /sysdeps/x86_64/strcmp.S | |
parent | c353689e49e72f3aafa1a9e68d4f7a4f33a79cbe (diff) | |
download | glibc-ceabdcd130ca7043b0fcf2676183d79431d10493.tar glibc-ceabdcd130ca7043b0fcf2676183d79431d10493.tar.gz glibc-ceabdcd130ca7043b0fcf2676183d79431d10493.tar.bz2 glibc-ceabdcd130ca7043b0fcf2676183d79431d10493.zip |
x86: Add support to build strcmp/strlen/strchr with explicit ISA level
1. Add default ISA level selection in non-multiarch/rtld
implementations.
2. Add ISA level build guards to different implementations.
- I.e strcmp-avx2.S which is ISA level 3 will only build if
compiled ISA level <= 3. Otherwise there is no reason to
include it as we will always use one of the ISA level 4
implementations (strcmp-evex.S).
3. Refactor the ifunc selector and ifunc implementation list to use
the ISA level aware wrapper macros that allow functions below the
compiled ISA level (with a guranteed replacement) to be skipped.
Tested with and without multiarch on x86_64 for ISA levels:
{generic, x86-64-v2, x86-64-v3, x86-64-v4}
And m32 with and without multiarch.
Diffstat (limited to 'sysdeps/x86_64/strcmp.S')
-rw-r--r-- | sysdeps/x86_64/strcmp.S | 13 |
1 files changed, 11 insertions, 2 deletions
diff --git a/sysdeps/x86_64/strcmp.S b/sysdeps/x86_64/strcmp.S index 19e54bd3a7..7c3cf87a42 100644 --- a/sysdeps/x86_64/strcmp.S +++ b/sysdeps/x86_64/strcmp.S @@ -1,4 +1,4 @@ -/* Highly optimized version for x86-64. +/* strcmp dispatch for RTLD and non-multiarch build Copyright (C) 1999-2022 Free Software Foundation, Inc. This file is part of the GNU C Library. @@ -18,5 +18,14 @@ /* Symbol = strcmp. */ -#include "multiarch/strcmp-sse2.S" +#define DEFAULT_IMPL_V1 "multiarch/strcmp-sse2.S" +/* strcmp-sse2-unaligned.S is often faster than strcmp-sse42.S and + doesn't have the drawback of using the `pcmpstri` instruction + which can be very slow on some CPUs. */ +#define DEFAULT_IMPL_V2 "multiarch/strcmp-sse2-unaligned.S" +#define DEFAULT_IMPL_V3 "multiarch/strcmp-avx2.S" +#define DEFAULT_IMPL_V4 "multiarch/strcmp-evex.S" + +#include "isa-default-impl.h" + libc_hidden_builtin_def (strcmp) |