]> git.karo-electronics.de Git - karo-tx-linux.git/commit
x86: Add optimized popcnt variants
authorBorislav Petkov <borislav.petkov@amd.com>
Fri, 5 Mar 2010 16:34:46 +0000 (17:34 +0100)
committerH. Peter Anvin <hpa@zytor.com>
Tue, 6 Apr 2010 22:52:11 +0000 (15:52 -0700)
commitd61931d89be506372d01a90d1755f6d0a9fafe2d
tree652c34238edcb6c558163abc3cd9d6ce7c5f91a5
parent1527bc8b928dd1399c3d3467dd47d9ede210978a
x86: Add optimized popcnt variants

Add support for the hardware version of the Hamming weight function,
popcnt, present in CPUs which advertize it under CPUID, Function
0x0000_0001_ECX[23]. On CPUs which don't support it, we fallback to the
default lib/hweight.c sw versions.

A synthetic benchmark comparing popcnt with __sw_hweight64 showed almost
a 3x speedup on a F10h machine.

Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
LKML-Reference: <20100318112015.GC11152@aftab>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
arch/x86/Kconfig
arch/x86/include/asm/alternative.h
arch/x86/include/asm/arch_hweight.h [new file with mode: 0644]
arch/x86/include/asm/bitops.h
include/asm-generic/bitops/arch_hweight.h
lib/Makefile
lib/hweight.c
scripts/Makefile.lib