Drop -funroll-loops for csum_partial_64.c
PATCH 1/1 Drop -funroll-loops for csum_partial_64.c
Drop -funroll-loops for csum_partial_64.c
I did some rebenchmarking with modern compilers and dropping -funroll-loops makes the function consistently go faster by a few percent. So drop that flag.
Thanks to Richard Guenther for a hint.
Signed-off-by Andi Kleen <[email protected]>
--- arch/x86/lib/Makefile | 3 --- 1 file changed, 3 deletions(-)
arch/x86/lib/Makefile +0 -3
@@ -17,9 +17,6 @@ ifeq ($(CONFIG_X86_32),y) lib-$(CONFIG_X86_USE_3DNOW) += mmx_32.o else obj-y += io_64.o iomap_copy_64.o - - CFLAGS_csum-partial_64.o := -funroll-loops - lib-y += csum-partial_64.o csum-copy_64.o csum-wrappers_64.o lib-y += thunk_64.o clear_page_64.o copy_page_64.o lib-y += memmove_64.o memset_64.o
H. Peter Anvin
AppliedQuoted 1 line · Andi Kleen
Drop -funroll-loops for
csum_partial_64.c
Applied to tip:x86/core. Thanks.
-hpa