SSE4.1 implementation of some alpha-processing functions

DispatchAlpha* functions are hard to speed up, compared to SSE2.
ExtractAlpha sees a ~15% speed-up though.

Change-Id: I8715c2defecbc832f469eed7e6ffd012146b52de
This commit is contained in:
Pascal Massimino
2015-06-19 14:23:38 +02:00
committed by James Zern
parent 7f9c98f21d
commit bfc300c7ff
6 changed files with 102 additions and 0 deletions

View File

@ -59,6 +59,7 @@ libwebpdsp_avx2_la_CPPFLAGS = $(libwebpdsp_la_CPPFLAGS)
libwebpdsp_avx2_la_CFLAGS = $(AM_CFLAGS) $(AVX2_FLAGS)
libwebpdspdecode_sse41_la_SOURCES =
libwebpdspdecode_sse41_la_SOURCES += alpha_processing_sse41.c
libwebpdspdecode_sse41_la_SOURCES += dec_sse41.c
libwebpdspdecode_sse41_la_CPPFLAGS = $(libwebpdsp_la_CPPFLAGS)
libwebpdspdecode_sse41_la_CFLAGS = $(AM_CFLAGS) $(SSE41_FLAGS)