annotate sh4/Makefile @ 12510:ef2f2db5b7be libavcodec

Unroll loop in h264_idct_add8_sse2(). This means we can inline scan8[] in the code directly also and remove loop setup. 20% faster in function, 0.8% overall. See "[PATCH] unroll loop in h264_idct_add8_sse2()" thread on ML.
author rbultje
date Fri, 24 Sep 2010 14:05:45 +0000
parents cd71b238e78b
children
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
rev   line source
11501
cd71b238e78b Simplify arch-specific object file lists
mru
parents: 11500
diff changeset
1 OBJS += sh4/dsputil_align.o \
11500
8ae93fdefcd7 Move arch-specific makefile parts into $arch/Makefile
mru
parents:
diff changeset
2 sh4/dsputil_sh4.o \
8ae93fdefcd7 Move arch-specific makefile parts into $arch/Makefile
mru
parents:
diff changeset
3 sh4/idct_sh4.o \