[FFmpeg-cvslog] avutil/mem: Optimize fill32() by unrolling and using 64bit
Michael Niedermayer
git at videolan.org
Thu Mar 28 18:33:42 EET 2019
ffmpeg | branch: release/3.4 | Michael Niedermayer <michael at niedermayer.cc> | Thu Jan 17 22:35:10 2019 +0100| [9e5cb0df494b8ca352187fbb44f76d75499ddafb] | committer: Michael Niedermayer
avutil/mem: Optimize fill32() by unrolling and using 64bit
Reviewed-by: Marton Balint <cus at passwd.hu>
Signed-off-by: Michael Niedermayer <michael at niedermayer.cc>
(cherry picked from commit 12b1338be376a3e5fb606d9fe41b58dc4a9e62c7)
Signed-off-by: Michael Niedermayer <michael at niedermayer.cc>
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=9e5cb0df494b8ca352187fbb44f76d75499ddafb
---
libavutil/mem.c | 12 ++++++++++++
1 file changed, 12 insertions(+)
diff --git a/libavutil/mem.c b/libavutil/mem.c
index 36740f1154..4f7ac75df1 100644
--- a/libavutil/mem.c
+++ b/libavutil/mem.c
@@ -385,6 +385,18 @@ static void fill32(uint8_t *dst, int len)
{
uint32_t v = AV_RN32(dst - 4);
+#if HAVE_FAST_64BIT
+ uint64_t v2= v + ((uint64_t)v<<32);
+ while (len >= 32) {
+ AV_WN64(dst , v2);
+ AV_WN64(dst+ 8, v2);
+ AV_WN64(dst+16, v2);
+ AV_WN64(dst+24, v2);
+ dst += 32;
+ len -= 32;
+ }
+#endif
+
while (len >= 4) {
AV_WN32(dst, v);
dst += 4;
More information about the ffmpeg-cvslog
mailing list