projects
/
platform
/
upstream
/
libav.git
/ commitdiff
commit
grep
author
committer
pickaxe
?
search:
re
summary
|
shortlog
|
log
|
commit
| commitdiff |
tree
raw
|
patch
| inline |
side by side
(parent:
8b03c01
)
100l, inv_zigzag_direct16 must be aligned to 16 bytes for dct_quantize_SSE2
author
Reimar Döffinger
<Reimar.Doeffinger@gmx.de>
Wed, 26 Aug 2009 09:10:51 +0000
(09:10 +0000)
committer
Reimar Döffinger
<Reimar.Doeffinger@gmx.de>
Wed, 26 Aug 2009 09:10:51 +0000
(09:10 +0000)
(which uses movdqa to xmm1 on inv_zigzag_direct16).
Originally committed as revision 19711 to svn://svn.ffmpeg.org/ffmpeg/trunk
libavcodec/dsputil.c
patch
|
blob
|
history
diff --git
a/libavcodec/dsputil.c
b/libavcodec/dsputil.c
index
2d5a9a2
..
f2b2fc8
100644
(file)
--- a/
libavcodec/dsputil.c
+++ b/
libavcodec/dsputil.c
@@
-86,7
+86,7
@@
const uint8_t ff_zigzag248_direct[64] = {
};
/* not permutated inverse zigzag_direct + 1 for MMX quantizer */
-DECLARE_ALIGNED_
8
(uint16_t, inv_zigzag_direct16[64]);
+DECLARE_ALIGNED_
16
(uint16_t, inv_zigzag_direct16[64]);
const uint8_t ff_alternate_horizontal_scan[64] = {
0, 1, 2, 3, 8, 9, 16, 17,