Commit bc92214e authored by Mans Rullgard's avatar Mans Rullgard

vc1dsp: mark put/avg_vc1_mspel_mc() always_inline

This ensures that these functions are inlined into the per-position
entry points, allowing constant propagation as needed for proper
optimisation.

18% faster VC1 decoding on Cortex-A9.
Signed-off-by: 's avatarMans Rullgard <mans@mansr.com>
parent a8656cd4
...@@ -561,7 +561,7 @@ static av_always_inline int vc1_mspel_filter(const uint8_t *src, int stride, int ...@@ -561,7 +561,7 @@ static av_always_inline int vc1_mspel_filter(const uint8_t *src, int stride, int
/** Function used to do motion compensation with bicubic interpolation /** Function used to do motion compensation with bicubic interpolation
*/ */
#define VC1_MSPEL_MC(OP, OPNAME)\ #define VC1_MSPEL_MC(OP, OPNAME)\
static void OPNAME ## vc1_mspel_mc(uint8_t *dst, const uint8_t *src, int stride, int hmode, int vmode, int rnd)\ static av_always_inline void OPNAME ## vc1_mspel_mc(uint8_t *dst, const uint8_t *src, int stride, int hmode, int vmode, int rnd)\
{\ {\
int i, j;\ int i, j;\
\ \
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment