Mercurial > libavcodec.hg
comparison x86/vp8dsp.asm @ 12194:80b142c2e9f7 libavcodec
Change function prototypes for width=8 inner and mbedge loopfilter functions
so that it does both U and V planes at the same time. This will have speed
advantages when using SSE2 (or higher) optimizations, since we can do both
the U and V rows together in a single xmm register.
This also renames filter16 to filter16y and filter8 to filter8uv so that it's
more obvious what each function is used for.
author | rbultje |
---|---|
date | Mon, 19 Jul 2010 21:18:04 +0000 |
parents | b24153464669 |
children | e7847fcff0f4 |
comparison
equal
deleted
inserted
replaced
12193:0a63bed2a00e | 12194:80b142c2e9f7 |
---|---|
1377 ; void vp8_h/v_loop_filter<size>_inner_<opt>(uint8_t *dst, int stride, | 1377 ; void vp8_h/v_loop_filter<size>_inner_<opt>(uint8_t *dst, int stride, |
1378 ; int flimE, int flimI, int hev_thr); | 1378 ; int flimE, int flimI, int hev_thr); |
1379 ;----------------------------------------------------------------------------- | 1379 ;----------------------------------------------------------------------------- |
1380 | 1380 |
1381 %macro INNER_LOOPFILTER 4 | 1381 %macro INNER_LOOPFILTER 4 |
1382 cglobal vp8_%2_loop_filter16_inner_%1, 5, %3, %4 | 1382 cglobal vp8_%2_loop_filter16y_inner_%1, 5, %3, %4 |
1383 %define dst_reg r0 | 1383 %define dst_reg r0 |
1384 %define mstride_reg r1 | 1384 %define mstride_reg r1 |
1385 %define E_reg r2 | 1385 %define E_reg r2 |
1386 %define I_reg r3 | 1386 %define I_reg r3 |
1387 %define hev_thr_reg r4 | 1387 %define hev_thr_reg r4 |