diff options
author | Michael Niedermayer <michaelni@gmx.at> | 2011-06-02 05:12:10 +0200 |
---|---|---|
committer | Michael Niedermayer <michaelni@gmx.at> | 2011-06-02 05:12:10 +0200 |
commit | cd8cb54990be4fbd23e28f9d7c2c6170d3667692 (patch) | |
tree | 28965f743e1e731b5a7616cbd7b0551c416da327 /libavcodec/x86/h264_idct.asm | |
parent | 06a9da736554197601e4062298fcf45a5e8d49ff (diff) | |
parent | 8e112df409061034202b98fcc6ce2c1c670c0dda (diff) | |
download | ffmpeg-cd8cb54990be4fbd23e28f9d7c2c6170d3667692.tar.gz |
Merge remote-tracking branch 'qatar/master'
* qatar/master:
ARM: ac3dsp: optimised update_bap_counts()
mpegaudiodec: Fix av_dlog() invocation.
h264/10bit: add HAVE_ALIGNED_STACK checks.
Update 8-bit H.264 IDCT function names to reflect bit-depth.
Add IDCT functions for 10-bit H.264.
mpegaudioenc: Fix broken av_dlog statement.
Employ correct printf format specifiers, mostly in debug output.
ARM: fix MUL64 inline asm for pre-armv6
Conflicts:
libavcodec/mpegaudioenc.c
libavformat/ape.c
libavformat/mxfdec.c
libavformat/r3d.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Diffstat (limited to 'libavcodec/x86/h264_idct.asm')
-rw-r--r-- | libavcodec/x86/h264_idct.asm | 38 |
1 files changed, 19 insertions, 19 deletions
diff --git a/libavcodec/x86/h264_idct.asm b/libavcodec/x86/h264_idct.asm index 4f6f1d7bf8..87fbbf2fd5 100644 --- a/libavcodec/x86/h264_idct.asm +++ b/libavcodec/x86/h264_idct.asm @@ -73,7 +73,7 @@ SECTION .text INIT_MMX ; ff_h264_idct_add_mmx(uint8_t *dst, int16_t *block, int stride) -cglobal h264_idct_add_mmx, 3, 3, 0 +cglobal h264_idct_add_8_mmx, 3, 3, 0 IDCT4_ADD r0, r1, r2 RET @@ -125,7 +125,7 @@ cglobal h264_idct_add_mmx, 3, 3, 0 SUMSUB_BA w, 0, 4 SUMSUB_BA w, 3, 2 SUMSUB_BA w, 1, 5 - SWAP 7, 6, 4, 5, 2, 3, 1, 0 ; 70315246 -> 01234567 + SWAP 7, 6, 4, 5, 2, 3, 1, 0 ; 70315246 -> 01234567 %endmacro %macro IDCT8_1D_FULL 1 @@ -177,7 +177,7 @@ cglobal h264_idct_add_mmx, 3, 3, 0 INIT_MMX ; ff_h264_idct8_add_mmx(uint8_t *dst, int16_t *block, int stride) -cglobal h264_idct8_add_mmx, 3, 4, 0 +cglobal h264_idct8_add_8_mmx, 3, 4, 0 %assign pad 128+4-(stack_offset&7) SUB rsp, pad @@ -237,7 +237,7 @@ cglobal h264_idct8_add_mmx, 3, 4, 0 INIT_XMM ; ff_h264_idct8_add_sse2(uint8_t *dst, int16_t *block, int stride) -cglobal h264_idct8_add_sse2, 3, 4, 10 +cglobal h264_idct8_add_8_sse2, 3, 4, 10 IDCT8_ADD_SSE r0, r1, r2, r3 RET @@ -261,7 +261,7 @@ cglobal h264_idct8_add_sse2, 3, 4, 10 packuswb m1, m1 %endmacro -%macro DC_ADD_MMX2_OP 3-4 +%macro DC_ADD_MMX2_OP 4 %1 m2, [%2 ] %1 m3, [%2+%3 ] %1 m4, [%2+%3*2] @@ -282,13 +282,13 @@ cglobal h264_idct8_add_sse2, 3, 4, 10 INIT_MMX ; ff_h264_idct_dc_add_mmx2(uint8_t *dst, int16_t *block, int stride) -cglobal h264_idct_dc_add_mmx2, 3, 3, 0 +cglobal h264_idct_dc_add_8_mmx2, 3, 3, 0 DC_ADD_MMX2_INIT r1, r2 DC_ADD_MMX2_OP movh, r0, r2, r1 RET ; ff_h264_idct8_dc_add_mmx2(uint8_t *dst, int16_t *block, int stride) -cglobal h264_idct8_dc_add_mmx2, 3, 3, 0 +cglobal h264_idct8_dc_add_8_mmx2, 3, 3, 0 DC_ADD_MMX2_INIT r1, r2 DC_ADD_MMX2_OP mova, r0, r2, r1 lea r0, [r0+r2*4] @@ -297,7 +297,7 @@ cglobal h264_idct8_dc_add_mmx2, 3, 3, 0 ; ff_h264_idct_add16_mmx(uint8_t *dst, const int *block_offset, ; DCTELEM *block, int stride, const uint8_t nnzc[6*8]) -cglobal h264_idct_add16_mmx, 5, 7, 0 +cglobal h264_idct_add16_8_mmx, 5, 7, 0 xor r5, r5 %ifdef PIC lea r11, [scan8_mem] @@ -319,7 +319,7 @@ cglobal h264_idct_add16_mmx, 5, 7, 0 ; ff_h264_idct8_add4_mmx(uint8_t *dst, const int *block_offset, ; DCTELEM *block, int stride, const uint8_t nnzc[6*8]) -cglobal h264_idct8_add4_mmx, 5, 7, 0 +cglobal h264_idct8_add4_8_mmx, 5, 7, 0 %assign pad 128+4-(stack_offset&7) SUB rsp, pad @@ -351,7 +351,7 @@ cglobal h264_idct8_add4_mmx, 5, 7, 0 ; ff_h264_idct_add16_mmx2(uint8_t *dst, const int *block_offset, ; DCTELEM *block, int stride, const uint8_t nnzc[6*8]) -cglobal h264_idct_add16_mmx2, 5, 7, 0 +cglobal h264_idct_add16_8_mmx2, 5, 7, 0 xor r5, r5 %ifdef PIC lea r11, [scan8_mem] @@ -398,7 +398,7 @@ cglobal h264_idct_add16_mmx2, 5, 7, 0 ; ff_h264_idct_add16intra_mmx(uint8_t *dst, const int *block_offset, ; DCTELEM *block, int stride, const uint8_t nnzc[6*8]) -cglobal h264_idct_add16intra_mmx, 5, 7, 0 +cglobal h264_idct_add16intra_8_mmx, 5, 7, 0 xor r5, r5 %ifdef PIC lea r11, [scan8_mem] @@ -421,7 +421,7 @@ cglobal h264_idct_add16intra_mmx, 5, 7, 0 ; ff_h264_idct_add16intra_mmx2(uint8_t *dst, const int *block_offset, ; DCTELEM *block, int stride, const uint8_t nnzc[6*8]) -cglobal h264_idct_add16intra_mmx2, 5, 7, 0 +cglobal h264_idct_add16intra_8_mmx2, 5, 7, 0 xor r5, r5 %ifdef PIC lea r11, [scan8_mem] @@ -466,7 +466,7 @@ cglobal h264_idct_add16intra_mmx2, 5, 7, 0 ; ff_h264_idct8_add4_mmx2(uint8_t *dst, const int *block_offset, ; DCTELEM *block, int stride, const uint8_t nnzc[6*8]) -cglobal h264_idct8_add4_mmx2, 5, 7, 0 +cglobal h264_idct8_add4_8_mmx2, 5, 7, 0 %assign pad 128+4-(stack_offset&7) SUB rsp, pad @@ -529,7 +529,7 @@ cglobal h264_idct8_add4_mmx2, 5, 7, 0 INIT_XMM ; ff_h264_idct8_add4_sse2(uint8_t *dst, const int *block_offset, ; DCTELEM *block, int stride, const uint8_t nnzc[6*8]) -cglobal h264_idct8_add4_sse2, 5, 7, 10 +cglobal h264_idct8_add4_8_sse2, 5, 7, 10 xor r5, r5 %ifdef PIC lea r11, [scan8_mem] @@ -607,7 +607,7 @@ h264_idct_add8_mmx_plane: ; ff_h264_idct_add8_mmx(uint8_t **dest, const int *block_offset, ; DCTELEM *block, int stride, const uint8_t nnzc[6*8]) -cglobal h264_idct_add8_mmx, 5, 7, 0 +cglobal h264_idct_add8_8_mmx, 5, 7, 0 mov r5, 16 add r2, 512 %ifdef PIC @@ -668,7 +668,7 @@ h264_idct_add8_mmx2_plane ; ff_h264_idct_add8_mmx2(uint8_t **dest, const int *block_offset, ; DCTELEM *block, int stride, const uint8_t nnzc[6*8]) -cglobal h264_idct_add8_mmx2, 5, 7, 0 +cglobal h264_idct_add8_8_mmx2, 5, 7, 0 mov r5, 16 add r2, 512 %ifdef ARCH_X86_64 @@ -744,7 +744,7 @@ x264_add8x4_idct_sse2: ; ff_h264_idct_add16_sse2(uint8_t *dst, const int *block_offset, ; DCTELEM *block, int stride, const uint8_t nnzc[6*8]) -cglobal h264_idct_add16_sse2, 5, 5, 8 +cglobal h264_idct_add16_8_sse2, 5, 5, 8 %ifdef ARCH_X86_64 mov r10, r0 %endif @@ -791,7 +791,7 @@ cglobal h264_idct_add16_sse2, 5, 5, 8 ; ff_h264_idct_add16intra_sse2(uint8_t *dst, const int *block_offset, ; DCTELEM *block, int stride, const uint8_t nnzc[6*8]) -cglobal h264_idct_add16intra_sse2, 5, 7, 8 +cglobal h264_idct_add16intra_8_sse2, 5, 7, 8 %ifdef ARCH_X86_64 mov r10, r0 %endif @@ -840,7 +840,7 @@ cglobal h264_idct_add16intra_sse2, 5, 7, 8 ; ff_h264_idct_add8_sse2(uint8_t **dest, const int *block_offset, ; DCTELEM *block, int stride, const uint8_t nnzc[6*8]) -cglobal h264_idct_add8_sse2, 5, 7, 8 +cglobal h264_idct_add8_8_sse2, 5, 7, 8 add r2, 512 %ifdef ARCH_X86_64 mov r10, r0 |