diff options
Diffstat (limited to 'libavcodec/x86/h264dsp_init.c')
-rw-r--r-- | libavcodec/x86/h264dsp_init.c | 150 |
1 files changed, 115 insertions, 35 deletions
diff --git a/libavcodec/x86/h264dsp_init.c b/libavcodec/x86/h264dsp_init.c index 134d594..08eb7ea 100644 --- a/libavcodec/x86/h264dsp_init.c +++ b/libavcodec/x86/h264dsp_init.c @@ -1,20 +1,20 @@ /* * Copyright (c) 2004-2005 Michael Niedermayer, Loren Merritt * - * This file is part of Libav. + * This file is part of FFmpeg. * - * Libav is free software; you can redistribute it and/or + * FFmpeg is free software; you can redistribute it and/or * modify it under the terms of the GNU Lesser General Public * License as published by the Free Software Foundation; either * version 2.1 of the License, or (at your option) any later version. * - * Libav is distributed in the hope that it will be useful, + * FFmpeg is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU * Lesser General Public License for more details. * * You should have received a copy of the GNU Lesser General Public - * License along with Libav; if not, write to the Free Software + * License along with FFmpeg; if not, write to the Free Software * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA */ @@ -32,8 +32,12 @@ void ff_h264_idct ## NUM ## _add_ ## DEPTH ## _ ## OPT(uint8_t *dst, \ int stride); IDCT_ADD_FUNC(, 8, mmx) +IDCT_ADD_FUNC(, 8, sse2) +IDCT_ADD_FUNC(, 8, avx) IDCT_ADD_FUNC(, 10, sse2) IDCT_ADD_FUNC(_dc, 8, mmxext) +IDCT_ADD_FUNC(_dc, 8, sse2) +IDCT_ADD_FUNC(_dc, 8, avx) IDCT_ADD_FUNC(_dc, 10, mmxext) IDCT_ADD_FUNC(8_dc, 8, mmxext) IDCT_ADD_FUNC(8_dc, 10, sse2) @@ -78,6 +82,11 @@ IDCT_ADD_REP_FUNC2(, 8, 8, sse2) IDCT_ADD_REP_FUNC2(, 8, 10, sse2) IDCT_ADD_REP_FUNC2(, 8, 10, avx) +IDCT_ADD_REP_FUNC2(, 8_422, 8, mmx) + +IDCT_ADD_REP_FUNC2(, 8_422, 10, sse2) +IDCT_ADD_REP_FUNC2(, 8_422, 10, avx) + void ff_h264_luma_dc_dequant_idct_mmx(int16_t *output, int16_t *input, int qmul); void ff_h264_luma_dc_dequant_idct_sse2(int16_t *output, int16_t *input, int qmul); @@ -92,39 +101,48 @@ void ff_h264_loop_filter_strength_mmxext(int16_t bS[2][4][4], uint8_t nnz[40], #define LF_FUNC(DIR, TYPE, DEPTH, OPT) \ void ff_deblock_ ## DIR ## _ ## TYPE ## _ ## DEPTH ## _ ## OPT(uint8_t *pix, \ - int stride, \ + ptrdiff_t stride, \ int alpha, \ int beta, \ int8_t *tc0); #define LF_IFUNC(DIR, TYPE, DEPTH, OPT) \ void ff_deblock_ ## DIR ## _ ## TYPE ## _ ## DEPTH ## _ ## OPT(uint8_t *pix, \ - int stride, \ + ptrdiff_t stride, \ int alpha, \ int beta); #define LF_FUNCS(type, depth) \ -LF_FUNC(h, chroma, depth, mmxext) \ -LF_IFUNC(h, chroma_intra, depth, mmxext) \ -LF_FUNC(v, chroma, depth, mmxext) \ -LF_IFUNC(v, chroma_intra, depth, mmxext) \ -LF_FUNC(h, luma, depth, mmxext) \ -LF_IFUNC(h, luma_intra, depth, mmxext) \ -LF_FUNC(h, luma, depth, sse2) \ -LF_IFUNC(h, luma_intra, depth, sse2) \ -LF_FUNC(v, luma, depth, sse2) \ -LF_IFUNC(v, luma_intra, depth, sse2) \ -LF_FUNC(h, chroma, depth, sse2) \ -LF_IFUNC(h, chroma_intra, depth, sse2) \ -LF_FUNC(v, chroma, depth, sse2) \ -LF_IFUNC(v, chroma_intra, depth, sse2) \ -LF_FUNC(h, luma, depth, avx) \ -LF_IFUNC(h, luma_intra, depth, avx) \ -LF_FUNC(v, luma, depth, avx) \ -LF_IFUNC(v, luma_intra, depth, avx) \ -LF_FUNC(h, chroma, depth, avx) \ -LF_IFUNC(h, chroma_intra, depth, avx) \ -LF_FUNC(v, chroma, depth, avx) \ -LF_IFUNC(v, chroma_intra, depth, avx) +LF_FUNC(h, chroma, depth, mmxext) \ +LF_IFUNC(h, chroma_intra, depth, mmxext) \ +LF_FUNC(h, chroma422, depth, mmxext) \ +LF_IFUNC(h, chroma422_intra, depth, mmxext) \ +LF_FUNC(v, chroma, depth, mmxext) \ +LF_IFUNC(v, chroma_intra, depth, mmxext) \ +LF_FUNC(h, luma, depth, mmxext) \ +LF_IFUNC(h, luma_intra, depth, mmxext) \ +LF_FUNC(h, luma, depth, sse2) \ +LF_IFUNC(h, luma_intra, depth, sse2) \ +LF_FUNC(v, luma, depth, sse2) \ +LF_IFUNC(v, luma_intra, depth, sse2) \ +LF_FUNC(h, chroma, depth, sse2) \ +LF_IFUNC(h, chroma_intra, depth, sse2) \ +LF_FUNC(h, chroma422, depth, sse2) \ +LF_IFUNC(h, chroma422_intra, depth, sse2) \ +LF_FUNC(v, chroma, depth, sse2) \ +LF_IFUNC(v, chroma_intra, depth, sse2) \ +LF_FUNC(h, luma, depth, avx) \ +LF_IFUNC(h, luma_intra, depth, avx) \ +LF_FUNC(v, luma, depth, avx) \ +LF_IFUNC(v, luma_intra, depth, avx) \ +LF_FUNC(h, chroma, depth, avx) \ +LF_IFUNC(h, chroma_intra, depth, avx) \ +LF_FUNC(h, chroma422, depth, avx) \ +LF_IFUNC(h, chroma422_intra, depth, avx) \ +LF_FUNC(v, chroma, depth, avx) \ +LF_IFUNC(v, chroma_intra, depth, avx) + +LF_FUNC(h, luma_mbaff, 8, sse2) +LF_FUNC(h, luma_mbaff, 8, avx) LF_FUNCS(uint8_t, 8) LF_FUNCS(uint16_t, 10) @@ -155,13 +173,13 @@ LF_IFUNC(v, luma_intra, 10, mmxext) /* weighted prediction */ #define H264_WEIGHT(W, OPT) \ -void ff_h264_weight_ ## W ## _ ## OPT(uint8_t *dst, int stride, \ +void ff_h264_weight_ ## W ## _ ## OPT(uint8_t *dst, ptrdiff_t stride, \ int height, int log2_denom, \ int weight, int offset); #define H264_BIWEIGHT(W, OPT) \ void ff_h264_biweight_ ## W ## _ ## OPT(uint8_t *dst, uint8_t *src, \ - int stride, int height, \ + ptrdiff_t stride, int height, \ int log2_denom, int weightd, \ int weights, int offset); @@ -181,7 +199,7 @@ H264_BIWEIGHT_MMX(4) #define H264_WEIGHT_10(W, DEPTH, OPT) \ void ff_h264_weight_ ## W ## _ ## DEPTH ## _ ## OPT(uint8_t *dst, \ - int stride, \ + ptrdiff_t stride, \ int height, \ int log2_denom, \ int weight, \ @@ -190,7 +208,7 @@ void ff_h264_weight_ ## W ## _ ## DEPTH ## _ ## OPT(uint8_t *dst, \ #define H264_BIWEIGHT_10(W, DEPTH, OPT) \ void ff_h264_biweight_ ## W ## _ ## DEPTH ## _ ## OPT(uint8_t *dst, \ uint8_t *src, \ - int stride, \ + ptrdiff_t stride, \ int height, \ int log2_denom, \ int weightd, \ @@ -210,6 +228,7 @@ H264_BIWEIGHT_10_SSE(4, 10) av_cold void ff_h264dsp_init_x86(H264DSPContext *c, const int bit_depth, const int chroma_format_idc) { +#if HAVE_X86ASM int cpu_flags = av_get_cpu_flags(); if (EXTERNAL_MMXEXT(cpu_flags) && chroma_format_idc <= 1) @@ -224,8 +243,11 @@ av_cold void ff_h264dsp_init_x86(H264DSPContext *c, const int bit_depth, c->h264_idct_add16 = ff_h264_idct_add16_8_mmx; c->h264_idct8_add4 = ff_h264_idct8_add4_8_mmx; - if (chroma_format_idc <= 1) + if (chroma_format_idc <= 1) { c->h264_idct_add8 = ff_h264_idct_add8_8_mmx; + } else { + c->h264_idct_add8 = ff_h264_idct_add8_422_8_mmx; + } c->h264_idct_add16intra = ff_h264_idct_add16intra_8_mmx; if (cpu_flags & AV_CPU_FLAG_CMOV) c->h264_luma_dc_dequant_idct = ff_h264_luma_dc_dequant_idct_mmx; @@ -244,6 +266,9 @@ av_cold void ff_h264dsp_init_x86(H264DSPContext *c, const int bit_depth, if (chroma_format_idc <= 1) { c->h264_h_loop_filter_chroma = ff_deblock_h_chroma_8_mmxext; c->h264_h_loop_filter_chroma_intra = ff_deblock_h_chroma_intra_8_mmxext; + } else { + c->h264_h_loop_filter_chroma = ff_deblock_h_chroma422_8_mmxext; + c->h264_h_loop_filter_chroma_intra = ff_deblock_h_chroma422_intra_8_mmxext; } #if ARCH_X86_32 && HAVE_MMXEXT_EXTERNAL c->h264_v_loop_filter_luma = deblock_v_luma_8_mmxext; @@ -279,6 +304,23 @@ av_cold void ff_h264dsp_init_x86(H264DSPContext *c, const int bit_depth, c->h264_h_loop_filter_luma = ff_deblock_h_luma_8_sse2; c->h264_v_loop_filter_luma_intra = ff_deblock_v_luma_intra_8_sse2; c->h264_h_loop_filter_luma_intra = ff_deblock_h_luma_intra_8_sse2; + +#if ARCH_X86_64 + c->h264_h_loop_filter_luma_mbaff = ff_deblock_h_luma_mbaff_8_sse2; +#endif + + c->h264_v_loop_filter_chroma = ff_deblock_v_chroma_8_sse2; + c->h264_v_loop_filter_chroma_intra = ff_deblock_v_chroma_intra_8_sse2; + if (chroma_format_idc <= 1) { + c->h264_h_loop_filter_chroma = ff_deblock_h_chroma_8_sse2; + c->h264_h_loop_filter_chroma_intra = ff_deblock_h_chroma_intra_8_sse2; + } else { + c->h264_h_loop_filter_chroma = ff_deblock_h_chroma422_8_sse2; + c->h264_h_loop_filter_chroma_intra = ff_deblock_h_chroma422_intra_8_sse2; + } + + c->h264_idct_add = ff_h264_idct_add_8_sse2; + c->h264_idct_dc_add = ff_h264_idct_dc_add_8_sse2; } if (EXTERNAL_SSSE3(cpu_flags)) { c->biweight_h264_pixels_tab[0] = ff_h264_biweight_16_ssse3; @@ -289,12 +331,33 @@ av_cold void ff_h264dsp_init_x86(H264DSPContext *c, const int bit_depth, c->h264_h_loop_filter_luma = ff_deblock_h_luma_8_avx; c->h264_v_loop_filter_luma_intra = ff_deblock_v_luma_intra_8_avx; c->h264_h_loop_filter_luma_intra = ff_deblock_h_luma_intra_8_avx; +#if ARCH_X86_64 + c->h264_h_loop_filter_luma_mbaff = ff_deblock_h_luma_mbaff_8_avx; +#endif + + c->h264_v_loop_filter_chroma = ff_deblock_v_chroma_8_avx; + c->h264_v_loop_filter_chroma_intra = ff_deblock_v_chroma_intra_8_avx; + if (chroma_format_idc <= 1) { + c->h264_h_loop_filter_chroma = ff_deblock_h_chroma_8_avx; + c->h264_h_loop_filter_chroma_intra = ff_deblock_h_chroma_intra_8_avx; + } else { + c->h264_h_loop_filter_chroma = ff_deblock_h_chroma422_8_avx; + c->h264_h_loop_filter_chroma_intra = ff_deblock_h_chroma422_intra_8_avx; + } + + c->h264_idct_add = ff_h264_idct_add_8_avx; + c->h264_idct_dc_add = ff_h264_idct_dc_add_8_avx; } } else if (bit_depth == 10) { if (EXTERNAL_MMXEXT(cpu_flags)) { #if ARCH_X86_32 c->h264_v_loop_filter_chroma = ff_deblock_v_chroma_10_mmxext; c->h264_v_loop_filter_chroma_intra = ff_deblock_v_chroma_intra_10_mmxext; + if (chroma_format_idc <= 1) { + c->h264_h_loop_filter_chroma = ff_deblock_h_chroma_10_mmxext; + } else { + c->h264_h_loop_filter_chroma = ff_deblock_h_chroma422_10_mmxext; + } c->h264_v_loop_filter_luma = ff_deblock_v_luma_10_mmxext; c->h264_h_loop_filter_luma = ff_deblock_h_luma_10_mmxext; c->h264_v_loop_filter_luma_intra = ff_deblock_v_luma_intra_10_mmxext; @@ -307,8 +370,11 @@ av_cold void ff_h264dsp_init_x86(H264DSPContext *c, const int bit_depth, c->h264_idct8_dc_add = ff_h264_idct8_dc_add_10_sse2; c->h264_idct_add16 = ff_h264_idct_add16_10_sse2; - if (chroma_format_idc <= 1) + if (chroma_format_idc <= 1) { c->h264_idct_add8 = ff_h264_idct_add8_10_sse2; + } else { + c->h264_idct_add8 = ff_h264_idct_add8_422_10_sse2; + } c->h264_idct_add16intra = ff_h264_idct_add16intra_10_sse2; #if HAVE_ALIGNED_STACK c->h264_idct8_add = ff_h264_idct8_add_10_sse2; @@ -325,6 +391,11 @@ av_cold void ff_h264dsp_init_x86(H264DSPContext *c, const int bit_depth, c->h264_v_loop_filter_chroma = ff_deblock_v_chroma_10_sse2; c->h264_v_loop_filter_chroma_intra = ff_deblock_v_chroma_intra_10_sse2; + if (chroma_format_idc <= 1) { + c->h264_h_loop_filter_chroma = ff_deblock_h_chroma_10_sse2; + } else { + c->h264_h_loop_filter_chroma = ff_deblock_h_chroma422_10_sse2; + } #if HAVE_ALIGNED_STACK c->h264_v_loop_filter_luma = ff_deblock_v_luma_10_sse2; c->h264_h_loop_filter_luma = ff_deblock_h_luma_10_sse2; @@ -347,8 +418,11 @@ av_cold void ff_h264dsp_init_x86(H264DSPContext *c, const int bit_depth, c->h264_idct8_dc_add = ff_h264_idct8_dc_add_10_avx; c->h264_idct_add16 = ff_h264_idct_add16_10_avx; - if (chroma_format_idc <= 1) + if (chroma_format_idc <= 1) { c->h264_idct_add8 = ff_h264_idct_add8_10_avx; + } else { + c->h264_idct_add8 = ff_h264_idct_add8_422_10_avx; + } c->h264_idct_add16intra = ff_h264_idct_add16intra_10_avx; #if HAVE_ALIGNED_STACK c->h264_idct8_add = ff_h264_idct8_add_10_avx; @@ -357,6 +431,11 @@ av_cold void ff_h264dsp_init_x86(H264DSPContext *c, const int bit_depth, c->h264_v_loop_filter_chroma = ff_deblock_v_chroma_10_avx; c->h264_v_loop_filter_chroma_intra = ff_deblock_v_chroma_intra_10_avx; + if (chroma_format_idc <= 1) { + c->h264_h_loop_filter_chroma = ff_deblock_h_chroma_10_avx; + } else { + c->h264_h_loop_filter_chroma = ff_deblock_h_chroma422_10_avx; + } #if HAVE_ALIGNED_STACK c->h264_v_loop_filter_luma = ff_deblock_v_luma_10_avx; c->h264_h_loop_filter_luma = ff_deblock_h_luma_10_avx; @@ -365,4 +444,5 @@ av_cold void ff_h264dsp_init_x86(H264DSPContext *c, const int bit_depth, #endif /* HAVE_ALIGNED_STACK */ } } +#endif } |