diff options
| -rw-r--r-- | libavcodec/x86/cavsdsp.c | 32 | ||||
| -rw-r--r-- | libavcodec/x86/h264_qpel.c | 9 | ||||
| -rw-r--r-- | libavcodec/x86/h264dsp_init.c | 12 | ||||
| -rw-r--r-- | libavcodec/x86/hpeldsp_avg_template.c | 32 | ||||
| -rw-r--r-- | libavcodec/x86/hpeldsp_init.c | 28 |
5 files changed, 60 insertions, 53 deletions
diff --git a/libavcodec/x86/cavsdsp.c b/libavcodec/x86/cavsdsp.c index d9cf85f1b6..37e7eec483 100644 --- a/libavcodec/x86/cavsdsp.c +++ b/libavcodec/x86/cavsdsp.c @@ -413,22 +413,22 @@ static void OPNAME ## cavs_qpel16_h_ ## MMX(uint8_t *dst, uint8_t *src, int dstS }\ #define CAVS_MC(OPNAME, SIZE, MMX) \ -static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc20_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\ +static void OPNAME ## cavs_qpel ## SIZE ## _mc20_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\ {\ OPNAME ## cavs_qpel ## SIZE ## _h_ ## MMX(dst, src, stride, stride);\ }\ \ -static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc01_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\ +static void OPNAME ## cavs_qpel ## SIZE ## _mc01_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\ {\ OPNAME ## cavs_qpel ## SIZE ## _v1_ ## MMX(dst, src, stride, stride);\ }\ \ -static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc02_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\ +static void OPNAME ## cavs_qpel ## SIZE ## _mc02_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\ {\ OPNAME ## cavs_qpel ## SIZE ## _v2_ ## MMX(dst, src, stride, stride);\ }\ \ -static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc03_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\ +static void OPNAME ## cavs_qpel ## SIZE ## _mc03_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\ {\ OPNAME ## cavs_qpel ## SIZE ## _v3_ ## MMX(dst, src, stride, stride);\ }\ @@ -459,11 +459,11 @@ static av_cold void cavsdsp_init_mmx(CAVSDSPContext *c, } #endif /* HAVE_MMX_INLINE */ -#define DSPFUNC(PFX, IDX, NUM, EXT) \ - c->PFX ## _cavs_qpel_pixels_tab[IDX][ 2] = ff_ ## PFX ## _cavs_qpel ## NUM ## _mc20_ ## EXT; \ - c->PFX ## _cavs_qpel_pixels_tab[IDX][ 4] = ff_ ## PFX ## _cavs_qpel ## NUM ## _mc01_ ## EXT; \ - c->PFX ## _cavs_qpel_pixels_tab[IDX][ 8] = ff_ ## PFX ## _cavs_qpel ## NUM ## _mc02_ ## EXT; \ - c->PFX ## _cavs_qpel_pixels_tab[IDX][12] = ff_ ## PFX ## _cavs_qpel ## NUM ## _mc03_ ## EXT; \ +#define DSPFUNC(PFX, IDX, NUM, EXT) \ + c->PFX ## _cavs_qpel_pixels_tab[IDX][ 2] = PFX ## _cavs_qpel ## NUM ## _mc20_ ## EXT; \ + c->PFX ## _cavs_qpel_pixels_tab[IDX][ 4] = PFX ## _cavs_qpel ## NUM ## _mc01_ ## EXT; \ + c->PFX ## _cavs_qpel_pixels_tab[IDX][ 8] = PFX ## _cavs_qpel ## NUM ## _mc02_ ## EXT; \ + c->PFX ## _cavs_qpel_pixels_tab[IDX][12] = PFX ## _cavs_qpel ## NUM ## _mc03_ ## EXT; \ #if HAVE_MMXEXT_INLINE QPEL_CAVS(put_, PUT_OP, mmxext) @@ -474,8 +474,8 @@ CAVS_MC(put_, 16, mmxext) CAVS_MC(avg_, 8, mmxext) CAVS_MC(avg_, 16, mmxext) -static av_cold void ff_cavsdsp_init_mmxext(CAVSDSPContext *c, - AVCodecContext *avctx) +static av_cold void cavsdsp_init_mmxext(CAVSDSPContext *c, + AVCodecContext *avctx) { DSPFUNC(put, 0, 16, mmxext); DSPFUNC(put, 1, 8, mmxext); @@ -493,8 +493,8 @@ CAVS_MC(put_, 16,3dnow) CAVS_MC(avg_, 8, 3dnow) CAVS_MC(avg_, 16,3dnow) -static av_cold void ff_cavsdsp_init_3dnow(CAVSDSPContext *c, - AVCodecContext *avctx) +static av_cold void cavsdsp_init_3dnow(CAVSDSPContext *c, + AVCodecContext *avctx) { DSPFUNC(put, 0, 16, 3dnow); DSPFUNC(put, 1, 8, 3dnow); @@ -512,9 +512,11 @@ av_cold void ff_cavsdsp_init_x86(CAVSDSPContext *c, AVCodecContext *avctx) cavsdsp_init_mmx(c, avctx); #endif /* HAVE_MMX_INLINE */ #if HAVE_MMXEXT_INLINE - if (mm_flags & AV_CPU_FLAG_MMXEXT) ff_cavsdsp_init_mmxext(c, avctx); + if (mm_flags & AV_CPU_FLAG_MMXEXT) + cavsdsp_init_mmxext(c, avctx); #endif /* HAVE_MMXEXT_INLINE */ #if HAVE_AMD3DNOW_INLINE - if (mm_flags & AV_CPU_FLAG_3DNOW) ff_cavsdsp_init_3dnow(c, avctx); + if (mm_flags & AV_CPU_FLAG_3DNOW) + cavsdsp_init_3dnow(c, avctx); #endif /* HAVE_AMD3DNOW_INLINE */ } diff --git a/libavcodec/x86/h264_qpel.c b/libavcodec/x86/h264_qpel.c index d1e8afb35b..ef1b59d92b 100644 --- a/libavcodec/x86/h264_qpel.c +++ b/libavcodec/x86/h264_qpel.c @@ -206,7 +206,12 @@ static av_always_inline void ff_ ## OPNAME ## h264_qpel16_v_lowpass_ ## MMX(uint ff_ ## OPNAME ## h264_qpel8or16_v_lowpass_ ## MMX(dst+8, src+8, dstStride, srcStride, 16);\ } -static av_always_inline void ff_put_h264_qpel8or16_hv1_lowpass_sse2(int16_t *tmp, uint8_t *src, int tmpStride, int srcStride, int size){ +static av_always_inline void put_h264_qpel8or16_hv1_lowpass_sse2(int16_t *tmp, + uint8_t *src, + int tmpStride, + int srcStride, + int size) +{ int w = (size+8)>>3; src -= 2*srcStride+2; while(w--){ @@ -218,7 +223,7 @@ static av_always_inline void ff_put_h264_qpel8or16_hv1_lowpass_sse2(int16_t *tmp #define QPEL_H264_HV_XMM(OPNAME, OP, MMX)\ static av_always_inline void ff_ ## OPNAME ## h264_qpel8or16_hv_lowpass_ ## MMX(uint8_t *dst, int16_t *tmp, uint8_t *src, int dstStride, int tmpStride, int srcStride, int size){\ - ff_put_h264_qpel8or16_hv1_lowpass_sse2(tmp, src, tmpStride, srcStride, size);\ + put_h264_qpel8or16_hv1_lowpass_sse2(tmp, src, tmpStride, srcStride, size);\ ff_ ## OPNAME ## h264_qpel8or16_hv2_lowpass_ ## MMX(dst, tmp, dstStride, tmpStride, size);\ }\ static av_always_inline void ff_ ## OPNAME ## h264_qpel8_hv_lowpass_ ## MMX(uint8_t *dst, int16_t *tmp, uint8_t *src, int dstStride, int tmpStride, int srcStride){\ diff --git a/libavcodec/x86/h264dsp_init.c b/libavcodec/x86/h264dsp_init.c index 15dceedc1a..a98ec0139f 100644 --- a/libavcodec/x86/h264dsp_init.c +++ b/libavcodec/x86/h264dsp_init.c @@ -132,8 +132,8 @@ LF_FUNCS(uint16_t, 10) #if ARCH_X86_32 && HAVE_MMXEXT_EXTERNAL LF_FUNC(v8, luma, 8, mmxext) -static void ff_deblock_v_luma_8_mmxext(uint8_t *pix, int stride, int alpha, - int beta, int8_t *tc0) +static void deblock_v_luma_8_mmxext(uint8_t *pix, int stride, int alpha, + int beta, int8_t *tc0) { if ((tc0[0] & tc0[1]) >= 0) ff_deblock_v8_luma_8_mmxext(pix + 0, stride, alpha, beta, tc0); @@ -141,8 +141,8 @@ static void ff_deblock_v_luma_8_mmxext(uint8_t *pix, int stride, int alpha, ff_deblock_v8_luma_8_mmxext(pix + 8, stride, alpha, beta, tc0 + 2); } LF_IFUNC(v8, luma_intra, 8, mmxext) -static void ff_deblock_v_luma_intra_8_mmxext(uint8_t *pix, int stride, - int alpha, int beta) +static void deblock_v_luma_intra_8_mmxext(uint8_t *pix, int stride, + int alpha, int beta) { ff_deblock_v8_luma_intra_8_mmxext(pix + 0, stride, alpha, beta); ff_deblock_v8_luma_intra_8_mmxext(pix + 8, stride, alpha, beta); @@ -247,9 +247,9 @@ av_cold void ff_h264dsp_init_x86(H264DSPContext *c, const int bit_depth, c->h264_h_loop_filter_chroma_intra = ff_deblock_h_chroma_intra_8_mmxext; } #if ARCH_X86_32 && HAVE_MMXEXT_EXTERNAL - c->h264_v_loop_filter_luma = ff_deblock_v_luma_8_mmxext; + c->h264_v_loop_filter_luma = deblock_v_luma_8_mmxext; c->h264_h_loop_filter_luma = ff_deblock_h_luma_8_mmxext; - c->h264_v_loop_filter_luma_intra = ff_deblock_v_luma_intra_8_mmxext; + c->h264_v_loop_filter_luma_intra = deblock_v_luma_intra_8_mmxext; c->h264_h_loop_filter_luma_intra = ff_deblock_h_luma_intra_8_mmxext; #endif /* ARCH_X86_32 && HAVE_MMXEXT_EXTERNAL */ c->weight_h264_pixels_tab[0] = ff_h264_weight_16_mmxext; diff --git a/libavcodec/x86/hpeldsp_avg_template.c b/libavcodec/x86/hpeldsp_avg_template.c index 2249ce6c97..f7b6cccb64 100644 --- a/libavcodec/x86/hpeldsp_avg_template.c +++ b/libavcodec/x86/hpeldsp_avg_template.c @@ -25,52 +25,52 @@ */ //FIXME the following could be optimized too ... -static void DEF(ff_put_no_rnd_pixels16_x2)(uint8_t *block, - const uint8_t *pixels, - ptrdiff_t line_size, int h) +static void DEF(put_no_rnd_pixels16_x2)(uint8_t *block, + const uint8_t *pixels, + ptrdiff_t line_size, int h) { DEF(ff_put_no_rnd_pixels8_x2)(block, pixels, line_size, h); DEF(ff_put_no_rnd_pixels8_x2)(block + 8, pixels + 8, line_size, h); } -static void DEF(ff_put_pixels16_y2)(uint8_t *block, const uint8_t *pixels, - ptrdiff_t line_size, int h) +static void DEF(put_pixels16_y2)(uint8_t *block, const uint8_t *pixels, + ptrdiff_t line_size, int h) { DEF(ff_put_pixels8_y2)(block, pixels, line_size, h); DEF(ff_put_pixels8_y2)(block + 8, pixels + 8, line_size, h); } -static void DEF(ff_put_no_rnd_pixels16_y2)(uint8_t *block, - const uint8_t *pixels, - ptrdiff_t line_size, int h) +static void DEF(put_no_rnd_pixels16_y2)(uint8_t *block, + const uint8_t *pixels, + ptrdiff_t line_size, int h) { DEF(ff_put_no_rnd_pixels8_y2)(block, pixels, line_size, h); DEF(ff_put_no_rnd_pixels8_y2)(block + 8, pixels + 8, line_size, h); } -static void DEF(ff_avg_pixels16)(uint8_t *block, const uint8_t *pixels, - ptrdiff_t line_size, int h) +static void DEF(avg_pixels16)(uint8_t *block, const uint8_t *pixels, + ptrdiff_t line_size, int h) { DEF(ff_avg_pixels8)(block, pixels, line_size, h); DEF(ff_avg_pixels8)(block + 8, pixels + 8, line_size, h); } -static void DEF(ff_avg_pixels16_x2)(uint8_t *block, const uint8_t *pixels, - ptrdiff_t line_size, int h) +static void DEF(avg_pixels16_x2)(uint8_t *block, const uint8_t *pixels, + ptrdiff_t line_size, int h) { DEF(ff_avg_pixels8_x2)(block, pixels, line_size, h); DEF(ff_avg_pixels8_x2)(block + 8, pixels + 8, line_size, h); } -static void DEF(ff_avg_pixels16_y2)(uint8_t *block, const uint8_t *pixels, - ptrdiff_t line_size, int h) +static void DEF(avg_pixels16_y2)(uint8_t *block, const uint8_t *pixels, + ptrdiff_t line_size, int h) { DEF(ff_avg_pixels8_y2)(block, pixels, line_size, h); DEF(ff_avg_pixels8_y2)(block + 8, pixels + 8, line_size, h); } -static void DEF(ff_avg_pixels16_xy2)(uint8_t *block, const uint8_t *pixels, - ptrdiff_t line_size, int h) +static void DEF(avg_pixels16_xy2)(uint8_t *block, const uint8_t *pixels, + ptrdiff_t line_size, int h) { DEF(ff_avg_pixels8_xy2)(block, pixels, line_size, h); DEF(ff_avg_pixels8_xy2)(block + 8, pixels + 8, line_size, h); diff --git a/libavcodec/x86/hpeldsp_init.c b/libavcodec/x86/hpeldsp_init.c index 73892ef400..824ea1c5eb 100644 --- a/libavcodec/x86/hpeldsp_init.c +++ b/libavcodec/x86/hpeldsp_init.c @@ -226,11 +226,11 @@ static void hpeldsp_init_mmxext(HpelDSPContext *c, int flags, int mm_flags) { #if HAVE_YASM c->put_pixels_tab[0][1] = ff_put_pixels16_x2_mmxext; - c->put_pixels_tab[0][2] = ff_put_pixels16_y2_mmxext; + c->put_pixels_tab[0][2] = put_pixels16_y2_mmxext; - c->avg_pixels_tab[0][0] = ff_avg_pixels16_mmxext; - c->avg_pixels_tab[0][1] = ff_avg_pixels16_x2_mmxext; - c->avg_pixels_tab[0][2] = ff_avg_pixels16_y2_mmxext; + c->avg_pixels_tab[0][0] = avg_pixels16_mmxext; + c->avg_pixels_tab[0][1] = avg_pixels16_x2_mmxext; + c->avg_pixels_tab[0][2] = avg_pixels16_y2_mmxext; c->put_pixels_tab[1][1] = ff_put_pixels8_x2_mmxext; c->put_pixels_tab[1][2] = ff_put_pixels8_y2_mmxext; @@ -240,12 +240,12 @@ static void hpeldsp_init_mmxext(HpelDSPContext *c, int flags, int mm_flags) c->avg_pixels_tab[1][2] = ff_avg_pixels8_y2_mmxext; if (!(flags & CODEC_FLAG_BITEXACT)) { - c->put_no_rnd_pixels_tab[0][1] = ff_put_no_rnd_pixels16_x2_mmxext; - c->put_no_rnd_pixels_tab[0][2] = ff_put_no_rnd_pixels16_y2_mmxext; + c->put_no_rnd_pixels_tab[0][1] = put_no_rnd_pixels16_x2_mmxext; + c->put_no_rnd_pixels_tab[0][2] = put_no_rnd_pixels16_y2_mmxext; c->put_no_rnd_pixels_tab[1][1] = ff_put_no_rnd_pixels8_x2_mmxext; c->put_no_rnd_pixels_tab[1][2] = ff_put_no_rnd_pixels8_y2_mmxext; - c->avg_pixels_tab[0][3] = ff_avg_pixels16_xy2_mmxext; + c->avg_pixels_tab[0][3] = avg_pixels16_xy2_mmxext; c->avg_pixels_tab[1][3] = ff_avg_pixels8_xy2_mmxext; } @@ -260,11 +260,11 @@ static void hpeldsp_init_3dnow(HpelDSPContext *c, int flags, int mm_flags) { #if HAVE_YASM c->put_pixels_tab[0][1] = ff_put_pixels16_x2_3dnow; - c->put_pixels_tab[0][2] = ff_put_pixels16_y2_3dnow; + c->put_pixels_tab[0][2] = put_pixels16_y2_3dnow; - c->avg_pixels_tab[0][0] = ff_avg_pixels16_3dnow; - c->avg_pixels_tab[0][1] = ff_avg_pixels16_x2_3dnow; - c->avg_pixels_tab[0][2] = ff_avg_pixels16_y2_3dnow; + c->avg_pixels_tab[0][0] = avg_pixels16_3dnow; + c->avg_pixels_tab[0][1] = avg_pixels16_x2_3dnow; + c->avg_pixels_tab[0][2] = avg_pixels16_y2_3dnow; c->put_pixels_tab[1][1] = ff_put_pixels8_x2_3dnow; c->put_pixels_tab[1][2] = ff_put_pixels8_y2_3dnow; @@ -274,12 +274,12 @@ static void hpeldsp_init_3dnow(HpelDSPContext *c, int flags, int mm_flags) c->avg_pixels_tab[1][2] = ff_avg_pixels8_y2_3dnow; if (!(flags & CODEC_FLAG_BITEXACT)){ - c->put_no_rnd_pixels_tab[0][1] = ff_put_no_rnd_pixels16_x2_3dnow; - c->put_no_rnd_pixels_tab[0][2] = ff_put_no_rnd_pixels16_y2_3dnow; + c->put_no_rnd_pixels_tab[0][1] = put_no_rnd_pixels16_x2_3dnow; + c->put_no_rnd_pixels_tab[0][2] = put_no_rnd_pixels16_y2_3dnow; c->put_no_rnd_pixels_tab[1][1] = ff_put_no_rnd_pixels8_x2_3dnow; c->put_no_rnd_pixels_tab[1][2] = ff_put_no_rnd_pixels8_y2_3dnow; - c->avg_pixels_tab[0][3] = ff_avg_pixels16_xy2_3dnow; + c->avg_pixels_tab[0][3] = avg_pixels16_xy2_3dnow; c->avg_pixels_tab[1][3] = ff_avg_pixels8_xy2_3dnow; } |