Merge remote branch 'qatar/master'

[ffmpeg] / libavcodec / x86 / dsputil_mmx.c

diff --git a/libavcodec/x86/dsputil_mmx.c b/libavcodec/x86/dsputil_mmx.c
index 985a15d2f1b4bddaaff05ac51b5c659eff75b1bd..d867dc3e6a2a18bca734921460a02f5c64597c5f 100644 (file)
--- a/libavcodec/x86/dsputil_mmx.c
+++ b/libavcodec/x86/dsputil_mmx.c
@@ -43,6 +43,7 @@ DECLARE_ALIGNED(16, const uint64_t, ff_pdw_80000000)[2] =
 {0x8000000080000000ULL, 0x8000000080000000ULL};
 
 DECLARE_ALIGNED(8,  const uint64_t, ff_pw_1  ) = 0x0001000100010001ULL;
+DECLARE_ALIGNED(16, const xmm_reg,  ff_pw_2  ) = {0x0002000200020002ULL, 0x0002000200020002ULL};
 DECLARE_ALIGNED(16, const xmm_reg,  ff_pw_3  ) = {0x0003000300030003ULL, 0x0003000300030003ULL};
 DECLARE_ALIGNED(16, const xmm_reg,  ff_pw_4  ) = {0x0004000400040004ULL, 0x0004000400040004ULL};
 DECLARE_ALIGNED(16, const xmm_reg,  ff_pw_5  ) = {0x0005000500050005ULL, 0x0005000500050005ULL};
@@ -2322,7 +2323,7 @@ float ff_scalarproduct_float_sse(const float *v1, const float *v2, int order);
 void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx)
 {
     int mm_flags = av_get_cpu_flags();
-    const int h264_high_depth = avctx->codec_id == CODEC_ID_H264 && avctx->bits_per_raw_sample > 8;
+    const int high_bit_depth = avctx->codec_id == CODEC_ID_H264 && avctx->bits_per_raw_sample > 8;
 
     if (avctx->dsp_mask) {
         if (avctx->dsp_mask & AV_CPU_FLAG_FORCE)
@@ -2404,7 +2405,7 @@ void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx)
         c->put_pixels_clamped = ff_put_pixels_clamped_mmx;
         c->put_signed_pixels_clamped = ff_put_signed_pixels_clamped_mmx;
         c->add_pixels_clamped = ff_add_pixels_clamped_mmx;
-        if (!h264_high_depth) {
+        if (!high_bit_depth) {
         c->clear_block  = clear_block_mmx;
         c->clear_blocks = clear_blocks_mmx;
         if ((mm_flags & AV_CPU_FLAG_SSE) &&
@@ -2421,7 +2422,7 @@ void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx)
         c->PFX ## _pixels_tab[IDX][2] = PFX ## _pixels ## SIZE ## _y2_ ## CPU; \
         c->PFX ## _pixels_tab[IDX][3] = PFX ## _pixels ## SIZE ## _xy2_ ## CPU
 
-        if (!h264_high_depth) {
+        if (!high_bit_depth) {
         SET_HPEL_FUNCS(put, 0, 16, mmx);
         SET_HPEL_FUNCS(put_no_rnd, 0, 16, mmx);
         SET_HPEL_FUNCS(avg, 0, 16, mmx);
@@ -2436,13 +2437,13 @@ void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx)
         c->gmc= gmc_mmx;
 #endif
 #if ARCH_X86_32 && HAVE_YASM
-        if (!h264_high_depth)
+        if (!high_bit_depth)
         c->emulated_edge_mc = emulated_edge_mc_mmx;
 #endif
 
         c->add_bytes= add_bytes_mmx;
 
-        if (!h264_high_depth)
+        if (!high_bit_depth)
         c->draw_edges = draw_edges_mmx;
 
         if (CONFIG_H263_DECODER || CONFIG_H263_ENCODER) {
@@ -2451,7 +2452,7 @@ void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx)
         }
 
 #if HAVE_YASM
-        if (!h264_high_depth) {
+        if (!high_bit_depth) {
         c->put_h264_chroma_pixels_tab[0]= ff_put_h264_chroma_mc8_mmx_rnd;
         c->put_h264_chroma_pixels_tab[1]= ff_put_h264_chroma_mc4_mmx;
         }
@@ -2463,7 +2464,7 @@ void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx)
         if (mm_flags & AV_CPU_FLAG_MMX2) {
             c->prefetch = prefetch_mmx2;
 
-            if (!h264_high_depth) {
+            if (!high_bit_depth) {
             c->put_pixels_tab[0][1] = put_pixels16_x2_mmx2;
             c->put_pixels_tab[0][2] = put_pixels16_y2_mmx2;
 
@@ -2480,7 +2481,7 @@ void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx)
             }
 
             if(!(avctx->flags & CODEC_FLAG_BITEXACT)){
-                if (!h264_high_depth) {
+                if (!high_bit_depth) {
                 c->put_no_rnd_pixels_tab[0][1] = put_no_rnd_pixels16_x2_mmx2;
                 c->put_no_rnd_pixels_tab[0][2] = put_no_rnd_pixels16_y2_mmx2;
                 c->put_no_rnd_pixels_tab[1][1] = put_no_rnd_pixels8_x2_mmx2;
@@ -2529,7 +2530,7 @@ void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx)
             SET_QPEL_FUNCS(avg_qpel, 0, 16, mmx2);
             SET_QPEL_FUNCS(avg_qpel, 1, 8, mmx2);
 
-            if (!h264_high_depth) {
+            if (!high_bit_depth) {
             SET_QPEL_FUNCS(put_h264_qpel, 0, 16, mmx2);
             SET_QPEL_FUNCS(put_h264_qpel, 1, 8, mmx2);
             SET_QPEL_FUNCS(put_h264_qpel, 2, 4, mmx2);
@@ -2547,7 +2548,7 @@ void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx)
             c->avg_rv40_chroma_pixels_tab[0]= ff_avg_rv40_chroma_mc8_mmx2;
             c->avg_rv40_chroma_pixels_tab[1]= ff_avg_rv40_chroma_mc4_mmx2;
 
-            if (!h264_high_depth) {
+            if (!high_bit_depth) {
             c->avg_h264_chroma_pixels_tab[0]= ff_avg_h264_chroma_mc8_mmx2_rnd;
             c->avg_h264_chroma_pixels_tab[1]= ff_avg_h264_chroma_mc4_mmx2;
             c->avg_h264_chroma_pixels_tab[2]= ff_avg_h264_chroma_mc2_mmx2;
@@ -2564,7 +2565,7 @@ void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx)
         } else if (mm_flags & AV_CPU_FLAG_3DNOW) {
             c->prefetch = prefetch_3dnow;
 
-            if (!h264_high_depth) {
+            if (!high_bit_depth) {
             c->put_pixels_tab[0][1] = put_pixels16_x2_3dnow;
             c->put_pixels_tab[0][2] = put_pixels16_y2_3dnow;
 
@@ -2602,7 +2603,7 @@ void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx)
             SET_QPEL_FUNCS(avg_qpel, 0, 16, 3dnow);
             SET_QPEL_FUNCS(avg_qpel, 1, 8, 3dnow);
 
-            if (!h264_high_depth) {
+            if (!high_bit_depth) {
             SET_QPEL_FUNCS(put_h264_qpel, 0, 16, 3dnow);
             SET_QPEL_FUNCS(put_h264_qpel, 1, 8, 3dnow);
             SET_QPEL_FUNCS(put_h264_qpel, 2, 4, 3dnow);
@@ -2617,7 +2618,7 @@ void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx)
             SET_QPEL_FUNCS(avg_2tap_qpel, 1, 8, 3dnow);
 
 #if HAVE_YASM
-            if (!h264_high_depth) {
+            if (!high_bit_depth) {
             c->avg_h264_chroma_pixels_tab[0]= ff_avg_h264_chroma_mc8_3dnow_rnd;
             c->avg_h264_chroma_pixels_tab[1]= ff_avg_h264_chroma_mc4_3dnow;
             }
@@ -2635,7 +2636,7 @@ void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx)
             c->avg_h264_qpel_pixels_tab[1][x+y*4] = avg_h264_qpel8_mc##x##y##_##CPU;
         if((mm_flags & AV_CPU_FLAG_SSE2) && !(mm_flags & AV_CPU_FLAG_3DNOW)){
             // these functions are slower than mmx on AMD, but faster on Intel
-            if (!h264_high_depth) {
+            if (!high_bit_depth) {
             c->put_pixels_tab[0][0] = put_pixels16_sse2;
             c->put_no_rnd_pixels_tab[0][0] = put_pixels16_sse2;
             c->avg_pixels_tab[0][0] = avg_pixels16_sse2;
@@ -2643,7 +2644,7 @@ void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx)
             }
         }
         if(mm_flags & AV_CPU_FLAG_SSE2){
-            if (!h264_high_depth) {
+            if (!high_bit_depth) {
             H264_QPEL_FUNCS(0, 1, sse2);
             H264_QPEL_FUNCS(0, 2, sse2);
             H264_QPEL_FUNCS(0, 3, sse2);
@@ -2660,7 +2661,7 @@ void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx)
         }
 #if HAVE_SSSE3
         if(mm_flags & AV_CPU_FLAG_SSSE3){
-            if (!h264_high_depth) {
+            if (!high_bit_depth) {
             H264_QPEL_FUNCS(1, 0, ssse3);
             H264_QPEL_FUNCS(1, 1, ssse3);
             H264_QPEL_FUNCS(1, 2, ssse3);
@@ -2675,7 +2676,7 @@ void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx)
             H264_QPEL_FUNCS(3, 3, ssse3);
             }
 #if HAVE_YASM
-            if (!h264_high_depth) {
+            if (!high_bit_depth) {
             c->put_h264_chroma_pixels_tab[0]= ff_put_h264_chroma_mc8_ssse3_rnd;
             c->avg_h264_chroma_pixels_tab[0]= ff_avg_h264_chroma_mc8_ssse3_rnd;
             c->put_h264_chroma_pixels_tab[1]= ff_put_h264_chroma_mc4_ssse3;
@@ -2737,7 +2738,7 @@ void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx)
                 }
             }
 
-            if (!h264_high_depth)
+            if (!high_bit_depth)
             c->emulated_edge_mc = emulated_edge_mc_sse;
             c->gmc= gmc_sse;
 #endif