Merge remote-tracking branch 'qatar/master'

* qatar/master:
  x86: Get rid of duplication between *_rnd_template.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>

2 files changed, 5 insertions, 197 deletions

diff --git a/libavcodec/x86/dsputil_rnd_template.c b/libavcodec/x86/dsputil_rnd_template.c
index 1a89b773e5..776530910e 100644
--- a/libavcodec/x86/dsputil_rnd_template.c
+++ b/libavcodec/x86/dsputil_rnd_template.c
@@ -91,6 +91,8 @@ static void DEF(put, pixels8_xy2)(uint8_t *block, const uint8_t *pixels, ptrdiff
         :REG_a, "memory");
 }
 
+// avg_pixels
+#ifndef NO_RND
 // in case more speed is needed - unroling would certainly help
 static void DEF(avg, pixels8)(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
 {
@@ -110,6 +112,7 @@ static void DEF(avg, pixels8)(uint8_t *block, const uint8_t *pixels, ptrdiff_t l
     }
     while (--h);
 }
+#endif /* NO_RND */
 
 static void DEF(avg, pixels16)(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
 {
diff --git a/libavcodec/x86/hpeldsp_rnd_template.c b/libavcodec/x86/hpeldsp_rnd_template.c
index 07de675d31..5b5866d0b6 100644
--- a/libavcodec/x86/hpeldsp_rnd_template.c
+++ b/libavcodec/x86/hpeldsp_rnd_template.c
@@ -24,6 +24,8 @@
  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
  */
 
+#include "dsputil_rnd_template.c"
+
 // put_pixels
 static void DEF(put, pixels8_x2)(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
 {
@@ -132,118 +134,6 @@ static void DEF(put, pixels8_y2)(uint8_t *block, const uint8_t *pixels, ptrdiff_
         :REG_a, "memory");
 }
 
-static void DEF(put, pixels8_xy2)(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
-{
-    MOVQ_ZERO(mm7);
-    SET_RND(mm6); // =2 for rnd  and  =1 for no_rnd version
-    __asm__ volatile(
-        "movq   (%1), %%mm0             \n\t"
-        "movq   1(%1), %%mm4            \n\t"
-        "movq   %%mm0, %%mm1            \n\t"
-        "movq   %%mm4, %%mm5            \n\t"
-        "punpcklbw %%mm7, %%mm0         \n\t"
-        "punpcklbw %%mm7, %%mm4         \n\t"
-        "punpckhbw %%mm7, %%mm1         \n\t"
-        "punpckhbw %%mm7, %%mm5         \n\t"
-        "paddusw %%mm0, %%mm4           \n\t"
-        "paddusw %%mm1, %%mm5           \n\t"
-        "xor    %%"REG_a", %%"REG_a"    \n\t"
-        "add    %3, %1                  \n\t"
-        ".p2align 3                     \n\t"
-        "1:                             \n\t"
-        "movq   (%1, %%"REG_a"), %%mm0  \n\t"
-        "movq   1(%1, %%"REG_a"), %%mm2 \n\t"
-        "movq   %%mm0, %%mm1            \n\t"
-        "movq   %%mm2, %%mm3            \n\t"
-        "punpcklbw %%mm7, %%mm0         \n\t"
-        "punpcklbw %%mm7, %%mm2         \n\t"
-        "punpckhbw %%mm7, %%mm1         \n\t"
-        "punpckhbw %%mm7, %%mm3         \n\t"
-        "paddusw %%mm2, %%mm0           \n\t"
-        "paddusw %%mm3, %%mm1           \n\t"
-        "paddusw %%mm6, %%mm4           \n\t"
-        "paddusw %%mm6, %%mm5           \n\t"
-        "paddusw %%mm0, %%mm4           \n\t"
-        "paddusw %%mm1, %%mm5           \n\t"
-        "psrlw  $2, %%mm4               \n\t"
-        "psrlw  $2, %%mm5               \n\t"
-        "packuswb  %%mm5, %%mm4         \n\t"
-        "movq   %%mm4, (%2, %%"REG_a")  \n\t"
-        "add    %3, %%"REG_a"           \n\t"
-
-        "movq   (%1, %%"REG_a"), %%mm2  \n\t" // 0 <-> 2   1 <-> 3
-        "movq   1(%1, %%"REG_a"), %%mm4 \n\t"
-        "movq   %%mm2, %%mm3            \n\t"
-        "movq   %%mm4, %%mm5            \n\t"
-        "punpcklbw %%mm7, %%mm2         \n\t"
-        "punpcklbw %%mm7, %%mm4         \n\t"
-        "punpckhbw %%mm7, %%mm3         \n\t"
-        "punpckhbw %%mm7, %%mm5         \n\t"
-        "paddusw %%mm2, %%mm4           \n\t"
-        "paddusw %%mm3, %%mm5           \n\t"
-        "paddusw %%mm6, %%mm0           \n\t"
-        "paddusw %%mm6, %%mm1           \n\t"
-        "paddusw %%mm4, %%mm0           \n\t"
-        "paddusw %%mm5, %%mm1           \n\t"
-        "psrlw  $2, %%mm0               \n\t"
-        "psrlw  $2, %%mm1               \n\t"
-        "packuswb  %%mm1, %%mm0         \n\t"
-        "movq   %%mm0, (%2, %%"REG_a")  \n\t"
-        "add    %3, %%"REG_a"           \n\t"
-
-        "subl   $2, %0                  \n\t"
-        "jnz    1b                      \n\t"
-        :"+g"(h), "+S"(pixels)
-        :"D"(block), "r"((x86_reg)line_size)
-        :REG_a, "memory");
-}
-
-// avg_pixels
-#ifndef NO_RND
-// in case more speed is needed - unroling would certainly help
-static void DEF(avg, pixels8)(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
-{
-    MOVQ_BFE(mm6);
-    JUMPALIGN();
-    do {
-        __asm__ volatile(
-             "movq  %0, %%mm0           \n\t"
-             "movq  %1, %%mm1           \n\t"
-             OP_AVG(%%mm0, %%mm1, %%mm2, %%mm6)
-             "movq  %%mm2, %0           \n\t"
-             :"+m"(*block)
-             :"m"(*pixels)
-             :"memory");
-        pixels += line_size;
-        block += line_size;
-    }
-    while (--h);
-}
-#endif // NO_RND
-
-static void DEF(avg, pixels16)(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
-{
-    MOVQ_BFE(mm6);
-    JUMPALIGN();
-    do {
-        __asm__ volatile(
-             "movq  %0, %%mm0           \n\t"
-             "movq  %1, %%mm1           \n\t"
-             OP_AVG(%%mm0, %%mm1, %%mm2, %%mm6)
-             "movq  %%mm2, %0           \n\t"
-             "movq  8%0, %%mm0          \n\t"
-             "movq  8%1, %%mm1          \n\t"
-             OP_AVG(%%mm0, %%mm1, %%mm2, %%mm6)
-             "movq  %%mm2, 8%0          \n\t"
-             :"+m"(*block)
-             :"m"(*pixels)
-             :"memory");
-        pixels += line_size;
-        block += line_size;
-    }
-    while (--h);
-}
-
 #ifndef NO_RND
 static void DEF(avg, pixels8_x2)(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
 {
@@ -331,98 +221,13 @@ static void DEF(avg, pixels8_y2)(uint8_t *block, const uint8_t *pixels, ptrdiff_
         :REG_a, "memory");
 }
 
-// this routine is 'slightly' suboptimal but mostly unused
-static void DEF(avg, pixels8_xy2)(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
-{
-    MOVQ_ZERO(mm7);
-    SET_RND(mm6); // =2 for rnd  and  =1 for no_rnd version
-    __asm__ volatile(
-        "movq   (%1), %%mm0             \n\t"
-        "movq   1(%1), %%mm4            \n\t"
-        "movq   %%mm0, %%mm1            \n\t"
-        "movq   %%mm4, %%mm5            \n\t"
-        "punpcklbw %%mm7, %%mm0         \n\t"
-        "punpcklbw %%mm7, %%mm4         \n\t"
-        "punpckhbw %%mm7, %%mm1         \n\t"
-        "punpckhbw %%mm7, %%mm5         \n\t"
-        "paddusw %%mm0, %%mm4           \n\t"
-        "paddusw %%mm1, %%mm5           \n\t"
-        "xor    %%"REG_a", %%"REG_a"    \n\t"
-        "add    %3, %1                  \n\t"
-        ".p2align 3                     \n\t"
-        "1:                             \n\t"
-        "movq   (%1, %%"REG_a"), %%mm0  \n\t"
-        "movq   1(%1, %%"REG_a"), %%mm2 \n\t"
-        "movq   %%mm0, %%mm1            \n\t"
-        "movq   %%mm2, %%mm3            \n\t"
-        "punpcklbw %%mm7, %%mm0         \n\t"
-        "punpcklbw %%mm7, %%mm2         \n\t"
-        "punpckhbw %%mm7, %%mm1         \n\t"
-        "punpckhbw %%mm7, %%mm3         \n\t"
-        "paddusw %%mm2, %%mm0           \n\t"
-        "paddusw %%mm3, %%mm1           \n\t"
-        "paddusw %%mm6, %%mm4           \n\t"
-        "paddusw %%mm6, %%mm5           \n\t"
-        "paddusw %%mm0, %%mm4           \n\t"
-        "paddusw %%mm1, %%mm5           \n\t"
-        "psrlw  $2, %%mm4               \n\t"
-        "psrlw  $2, %%mm5               \n\t"
-                "movq   (%2, %%"REG_a"), %%mm3  \n\t"
-        "packuswb  %%mm5, %%mm4         \n\t"
-                "pcmpeqd %%mm2, %%mm2   \n\t"
-                "paddb %%mm2, %%mm2     \n\t"
-                OP_AVG(%%mm3, %%mm4, %%mm5, %%mm2)
-                "movq   %%mm5, (%2, %%"REG_a")  \n\t"
-        "add    %3, %%"REG_a"                \n\t"
-
-        "movq   (%1, %%"REG_a"), %%mm2  \n\t" // 0 <-> 2   1 <-> 3
-        "movq   1(%1, %%"REG_a"), %%mm4 \n\t"
-        "movq   %%mm2, %%mm3            \n\t"
-        "movq   %%mm4, %%mm5            \n\t"
-        "punpcklbw %%mm7, %%mm2         \n\t"
-        "punpcklbw %%mm7, %%mm4         \n\t"
-        "punpckhbw %%mm7, %%mm3         \n\t"
-        "punpckhbw %%mm7, %%mm5         \n\t"
-        "paddusw %%mm2, %%mm4           \n\t"
-        "paddusw %%mm3, %%mm5           \n\t"
-        "paddusw %%mm6, %%mm0           \n\t"
-        "paddusw %%mm6, %%mm1           \n\t"
-        "paddusw %%mm4, %%mm0           \n\t"
-        "paddusw %%mm5, %%mm1           \n\t"
-        "psrlw  $2, %%mm0               \n\t"
-        "psrlw  $2, %%mm1               \n\t"
-                "movq   (%2, %%"REG_a"), %%mm3  \n\t"
-        "packuswb  %%mm1, %%mm0         \n\t"
-                "pcmpeqd %%mm2, %%mm2   \n\t"
-                "paddb %%mm2, %%mm2     \n\t"
-                OP_AVG(%%mm3, %%mm0, %%mm1, %%mm2)
-                "movq   %%mm1, (%2, %%"REG_a")  \n\t"
-        "add    %3, %%"REG_a"           \n\t"
-
-        "subl   $2, %0                  \n\t"
-        "jnz    1b                      \n\t"
-        :"+g"(h), "+S"(pixels)
-        :"D"(block), "r"((x86_reg)line_size)
-        :REG_a, "memory");
-}
-
 //FIXME optimize
 static void DEF(put, pixels16_y2)(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h){
     DEF(put, pixels8_y2)(block  , pixels  , line_size, h);
     DEF(put, pixels8_y2)(block+8, pixels+8, line_size, h);
 }
 
-static void DEF(put, pixels16_xy2)(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h){
-    DEF(put, pixels8_xy2)(block  , pixels  , line_size, h);
-    DEF(put, pixels8_xy2)(block+8, pixels+8, line_size, h);
-}
-
 static void DEF(avg, pixels16_y2)(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h){
     DEF(avg, pixels8_y2)(block  , pixels  , line_size, h);
     DEF(avg, pixels8_y2)(block+8, pixels+8, line_size, h);
 }
-
-static void DEF(avg, pixels16_xy2)(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h){
-    DEF(avg, pixels8_xy2)(block  , pixels  , line_size, h);
-    DEF(avg, pixels8_xy2)(block+8, pixels+8, line_size, h);
-}