avcodec/x86/hpeldsp: Remove obsolete MMX/3dnow functions

x64 always has MMX, MMXEXT, SSE and SSE2 and this means
that some functions for MMX, MMXEXT and 3dnow are always
overridden by other functions (unless one e.g. explicitly
disables SSE2) for x64. So given that the only systems that
benefit from these functions are truely ancient 32bit x86s
they are removed.

Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>

5 files changed, 22 insertions, 163 deletions

diff --git a/libavcodec/x86/fpel.asm b/libavcodec/x86/fpel.asm
index d38a1b1035..ebe8e43750 100644
--- a/libavcodec/x86/fpel.asm
+++ b/libavcodec/x86/fpel.asm
@@ -91,7 +91,6 @@ cglobal %1_pixels%2, 4,5,4
 INIT_MMX mmx
 OP_PIXELS put, 4
 OP_PIXELS put, 8
-OP_PIXELS avg, 8
 OP_PIXELS put, 16
 OP_PIXELS avg, 16
 
diff --git a/libavcodec/x86/hpeldsp.asm b/libavcodec/x86/hpeldsp.asm
index ce5d7a4e28..b3a270a173 100644
--- a/libavcodec/x86/hpeldsp.asm
+++ b/libavcodec/x86/hpeldsp.asm
@@ -83,8 +83,6 @@ cglobal put_pixels8_x2, 4,5
 
 INIT_MMX mmxext
 PUT_PIXELS8_X2
-INIT_MMX 3dnow
-PUT_PIXELS8_X2
 
 
 ; void ff_put_pixels16_x2(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
@@ -127,15 +125,13 @@ cglobal put_pixels16_x2, 4,5
 
 INIT_MMX mmxext
 PUT_PIXELS_16
-INIT_MMX 3dnow
-PUT_PIXELS_16
 ; The 8_X2 macro can easily be used here
 INIT_XMM sse2
 PUT_PIXELS8_X2
 
 
 ; void ff_put_no_rnd_pixels8_x2(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
-%macro PUT_NO_RND_PIXELS8_X2 0
+INIT_MMX mmxext
 cglobal put_no_rnd_pixels8_x2, 4,5
     mova         m6, [pb_1]
     lea          r4, [r2*2]
@@ -167,12 +163,6 @@ cglobal put_no_rnd_pixels8_x2, 4,5
     sub         r3d, 4
     jne .loop
     REP_RET
-%endmacro
-
-INIT_MMX mmxext
-PUT_NO_RND_PIXELS8_X2
-INIT_MMX 3dnow
-PUT_NO_RND_PIXELS8_X2
 
 
 ; void ff_put_pixels8_y2(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
@@ -209,15 +199,13 @@ cglobal put_pixels8_y2, 4,5
 
 INIT_MMX mmxext
 PUT_PIXELS8_Y2
-INIT_MMX 3dnow
-PUT_PIXELS8_Y2
 ; actually, put_pixels16_y2_sse2
 INIT_XMM sse2
 PUT_PIXELS8_Y2
 
 
 ; void ff_put_no_rnd_pixels8_y2(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
-%macro PUT_NO_RND_PIXELS8_Y2 0
+INIT_MMX mmxext
 cglobal put_no_rnd_pixels8_y2, 4,5
     mova         m6, [pb_1]
     lea          r4, [r2+r2]
@@ -245,42 +233,6 @@ cglobal put_no_rnd_pixels8_y2, 4,5
     sub         r3d, 4
     jne .loop
     REP_RET
-%endmacro
-
-INIT_MMX mmxext
-PUT_NO_RND_PIXELS8_Y2
-INIT_MMX 3dnow
-PUT_NO_RND_PIXELS8_Y2
-
-
-; void ff_avg_pixels8(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
-%macro AVG_PIXELS8 0
-cglobal avg_pixels8, 4,5
-    lea          r4, [r2*2]
-.loop:
-    mova         m0, [r0]
-    mova         m1, [r0+r2]
-    PAVGB        m0, [r1]
-    PAVGB        m1, [r1+r2]
-    mova       [r0], m0
-    mova    [r0+r2], m1
-    add          r1, r4
-    add          r0, r4
-    mova         m0, [r0]
-    mova         m1, [r0+r2]
-    PAVGB        m0, [r1]
-    PAVGB        m1, [r1+r2]
-    add          r1, r4
-    mova       [r0], m0
-    mova    [r0+r2], m1
-    add          r0, r4
-    sub         r3d, 4
-    jne .loop
-    REP_RET
-%endmacro
-
-INIT_MMX 3dnow
-AVG_PIXELS8
 
 
 ; void ff_avg_pixels8_x2(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
@@ -291,10 +243,6 @@ cglobal avg_pixels16_x2, 4,5,4
 cglobal avg_pixels8_x2, 4,5
 %endif
     lea          r4, [r2*2]
-%if notcpuflag(mmxext)
-    pcmpeqd      m5, m5
-    paddb        m5, m5
-%endif
 .loop:
     movu         m0, [r1]
     movu         m2, [r1+r2]
@@ -335,12 +283,8 @@ cglobal avg_pixels8_x2, 4,5
     REP_RET
 %endmacro
 
-INIT_MMX mmx
-AVG_PIXELS8_X2
 INIT_MMX mmxext
 AVG_PIXELS8_X2
-INIT_MMX 3dnow
-AVG_PIXELS8_X2
 ; actually avg_pixels16_x2
 INIT_XMM sse2
 AVG_PIXELS8_X2
@@ -384,8 +328,6 @@ cglobal avg_pixels8_y2, 4,5
 
 INIT_MMX mmxext
 AVG_PIXELS8_Y2
-INIT_MMX 3dnow
-AVG_PIXELS8_Y2
 ; actually avg_pixels16_y2
 INIT_XMM sse2
 AVG_PIXELS8_Y2
@@ -394,7 +336,7 @@ AVG_PIXELS8_Y2
 ; void ff_avg_pixels8_xy2(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
 ; Note this is not correctly rounded, and is therefore used for
 ; not-bitexact output
-%macro AVG_APPROX_PIXELS8_XY2 0
+INIT_MMX mmxext
 cglobal avg_approx_pixels8_xy2, 4,5
     mova         m6, [pb_1]
     lea          r4, [r2*2]
@@ -429,12 +371,6 @@ cglobal avg_approx_pixels8_xy2, 4,5
     sub         r3d, 4
     jne .loop
     REP_RET
-%endmacro
-
-INIT_MMX mmxext
-AVG_APPROX_PIXELS8_XY2
-INIT_MMX 3dnow
-AVG_APPROX_PIXELS8_XY2
 
 
 ; void ff_avg_pixels16_xy2(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
@@ -517,8 +453,6 @@ cglobal %1_pixels8_xy2, 4,5
 
 INIT_MMX mmxext
 SET_PIXELS_XY2 avg
-INIT_MMX 3dnow
-SET_PIXELS_XY2 avg
 INIT_XMM sse2
 SET_PIXELS_XY2 put
 SET_PIXELS_XY2 avg
diff --git a/libavcodec/x86/hpeldsp.h b/libavcodec/x86/hpeldsp.h
index bf97029b57..fd740da72e 100644
--- a/libavcodec/x86/hpeldsp.h
+++ b/libavcodec/x86/hpeldsp.h
@@ -45,8 +45,6 @@ void ff_put_pixels8_xy2_mmx(uint8_t *block, const uint8_t *pixels,
                             ptrdiff_t line_size, int h);
 void ff_put_pixels8_xy2_ssse3(uint8_t *block, const uint8_t *pixels,
                               ptrdiff_t line_size, int h);
-void ff_put_pixels16_xy2_mmx(uint8_t *block, const uint8_t *pixels,
-                             ptrdiff_t line_size, int h);
 void ff_put_pixels16_xy2_sse2(uint8_t *block, const uint8_t *pixels,
                               ptrdiff_t line_size, int h);
 void ff_put_pixels16_xy2_ssse3(uint8_t *block, const uint8_t *pixels,
diff --git a/libavcodec/x86/hpeldsp_init.c b/libavcodec/x86/hpeldsp_init.c
index 6336587281..09c48c341e 100644
--- a/libavcodec/x86/hpeldsp_init.c
+++ b/libavcodec/x86/hpeldsp_init.c
@@ -35,12 +35,8 @@
 
 void ff_put_pixels8_x2_mmxext(uint8_t *block, const uint8_t *pixels,
                               ptrdiff_t line_size, int h);
-void ff_put_pixels8_x2_3dnow(uint8_t *block, const uint8_t *pixels,
-                             ptrdiff_t line_size, int h);
 void ff_put_pixels16_x2_mmxext(uint8_t *block, const uint8_t *pixels,
                                ptrdiff_t line_size, int h);
-void ff_put_pixels16_x2_3dnow(uint8_t *block, const uint8_t *pixels,
-                              ptrdiff_t line_size, int h);
 void ff_put_pixels16_x2_sse2(uint8_t *block, const uint8_t *pixels,
                              ptrdiff_t line_size, int h);
 void ff_avg_pixels16_x2_sse2(uint8_t *block, const uint8_t *pixels,
@@ -51,42 +47,21 @@ void ff_avg_pixels16_y2_sse2(uint8_t *block, const uint8_t *pixels,
                              ptrdiff_t line_size, int h);
 void ff_put_no_rnd_pixels8_x2_mmxext(uint8_t *block, const uint8_t *pixels,
                                      ptrdiff_t line_size, int h);
-void ff_put_no_rnd_pixels8_x2_3dnow(uint8_t *block, const uint8_t *pixels,
-                                    ptrdiff_t line_size, int h);
 void ff_put_pixels8_y2_mmxext(uint8_t *block, const uint8_t *pixels,
                               ptrdiff_t line_size, int h);
-void ff_put_pixels8_y2_3dnow(uint8_t *block, const uint8_t *pixels,
-                             ptrdiff_t line_size, int h);
 void ff_put_no_rnd_pixels8_y2_mmxext(uint8_t *block, const uint8_t *pixels,
                                      ptrdiff_t line_size, int h);
-void ff_put_no_rnd_pixels8_y2_3dnow(uint8_t *block, const uint8_t *pixels,
-                                    ptrdiff_t line_size, int h);
-void ff_avg_pixels8_3dnow(uint8_t *block, const uint8_t *pixels,
-                          ptrdiff_t line_size, int h);
 void ff_avg_pixels8_x2_mmxext(uint8_t *block, const uint8_t *pixels,
                               ptrdiff_t line_size, int h);
-void ff_avg_pixels8_x2_3dnow(uint8_t *block, const uint8_t *pixels,
-                             ptrdiff_t line_size, int h);
 void ff_avg_pixels8_y2_mmxext(uint8_t *block, const uint8_t *pixels,
                               ptrdiff_t line_size, int h);
-void ff_avg_pixels8_y2_3dnow(uint8_t *block, const uint8_t *pixels,
-                             ptrdiff_t line_size, int h);
-void ff_avg_pixels8_xy2_3dnow(uint8_t *block, const uint8_t *pixels,
-                              ptrdiff_t line_size, int h);
 void ff_avg_approx_pixels8_xy2_mmxext(uint8_t *block, const uint8_t *pixels,
                                       ptrdiff_t line_size, int h);
-void ff_avg_approx_pixels8_xy2_3dnow(uint8_t *block, const uint8_t *pixels,
-                                     ptrdiff_t line_size, int h);
 
-#define avg_pixels8_mmx         ff_avg_pixels8_mmx
-#define avg_pixels8_x2_mmx      ff_avg_pixels8_x2_mmx
 #define avg_pixels16_mmx        ff_avg_pixels16_mmx
-#define avg_pixels8_xy2_mmx     ff_avg_pixels8_xy2_mmx
-#define avg_pixels16_xy2_mmx    ff_avg_pixels16_xy2_mmx
 #define put_pixels8_mmx         ff_put_pixels8_mmx
 #define put_pixels16_mmx        ff_put_pixels16_mmx
 #define put_pixels8_xy2_mmx     ff_put_pixels8_xy2_mmx
-#define put_pixels16_xy2_mmx    ff_put_pixels16_xy2_mmx
 #define avg_no_rnd_pixels16_mmx ff_avg_pixels16_mmx
 #define put_no_rnd_pixels8_mmx  ff_put_pixels8_mmx
 #define put_no_rnd_pixels16_mmx ff_put_pixels16_mmx
@@ -121,30 +96,19 @@ CALL_2X_PIXELS(put_no_rnd_pixels16_xy2_mmx, put_no_rnd_pixels8_xy2_mmx, 8)
 /***********************************/
 /* MMX rounding */
 
-#define DEF(x, y) x ## _ ## y ## _mmx
 #define SET_RND  MOVQ_WTWO
-#define PAVGBP(a, b, c, d, e, f)        PAVGBP_MMX(a, b, c, d, e, f)
-#define PAVGB(a, b, c, e)               PAVGB_MMX(a, b, c, e)
-
-#include "hpeldsp_rnd_template.c"
-
-#undef DEF
 #define DEF(x, y) ff_ ## x ## _ ## y ## _mmx
 #define STATIC
+#define NO_AVG
 
 #include "rnd_template.c"
 
+#undef NO_AVG
 #undef DEF
 #undef SET_RND
-#undef PAVGBP
-#undef PAVGB
 
 #if HAVE_MMX
-CALL_2X_PIXELS(avg_pixels16_y2_mmx, avg_pixels8_y2_mmx, 8)
-CALL_2X_PIXELS(put_pixels16_y2_mmx, put_pixels8_y2_mmx, 8)
-
-CALL_2X_PIXELS_EXPORT(ff_avg_pixels16_xy2_mmx, ff_avg_pixels8_xy2_mmx, 8)
-CALL_2X_PIXELS_EXPORT(ff_put_pixels16_xy2_mmx, ff_put_pixels8_xy2_mmx, 8)
+CALL_2X_PIXELS(put_pixels16_xy2_mmx, ff_put_pixels8_xy2_mmx, 8)
 #endif
 
 #endif /* HAVE_INLINE_ASM */
@@ -162,46 +126,42 @@ CALL_2X_PIXELS_EXPORT(ff_put_pixels16_xy2_mmx, ff_put_pixels8_xy2_mmx, 8)
     CALL_2X_PIXELS(avg_pixels16_xy2       ## CPUEXT, ff_avg_pixels8_xy2       ## CPUEXT, 8) \
     CALL_2X_PIXELS(avg_approx_pixels16_xy2## CPUEXT, ff_avg_approx_pixels8_xy2## CPUEXT, 8)
 
-HPELDSP_AVG_PIXELS16(_3dnow)
 HPELDSP_AVG_PIXELS16(_mmxext)
 
 #endif /* HAVE_X86ASM */
 
 #define SET_HPEL_FUNCS_EXT(PFX, IDX, SIZE, CPU)                             \
     if (HAVE_MMX_EXTERNAL)                                                  \
-    c->PFX ## _pixels_tab IDX [0] = PFX ## _pixels ## SIZE ## _     ## CPU;
+        c->PFX ## _pixels_tab IDX [0] = PFX ## _pixels ## SIZE ## _ ## CPU
 
 #if HAVE_MMX_INLINE
-#define SET_HPEL_FUNCS(PFX, IDX, SIZE, CPU)                                     \
+#define SET_HPEL_FUNCS03(PFX, IDX, SIZE, CPU)                                   \
+    do {                                                                        \
+        SET_HPEL_FUNCS_EXT(PFX, IDX, SIZE, CPU);                                \
+        c->PFX ## _pixels_tab IDX [3] = PFX ## _pixels ## SIZE ## _xy2_ ## CPU; \
+    } while (0)
+#define SET_HPEL_FUNCS12(PFX, IDX, SIZE, CPU)                                   \
     do {                                                                        \
-        SET_HPEL_FUNCS_EXT(PFX, IDX, SIZE, CPU)                                 \
         c->PFX ## _pixels_tab IDX [1] = PFX ## _pixels ## SIZE ## _x2_  ## CPU; \
         c->PFX ## _pixels_tab IDX [2] = PFX ## _pixels ## SIZE ## _y2_  ## CPU; \
-        c->PFX ## _pixels_tab IDX [3] = PFX ## _pixels ## SIZE ## _xy2_ ## CPU; \
     } while (0)
 #else
+#define SET_HPEL_FUNCS03(PFX, IDX, SIZE, CPU) SET_HPEL_FUNCS_EXT(PFX, IDX, SIZE, CPU)
+#define SET_HPEL_FUNCS12(PFX, IDX, SIZE, CPU) ((void)0)
+#endif
 #define SET_HPEL_FUNCS(PFX, IDX, SIZE, CPU)                                     \
     do {                                                                        \
-        SET_HPEL_FUNCS_EXT(PFX, IDX, SIZE, CPU)                                 \
+        SET_HPEL_FUNCS03(PFX, IDX, SIZE, CPU);                                  \
+        SET_HPEL_FUNCS12(PFX, IDX, SIZE, CPU);                                  \
     } while (0)
-#endif
 
 static void hpeldsp_init_mmx(HpelDSPContext *c, int flags)
 {
-    SET_HPEL_FUNCS(put,        [0], 16, mmx);
+    SET_HPEL_FUNCS03(put,      [0], 16, mmx);
     SET_HPEL_FUNCS(put_no_rnd, [0], 16, mmx);
-    SET_HPEL_FUNCS(avg,        [0], 16, mmx);
     SET_HPEL_FUNCS(avg_no_rnd,    , 16, mmx);
-    SET_HPEL_FUNCS(put,        [1],  8, mmx);
+    SET_HPEL_FUNCS03(put,      [1],  8, mmx);
     SET_HPEL_FUNCS(put_no_rnd, [1],  8, mmx);
-    if (HAVE_MMX_EXTERNAL) {
-        c->avg_pixels_tab[1][0] = ff_avg_pixels8_mmx;
-        c->avg_pixels_tab[1][1] = ff_avg_pixels8_x2_mmx;
-    }
-#if HAVE_MMX_INLINE
-    c->avg_pixels_tab[1][2] = avg_pixels8_y2_mmx;
-    c->avg_pixels_tab[1][3] = ff_avg_pixels8_xy2_mmx;
-#endif
 }
 
 static void hpeldsp_init_mmxext(HpelDSPContext *c, int flags)
@@ -235,37 +195,6 @@ static void hpeldsp_init_mmxext(HpelDSPContext *c, int flags)
 #endif /* HAVE_MMXEXT_EXTERNAL */
 }
 
-static void hpeldsp_init_3dnow(HpelDSPContext *c, int flags)
-{
-#if HAVE_AMD3DNOW_EXTERNAL
-    c->put_pixels_tab[0][1] = ff_put_pixels16_x2_3dnow;
-    c->put_pixels_tab[0][2] = put_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->avg_pixels_tab[0][3] = avg_pixels16_xy2_3dnow;
-
-    c->put_pixels_tab[1][1] = ff_put_pixels8_x2_3dnow;
-    c->put_pixels_tab[1][2] = ff_put_pixels8_y2_3dnow;
-
-    c->avg_pixels_tab[1][0] = ff_avg_pixels8_3dnow;
-    c->avg_pixels_tab[1][1] = ff_avg_pixels8_x2_3dnow;
-    c->avg_pixels_tab[1][2] = ff_avg_pixels8_y2_3dnow;
-    c->avg_pixels_tab[1][3] = ff_avg_pixels8_xy2_3dnow;
-
-    if (!(flags & AV_CODEC_FLAG_BITEXACT)){
-        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] = avg_approx_pixels16_xy2_3dnow;
-        c->avg_pixels_tab[1][3] = ff_avg_approx_pixels8_xy2_3dnow;
-    }
-#endif /* HAVE_AMD3DNOW_EXTERNAL */
-}
-
 static void hpeldsp_init_sse2_fast(HpelDSPContext *c, int flags)
 {
 #if HAVE_SSE2_EXTERNAL
@@ -298,9 +227,6 @@ av_cold void ff_hpeldsp_init_x86(HpelDSPContext *c, int flags)
     if (INLINE_MMX(cpu_flags))
         hpeldsp_init_mmx(c, flags);
 
-    if (EXTERNAL_AMD3DNOW(cpu_flags))
-        hpeldsp_init_3dnow(c, flags);
-
     if (EXTERNAL_MMXEXT(cpu_flags))
         hpeldsp_init_mmxext(c, flags);
 
diff --git a/libavcodec/x86/rnd_template.c b/libavcodec/x86/rnd_template.c
index 09946bd23f..b825eeba6e 100644
--- a/libavcodec/x86/rnd_template.c
+++ b/libavcodec/x86/rnd_template.c
@@ -97,6 +97,7 @@ av_unused STATIC void DEF(put, pixels8_xy2)(uint8_t *block, const uint8_t *pixel
         :FF_REG_a, "memory");
 }
 
+#ifndef NO_AVG
 // avg_pixels
 // this routine is 'slightly' suboptimal but mostly unused
 av_unused STATIC void DEF(avg, pixels8_xy2)(uint8_t *block, const uint8_t *pixels,
@@ -173,3 +174,4 @@ av_unused STATIC void DEF(avg, pixels8_xy2)(uint8_t *block, const uint8_t *pixel
         :"D"(block), "r"((x86_reg)line_size)
         :FF_REG_a, "memory");
 }
+#endif