avcodec/x86/hpeldsp_vp3: Merge into hpeldsp

Once upon a time, 413abbe16465a7b49472ac110e42939e853e24a1
added versions of some put_no_rnd_pixels functions for use
in VP3 and Theora (with an explicit check so that they are
only used for VP3 and Theora). When this was moved to hpeldsp
(from dsputil) in 3ced55d51c2e65b37e50d500dff88bcd80e01b9c,
the check was replaced by a check for the bitexact flag
(and a CONFIG_VP3_DECODER compile-time check), so that
these functions were now used for other codecs as well.

Later commit 1dfc3cf89d0eb026af28be46294b85d79499ffb5
split off the "VP3-specific bits into a separate file",
yet these bits were not really VP3-specific bits at all
any more. (The error was repeated in commit
0a39c9ac0bfd7345fe676b4e2707d9cec3cbb553.) This commit
has not been reverted, because this would make future
changes from Libav (from where it originated) harder,
yet Libav is no more, so this commit effectively reverts
1dfc3cf89d0eb026af28be46294b85d79499ffb5.

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

6 files changed, 86 insertions, 153 deletions

diff --git a/libavcodec/x86/Makefile b/libavcodec/x86/Makefile
index 118daca333..b4cc5e0d08 100644
--- a/libavcodec/x86/Makefile
+++ b/libavcodec/x86/Makefile
@@ -75,7 +75,6 @@ OBJS-$(CONFIG_UTVIDEO_DECODER)         += x86/utvideodsp_init.o
 OBJS-$(CONFIG_V210_DECODER)            += x86/v210-init.o
 OBJS-$(CONFIG_V210_ENCODER)            += x86/v210enc_init.o
 OBJS-$(CONFIG_VORBIS_DECODER)          += x86/vorbisdsp_init.o
-OBJS-$(CONFIG_VP3_DECODER)             += x86/hpeldsp_vp3_init.o
 OBJS-$(CONFIG_VP6_DECODER)             += x86/vp6dsp_init.o
 OBJS-$(CONFIG_VP9_DECODER)             += x86/vp9dsp_init.o            \
                                           x86/vp9dsp_init_10bpp.o      \
@@ -192,7 +191,6 @@ X86ASM-OBJS-$(CONFIG_UTVIDEO_DECODER)  += x86/utvideodsp.o
 X86ASM-OBJS-$(CONFIG_V210_ENCODER)     += x86/v210enc.o
 X86ASM-OBJS-$(CONFIG_V210_DECODER)     += x86/v210.o
 X86ASM-OBJS-$(CONFIG_VORBIS_DECODER)   += x86/vorbisdsp.o
-X86ASM-OBJS-$(CONFIG_VP3_DECODER)      += x86/hpeldsp_vp3.o
 X86ASM-OBJS-$(CONFIG_VP6_DECODER)      += x86/vp6dsp.o
 X86ASM-OBJS-$(CONFIG_VP9_DECODER)      += x86/vp9intrapred.o            \
                                           x86/vp9intrapred_16bpp.o      \
diff --git a/libavcodec/x86/hpeldsp.asm b/libavcodec/x86/hpeldsp.asm
index 7a2b7135d8..3bc278618c 100644
--- a/libavcodec/x86/hpeldsp.asm
+++ b/libavcodec/x86/hpeldsp.asm
@@ -165,6 +165,47 @@ cglobal put_no_rnd_pixels8_x2, 4,5
     RET
 
 
+; void ff_put_no_rnd_pixels8_x2_exact(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
+INIT_MMX mmxext
+cglobal put_no_rnd_pixels8_x2_exact, 4,5
+    lea          r4, [r2*3]
+    pcmpeqb      m6, m6
+.loop:
+    mova         m0, [r1]
+    mova         m2, [r1+r2]
+    mova         m1, [r1+1]
+    mova         m3, [r1+r2+1]
+    pxor         m0, m6
+    pxor         m2, m6
+    pxor         m1, m6
+    pxor         m3, m6
+    PAVGB        m0, m1
+    PAVGB        m2, m3
+    pxor         m0, m6
+    pxor         m2, m6
+    mova       [r0], m0
+    mova    [r0+r2], m2
+    mova         m0, [r1+r2*2]
+    mova         m1, [r1+r2*2+1]
+    mova         m2, [r1+r4]
+    mova         m3, [r1+r4+1]
+    pxor         m0, m6
+    pxor         m1, m6
+    pxor         m2, m6
+    pxor         m3, m6
+    PAVGB        m0, m1
+    PAVGB        m2, m3
+    pxor         m0, m6
+    pxor         m2, m6
+    mova  [r0+r2*2], m0
+    mova    [r0+r4], m2
+    lea          r1, [r1+r2*4]
+    lea          r0, [r0+r2*4]
+    sub         r3d, 4
+    jg .loop
+    RET
+
+
 ; void ff_put_pixels8_y2(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
 %macro PUT_PIXELS8_Y2 0
 %if cpuflag(sse2)
@@ -235,6 +276,42 @@ cglobal put_no_rnd_pixels8_y2, 4,5
     RET
 
 
+; void ff_put_no_rnd_pixels8_y2_exact(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
+INIT_MMX mmxext
+cglobal put_no_rnd_pixels8_y2_exact, 4,5
+    lea          r4, [r2*3]
+    mova         m0, [r1]
+    pcmpeqb      m6, m6
+    add          r1, r2
+    pxor         m0, m6
+.loop:
+    mova         m1, [r1]
+    mova         m2, [r1+r2]
+    pxor         m1, m6
+    pxor         m2, m6
+    PAVGB        m0, m1
+    PAVGB        m1, m2
+    pxor         m0, m6
+    pxor         m1, m6
+    mova       [r0], m0
+    mova    [r0+r2], m1
+    mova         m1, [r1+r2*2]
+    mova         m0, [r1+r4]
+    pxor         m1, m6
+    pxor         m0, m6
+    PAVGB        m2, m1
+    PAVGB        m1, m0
+    pxor         m2, m6
+    pxor         m1, m6
+    mova  [r0+r2*2], m2
+    mova    [r0+r4], m1
+    lea          r1, [r1+r2*4]
+    lea          r0, [r0+r2*4]
+    sub         r3d, 4
+    jg .loop
+    RET
+
+
 ; void ff_avg_pixels8_x2(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
 %macro AVG_PIXELS8_X2 0
 %if cpuflag(sse2)
diff --git a/libavcodec/x86/hpeldsp.h b/libavcodec/x86/hpeldsp.h
index fd740da72e..ac7e625fda 100644
--- a/libavcodec/x86/hpeldsp.h
+++ b/libavcodec/x86/hpeldsp.h
@@ -22,8 +22,6 @@
 #include <stddef.h>
 #include <stdint.h>
 
-#include "libavcodec/hpeldsp.h"
-
 void ff_avg_pixels8_x2_mmx(uint8_t *block, const uint8_t *pixels,
                            ptrdiff_t line_size, int h);
 
@@ -50,6 +48,4 @@ void ff_put_pixels16_xy2_sse2(uint8_t *block, const uint8_t *pixels,
 void ff_put_pixels16_xy2_ssse3(uint8_t *block, const uint8_t *pixels,
                                ptrdiff_t line_size, int h);
 
-void ff_hpeldsp_vp3_init_x86(HpelDSPContext *c, int cpu_flags, int flags);
-
 #endif /* AVCODEC_X86_HPELDSP_H */
diff --git a/libavcodec/x86/hpeldsp_init.c b/libavcodec/x86/hpeldsp_init.c
index 09c48c341e..f08c66f5c8 100644
--- a/libavcodec/x86/hpeldsp_init.c
+++ b/libavcodec/x86/hpeldsp_init.c
@@ -22,8 +22,6 @@
  * MMX optimization by Nick Kurshev <nickols_k@mail.ru>
  */
 
-#include "config_components.h"
-
 #include "libavutil/attributes.h"
 #include "libavutil/cpu.h"
 #include "libavutil/x86/cpu.h"
@@ -47,10 +45,16 @@ 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_exact_mmxext(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_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_exact_mmxext(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_y2_mmxext(uint8_t *block, const uint8_t *pixels,
@@ -183,6 +187,9 @@ static void hpeldsp_init_mmxext(HpelDSPContext *c, int flags)
     c->avg_pixels_tab[1][2] = ff_avg_pixels8_y2_mmxext;
     c->avg_pixels_tab[1][3] = ff_avg_pixels8_xy2_mmxext;
 
+    c->put_no_rnd_pixels_tab[1][1] = ff_put_no_rnd_pixels8_x2_exact_mmxext;
+    c->put_no_rnd_pixels_tab[1][2] = ff_put_no_rnd_pixels8_y2_exact_mmxext;
+
     if (!(flags & AV_CODEC_FLAG_BITEXACT)) {
         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;
@@ -235,7 +242,4 @@ av_cold void ff_hpeldsp_init_x86(HpelDSPContext *c, int flags)
 
     if (EXTERNAL_SSSE3(cpu_flags))
         hpeldsp_init_ssse3(c, flags);
-
-    if (CONFIG_VP3_DECODER)
-        ff_hpeldsp_vp3_init_x86(c, cpu_flags, flags);
 }
diff --git a/libavcodec/x86/hpeldsp_vp3.asm b/libavcodec/x86/hpeldsp_vp3.asm
deleted file mode 100644
index e580133e45..0000000000
--- a/libavcodec/x86/hpeldsp_vp3.asm
+++ /dev/null
@@ -1,99 +0,0 @@
-;******************************************************************************
-;* SIMD-optimized halfpel functions for VP3
-;*
-;* This file is part of FFmpeg.
-;*
-;* 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.
-;*
-;* 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 FFmpeg; if not, write to the Free Software
-;* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
-;******************************************************************************
-
-%include "libavutil/x86/x86util.asm"
-
-SECTION .text
-
-; void ff_put_no_rnd_pixels8_x2_exact(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
-INIT_MMX mmxext
-cglobal put_no_rnd_pixels8_x2_exact, 4,5
-    lea          r4, [r2*3]
-    pcmpeqb      m6, m6
-.loop:
-    mova         m0, [r1]
-    mova         m2, [r1+r2]
-    mova         m1, [r1+1]
-    mova         m3, [r1+r2+1]
-    pxor         m0, m6
-    pxor         m2, m6
-    pxor         m1, m6
-    pxor         m3, m6
-    PAVGB        m0, m1
-    PAVGB        m2, m3
-    pxor         m0, m6
-    pxor         m2, m6
-    mova       [r0], m0
-    mova    [r0+r2], m2
-    mova         m0, [r1+r2*2]
-    mova         m1, [r1+r2*2+1]
-    mova         m2, [r1+r4]
-    mova         m3, [r1+r4+1]
-    pxor         m0, m6
-    pxor         m1, m6
-    pxor         m2, m6
-    pxor         m3, m6
-    PAVGB        m0, m1
-    PAVGB        m2, m3
-    pxor         m0, m6
-    pxor         m2, m6
-    mova  [r0+r2*2], m0
-    mova    [r0+r4], m2
-    lea          r1, [r1+r2*4]
-    lea          r0, [r0+r2*4]
-    sub         r3d, 4
-    jg .loop
-    RET
-
-
-; void ff_put_no_rnd_pixels8_y2_exact(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
-INIT_MMX mmxext
-cglobal put_no_rnd_pixels8_y2_exact, 4,5
-    lea          r4, [r2*3]
-    mova         m0, [r1]
-    pcmpeqb      m6, m6
-    add          r1, r2
-    pxor         m0, m6
-.loop:
-    mova         m1, [r1]
-    mova         m2, [r1+r2]
-    pxor         m1, m6
-    pxor         m2, m6
-    PAVGB        m0, m1
-    PAVGB        m1, m2
-    pxor         m0, m6
-    pxor         m1, m6
-    mova       [r0], m0
-    mova    [r0+r2], m1
-    mova         m1, [r1+r2*2]
-    mova         m0, [r1+r4]
-    pxor         m1, m6
-    pxor         m0, m6
-    PAVGB        m2, m1
-    PAVGB        m1, m0
-    pxor         m2, m6
-    pxor         m1, m6
-    mova  [r0+r2*2], m2
-    mova    [r0+r4], m1
-    lea          r1, [r1+r2*4]
-    lea          r0, [r0+r2*4]
-    sub         r3d, 4
-    jg .loop
-    RET
diff --git a/libavcodec/x86/hpeldsp_vp3_init.c b/libavcodec/x86/hpeldsp_vp3_init.c
deleted file mode 100644
index 1dbd1ba6f9..0000000000
--- a/libavcodec/x86/hpeldsp_vp3_init.c
+++ /dev/null
@@ -1,43 +0,0 @@
-/*
- * This file is part of FFmpeg.
- *
- * 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.
- *
- * 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 FFmpeg; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
- */
-
-#include "libavutil/attributes.h"
-#include "libavutil/cpu.h"
-#include "libavutil/x86/cpu.h"
-
-#include "libavcodec/avcodec.h"
-#include "libavcodec/hpeldsp.h"
-
-#include "hpeldsp.h"
-
-void ff_put_no_rnd_pixels8_x2_exact_mmxext(uint8_t *block,
-                                           const uint8_t *pixels,
-                                           ptrdiff_t line_size, int h);
-void ff_put_no_rnd_pixels8_y2_exact_mmxext(uint8_t *block,
-                                           const uint8_t *pixels,
-                                           ptrdiff_t line_size, int h);
-
-av_cold void ff_hpeldsp_vp3_init_x86(HpelDSPContext *c, int cpu_flags, int flags)
-{
-    if (EXTERNAL_MMXEXT(cpu_flags)) {
-        if (flags & AV_CODEC_FLAG_BITEXACT) {
-            c->put_no_rnd_pixels_tab[1][1] = ff_put_no_rnd_pixels8_x2_exact_mmxext;
-            c->put_no_rnd_pixels_tab[1][2] = ff_put_no_rnd_pixels8_y2_exact_mmxext;
-        }
-    }
-}