Merge remote-tracking branch 'qatar/master'

* qatar/master:
  fate: Work around non-standard wc implementations at more places
  fate: work around non-standard wc implementations
  x86: rv40: Mark rv40_weight functions as MMX2; they use MMX2 instructions.
  ac3dsp: simplify x86 versions of ac3_max_msb_abs_int16
  fate: use standard diff options
  tta: Fix comment about channel number; TTA supports >2 channels.
  avfilter: Move ff_get_ref_perms_string() to where it is used.
  build: Add 'check' target to run all compile and test targets.
  indeo3: validate new frame size before resetting decoder
  indeo3: when freeing buffers, set pointers referencing them to NULL as well
  indeo3: initialise pixel planes on allocation
  indeo3: ensure that decoded cell data is in 7-bit range as presumed by decoder
  fate: rename psx-str-v3-mdec to mdec-v3
  fate: convert psx-str to a demuxer test
  lavf: add mdec to is_intra_only() list

Conflicts:
	doc/developer.texi
	libavcodec/indeo3.c
	libavfilter/video.c
	libavformat/utils.c
	tests/fate/demux.mak
	tests/fate/video.mak
	tests/lavf-regression.sh
	tests/ref/vsynth1/cljr
	tests/ref/vsynth1/ffvhuff
	tests/ref/vsynth2/cljr
	tests/ref/vsynth2/ffvhuff

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

4 files changed, 49 insertions, 32 deletions

diff --git a/libavcodec/x86/ac3dsp.asm b/libavcodec/x86/ac3dsp.asm
index 99e121ec68..4e0fce35d1 100644
--- a/libavcodec/x86/ac3dsp.asm
+++ b/libavcodec/x86/ac3dsp.asm
@@ -91,12 +91,36 @@ AC3_EXPONENT_MIN sse2
 ;        This is used for mmxext and sse2 because they have pminsw/pmaxsw.
 ;-----------------------------------------------------------------------------
 
-%macro AC3_MAX_MSB_ABS_INT16 2
-cglobal ac3_max_msb_abs_int16_%1, 2,2,5, src, len
+; logical 'or' of 4 or 8 words in an mmx or xmm register into the low word
+%macro OR_WORDS_HORIZ 2 ; src, tmp
+%if cpuflag(sse2)
+    movhlps     %2, %1
+    por         %1, %2
+    pshuflw     %2, %1, q0032
+    por         %1, %2
+    pshuflw     %2, %1, q0001
+    por         %1, %2
+%elif cpuflag(mmx2)
+    pshufw      %2, %1, q0032
+    por         %1, %2
+    pshufw      %2, %1, q0001
+    por         %1, %2
+%else ; mmx
+    movq        %2, %1
+    psrlq       %2, 32
+    por         %1, %2
+    movq        %2, %1
+    psrlq       %2, 16
+    por         %1, %2
+%endif
+%endmacro
+
+%macro AC3_MAX_MSB_ABS_INT16 1
+cglobal ac3_max_msb_abs_int16, 2,2,5, src, len
     pxor        m2, m2
     pxor        m3, m3
 .loop:
-%ifidn %2, min_max
+%ifidn %1, min_max
     mova        m0, [srcq]
     mova        m1, [srcq+mmsize]
     pminsw      m2, m0
@@ -104,7 +128,7 @@ cglobal ac3_max_msb_abs_int16_%1, 2,2,5, src, len
     pmaxsw      m3, m0
     pmaxsw      m3, m1
 %else ; or_abs
-%ifidn %1, mmx
+%if notcpuflag(ssse3)
     mova        m0, [srcq]
     mova        m1, [srcq+mmsize]
     ABS2        m0, m1, m3, m4
@@ -119,34 +143,27 @@ cglobal ac3_max_msb_abs_int16_%1, 2,2,5, src, len
     add       srcq, mmsize*2
     sub       lend, mmsize
     ja .loop
-%ifidn %2, min_max
+%ifidn %1, min_max
     ABS2        m2, m3, m0, m1
     por         m2, m3
 %endif
-%ifidn mmsize, 16
-    movhlps     m0, m2
-    por         m2, m0
-%endif
-    PSHUFLW     m0, m2, 0xe
-    por         m2, m0
-    PSHUFLW     m0, m2, 0x1
-    por         m2, m0
+    OR_WORDS_HORIZ m2, m0
     movd       eax, m2
     and        eax, 0xFFFF
     RET
 %endmacro
 
-INIT_MMX
+INIT_MMX mmx
 %define ABS2 ABS2_MMX
-%define PSHUFLW pshufw
-AC3_MAX_MSB_ABS_INT16 mmx, or_abs
+AC3_MAX_MSB_ABS_INT16 or_abs
+INIT_MMX mmx2
 %define ABS2 ABS2_MMX2
-AC3_MAX_MSB_ABS_INT16 mmxext, min_max
-INIT_XMM
-%define PSHUFLW pshuflw
-AC3_MAX_MSB_ABS_INT16 sse2, min_max
+AC3_MAX_MSB_ABS_INT16 min_max
+INIT_XMM sse2
+AC3_MAX_MSB_ABS_INT16 min_max
+INIT_XMM ssse3
 %define ABS2 ABS2_SSSE3
-AC3_MAX_MSB_ABS_INT16 ssse3, or_abs
+AC3_MAX_MSB_ABS_INT16 or_abs
 
 ;-----------------------------------------------------------------------------
 ; macro used for ff_ac3_lshift_int16() and ff_ac3_rshift_int32()
diff --git a/libavcodec/x86/ac3dsp_mmx.c b/libavcodec/x86/ac3dsp_mmx.c
index 9578e98d8b..54fa380130 100644
--- a/libavcodec/x86/ac3dsp_mmx.c
+++ b/libavcodec/x86/ac3dsp_mmx.c
@@ -27,10 +27,10 @@ extern void ff_ac3_exponent_min_mmx   (uint8_t *exp, int num_reuse_blocks, int n
 extern void ff_ac3_exponent_min_mmxext(uint8_t *exp, int num_reuse_blocks, int nb_coefs);
 extern void ff_ac3_exponent_min_sse2  (uint8_t *exp, int num_reuse_blocks, int nb_coefs);
 
-extern int ff_ac3_max_msb_abs_int16_mmx   (const int16_t *src, int len);
-extern int ff_ac3_max_msb_abs_int16_mmxext(const int16_t *src, int len);
-extern int ff_ac3_max_msb_abs_int16_sse2  (const int16_t *src, int len);
-extern int ff_ac3_max_msb_abs_int16_ssse3 (const int16_t *src, int len);
+extern int ff_ac3_max_msb_abs_int16_mmx  (const int16_t *src, int len);
+extern int ff_ac3_max_msb_abs_int16_mmx2 (const int16_t *src, int len);
+extern int ff_ac3_max_msb_abs_int16_sse2 (const int16_t *src, int len);
+extern int ff_ac3_max_msb_abs_int16_ssse3(const int16_t *src, int len);
 
 extern void ff_ac3_lshift_int16_mmx (int16_t *src, unsigned int len, unsigned int shift);
 extern void ff_ac3_lshift_int16_sse2(int16_t *src, unsigned int len, unsigned int shift);
@@ -67,7 +67,7 @@ av_cold void ff_ac3dsp_init_x86(AC3DSPContext *c, int bit_exact)
     }
     if (mm_flags & AV_CPU_FLAG_MMX2 && HAVE_MMX2) {
         c->ac3_exponent_min = ff_ac3_exponent_min_mmxext;
-        c->ac3_max_msb_abs_int16 = ff_ac3_max_msb_abs_int16_mmxext;
+        c->ac3_max_msb_abs_int16 = ff_ac3_max_msb_abs_int16_mmx2;
     }
     if (mm_flags & AV_CPU_FLAG_SSE && HAVE_SSE) {
         c->float_to_fixed24 = ff_float_to_fixed24_sse;
diff --git a/libavcodec/x86/rv40dsp.asm b/libavcodec/x86/rv40dsp.asm
index 9b50940d4d..70c0c0400f 100644
--- a/libavcodec/x86/rv40dsp.asm
+++ b/libavcodec/x86/rv40dsp.asm
@@ -487,7 +487,7 @@ cglobal rv40_weight_func_%1_%2, 6, 7, 8
     REP_RET
 %endmacro
 
-INIT_MMX mmx
+INIT_MMX mmx2
 RV40_WEIGHT   rnd,    8, 3
 RV40_WEIGHT   rnd,   16, 4
 RV40_WEIGHT   nornd,  8, 3
diff --git a/libavcodec/x86/rv40dsp_init.c b/libavcodec/x86/rv40dsp_init.c
index 3f42363e4e..cc1ea45e88 100644
--- a/libavcodec/x86/rv40dsp_init.c
+++ b/libavcodec/x86/rv40dsp_init.c
@@ -52,7 +52,7 @@ void ff_rv40_weight_func_nornd_16_##opt(uint8_t *dst, uint8_t *src1, uint8_t *sr
                                         int w1, int w2, ptrdiff_t stride); \
 void ff_rv40_weight_func_nornd_8_##opt (uint8_t *dst, uint8_t *src1, uint8_t *src2, \
                                         int w1, int w2, ptrdiff_t stride);
-DECLARE_WEIGHT(mmx)
+DECLARE_WEIGHT(mmx2)
 DECLARE_WEIGHT(sse2)
 DECLARE_WEIGHT(ssse3)
 
@@ -190,10 +190,6 @@ void ff_rv40dsp_init_x86(RV34DSPContext *c, DSPContext *dsp)
     if (mm_flags & AV_CPU_FLAG_MMX) {
         c->put_chroma_pixels_tab[0] = ff_put_rv40_chroma_mc8_mmx;
         c->put_chroma_pixels_tab[1] = ff_put_rv40_chroma_mc4_mmx;
-        c->rv40_weight_pixels_tab[0][0] = ff_rv40_weight_func_rnd_16_mmx;
-        c->rv40_weight_pixels_tab[0][1] = ff_rv40_weight_func_rnd_8_mmx;
-        c->rv40_weight_pixels_tab[1][0] = ff_rv40_weight_func_nornd_16_mmx;
-        c->rv40_weight_pixels_tab[1][1] = ff_rv40_weight_func_nornd_8_mmx;
         c->put_pixels_tab[0][15] = ff_put_rv40_qpel16_mc33_mmx;
         c->put_pixels_tab[1][15] = ff_put_rv40_qpel8_mc33_mmx;
         c->avg_pixels_tab[0][15] = ff_avg_rv40_qpel16_mc33_mmx;
@@ -205,6 +201,10 @@ void ff_rv40dsp_init_x86(RV34DSPContext *c, DSPContext *dsp)
     if (mm_flags & AV_CPU_FLAG_MMX2) {
         c->avg_chroma_pixels_tab[0] = ff_avg_rv40_chroma_mc8_mmx2;
         c->avg_chroma_pixels_tab[1] = ff_avg_rv40_chroma_mc4_mmx2;
+        c->rv40_weight_pixels_tab[0][0] = ff_rv40_weight_func_rnd_16_mmx2;
+        c->rv40_weight_pixels_tab[0][1] = ff_rv40_weight_func_rnd_8_mmx2;
+        c->rv40_weight_pixels_tab[1][0] = ff_rv40_weight_func_nornd_16_mmx2;
+        c->rv40_weight_pixels_tab[1][1] = ff_rv40_weight_func_nornd_8_mmx2;
 #if ARCH_X86_32
         QPEL_MC_SET(avg_, _mmx2)
 #endif