Merge remote-tracking branch 'qatar/master'

* qatar/master:
  proresdsp: port x86 assembly to cpuflags.
  lavr: x86: improve non-SSE4 version of S16_TO_S32_SX macro
  lavfi: better channel layout negotiation
  alac: check for truncated packets
  alac: reverse lpc coeff order, simplify filter
  lavr: add x86-optimized mixing functions
  x86: add support for fmaddps fma4 instruction with abstraction to avx/sse
  tscc2: fix typo in array index
  build: use COMPILE template for HOSTOBJS
  build: do full flag handling for all compiler-type tools
  eval: fix printing of NaN in eval fate test.
  build: Rename aandct component to more descriptive aandcttables
  mpegaudio: bury inline asm under HAVE_INLINE_ASM.
  x86inc: automatically insert vzeroupper for YMM functions.
  rtmp: Check the buffer length of ping packets
  rtmp: Allow having more unknown data at the end of a chunk size packet without failing
  rtmp: Prevent reading outside of an allocate buffer when receiving server bandwidth packets

Conflicts:
	Makefile
	configure
	libavcodec/x86/proresdsp.asm
	libavutil/eval.c

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

3 files changed, 25 insertions, 21 deletions

diff --git a/libavutil/eval.c b/libavutil/eval.c
index fa76c6c949..6aa257efc4 100644
--- a/libavutil/eval.c
+++ b/libavutil/eval.c
@@ -797,11 +797,10 @@ int main(int argc, char **argv)
         av_expr_parse_and_eval(&d, *expr,
                                const_names, const_values,
                                NULL, NULL, NULL, NULL, NULL, 0, NULL);
-        if(isnan(d)){
+        if (isnan(d))
             printf("'%s' -> nan\n\n", *expr);
-        }else{
+        else
             printf("'%s' -> %f\n\n", *expr, d);
-        }
     }
 
     av_expr_parse_and_eval(&d, "1+(5-2)^(3-1)+1/2+sin(PI)-max(-2.2,-3.1)",
diff --git a/libavutil/x86/float_dsp.asm b/libavutil/x86/float_dsp.asm
index f68e0bfe2d..7a18a20aca 100644
--- a/libavutil/x86/float_dsp.asm
+++ b/libavutil/x86/float_dsp.asm
@@ -42,12 +42,7 @@ ALIGN 16
 
     sub       lenq, 2*mmsize
     jge       .loop
-%if mmsize == 32
-    vzeroupper
-    RET
-%else
     REP_RET
-%endif
 %endmacro
 
 INIT_XMM sse
@@ -88,12 +83,7 @@ cglobal vector_fmac_scalar, 4,4,3, dst, src, mul, len
     mova  [dstq+lenq+mmsize], m2
     sub    lenq, 2*mmsize
     jge .loop
-%if mmsize == 32
-    vzeroupper
-    RET
-%else
     REP_RET
-%endif
 %endmacro
 
 INIT_XMM sse
diff --git a/libavutil/x86/x86inc.asm b/libavutil/x86/x86inc.asm
index 9a39df6ec8..c80e0a1c1a 100644
--- a/libavutil/x86/x86inc.asm
+++ b/libavutil/x86/x86inc.asm
@@ -392,11 +392,14 @@ DECLARE_REG 14, R15, R15D, R15W, R15B, 120
 %macro RET 0
     WIN64_RESTORE_XMM_INTERNAL rsp
     POP_IF_USED 14, 13, 12, 11, 10, 9, 8, 7
+%if mmsize == 32
+    vzeroupper
+%endif
     ret
 %endmacro
 
 %macro REP_RET 0
-    %if regs_used > 7 || xmm_regs_used > 6
+    %if regs_used > 7 || xmm_regs_used > 6 || mmsize == 32
         RET
     %else
         rep ret
@@ -433,11 +436,14 @@ DECLARE_REG 14, R15, R15D, R15W, R15B, 72
 
 %macro RET 0
     POP_IF_USED 14, 13, 12, 11, 10, 9
+%if mmsize == 32
+    vzeroupper
+%endif
     ret
 %endmacro
 
 %macro REP_RET 0
-    %if regs_used > 9
+    %if regs_used > 9 || mmsize == 32
         RET
     %else
         rep ret
@@ -479,11 +485,14 @@ DECLARE_ARG 7, 8, 9, 10, 11, 12, 13, 14
 
 %macro RET 0
     POP_IF_USED 6, 5, 4, 3
+%if mmsize == 32
+    vzeroupper
+%endif
     ret
 %endmacro
 
 %macro REP_RET 0
-    %if regs_used > 3
+    %if regs_used > 3 || mmsize == 32
         RET
     %else
         rep ret
@@ -1126,16 +1135,22 @@ AVX_INSTR pfmul, 1, 0, 1
 %undef j
 
 %macro FMA_INSTR 3
-    %macro %1 4-7 %1, %2, %3
-        %if cpuflag(xop)
-            v%5 %1, %2, %3, %4
+    %macro %1 5-8 %1, %2, %3
+        %if cpuflag(xop) || cpuflag(fma4)
+            v%6 %1, %2, %3, %4
         %else
-            %6 %1, %2, %3
-            %7 %1, %4
+            %ifidn %1, %4
+                %7 %5, %2, %3
+                %8 %1, %4, %5
+            %else
+                %7 %1, %2, %3
+                %8 %1, %4
+            %endif
         %endif
     %endmacro
 %endmacro
 
+FMA_INSTR  fmaddps,   mulps, addps
 FMA_INSTR  pmacsdd,  pmulld, paddd
 FMA_INSTR  pmacsww,  pmullw, paddw
 FMA_INSTR pmadcswd, pmaddwd, paddd