Merge remote-tracking branch 'qatar/master'

* qatar/master: (24 commits)
  flvdec: remove incomplete, disabled seeking code
  mem: add support for _aligned_malloc() as found on Windows
  lavc: Extend the documentation for avcodec_init_packet
  flvdec: remove incomplete, disabled seeking code
  http: replace atoll() with strtoll()
  mpegts: remove unused/incomplete/broken seeking code
  af_amix: allow float planar sample format as input
  af_amix: use AVFloatDSPContext.vector_fmac_scalar()
  float_dsp: add x86-optimized functions for vector_fmac_scalar()
  float_dsp: Move vector_fmac_scalar() from libavcodec to libavutil
  lavr: Add x86-optimized function for flt to s32 conversion
  lavr: Add x86-optimized function for flt to s16 conversion
  lavr: Add x86-optimized functions for s32 to flt conversion
  lavr: Add x86-optimized functions for s32 to s16 conversion
  lavr: Add x86-optimized functions for s16 to flt conversion
  lavr: Add x86-optimized function for s16 to s32 conversion
  rtpenc: Support packetizing iLBC
  rtpdec: Add a depacketizer for iLBC
  Implement the iLBC storage file format
  mov: Support muxing/demuxing iLBC
  ...

Conflicts:
	Changelog
	configure
	libavcodec/avcodec.h
	libavcodec/dsputil.c
	libavcodec/version.h
	libavformat/movenc.c
	libavformat/mpegts.c
	libavformat/version.h
	libavutil/mem.c

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

7 files changed, 137 insertions, 0 deletions

diff --git a/libavutil/arm/float_dsp_init_neon.c b/libavutil/arm/float_dsp_init_neon.c
index fa6d0d7d15..3ca0288b31 100644
--- a/libavutil/arm/float_dsp_init_neon.c
+++ b/libavutil/arm/float_dsp_init_neon.c
@@ -26,7 +26,11 @@
 
 void ff_vector_fmul_neon(float *dst, const float *src0, const float *src1, int len);
 
+void ff_vector_fmac_scalar_neon(float *dst, const float *src, float mul,
+                                int len);
+
 void ff_float_dsp_init_neon(AVFloatDSPContext *fdsp)
 {
     fdsp->vector_fmul = ff_vector_fmul_neon;
+    fdsp->vector_fmac_scalar = ff_vector_fmac_scalar_neon;
 }
diff --git a/libavutil/arm/float_dsp_neon.S b/libavutil/arm/float_dsp_neon.S
index d66fa09424..03b164388f 100644
--- a/libavutil/arm/float_dsp_neon.S
+++ b/libavutil/arm/float_dsp_neon.S
@@ -62,3 +62,51 @@ function ff_vector_fmul_neon, export=1
 3:      vst1.32         {d16-d19},[r0,:128]!
         bx              lr
 endfunc
+
+function ff_vector_fmac_scalar_neon, export=1
+VFP     len .req r2
+VFP     acc .req r3
+NOVFP   len .req r3
+NOVFP   acc .req r2
+VFP     vdup.32         q15, d0[0]
+NOVFP   vdup.32         q15, r2
+        bics            r12, len, #15
+        mov             acc, r0
+        beq             3f
+        vld1.32         {q0},     [r1,:128]!
+        vld1.32         {q8},     [acc,:128]!
+        vld1.32         {q1},     [r1,:128]!
+        vld1.32         {q9},     [acc,:128]!
+1:      vmla.f32        q8,  q0,  q15
+        vld1.32         {q2},     [r1,:128]!
+        vld1.32         {q10},    [acc,:128]!
+        vmla.f32        q9,  q1,  q15
+        vld1.32         {q3},     [r1,:128]!
+        vld1.32         {q11},    [acc,:128]!
+        vmla.f32        q10, q2,  q15
+        vst1.32         {q8},     [r0,:128]!
+        vmla.f32        q11, q3,  q15
+        vst1.32         {q9},     [r0,:128]!
+        subs            r12, r12, #16
+        beq             2f
+        vld1.32         {q0},     [r1,:128]!
+        vld1.32         {q8},     [acc,:128]!
+        vst1.32         {q10},    [r0,:128]!
+        vld1.32         {q1},     [r1,:128]!
+        vld1.32         {q9},     [acc,:128]!
+        vst1.32         {q11},    [r0,:128]!
+        b               1b
+2:      vst1.32         {q10},    [r0,:128]!
+        vst1.32         {q11},    [r0,:128]!
+        ands            len, len, #15
+        it              eq
+        bxeq            lr
+3:      vld1.32         {q0},     [r1,:128]!
+        vld1.32         {q8},     [acc,:128]!
+        vmla.f32        q8,  q0,  q15
+        vst1.32         {q8},     [r0,:128]!
+        subs            len, len, #4
+        bgt             3b
+        bx              lr
+        .unreq          len
+endfunc
diff --git a/libavutil/float_dsp.c b/libavutil/float_dsp.c
index 87cfd88268..f5a8360c86 100644
--- a/libavutil/float_dsp.c
+++ b/libavutil/float_dsp.c
@@ -31,9 +31,18 @@ static void vector_fmul_c(float *dst, const float *src0, const float *src1,
         dst[i] = src0[i] * src1[i];
 }
 
+static void vector_fmac_scalar_c(float *dst, const float *src, float mul,
+                                 int len)
+{
+    int i;
+    for (i = 0; i < len; i++)
+        dst[i] += src[i] * mul;
+}
+
 void avpriv_float_dsp_init(AVFloatDSPContext *fdsp, int bit_exact)
 {
     fdsp->vector_fmul = vector_fmul_c;
+    fdsp->vector_fmac_scalar = vector_fmac_scalar_c;
 
 #if ARCH_ARM
     ff_float_dsp_init_arm(fdsp);
diff --git a/libavutil/float_dsp.h b/libavutil/float_dsp.h
index 02c4ab7bde..735eb34c36 100644
--- a/libavutil/float_dsp.h
+++ b/libavutil/float_dsp.h
@@ -35,6 +35,22 @@ typedef struct AVFloatDSPContext {
      */
     void (*vector_fmul)(float *dst, const float *src0, const float *src1,
                         int len);
+
+    /**
+     * Multiply a vector of floats by a scalar float and add to
+     * destination vector.  Source and destination vectors must
+     * overlap exactly or not at all.
+     *
+     * @param dst result vector
+     *            constraints: 32-byte aligned
+     * @param src input vector
+     *            constraints: 32-byte aligned
+     * @param mul scalar value
+     * @param len length of vector
+     *            constraints: multiple of 16
+     */
+    void (*vector_fmac_scalar)(float *dst, const float *src, float mul,
+                               int len);
 } AVFloatDSPContext;
 
 /**
diff --git a/libavutil/mem.c b/libavutil/mem.c
index de22ad8db8..385ace0702 100644
--- a/libavutil/mem.c
+++ b/libavutil/mem.c
@@ -94,6 +94,8 @@ void *av_malloc(size_t size)
     if (size) //OS X on SDK 10.6 has a broken posix_memalign implementation
     if (posix_memalign(&ptr,ALIGN,size))
         ptr = NULL;
+#elif HAVE_ALIGNED_MALLOC
+    ptr = _aligned_malloc(size, ALIGN);
 #elif HAVE_MEMALIGN
     ptr = memalign(ALIGN,size);
     /* Why 64?
@@ -145,6 +147,8 @@ void *av_realloc(void *ptr, size_t size)
     ptr= realloc((char*)ptr - diff, size + diff);
     if(ptr) ptr = (char*)ptr + diff;
     return ptr;
+#elif HAVE_ALIGNED_MALLOC
+    return _aligned_realloc(ptr, size + !size, ALIGN);
 #else
     return realloc(ptr, size + !size);
 #endif
@@ -170,6 +174,8 @@ void av_free(void *ptr)
 #if CONFIG_MEMALIGN_HACK
     if (ptr)
         free((char*)ptr - ((char*)ptr)[-1]);
+#elif HAVE_ALIGNED_MALLOC
+    _aligned_free(ptr);
 #else
     free(ptr);
 #endif
diff --git a/libavutil/x86/float_dsp.asm b/libavutil/x86/float_dsp.asm
index 6ed716c026..f68e0bfe2d 100644
--- a/libavutil/x86/float_dsp.asm
+++ b/libavutil/x86/float_dsp.asm
@@ -21,6 +21,7 @@
 ;******************************************************************************
 
 %include "x86inc.asm"
+%include "x86util.asm"
 
 SECTION .text
 
@@ -55,3 +56,49 @@ VECTOR_FMUL
 INIT_YMM avx
 VECTOR_FMUL
 %endif
+
+;------------------------------------------------------------------------------
+; void ff_vector_fmac_scalar(float *dst, const float *src, float mul, int len)
+;------------------------------------------------------------------------------
+
+%macro VECTOR_FMAC_SCALAR 0
+%if UNIX64
+cglobal vector_fmac_scalar, 3,3,3, dst, src, len
+%else
+cglobal vector_fmac_scalar, 4,4,3, dst, src, mul, len
+%endif
+%if WIN64
+    SWAP 0, 2
+%endif
+%if ARCH_X86_32
+    VBROADCASTSS m0, mulm
+%else
+    shufps     xmm0, xmm0, 0
+%if cpuflag(avx)
+    vinsertf128  m0, m0, xmm0, 1
+%endif
+%endif
+    lea    lenq, [lend*4-2*mmsize]
+.loop
+    mulps    m1, m0, [srcq+lenq       ]
+    mulps    m2, m0, [srcq+lenq+mmsize]
+    addps    m1, m1, [dstq+lenq       ]
+    addps    m2, m2, [dstq+lenq+mmsize]
+    mova  [dstq+lenq       ], m1
+    mova  [dstq+lenq+mmsize], m2
+    sub    lenq, 2*mmsize
+    jge .loop
+%if mmsize == 32
+    vzeroupper
+    RET
+%else
+    REP_RET
+%endif
+%endmacro
+
+INIT_XMM sse
+VECTOR_FMAC_SCALAR
+%if HAVE_AVX
+INIT_YMM avx
+VECTOR_FMAC_SCALAR
+%endif
diff --git a/libavutil/x86/float_dsp_init.c b/libavutil/x86/float_dsp_init.c
index 8f6980cbc2..3e05b9d4ca 100644
--- a/libavutil/x86/float_dsp_init.c
+++ b/libavutil/x86/float_dsp_init.c
@@ -26,6 +26,11 @@ extern void ff_vector_fmul_sse(float *dst, const float *src0, const float *src1,
 extern void ff_vector_fmul_avx(float *dst, const float *src0, const float *src1,
                                int len);
 
+extern void ff_vector_fmac_scalar_sse(float *dst, const float *src, float mul,
+                                      int len);
+extern void ff_vector_fmac_scalar_avx(float *dst, const float *src, float mul,
+                                      int len);
+
 void ff_float_dsp_init_x86(AVFloatDSPContext *fdsp)
 {
 #if HAVE_YASM
@@ -33,9 +38,11 @@ void ff_float_dsp_init_x86(AVFloatDSPContext *fdsp)
 
     if (mm_flags & AV_CPU_FLAG_SSE && HAVE_SSE) {
         fdsp->vector_fmul = ff_vector_fmul_sse;
+        fdsp->vector_fmac_scalar = ff_vector_fmac_scalar_sse;
     }
     if (mm_flags & AV_CPU_FLAG_AVX && HAVE_AVX) {
         fdsp->vector_fmul = ff_vector_fmul_avx;
+        fdsp->vector_fmac_scalar = ff_vector_fmac_scalar_avx;
     }
 #endif
 }