Merge remote-tracking branch 'qatar/master'

* qatar/master:
  x86: fft: fix imdct_half() for AVX
  rtmppkt: Add missing libavcodec/bytestream.h include.
  rtmp: add functions for reading AMF values
  vc1dec: remove useless #include simple_idct.h
  dct-test: always link with aandcttab.o
  vp8: pack struct VP8ThreadData more efficiently
  x86: remove libmpeg2 mmx(ext) idct functions
  eamad: Use dsputils instead of a custom bswap16_buf
  Canopus Lossless decoder

Conflicts:
	Changelog
	LICENSE
	libavcodec/avcodec.h
	libavcodec/cllc.c
	libavcodec/eamad.c
	libavcodec/version.h

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

6 files changed, 30 insertions, 24 deletions

diff --git a/libavcodec/Makefile b/libavcodec/Makefile
index b21819bc20..793a9c8312 100644
--- a/libavcodec/Makefile
+++ b/libavcodec/Makefile
@@ -829,7 +829,7 @@ HOSTPROGS = aac_tablegen                                                \
 
 CLEANFILES = *_tables.c *_tables.h *_tablegen$(HOSTEXESUF)
 
-$(SUBDIR)dct-test$(EXESUF): $(SUBDIR)dctref.o
+$(SUBDIR)dct-test$(EXESUF): $(SUBDIR)dctref.o $(SUBDIR)aandcttab.o
 
 TRIG_TABLES  = cos cos_fixed sin
 TRIG_TABLES := $(TRIG_TABLES:%=$(SUBDIR)%_tables.c)
diff --git a/libavcodec/avcodec.h b/libavcodec/avcodec.h
index 7a7400d6c6..cd933fd7cb 100644
--- a/libavcodec/avcodec.h
+++ b/libavcodec/avcodec.h
@@ -261,6 +261,7 @@ enum CodecID {
     CODEC_ID_MSA1,
     CODEC_ID_TSCC2,
     CODEC_ID_MTS2,
+    CODEC_ID_CLLC,
     CODEC_ID_Y41P       = MKBETAG('Y','4','1','P'),
     CODEC_ID_ESCAPE130  = MKBETAG('E','1','3','0'),
     CODEC_ID_EXR        = MKBETAG('0','E','X','R'),
@@ -274,7 +275,6 @@ enum CodecID {
     CODEC_ID_YUV4       = MKBETAG('Y','U','V','4'),
     CODEC_ID_SANM       = MKBETAG('S','A','N','M'),
     CODEC_ID_PAF_VIDEO  = MKBETAG('P','A','F','V'),
-    CODEC_ID_CLLC       = MKBETAG('C','L','L','C'),
 
     /* various PCM "codecs" */
     CODEC_ID_FIRST_AUDIO = 0x10000,     ///< A dummy id pointing at the start of audio codecs
diff --git a/libavcodec/vc1dec.c b/libavcodec/vc1dec.c
index f5fbe54c82..55d695bee2 100644
--- a/libavcodec/vc1dec.c
+++ b/libavcodec/vc1dec.c
@@ -36,7 +36,6 @@
 #include "vc1acdata.h"
 #include "msmpeg4data.h"
 #include "unary.h"
-#include "simple_idct.h"
 #include "mathops.h"
 #include "vdpau_internal.h"
 #include "libavutil/avassert.h"
diff --git a/libavcodec/version.h b/libavcodec/version.h
index 48096ecb32..cb26975eed 100644
--- a/libavcodec/version.h
+++ b/libavcodec/version.h
@@ -27,7 +27,7 @@
  */
 
 #define LIBAVCODEC_VERSION_MAJOR 54
-#define LIBAVCODEC_VERSION_MINOR 47
+#define LIBAVCODEC_VERSION_MINOR 48
 #define LIBAVCODEC_VERSION_MICRO 100
 
 #define LIBAVCODEC_VERSION_INT  AV_VERSION_INT(LIBAVCODEC_VERSION_MAJOR, \
diff --git a/libavcodec/vp8.h b/libavcodec/vp8.h
index 23d62af7fa..4f026dfe8d 100644
--- a/libavcodec/vp8.h
+++ b/libavcodec/vp8.h
@@ -94,21 +94,8 @@ typedef struct {
 } VP8Macroblock;
 
 typedef struct {
-#if HAVE_THREADS
-    pthread_mutex_t lock;
-    pthread_cond_t  cond;
-#endif
-    int thread_nr;
-    int thread_mb_pos; // (mb_y << 16) | (mb_x & 0xFFFF)
-    int wait_mb_pos; // What the current thread is waiting on.
-    uint8_t *edge_emu_buffer;
-    /**
-     * For coeff decode, we need to know whether the above block had non-zero
-     * coefficients. This means for each macroblock, we need data for 4 luma
-     * blocks, 2 u blocks, 2 v blocks, and the luma dc block, for a total of 9
-     * per macroblock. We keep the last row in top_nnz.
-     */
-    DECLARE_ALIGNED(8, uint8_t, left_nnz)[9];
+    DECLARE_ALIGNED(16, DCTELEM, block)[6][4][16];
+    DECLARE_ALIGNED(16, DCTELEM, block_dc)[16];
     /**
      * This is the index plus one of the last non-zero coeff
      * for each of the blocks in the current macroblock.
@@ -117,8 +104,21 @@ typedef struct {
      *     2+-> full transform
      */
     DECLARE_ALIGNED(16, uint8_t, non_zero_count_cache)[6][4];
-    DECLARE_ALIGNED(16, DCTELEM, block)[6][4][16];
-    DECLARE_ALIGNED(16, DCTELEM, block_dc)[16];
+    /**
+     * For coeff decode, we need to know whether the above block had non-zero
+     * coefficients. This means for each macroblock, we need data for 4 luma
+     * blocks, 2 u blocks, 2 v blocks, and the luma dc block, for a total of 9
+     * per macroblock. We keep the last row in top_nnz.
+     */
+    DECLARE_ALIGNED(8, uint8_t, left_nnz)[9];
+    int thread_nr;
+#if HAVE_THREADS
+    pthread_mutex_t lock;
+    pthread_cond_t  cond;
+#endif
+    int thread_mb_pos; // (mb_y << 16) | (mb_x & 0xFFFF)
+    int wait_mb_pos; // What the current thread is waiting on.
+    uint8_t *edge_emu_buffer;
     VP8FilterStrength *filter_strength;
 } VP8ThreadData;
 
diff --git a/libavcodec/x86/fft_mmx.asm b/libavcodec/x86/fft_mmx.asm
index 8392817bed..73234388ab 100644
--- a/libavcodec/x86/fft_mmx.asm
+++ b/libavcodec/x86/fft_mmx.asm
@@ -1010,7 +1010,11 @@ cglobal imdct_half, 3,12,8; FFTContext *s, FFTSample *output, const FFTSample *i
     push  rrevtab
 %endif
 
-    sub   r3, mmsize/4
+%if mmsize == 8
+    sub   r3, 2
+%else
+    sub   r3, 4
+%endif
 %if ARCH_X86_64 || mmsize == 8
     xor   r4, r4
     sub   r4, r3
@@ -1037,7 +1041,9 @@ cglobal imdct_half, 3,12,8; FFTContext *s, FFTSample *output, const FFTSample *i
     mova [r1+r5*8], m0
     mova [r1+r6*8], m2
     add    r4, 2
-%elif ARCH_X86_64
+    sub    r4, 2
+%else
+%if ARCH_X86_64
     movzx  r5,  word [rrevtab+r4-4]
     movzx  r6,  word [rrevtab+r4-2]
     movzx  r10, word [rrevtab+r3]
@@ -1058,7 +1064,8 @@ cglobal imdct_half, 3,12,8; FFTContext *s, FFTSample *output, const FFTSample *i
     movlps [r1+r5*8], xmm1
     movhps [r1+r4*8], xmm1
 %endif
-    sub    r3, mmsize/4
+    sub    r3, 4
+%endif
     jns    .pre
 
     mov  r5, r0