Merge remote-tracking branch 'qatar/master'

* qatar/master: (29 commits)
  fate: add golomb-test
  golomb-test: K&R formatting cosmetics
  h264: Split h264-test off into a separate file - golomb-test.c.
  h264-test: cleanup: drop timer invocations, commented out code and other cruft
  h264-test: Remove unused DSP and AVCodec contexts and related init calls.
  adpcm: Add missing stdint.h #include to fix standalone header compilation.
  lavf: add functions for accessing the fourcc<->CodecID mapping tables.
  lavc: set AVCodecContext.codec in avcodec_get_context_defaults3().
  lavc: make avcodec_close() work properly on unopened codecs.
  lavc: add avcodec_is_open().
  lavf: rename AVInputFormat.value to raw_codec_id.
  lavf: remove the pointless value field from flv and iv8
  lavc/lavf: remove unnecessary symbols from the symbol version script.
  lavc: reorder AVCodec fields.
  lavf: reorder AVInput/OutputFormat fields.
  mp3dec: Fix a heap-buffer-overflow
  adpcmenc: remove some unneeded casts
  adpcmenc: use int16_t and uint8_t instead of short and unsigned char.
  adpcmenc: fix adpcm_ms extradata allocation
  adpcmenc: return proper AVERROR codes instead of -1
  ...

Conflicts:
	doc/APIchanges
	libavcodec/Makefile
	libavcodec/adpcmenc.c
	libavcodec/avcodec.h
	libavcodec/h264.c
	libavcodec/libavcodec.v
	libavcodec/mpc7.c
	libavcodec/mpegaudiodec.c
	libavcodec/options.c
	libavformat/Makefile
	libavformat/avformat.h
	libavformat/flvdec.c
	libavformat/libavformat.v

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

14 files changed, 436 insertions, 157 deletions

diff --git a/libavcodec/Makefile b/libavcodec/Makefile
index b8128efd4e..2173855e5b 100644
--- a/libavcodec/Makefile
+++ b/libavcodec/Makefile
@@ -737,7 +737,7 @@ SKIPHEADERS-$(CONFIG_VDPAU)            += vdpau.h
 SKIPHEADERS-$(CONFIG_XVMC)             += xvmc.h
 SKIPHEADERS-$(HAVE_W32THREADS)         += w32pthreads.h
 
-TESTPROGS = cabac dct fft fft-fixed h264 iirfilter rangecoder snowenc
+TESTPROGS = cabac dct fft fft-fixed golomb iirfilter rangecoder snowenc
 TESTPROGS-$(HAVE_MMX) += motion
 TESTOBJS = dctref.o
 
diff --git a/libavcodec/adpcm.h b/libavcodec/adpcm.h
index 475719adf5..08fd23f87b 100644
--- a/libavcodec/adpcm.h
+++ b/libavcodec/adpcm.h
@@ -26,18 +26,20 @@
 #ifndef AVCODEC_ADPCM_H
 #define AVCODEC_ADPCM_H
 
+#include <stdint.h>
+
 #define BLKSIZE 1024
 
 typedef struct ADPCMChannelStatus {
     int predictor;
-    short int step_index;
+    int16_t step_index;
     int step;
     /* for encoding */
     int prev_sample;
 
     /* MS version */
-    short sample1;
-    short sample2;
+    int16_t sample1;
+    int16_t sample2;
     int coeff1;
     int coeff2;
     int idelta;
diff --git a/libavcodec/adpcmenc.c b/libavcodec/adpcmenc.c
index d40b8a78d9..de85054288 100644
--- a/libavcodec/adpcmenc.c
+++ b/libavcodec/adpcmenc.c
@@ -65,12 +65,16 @@ static av_cold int adpcm_encode_init(AVCodecContext *avctx)
     ADPCMEncodeContext *s = avctx->priv_data;
     uint8_t *extradata;
     int i;
-    if (avctx->channels > 2)
-        return -1; /* only stereo or mono =) */
+    int ret = AVERROR(ENOMEM);
+
+    if (avctx->channels > 2) {
+        av_log(avctx, AV_LOG_ERROR, "only stereo or mono is supported\n");
+        return AVERROR(EINVAL);
+    }
 
     if (avctx->trellis && (unsigned)avctx->trellis > 16U) {
         av_log(avctx, AV_LOG_ERROR, "invalid trellis size\n");
-        return -1;
+        return AVERROR(EINVAL);
     }
 
     if (avctx->trellis) {
@@ -107,12 +111,12 @@ static av_cold int adpcm_encode_init(AVCodecContext *avctx)
         /* each 16 bits sample gives one nibble
            and we have 7 bytes per channel overhead */
         avctx->frame_size = (BLKSIZE - 7 * avctx->channels) * 2 / avctx->channels + 2;
-        avctx->block_align = BLKSIZE;
         avctx->bits_per_coded_sample = 4;
+        avctx->block_align    = BLKSIZE;
+        if (!(avctx->extradata = av_malloc(32 + FF_INPUT_BUFFER_PADDING_SIZE)))
+            goto error;
         avctx->extradata_size = 32;
-        extradata = avctx->extradata = av_malloc(avctx->extradata_size);
-        if (!extradata)
-            return AVERROR(ENOMEM);
+        extradata = avctx->extradata;
         bytestream_put_le16(&extradata, avctx->frame_size);
         bytestream_put_le16(&extradata, 7); /* wNumCoef */
         for (i = 0; i < 7; i++) {
@@ -130,22 +134,23 @@ static av_cold int adpcm_encode_init(AVCodecContext *avctx)
             avctx->sample_rate != 44100) {
             av_log(avctx, AV_LOG_ERROR, "Sample rate must be 11025, "
                    "22050 or 44100\n");
+            ret = AVERROR(EINVAL);
             goto error;
         }
         avctx->frame_size = 512 * (avctx->sample_rate / 11025);
         break;
     default:
+        ret = AVERROR(EINVAL);
         goto error;
     }
 
-    avctx->coded_frame = avcodec_alloc_frame();
-    if (!avctx->coded_frame)
+    if (!(avctx->coded_frame = avcodec_alloc_frame()))
         goto error;
 
     return 0;
 error:
     adpcm_encode_close(avctx);
-    return -1;
+    return ret;
 }
 
 static av_cold int adpcm_encode_close(AVCodecContext *avctx)
@@ -161,8 +166,8 @@ static av_cold int adpcm_encode_close(AVCodecContext *avctx)
 }
 
 
-static inline unsigned char adpcm_ima_compress_sample(ADPCMChannelStatus *c,
-                                                      short sample)
+static inline uint8_t adpcm_ima_compress_sample(ADPCMChannelStatus *c,
+                                                int16_t sample)
 {
     int delta  = sample - c->prev_sample;
     int nibble = FFMIN(7, abs(delta) * 4 /
@@ -174,8 +179,8 @@ static inline unsigned char adpcm_ima_compress_sample(ADPCMChannelStatus *c,
     return nibble;
 }
 
-static inline unsigned char adpcm_ima_qt_compress_sample(ADPCMChannelStatus *c,
-                                                         short sample)
+static inline uint8_t adpcm_ima_qt_compress_sample(ADPCMChannelStatus *c,
+                                                   int16_t sample)
 {
     int delta  = sample - c->prev_sample;
     int diff, step = ff_adpcm_step_table[c->step_index];
@@ -211,8 +216,8 @@ static inline unsigned char adpcm_ima_qt_compress_sample(ADPCMChannelStatus *c,
     return nibble;
 }
 
-static inline unsigned char adpcm_ms_compress_sample(ADPCMChannelStatus *c,
-                                                     short sample)
+static inline uint8_t adpcm_ms_compress_sample(ADPCMChannelStatus *c,
+                                               int16_t sample)
 {
     int predictor, nibble, bias;
 
@@ -228,20 +233,20 @@ static inline unsigned char adpcm_ms_compress_sample(ADPCMChannelStatus *c,
     nibble = (nibble + bias) / c->idelta;
     nibble = av_clip(nibble, -8, 7) & 0x0F;
 
-    predictor += (signed)((nibble & 0x08) ? (nibble - 0x10) : nibble) * c->idelta;
+    predictor += ((nibble & 0x08) ? (nibble - 0x10) : nibble) * c->idelta;
 
     c->sample2 = c->sample1;
     c->sample1 = av_clip_int16(predictor);
 
-    c->idelta = (ff_adpcm_AdaptationTable[(int)nibble] * c->idelta) >> 8;
+    c->idelta = (ff_adpcm_AdaptationTable[nibble] * c->idelta) >> 8;
     if (c->idelta < 16)
         c->idelta = 16;
 
     return nibble;
 }
 
-static inline unsigned char adpcm_yamaha_compress_sample(ADPCMChannelStatus *c,
-                                                         short sample)
+static inline uint8_t adpcm_yamaha_compress_sample(ADPCMChannelStatus *c,
+                                                   int16_t sample)
 {
     int nibble, delta;
 
@@ -262,8 +267,9 @@ static inline unsigned char adpcm_yamaha_compress_sample(ADPCMChannelStatus *c,
     return nibble;
 }
 
-static void adpcm_compress_trellis(AVCodecContext *avctx, const short *samples,
-                                   uint8_t *dst, ADPCMChannelStatus *c, int n)
+static void adpcm_compress_trellis(AVCodecContext *avctx,
+                                   const int16_t *samples, uint8_t *dst,
+                                   ADPCMChannelStatus *c, int n)
 {
     //FIXME 6% faster if frontier is a compile-time constant
     ADPCMEncodeContext *s = avctx->priv_data;
@@ -467,35 +473,35 @@ static void adpcm_compress_trellis(AVCodecContext *avctx, const short *samples,
     c->idelta     = nodes[0]->step;
 }
 
-static int adpcm_encode_frame(AVCodecContext *avctx,
-                              unsigned char *frame, int buf_size, void *data)
+static int adpcm_encode_frame(AVCodecContext *avctx, uint8_t *frame,
+                              int buf_size, void *data)
 {
     int n, i, st;
-    short *samples;
-    unsigned char *dst;
+    int16_t *samples;
+    uint8_t *dst;
     ADPCMEncodeContext *c = avctx->priv_data;
     uint8_t *buf;
 
     dst = frame;
-    samples = (short *)data;
+    samples = data;
     st = avctx->channels == 2;
     /* n = (BLKSIZE - 4 * avctx->channels) / (2 * 8 * avctx->channels); */
 
     switch(avctx->codec->id) {
     case CODEC_ID_ADPCM_IMA_WAV:
         n = avctx->frame_size / 8;
-        c->status[0].prev_sample = (signed short)samples[0]; /* XXX */
+        c->status[0].prev_sample = samples[0];
         /* c->status[0].step_index = 0;
         XXX: not sure how to init the state machine */
         bytestream_put_le16(&dst, c->status[0].prev_sample);
-        *dst++ = (unsigned char)c->status[0].step_index;
+        *dst++ = c->status[0].step_index;
         *dst++ = 0; /* unknown */
         samples++;
         if (avctx->channels == 2) {
-            c->status[1].prev_sample = (signed short)samples[0];
+            c->status[1].prev_sample = samples[0];
             /* c->status[1].step_index = 0; */
             bytestream_put_le16(&dst, c->status[1].prev_sample);
-            *dst++ = (unsigned char)c->status[1].step_index;
+            *dst++ = c->status[1].step_index;
             *dst++ = 0;
             samples++;
         }
@@ -595,7 +601,7 @@ static int adpcm_encode_frame(AVCodecContext *avctx,
             c->status[i].step_index = av_clip(c->status[i].step_index, 0, 63);
             put_sbits(&pb, 16, samples[i]);
             put_bits(&pb, 6, c->status[i].step_index);
-            c->status[i].prev_sample = (signed short)samples[i];
+            c->status[i].prev_sample = samples[i];
         }
 
         if (avctx->trellis > 0) {
@@ -692,10 +698,11 @@ static int adpcm_encode_frame(AVCodecContext *avctx,
             }
         break;
     default:
-    error:
-        return -1;
+        return AVERROR(EINVAL);
     }
     return dst - frame;
+error:
+    return AVERROR(ENOMEM);
 }
 
 
diff --git a/libavcodec/avcodec.h b/libavcodec/avcodec.h
index 4c5800b432..43c2128042 100644
--- a/libavcodec/avcodec.h
+++ b/libavcodec/avcodec.h
@@ -2848,31 +2848,20 @@ typedef struct AVCodec {
      * This is the primary way to find a codec from the user perspective.
      */
     const char *name;
+    /**
+     * Descriptive name for the codec, meant to be more human readable than name.
+     * You should use the NULL_IF_CONFIG_SMALL() macro to define it.
+     */
+    const char *long_name;
     enum AVMediaType type;
     enum CodecID id;
-    int priv_data_size;
-    int (*init)(AVCodecContext *);
-    int (*encode)(AVCodecContext *, uint8_t *buf, int buf_size, void *data);
-    int (*close)(AVCodecContext *);
-    int (*decode)(AVCodecContext *, void *outdata, int *outdata_size, AVPacket *avpkt);
     /**
      * Codec capabilities.
      * see CODEC_CAP_*
      */
     int capabilities;
-    struct AVCodec *next;
-    /**
-     * Flush buffers.
-     * Will be called when seeking
-     */
-    void (*flush)(AVCodecContext *);
     const AVRational *supported_framerates; ///< array of supported framerates, or NULL if any, array is terminated by {0,0}
     const enum PixelFormat *pix_fmts;       ///< array of supported pixel formats, or NULL if unknown, array is terminated by -1
-    /**
-     * Descriptive name for the codec, meant to be more human readable than name.
-     * You should use the NULL_IF_CONFIG_SMALL() macro to define it.
-     */
-    const char *long_name;
     const int *supported_samplerates;       ///< array of supported audio samplerates, or NULL if unknown, array is terminated by 0
     const enum AVSampleFormat *sample_fmts; ///< array of supported sample formats, or NULL if unknown, array is terminated by -1
     const uint64_t *channel_layouts;         ///< array of support channel layouts, or NULL if unknown. array is terminated by 0
@@ -2880,6 +2869,15 @@ typedef struct AVCodec {
     const AVClass *priv_class;              ///< AVClass for the private context
     const AVProfile *profiles;              ///< array of recognized profiles, or NULL if unknown, array is terminated by {FF_PROFILE_UNKNOWN}
 
+    /*****************************************************************
+     * No fields below this line are part of the public API. They
+     * may not be used outside of libavcodec and can be changed and
+     * removed at will.
+     * New public fields should be added right above.
+     *****************************************************************
+     */
+    int priv_data_size;
+    struct AVCodec *next;
     /**
      * @name Frame-level threading support functions
      * @{
@@ -2910,6 +2908,8 @@ typedef struct AVCodec {
      */
     void (*init_static_data)(struct AVCodec *codec);
 
+    int (*init)(AVCodecContext *);
+    int (*encode)(AVCodecContext *, uint8_t *buf, int buf_size, void *data);
     /**
      * Encode data to an AVPacket.
      *
@@ -2922,6 +2922,13 @@ typedef struct AVCodec {
      */
     int (*encode2)(AVCodecContext *avctx, AVPacket *avpkt, const AVFrame *frame,
                    int *got_packet_ptr);
+    int (*decode)(AVCodecContext *, void *outdata, int *outdata_size, AVPacket *avpkt);
+    int (*close)(AVCodecContext *);
+    /**
+     * Flush buffers.
+     * Will be called when seeking
+     */
+    void (*flush)(AVCodecContext *);
 } AVCodec;
 
 /**
@@ -3561,7 +3568,8 @@ AVCodecContext *avcodec_alloc_context2(enum AVMediaType);
 
 /**
  * Allocate an AVCodecContext and set its fields to default values.  The
- * resulting struct can be deallocated by simply calling av_free().
+ * resulting struct can be deallocated by calling avcodec_close() on it followed
+ * by av_free().
  *
  * @param codec if non-NULL, allocate private data and initialize defaults
  *              for the given codec. It is illegal to then call avcodec_open2()
@@ -3702,6 +3710,11 @@ int avcodec_open(AVCodecContext *avctx, AVCodec *codec);
  * @endcode
  *
  * @param avctx The context to initialize.
+ * @param codec The codec to open this context for. If a non-NULL codec has been
+ *              previously passed to avcodec_alloc_context3() or
+ *              avcodec_get_context_defaults3() for this context, then this
+ *              parameter MUST be either NULL or equal to the previously passed
+ *              codec.
  * @param options A dictionary filled with AVCodecContext and codec-private options.
  *                On return this object will be filled with options that were not found.
  *
@@ -3987,6 +4000,15 @@ int avcodec_encode_video(AVCodecContext *avctx, uint8_t *buf, int buf_size,
 int avcodec_encode_subtitle(AVCodecContext *avctx, uint8_t *buf, int buf_size,
                             const AVSubtitle *sub);
 
+/**
+ * Close a given AVCodecContext and free all the data associated with it
+ * (but not the AVCodecContext itself).
+ *
+ * Calling this function on an AVCodecContext that hasn't been opened will free
+ * the codec-specific data allocated in avcodec_alloc_context3() /
+ * avcodec_get_context_defaults3() with a non-NULL codec. Subsequent calls will
+ * do nothing.
+ */
 int avcodec_close(AVCodecContext *avctx);
 
 /**
@@ -4378,4 +4400,10 @@ const AVClass *avcodec_get_class(void);
  */
 const AVClass *avcodec_get_frame_class(void);
 
+/**
+ * @return a positive value if s is open (i.e. avcodec_open2() was called on it
+ * with no corresponding avcodec_close()), 0 otherwise.
+ */
+int avcodec_is_open(AVCodecContext *s);
+
 #endif /* AVCODEC_AVCODEC_H */
diff --git a/libavcodec/golomb-test.c b/libavcodec/golomb-test.c
new file mode 100644
index 0000000000..3dbf9d14cb
--- /dev/null
+++ b/libavcodec/golomb-test.c
@@ -0,0 +1,70 @@
+/*
+ * This file is part of Libav.
+ *
+ * Libav 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.
+ *
+ * Libav 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 Libav; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+#include <stdint.h>
+#include <stdio.h>
+
+#include "avcodec.h"
+#include "dsputil.h"
+#include "get_bits.h"
+#include "golomb.h"
+#include "put_bits.h"
+
+#undef printf
+#define COUNT 8000
+#define SIZE (COUNT * 40)
+
+int main(void)
+{
+    int i;
+    uint8_t temp[SIZE];
+    PutBitContext pb;
+    GetBitContext gb;
+
+    init_put_bits(&pb, temp, SIZE);
+    printf("testing unsigned exp golomb\n");
+    for (i = 0; i < COUNT; i++)
+        set_ue_golomb(&pb, i);
+    flush_put_bits(&pb);
+
+    init_get_bits(&gb, temp, 8 * SIZE);
+    for (i = 0; i < COUNT; i++) {
+        int j, s = show_bits(&gb, 24);
+
+        j = get_ue_golomb(&gb);
+        if (j != i)
+            printf("mismatch at %d (%d should be %d) bits: %6X\n", i, j, i, s);
+    }
+
+    init_put_bits(&pb, temp, SIZE);
+    printf("testing signed exp golomb\n");
+    for (i = 0; i < COUNT; i++)
+        set_se_golomb(&pb, i - COUNT / 2);
+    flush_put_bits(&pb);
+
+    init_get_bits(&gb, temp, 8 * SIZE);
+    for (i = 0; i < COUNT; i++) {
+        int j, s = show_bits(&gb, 24);
+
+        j = get_se_golomb(&gb);
+        if (j != i - COUNT / 2)
+            printf("mismatch at %d (%d should be %d) bits: %6X\n", i, j, i, s);
+    }
+
+    return 0;
+}
diff --git a/libavcodec/h264.c b/libavcodec/h264.c
index a3343da6e4..7a16bdaedc 100644
--- a/libavcodec/h264.c
+++ b/libavcodec/h264.c
@@ -4162,77 +4162,6 @@ static inline void fill_mb_avail(H264Context *h){
 }
 #endif
 
-#ifdef TEST
-#undef printf
-#undef random
-#define COUNT 8000
-#define SIZE (COUNT*40)
-extern AVCodec ff_h264_decoder;
-int main(void){
-    int i;
-    uint8_t temp[SIZE];
-    PutBitContext pb;
-    GetBitContext gb;
-    DSPContext dsp;
-    AVCodecContext avctx;
-
-    avcodec_get_context_defaults3(&avctx, &ff_h264_decoder);
-
-    dsputil_init(&dsp, &avctx);
-
-    init_put_bits(&pb, temp, SIZE);
-    printf("testing unsigned exp golomb\n");
-    for(i=0; i<COUNT; i++){
-        START_TIMER
-        set_ue_golomb(&pb, i);
-        STOP_TIMER("set_ue_golomb");
-    }
-    flush_put_bits(&pb);
-
-    init_get_bits(&gb, temp, 8*SIZE);
-    for(i=0; i<COUNT; i++){
-        int j, s = show_bits(&gb, 24);
-
-        {START_TIMER
-        j= get_ue_golomb(&gb);
-        if(j != i){
-            printf("mismatch! at %d (%d should be %d) bits:%6X\n", i, j, i, s);
-//            return -1;
-        }
-        STOP_TIMER("get_ue_golomb");}
-    }
-
-
-    init_put_bits(&pb, temp, SIZE);
-    printf("testing signed exp golomb\n");
-    for(i=0; i<COUNT; i++){
-        START_TIMER
-        set_se_golomb(&pb, i - COUNT/2);
-        STOP_TIMER("set_se_golomb");
-    }
-    flush_put_bits(&pb);
-
-    init_get_bits(&gb, temp, 8*SIZE);
-    for(i=0; i<COUNT; i++){
-        int j, s = show_bits(&gb, 24);
-
-        {START_TIMER
-        j= get_se_golomb(&gb);
-        if(j != i - COUNT/2){
-            printf("mismatch! at %d (%d should be %d) bits:%6X\n", i, j, i, s);
-//            return -1;
-        }
-        STOP_TIMER("get_se_golomb");}
-    }
-
-    printf("Testing RBSP\n");
-
-
-    return 0;
-}
-#endif /* TEST */
-
-
 av_cold void ff_h264_free_context(H264Context *h)
 {
     int i;
diff --git a/libavcodec/mpegaudiodec.c b/libavcodec/mpegaudiodec.c
index da34d84878..19d6f0d5eb 100644
--- a/libavcodec/mpegaudiodec.c
+++ b/libavcodec/mpegaudiodec.c
@@ -1380,18 +1380,17 @@ static int mp_decode_layer3(MPADecodeContext *s)
     if (!s->adu_mode) {
         int skip;
         const uint8_t *ptr = s->gb.buffer + (get_bits_count(&s->gb)>>3);
+        int extrasize = av_clip(get_bits_left(&s->gb) >> 3, 0, EXTRABYTES);
         assert((get_bits_count(&s->gb) & 7) == 0);
         /* now we get bits from the main_data_begin offset */
         av_dlog(s->avctx, "seekback: %d\n", main_data_begin);
     //av_log(NULL, AV_LOG_ERROR, "backstep:%d, lastbuf:%d\n", main_data_begin, s->last_buf_size);
 
-        if (s->gb.size_in_bits > get_bits_count(&s->gb))
-            memcpy(s->last_buf + s->last_buf_size, ptr,
-               FFMIN(EXTRABYTES, (s->gb.size_in_bits - get_bits_count(&s->gb))>>3));
+        memcpy(s->last_buf + s->last_buf_size, ptr, extrasize);
         s->in_gb = s->gb;
         init_get_bits(&s->gb, s->last_buf, s->last_buf_size*8);
 #if !UNCHECKED_BITSTREAM_READER
-        s->gb.size_in_bits_plus8 += EXTRABYTES * 8;
+        s->gb.size_in_bits_plus8 += extrasize * 8;
 #endif
         skip_bits_long(&s->gb, 8*(s->last_buf_size - main_data_begin));
     }
diff --git a/libavcodec/options.c b/libavcodec/options.c
index e1b8084bc3..d154da765b 100644
--- a/libavcodec/options.c
+++ b/libavcodec/options.c
@@ -526,7 +526,7 @@ AVCodecContext *avcodec_alloc_context(void){
 
 int avcodec_copy_context(AVCodecContext *dest, const AVCodecContext *src)
 {
-    if (dest->codec) { // check that the dest context is uninitialized
+    if (avcodec_is_open(dest)) { // check that the dest context is uninitialized
         av_log(dest, AV_LOG_ERROR,
                "Tried to copy AVCodecContext %p into already-initialized %p\n",
                src, dest);
diff --git a/libavcodec/utils.c b/libavcodec/utils.c
index a41c8a5596..0df3d7f8b0 100644
--- a/libavcodec/utils.c
+++ b/libavcodec/utils.c
@@ -700,6 +700,21 @@ int attribute_align_arg avcodec_open2(AVCodecContext *avctx, AVCodec *codec, AVD
     int ret = 0;
     AVDictionary *tmp = NULL;
 
+    if (avcodec_is_open(avctx))
+        return 0;
+
+    if ((!codec && !avctx->codec)) {
+        av_log(avctx, AV_LOG_ERROR, "No codec provided to avcodec_open2().\n");
+        return AVERROR(EINVAL);
+    }
+    if ((codec && avctx->codec && codec != avctx->codec)) {
+        av_log(avctx, AV_LOG_ERROR, "This AVCodecContext was allocated for %s, "
+               "but %s passed to avcodec_open2().\n", avctx->codec->name, codec->name);
+        return AVERROR(EINVAL);
+    }
+    if (!codec)
+        codec = avctx->codec;
+
     if (avctx->extradata_size < 0 || avctx->extradata_size >= FF_MAX_EXTRADATA_SIZE)
         return AVERROR(EINVAL);
 
@@ -719,11 +734,6 @@ int attribute_align_arg avcodec_open2(AVCodecContext *avctx, AVCodec *codec, AVD
         goto end;
     }
 
-    if(avctx->codec || !codec) {
-        ret = AVERROR(EINVAL);
-        goto end;
-    }
-
     avctx->internal = av_mallocz(sizeof(AVCodecInternal));
     if (!avctx->internal) {
         ret = AVERROR(ENOMEM);
@@ -1408,14 +1418,17 @@ av_cold int avcodec_close(AVCodecContext *avctx)
         return -1;
     }
 
-    if (HAVE_THREADS && avctx->thread_opaque)
-        ff_thread_free(avctx);
-    if (avctx->codec && avctx->codec->close)
-        avctx->codec->close(avctx);
-    avcodec_default_free_buffers(avctx);
-    avctx->coded_frame = NULL;
-    av_freep(&avctx->internal);
-    if (avctx->codec && avctx->codec->priv_class)
+    if (avcodec_is_open(avctx)) {
+        if (HAVE_THREADS && avctx->thread_opaque)
+            ff_thread_free(avctx);
+        if (avctx->codec && avctx->codec->close)
+            avctx->codec->close(avctx);
+        avcodec_default_free_buffers(avctx);
+        avctx->coded_frame = NULL;
+        av_freep(&avctx->internal);
+    }
+
+    if (avctx->priv_data && avctx->codec && avctx->codec->priv_class)
         av_opt_free(avctx->priv_data);
     av_opt_free(avctx);
     av_freep(&avctx->priv_data);
@@ -1976,3 +1989,8 @@ enum AVMediaType avcodec_get_type(enum CodecID codec_id)
 
     return AVMEDIA_TYPE_UNKNOWN;
 }
+
+int avcodec_is_open(AVCodecContext *s)
+{
+    return !!s->internal;
+}
diff --git a/libavcodec/wmalosslessdec.c b/libavcodec/wmalosslessdec.c
index 4b0a68b2d5..e6d202af3b 100644
--- a/libavcodec/wmalosslessdec.c
+++ b/libavcodec/wmalosslessdec.c
@@ -1550,15 +1550,14 @@ static void flush(AVCodecContext *avctx)
  *@brief wmall decoder
  */
 AVCodec ff_wmalossless_decoder = {
-    "wmalossless",
-    AVMEDIA_TYPE_AUDIO,
-    CODEC_ID_WMALOSSLESS,
-    sizeof(WmallDecodeCtx),
-    decode_init,
-    NULL,
-    decode_end,
-    decode_packet,
+    .name           = "wmalossless",
+    .type           = AVMEDIA_TYPE_AUDIO,
+    .id             = CODEC_ID_WMALOSSLESS,
+    .priv_data_size = sizeof(WmallDecodeCtx),
+    .init           = decode_init,
+    .close          = decode_end,
+    .decode         = decode_packet,
+    .flush          = flush,
     .capabilities = CODEC_CAP_SUBFRAMES | CODEC_CAP_EXPERIMENTAL,
-    .flush= flush,
     .long_name = NULL_IF_CONFIG_SMALL("Windows Media Audio 9 Lossless"),
 };
diff --git a/libavcodec/x86/Makefile b/libavcodec/x86/Makefile
index 282bc916bd..3b8ee56a49 100644
--- a/libavcodec/x86/Makefile
+++ b/libavcodec/x86/Makefile
@@ -29,8 +29,9 @@ MMX-OBJS-$(CONFIG_H264PRED)            += x86/h264_intrapred_init.o
 MMX-OBJS-$(CONFIG_RV30_DECODER)        += x86/rv34dsp_init.o
 YASM-OBJS-$(CONFIG_RV30_DECODER)       += x86/rv34dsp.o
 MMX-OBJS-$(CONFIG_RV40_DECODER)        += x86/rv34dsp_init.o            \
+                                          x86/rv40dsp_init.o
+YASM-OBJS-$(CONFIG_RV40_DECODER)       += x86/rv34dsp.o                 \
                                           x86/rv40dsp.o
-YASM-OBJS-$(CONFIG_RV40_DECODER)       += x86/rv34dsp.o
 
 YASM-OBJS-$(CONFIG_VC1_DECODER)        += x86/vc1dsp_yasm.o
 
diff --git a/libavcodec/x86/fmtconvert_mmx.c b/libavcodec/x86/fmtconvert_mmx.c
index a3d8f89816..ca0b29344a 100644
--- a/libavcodec/x86/fmtconvert_mmx.c
+++ b/libavcodec/x86/fmtconvert_mmx.c
@@ -110,9 +110,9 @@ static void float_interleave_sse(float *dst, const float **src,
 
 void ff_fmt_convert_init_x86(FmtConvertContext *c, AVCodecContext *avctx)
 {
+#if HAVE_YASM
     int mm_flags = av_get_cpu_flags();
 
-#if HAVE_YASM
     if (mm_flags & AV_CPU_FLAG_MMX) {
         c->float_interleave = float_interleave_mmx;
 
diff --git a/libavcodec/x86/rv40dsp.asm b/libavcodec/x86/rv40dsp.asm
new file mode 100644
index 0000000000..bff3e7b96a
--- /dev/null
+++ b/libavcodec/x86/rv40dsp.asm
@@ -0,0 +1,207 @@
+;******************************************************************************
+;* MMX/SSE2-optimized functions for the RV40 decoder
+;* Copyright (C) 2012 Christophe Gisquet <christophe.gisquet@gmail.com>
+;*
+;* This file is part of Libav.
+;*
+;* Libav 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.
+;*
+;* Libav 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 Libav; if not, write to the Free Software
+;* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+;******************************************************************************
+
+%include "x86inc.asm"
+%include "x86util.asm"
+
+SECTION_RODATA
+
+align 16
+shift_round:   times 8 dw 1 << (16 - 6)
+cextern pw_16
+
+SECTION .text
+
+; %1=5bits weights?, %2=dst %3=src1 %4=src3 %5=stride if sse2
+%macro RV40_WCORE  4-5
+    movh       m4, [%3 + 0]
+    movh       m5, [%4 + 0]
+%if %0 == 4
+%define OFFSET mmsize / 2
+%else
+    ; 8x8 block and sse2, stride was provided
+%define OFFSET %5
+%endif
+    movh       m6, [%3 + OFFSET]
+    movh       m7, [%4 + OFFSET]
+
+%if %1 == 0
+    ; 14bits weights
+    punpcklbw  m4, m0
+    punpcklbw  m5, m0
+    punpcklbw  m6, m0
+    punpcklbw  m7, m0
+
+    psllw      m4, 7
+    psllw      m5, 7
+    psllw      m6, 7
+    psllw      m7, 7
+    pmulhw     m4, m3
+    pmulhw     m5, m2
+    pmulhw     m6, m3
+    pmulhw     m7, m2
+
+    paddw      m4, m5
+    paddw      m6, m7
+%else
+    ; 5bits weights
+%if cpuflag(ssse3)
+    punpcklbw  m4, m5
+    punpcklbw  m6, m7
+
+    pmaddubsw  m4, m3
+    pmaddubsw  m6, m3
+%else
+    punpcklbw  m4, m0
+    punpcklbw  m5, m0
+    punpcklbw  m6, m0
+    punpcklbw  m7, m0
+
+    pmullw     m4, m3
+    pmullw     m5, m2
+    pmullw     m6, m3
+    pmullw     m7, m2
+    paddw      m4, m5
+    paddw      m6, m7
+%endif
+
+%endif
+
+    ; bias and shift down
+%if cpuflag(ssse3)
+    pmulhrsw   m4, m1
+    pmulhrsw   m6, m1
+%else
+    paddw      m4, m1
+    paddw      m6, m1
+    psrlw      m4, 5
+    psrlw      m6, 5
+%endif
+
+    packuswb   m4, m6
+%if %0 == 5
+    ; Only called for 8x8 blocks and sse2
+    movh       [%2 +  0], m4
+    movhps     [%2 + %5], m4
+%else
+    mova       [%2], m4
+%endif
+%endmacro
+
+
+%macro MAIN_LOOP   2
+%if mmsize == 8
+    RV40_WCORE %2, r0, r1, r2
+%if %1 == 16
+    RV40_WCORE %2, r0 + 8, r1 + 8, r2 + 8
+%endif
+
+    ; Prepare for next loop
+    add        r0, r5
+    add        r1, r5
+    add        r2, r5
+%else
+%ifidn %1, 8
+    RV40_WCORE %2, r0, r1, r2, r5
+    ; Prepare 2 next lines
+    lea        r0, [r0 + 2 * r5]
+    lea        r1, [r1 + 2 * r5]
+    lea        r2, [r2 + 2 * r5]
+%else
+    RV40_WCORE %2, r0, r1, r2
+    ; Prepare single next line
+    add        r0, r5
+    add        r1, r5
+    add        r2, r5
+%endif
+%endif
+
+    dec        r6
+%endmacro
+
+; rv40_weight_func_%1(uint8_t *dst, uint8_t *src1, uint8_t *src2, int w1, int w2, int stride)
+; %1=size  %2=num of xmm regs
+%macro RV40_WEIGHT  2
+cglobal rv40_weight_func_%1, 6, 7, %2
+%if cpuflag(ssse3)
+    mova       m1, [shift_round]
+%else
+    mova       m1, [pw_16]
+%endif
+    pxor       m0, m0
+    mov        r6, r3
+    or         r6, r4
+    ; The weights are FP0.14 notation of fractions depending on pts.
+    ; For timebases without rounding error (i.e. PAL), the fractions
+    ; can be simplified, and several operations can be avoided.
+    ; Therefore, we check here whether they are multiples of 2^9 for
+    ; those simplifications to occur.
+    and        r6, 0x1FF
+    ; Set loop counter and increments
+%if mmsize == 8
+    mov        r6, %1
+%else
+    mov        r6, (%1 * %1) / mmsize
+%endif
+
+    ; Use result of test now
+    jz .loop_512
+    movd       m2, r3
+    movd       m3, r4
+    SPLATW     m2, m2
+    SPLATW     m3, m3
+
+.loop:
+    MAIN_LOOP  %1, 0
+    jnz        .loop
+    REP_RET
+
+    ; Weights are multiple of 512, which allows some shortcuts
+.loop_512:
+    sar        r3, 9
+    sar        r4, 9
+    movd       m2, r3
+    movd       m3, r4
+%if cpuflag(ssse3)
+    punpcklbw  m3, m2
+    SPLATW     m3, m3
+%else
+    SPLATW     m2, m2
+    SPLATW     m3, m3
+%endif
+.loop2:
+    MAIN_LOOP  %1, 1
+    jnz        .loop2
+    REP_RET
+
+%endmacro
+
+INIT_MMX mmx
+RV40_WEIGHT    8, 0
+RV40_WEIGHT   16, 0
+
+INIT_XMM sse2
+RV40_WEIGHT    8, 8
+RV40_WEIGHT   16, 8
+
+INIT_XMM ssse3
+RV40_WEIGHT    8, 8
+RV40_WEIGHT   16, 8
diff --git a/libavcodec/x86/rv40dsp.c b/libavcodec/x86/rv40dsp_init.c
index 9f90ad8bb6..3d6c6f0fa0 100644
--- a/libavcodec/x86/rv40dsp.c
+++ b/libavcodec/x86/rv40dsp_init.c
@@ -40,14 +40,25 @@ void ff_avg_rv40_chroma_mc4_mmx2 (uint8_t *dst, uint8_t *src,
 void ff_avg_rv40_chroma_mc4_3dnow(uint8_t *dst, uint8_t *src,
                                   int stride, int h, int x, int y);
 
+#define DECLARE_WEIGHT(opt) \
+void ff_rv40_weight_func_16_##opt(uint8_t *dst, uint8_t *src1, uint8_t *src2, \
+                                  int w1, int w2, int stride); \
+void ff_rv40_weight_func_8_##opt (uint8_t *dst, uint8_t *src1, uint8_t *src2, \
+                                  int w1, int w2, int stride);
+DECLARE_WEIGHT(mmx)
+DECLARE_WEIGHT(sse2)
+DECLARE_WEIGHT(ssse3)
+
 void ff_rv40dsp_init_x86(RV34DSPContext *c, DSPContext *dsp)
 {
-    av_unused int mm_flags = av_get_cpu_flags();
-
 #if HAVE_YASM
+    int mm_flags = av_get_cpu_flags();
+
     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] = ff_rv40_weight_func_16_mmx;
+        c->rv40_weight_pixels_tab[1] = ff_rv40_weight_func_8_mmx;
     }
     if (mm_flags & AV_CPU_FLAG_MMX2) {
         c->avg_chroma_pixels_tab[0] = ff_avg_rv40_chroma_mc8_mmx2;
@@ -56,5 +67,13 @@ void ff_rv40dsp_init_x86(RV34DSPContext *c, DSPContext *dsp)
         c->avg_chroma_pixels_tab[0] = ff_avg_rv40_chroma_mc8_3dnow;
         c->avg_chroma_pixels_tab[1] = ff_avg_rv40_chroma_mc4_3dnow;
     }
+    if (mm_flags & AV_CPU_FLAG_SSE2) {
+        c->rv40_weight_pixels_tab[0] = ff_rv40_weight_func_16_sse2;
+        c->rv40_weight_pixels_tab[1] = ff_rv40_weight_func_8_sse2;
+    }
+    if (mm_flags & AV_CPU_FLAG_SSSE3) {
+        c->rv40_weight_pixels_tab[0] = ff_rv40_weight_func_16_ssse3;
+        c->rv40_weight_pixels_tab[1] = ff_rv40_weight_func_8_ssse3;
+    }
 #endif
 }