Merge remote-tracking branch 'qatar/master'

* qatar/master:
  drawtext: remove typo
  pcm-mpeg: implement new audio decoding api
  w32thread: port fixes to pthread_cond_broadcast() from x264.
  doc: add editor configuration section with Vim and Emacs settings
  dxva2.h: include d3d9.h to define LPDIRECT3DSURFACE9
  avformat/utils: Drop unused goto label.
  doxygen: Replace '\' by '@' in Doxygen markup tags.
  cosmetics: drop some completely pointless parentheses
  cljr: simplify CLJRContext
  drawtext: introduce rand(min, max)
  drawtext: introduce explicit draw/hide variable
  rtmp: Use nb_invokes for all invoke commands

Conflicts:
	libavcodec/mpegvideo.c
	libavfilter/vf_drawtext.c

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

12 files changed, 117 insertions, 54 deletions

diff --git a/libavcodec/atrac3.c b/libavcodec/atrac3.c
index ccbb718e00..b9f1f516cc 100644
--- a/libavcodec/atrac3.c
+++ b/libavcodec/atrac3.c
@@ -708,9 +708,10 @@ static int decodeChannelSoundUnit (ATRAC3Context *q, GetBitContext *gb, channel_
             memset(pSnd->IMDCT_buf, 0, 512 * sizeof(float));
 
         /* gain compensation and overlapping */
-        gainCompensateAndOverlap (pSnd->IMDCT_buf, &(pSnd->prevFrame[band*256]), &(pOut[band*256]),
-                                    &((pSnd->gainBlock[1 - (pSnd->gcBlkSwitch)]).gBlock[band]),
-                                    &((pSnd->gainBlock[pSnd->gcBlkSwitch]).gBlock[band]));
+        gainCompensateAndOverlap(pSnd->IMDCT_buf, &pSnd->prevFrame[band * 256],
+                                 &pOut[band * 256],
+                                 &pSnd->gainBlock[1 - pSnd->gcBlkSwitch].gBlock[band],
+                                 &pSnd->gainBlock[    pSnd->gcBlkSwitch].gBlock[band]);
     }
 
     /* Swap the gain control buffers for the next frame. */
@@ -795,7 +796,9 @@ static int decodeFrame(ATRAC3Context *q, const uint8_t* databuf,
         for (i=0 ; i<q->channels ; i++) {
 
             /* Set the bitstream reader at the start of a channel sound unit. */
-            init_get_bits(&q->gb, databuf+((i*q->bytes_per_frame)/q->channels), (q->bits_per_frame)/q->channels);
+            init_get_bits(&q->gb,
+                          databuf + i * q->bytes_per_frame / q->channels,
+                          q->bits_per_frame / q->channels);
 
             result = decodeChannelSoundUnit(q,&q->gb, &q->pUnits[i], out_samples[i], i, q->codingMode);
             if (result != 0)
diff --git a/libavcodec/cljr.c b/libavcodec/cljr.c
index 6e352b8e61..53ab40cbe2 100644
--- a/libavcodec/cljr.c
+++ b/libavcodec/cljr.c
@@ -35,9 +35,6 @@
 typedef struct CLJRContext{
     AVCodecContext *avctx;
     AVFrame picture;
-    int delta[16];
-    int offset[4];
-    GetBitContext gb;
 } CLJRContext;
 
 static int decode_frame(AVCodecContext *avctx,
@@ -47,6 +44,7 @@ static int decode_frame(AVCodecContext *avctx,
     const uint8_t *buf = avpkt->data;
     int buf_size = avpkt->size;
     CLJRContext * const a = avctx->priv_data;
+    GetBitContext gb;
     AVFrame *picture = data;
     AVFrame * const p= (AVFrame*)&a->picture;
     int x, y;
@@ -67,20 +65,20 @@ static int decode_frame(AVCodecContext *avctx,
     p->pict_type= AV_PICTURE_TYPE_I;
     p->key_frame= 1;
 
-    init_get_bits(&a->gb, buf, buf_size * 8);
+    init_get_bits(&gb, buf, buf_size * 8);
 
     for(y=0; y<avctx->height; y++){
         uint8_t *luma= &a->picture.data[0][ y*a->picture.linesize[0] ];
         uint8_t *cb= &a->picture.data[1][ y*a->picture.linesize[1] ];
         uint8_t *cr= &a->picture.data[2][ y*a->picture.linesize[2] ];
         for(x=0; x<avctx->width; x+=4){
-                luma[3] = get_bits(&a->gb, 5) << 3;
-            luma[2] = get_bits(&a->gb, 5) << 3;
-            luma[1] = get_bits(&a->gb, 5) << 3;
-            luma[0] = get_bits(&a->gb, 5) << 3;
+            luma[3] = get_bits(&gb, 5) << 3;
+            luma[2] = get_bits(&gb, 5) << 3;
+            luma[1] = get_bits(&gb, 5) << 3;
+            luma[0] = get_bits(&gb, 5) << 3;
             luma+= 4;
-            *(cb++) = get_bits(&a->gb, 6) << 2;
-            *(cr++) = get_bits(&a->gb, 6) << 2;
+            *(cb++) = get_bits(&gb, 6) << 2;
+            *(cr++) = get_bits(&gb, 6) << 2;
         }
     }
 
diff --git a/libavcodec/dxva2.h b/libavcodec/dxva2.h
index 6eb494b9fe..fc99560830 100644
--- a/libavcodec/dxva2.h
+++ b/libavcodec/dxva2.h
@@ -25,6 +25,7 @@
 
 #include <stdint.h>
 
+#include <d3d9.h>
 #include <dxva2api.h>
 
 #define FF_DXVA2_WORKAROUND_SCALING_LIST_ZIGZAG 1 ///< Work around for DXVA2 and old UVD/UVD+ ATI video cards
diff --git a/libavcodec/mpegvideo.c b/libavcodec/mpegvideo.c
index 9c55af725e..666cad8151 100644
--- a/libavcodec/mpegvideo.c
+++ b/libavcodec/mpegvideo.c
@@ -840,10 +840,14 @@ av_cold int MPV_common_init(MpegEncContext *s)
         FF_ALLOCZ_OR_GOTO(s->avctx, s->prev_pict_types, PREV_PICT_TYPES_BUFFER_SIZE, fail);
 
         s->parse_context.state = -1;
-        if ((s->avctx->debug & (FF_DEBUG_VIS_QP | FF_DEBUG_VIS_MB_TYPE)) || (s->avctx->debug_mv)) {
-            s->visualization_buffer[0] = av_malloc((s->mb_width * 16 + 2 * EDGE_WIDTH) * s->mb_height * 16 + 2 * EDGE_WIDTH);
-            s->visualization_buffer[1] = av_malloc((s->mb_width * 16 + 2 * EDGE_WIDTH) * s->mb_height * 16 + 2 * EDGE_WIDTH);
-            s->visualization_buffer[2] = av_malloc((s->mb_width * 16 + 2 * EDGE_WIDTH) * s->mb_height * 16 + 2 * EDGE_WIDTH);
+        if ((s->avctx->debug & (FF_DEBUG_VIS_QP | FF_DEBUG_VIS_MB_TYPE)) ||
+            s->avctx->debug_mv) {
+            s->visualization_buffer[0] = av_malloc((s->mb_width * 16 +
+                        2 * EDGE_WIDTH) * s->mb_height * 16 + 2 * EDGE_WIDTH);
+            s->visualization_buffer[1] = av_malloc((s->mb_width * 16 +
+                        2 * EDGE_WIDTH) * s->mb_height * 16 + 2 * EDGE_WIDTH);
+            s->visualization_buffer[2] = av_malloc((s->mb_width * 16 +
+                        2 * EDGE_WIDTH) * s->mb_height * 16 + 2 * EDGE_WIDTH);
         }
 
         s->context_initialized = 1;
@@ -1512,7 +1516,8 @@ void ff_print_debug_info(MpegEncContext *s, AVFrame *pict){
         }
     }
 
-    if((s->avctx->debug&(FF_DEBUG_VIS_QP|FF_DEBUG_VIS_MB_TYPE)) || (s->avctx->debug_mv)){
+    if ((s->avctx->debug & (FF_DEBUG_VIS_QP | FF_DEBUG_VIS_MB_TYPE)) ||
+        s->avctx->debug_mv) {
         const int shift= 1 + s->quarter_sample;
         int mb_y;
         uint8_t *ptr;
@@ -1538,7 +1543,7 @@ void ff_print_debug_info(MpegEncContext *s, AVFrame *pict){
             int mb_x;
             for(mb_x=0; mb_x<s->mb_width; mb_x++){
                 const int mb_index= mb_x + mb_y*s->mb_stride;
-                if((s->avctx->debug_mv) && pict->motion_val){
+                if (s->avctx->debug_mv && pict->motion_val) {
                   int type;
                   for(type=0; type<3; type++){
                     int direction = 0;
diff --git a/libavcodec/mpegvideo_common.h b/libavcodec/mpegvideo_common.h
index bf6c4db580..a7badd1ffb 100644
--- a/libavcodec/mpegvideo_common.h
+++ b/libavcodec/mpegvideo_common.h
@@ -585,7 +585,7 @@ static inline void chroma_4mv_motion(MpegEncContext *s,
     if (src_y == (s->height >> 1))
         dxy &= ~2;
 
-    offset = (src_y * (s->uvlinesize)) + src_x;
+    offset = src_y * s->uvlinesize + src_x;
     ptr = ref_picture[1] + offset;
     if(s->flags&CODEC_FLAG_EMU_EDGE){
         if(   (unsigned)src_x > (s->h_edge_pos>>1) - (dxy &1) - 8
diff --git a/libavcodec/msrledec.c b/libavcodec/msrledec.c
index 97510830d5..db8de7032d 100644
--- a/libavcodec/msrledec.c
+++ b/libavcodec/msrledec.c
@@ -138,8 +138,8 @@ static int msrle_decode_8_16_24_32(AVCodecContext *avctx, AVPicture *pic, int de
     uint32_t av_uninit(pix32);
     unsigned int width= FFABS(pic->linesize[0]) / (depth >> 3);
 
-    output = pic->data[0] + (avctx->height - 1) * pic->linesize[0];
-    output_end = pic->data[0] + (avctx->height) * pic->linesize[0];
+    output     = pic->data[0] + (avctx->height - 1) * pic->linesize[0];
+    output_end = pic->data[0] +  avctx->height      * pic->linesize[0];
     while(src < data + srcsize) {
         p1 = *src++;
         if(p1 == 0) { //Escape code
diff --git a/libavcodec/pcm-mpeg.c b/libavcodec/pcm-mpeg.c
index 5d3dd7036a..b9417733ca 100644
--- a/libavcodec/pcm-mpeg.c
+++ b/libavcodec/pcm-mpeg.c
@@ -121,17 +121,30 @@ static int pcm_bluray_parse_header(AVCodecContext *avctx,
     return 0;
 }
 
-static int pcm_bluray_decode_frame(AVCodecContext *avctx,
-                                   void *data,
-                                   int *data_size,
-                                   AVPacket *avpkt)
+typedef struct PCMBRDecode {
+    AVFrame frame;
+} PCMBRDecode;
+
+static av_cold int pcm_bluray_decode_init(AVCodecContext * avctx)
+{
+    PCMBRDecode *s = avctx->priv_data;
+
+    avcodec_get_frame_defaults(&s->frame);
+    avctx->coded_frame = &s->frame;
+
+    return 0;
+}
+
+static int pcm_bluray_decode_frame(AVCodecContext *avctx, void *data,
+                                   int *got_frame_ptr, AVPacket *avpkt)
 {
     const uint8_t *src = avpkt->data;
     int buf_size = avpkt->size;
+    PCMBRDecode *s = avctx->priv_data;
     int num_source_channels, channel, retval;
-    int sample_size, samples, output_size;
-    int16_t *dst16 = data;
-    int32_t *dst32 = data;
+    int sample_size, samples;
+    int16_t *dst16;
+    int32_t *dst32;
 
     if (buf_size < 4) {
         av_log(avctx, AV_LOG_ERROR, "PCM packet too small\n");
@@ -148,15 +161,14 @@ static int pcm_bluray_decode_frame(AVCodecContext *avctx,
     sample_size = (num_source_channels * avctx->bits_per_coded_sample) >> 3;
     samples = buf_size / sample_size;
 
-    output_size = samples * avctx->channels *
-                  (avctx->sample_fmt == AV_SAMPLE_FMT_S32 ? 4 : 2);
-    if (output_size > *data_size) {
-        av_log(avctx, AV_LOG_ERROR,
-               "Insufficient output buffer space (%d bytes, needed %d bytes)\n",
-               *data_size, output_size);
-        return -1;
+    /* get output buffer */
+    s->frame.nb_samples = samples;
+    if ((retval = avctx->get_buffer(avctx, &s->frame)) < 0) {
+        av_log(avctx, AV_LOG_ERROR, "get_buffer() failed\n");
+        return retval;
     }
-    *data_size = output_size;
+    dst16 = (int16_t *)s->frame.data[0];
+    dst32 = (int32_t *)s->frame.data[0];
 
     if (samples) {
         switch (avctx->channel_layout) {
@@ -167,7 +179,7 @@ static int pcm_bluray_decode_frame(AVCodecContext *avctx,
             samples *= num_source_channels;
             if (AV_SAMPLE_FMT_S16 == avctx->sample_fmt) {
 #if HAVE_BIGENDIAN
-                memcpy(dst16, src, output_size);
+                memcpy(dst16, src, buf_size);
 #else
                 do {
                     *dst16++ = bytestream_get_be16(&src);
@@ -291,10 +303,13 @@ static int pcm_bluray_decode_frame(AVCodecContext *avctx,
         }
     }
 
+    *got_frame_ptr   = 1;
+    *(AVFrame *)data = s->frame;
+
     retval = src - avpkt->data;
     if (avctx->debug & FF_DEBUG_BITSTREAM)
         av_dlog(avctx, "pcm_bluray_decode_frame: decoded %d -> %d bytes\n",
-                retval, *data_size);
+                retval, buf_size);
     return retval;
 }
 
@@ -302,7 +317,10 @@ AVCodec ff_pcm_bluray_decoder = {
     .name           = "pcm_bluray",
     .type           = AVMEDIA_TYPE_AUDIO,
     .id             = CODEC_ID_PCM_BLURAY,
+    .priv_data_size = sizeof(PCMBRDecode),
+    .init           = pcm_bluray_decode_init,
     .decode         = pcm_bluray_decode_frame,
+    .capabilities   = CODEC_CAP_DR1,
     .sample_fmts = (const enum AVSampleFormat[]){AV_SAMPLE_FMT_S16, AV_SAMPLE_FMT_S32,
                                          AV_SAMPLE_FMT_NONE},
     .long_name = NULL_IF_CONFIG_SMALL("PCM signed 16|20|24-bit big-endian for Blu-ray media"),
diff --git a/libavcodec/ppc/h264_altivec.c b/libavcodec/ppc/h264_altivec.c
index bf4fd0f016..9b3622a673 100644
--- a/libavcodec/ppc/h264_altivec.c
+++ b/libavcodec/ppc/h264_altivec.c
@@ -616,8 +616,8 @@ static inline void write16x4(uint8_t *dst, int dst_stride,
     *(dst_int+15*int_dst_stride) = *(src_int + 15);
 }
 
-/** \brief performs a 6x16 transpose of data in src, and stores it to dst
-    \todo FIXME: see if we can't spare some vec_lvsl() by them factorizing
+/** @brief performs a 6x16 transpose of data in src, and stores it to dst
+    @todo FIXME: see if we can't spare some vec_lvsl() by them factorizing
     out of unaligned_load() */
 #define readAndTranspose16x6(src, src_stride, r8, r9, r10, r11, r12, r13) {\
     register vec_u8 r0  = unaligned_load(0,             src);            \
diff --git a/libavcodec/ppc/util_altivec.h b/libavcodec/ppc/util_altivec.h
index cb638dff08..6800348570 100644
--- a/libavcodec/ppc/util_altivec.h
+++ b/libavcodec/ppc/util_altivec.h
@@ -94,7 +94,7 @@ do { \
 } while (0)
 
 
-/** \brief loads unaligned vector \a *src with offset \a offset
+/** @brief loads unaligned vector @a *src with offset @a offset
     and returns it */
 static inline vector unsigned char unaligned_load(int offset, uint8_t *src)
 {
diff --git a/libavcodec/tableprint.h b/libavcodec/tableprint.h
index e126a72afb..1b39dc6ba1 100644
--- a/libavcodec/tableprint.h
+++ b/libavcodec/tableprint.h
@@ -58,7 +58,7 @@ void write_##type##_2d_array(const void *arg, int len, int len2)\
 /**
  * @name Predefined functions for printing tables
  *
- * \{
+ * @{
  */
 void write_int8_t_array     (const int8_t   *, int);
 void write_uint8_t_array    (const uint8_t  *, int);
@@ -69,7 +69,7 @@ void write_int8_t_2d_array  (const void *, int, int);
 void write_uint8_t_2d_array (const void *, int, int);
 void write_uint32_t_2d_array(const void *, int, int);
 void write_float_2d_array   (const void *, int, int);
-/** \} */ // end of printfuncs group
+/** @} */ // end of printfuncs group
 
 #define WRITE_ARRAY(prefix, type, name)                 \
     do {                                                \
diff --git a/libavcodec/vc1dec.c b/libavcodec/vc1dec.c
index 6f1af89e22..20b7f72291 100644
--- a/libavcodec/vc1dec.c
+++ b/libavcodec/vc1dec.c
@@ -236,7 +236,7 @@ static void vc1_put_signed_blocks_clamped(VC1Context *v)
         if (s->mb_x) {
             topleft_mb_pos = (s->mb_y - 1) * s->mb_stride + s->mb_x - 1;
             fieldtx        = v->fieldtx_plane[topleft_mb_pos];
-            stride_y       = (s->linesize) << fieldtx;
+            stride_y       = s->linesize << fieldtx;
             v_dist         = (16 - fieldtx) >> (fieldtx == 0);
             s->dsp.put_signed_pixels_clamped(v->block[v->topleft_blk_idx][0],
                                              s->dest[0] - 16 * s->linesize - 16,
diff --git a/libavcodec/w32pthreads.h b/libavcodec/w32pthreads.h
index c015b87a42..2d1470c5ee 100644
--- a/libavcodec/w32pthreads.h
+++ b/libavcodec/w32pthreads.h
@@ -115,9 +115,12 @@ static inline int pthread_mutex_unlock(pthread_mutex_t *m)
 /* for pre-Windows 6.0 platforms we need to define and use our own condition
  * variable and api */
 typedef struct {
+    pthread_mutex_t mtx_broadcast;
     pthread_mutex_t mtx_waiter_count;
     volatile int waiter_count;
     HANDLE semaphore;
+    HANDLE waiters_done;
+    int is_broadcast;
 } win32_cond_t;
 
 static void pthread_cond_init(pthread_cond_t *cond, const void *unused_attr)
@@ -136,8 +139,12 @@ static void pthread_cond_init(pthread_cond_t *cond, const void *unused_attr)
     win32_cond->semaphore = CreateSemaphore(NULL, 0, 0x7fffffff, NULL);
     if (!win32_cond->semaphore)
         return;
+    win32_cond->waiters_done = CreateEvent(NULL, FALSE, FALSE, NULL);
+    if (!win32_cond->waiters_done)
+        return;
 
     pthread_mutex_init(&win32_cond->mtx_waiter_count, NULL);
+    pthread_mutex_init(&win32_cond->mtx_broadcast, NULL);
 }
 
 static void pthread_cond_destroy(pthread_cond_t *cond)
@@ -149,7 +156,9 @@ static void pthread_cond_destroy(pthread_cond_t *cond)
 
     /* non native condition variables */
     CloseHandle(win32_cond->semaphore);
+    CloseHandle(win32_cond->waiters_done);
     pthread_mutex_destroy(&win32_cond->mtx_waiter_count);
+    pthread_mutex_destroy(&win32_cond->mtx_broadcast);
     av_freep(&win32_cond);
     cond->ptr = NULL;
 }
@@ -157,41 +166,70 @@ static void pthread_cond_destroy(pthread_cond_t *cond)
 static void pthread_cond_broadcast(pthread_cond_t *cond)
 {
     win32_cond_t *win32_cond = cond->ptr;
+    int have_waiter;
+
     if (cond_broadcast) {
         cond_broadcast(cond);
         return;
     }
 
     /* non native condition variables */
+    pthread_mutex_lock(&win32_cond->mtx_broadcast);
     pthread_mutex_lock(&win32_cond->mtx_waiter_count);
+    have_waiter = 0;
+
     if (win32_cond->waiter_count) {
-        ReleaseSemaphore(win32_cond->semaphore, win32_cond->waiter_count, NULL);
-        win32_cond->waiter_count = 0;
+        win32_cond->is_broadcast = 1;
+        have_waiter = 1;
     }
-    pthread_mutex_unlock(&win32_cond->mtx_waiter_count);
+
+    if (have_waiter) {
+        ReleaseSemaphore(win32_cond->semaphore, win32_cond->waiter_count, NULL);
+        pthread_mutex_unlock(&win32_cond->mtx_waiter_count);
+        WaitForSingleObject(win32_cond->waiters_done, INFINITE);
+        win32_cond->is_broadcast = 0;
+    } else
+        pthread_mutex_unlock(&win32_cond->mtx_waiter_count);
+    pthread_mutex_unlock(&win32_cond->mtx_broadcast);
 }
 
 static void pthread_cond_wait(pthread_cond_t *cond, pthread_mutex_t *mutex)
 {
     win32_cond_t *win32_cond = cond->ptr;
+    int last_waiter;
     if (cond_wait) {
         cond_wait(cond, mutex, INFINITE);
         return;
     }
 
     /* non native condition variables */
+    pthread_mutex_lock(&win32_cond->mtx_broadcast);
+    pthread_mutex_unlock(&win32_cond->mtx_broadcast);
+
     pthread_mutex_lock(&win32_cond->mtx_waiter_count);
     win32_cond->waiter_count++;
     pthread_mutex_unlock(&win32_cond->mtx_waiter_count);
 
+    // unlock the external mutex
     pthread_mutex_unlock(mutex);
     WaitForSingleObject(win32_cond->semaphore, INFINITE);
-    pthread_mutex_lock(mutex);
+
+    pthread_mutex_lock(&win32_cond->mtx_waiter_count);
+    win32_cond->waiter_count--;
+    last_waiter = !win32_cond->waiter_count && win32_cond->is_broadcast;
+    pthread_mutex_unlock(&win32_cond->mtx_waiter_count);
+
+    if (last_waiter)
+        SetEvent(win32_cond->waiters_done);
+
+    // lock the external mutex
+    return pthread_mutex_lock(mutex);
 }
 
 static void pthread_cond_signal(pthread_cond_t *cond)
 {
     win32_cond_t *win32_cond = cond->ptr;
+    int have_waiter;
     if (cond_signal) {
         cond_signal(cond);
         return;
@@ -199,11 +237,11 @@ static void pthread_cond_signal(pthread_cond_t *cond)
 
     /* non-native condition variables */
     pthread_mutex_lock(&win32_cond->mtx_waiter_count);
-    if (win32_cond->waiter_count) {
-        ReleaseSemaphore(win32_cond->semaphore, 1, NULL);
-        win32_cond->waiter_count--;
-    }
+    have_waiter = win32_cond->waiter_count;
     pthread_mutex_unlock(&win32_cond->mtx_waiter_count);
+
+    if (have_waiter)
+        ReleaseSemaphore(win32_cond->semaphore, 1, NULL);
 }
 
 static void w32thread_init(void)