Merge branch 'master' into oldabi

* master:
  release_notes: document not fully understood mingw-sdl issue
  release_notes: some updates
  presets: forgotten libvpx presets
  release_notes: fix version
  release_notes: mention more codecs Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
  release_notes: there will be 2 releases each for one ABI/API.
  release_notes: suggest git log instead of the poorly maintained APIChanges
  release_notes: we do support releases
  build system: disable memalign on haiku, its not reliable there.
  ffprobe: remove duplicate avformat_alloc_context()
  Fix segmentation fault in ffprobe
  wma: fix infinite loop
  Fix H.264 4:4:4 lossless decoding.

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

configure

@@ -3148,6 +3148,11 @@ check_deps $CONFIG_LIST       \
 
 enabled asm || { arch=c; disable $ARCH_LIST $ARCH_EXT_LIST; }
 
+if test $target_os == "haiku"; then
+    disable memalign
+    disable posix_memalign
+fi
+
 ! enabled_any memalign posix_memalign malloc_aligned &&
     enabled_any $need_memalign && enable memalign_hack
 

doc/RELEASE_NOTES

@@ -1,45 +1,58 @@
 Release Notes
 =============
 
-* 0.7 "Love and Peace" June, 2011
+* 0.8 "Love"  June, 2011
+* 0.7 "Peace" June, 2011 (identical to 0.8 but using 0.6 ABI/API)
+
 
 General notes
 -------------
 
 This release enables frame-based multithreaded decoding for a number of codecs,
-including VP8, H.263 and H.264. Additionally, there has been a major cleanup of
+including theora, huffyuv, VP8, H.263, mpeg4 and H.264. Additionally, there has
+been a major cleanup of
 both internal and external APIs. For this reason, the major versions of all
-libraries have been bumped. On the one hand, this means that 0.7 can be installed
+libraries except libpostproc have been bumped. This means that 0.8 can be installed
-side-by-side with previous releases, on the other hand, in order to benefit
+side-by-side with previous releases, on the other hand applications need to be
-from the new features, applications need to be recompiled.
+recompiled to use 0.8.
 
-Other important changes are additions of decoders including, but not limited to,
+Other important changes are more than 200 bugfixes, known regressions were fixed
+w.r.t 0.5 and 0.6, additions of decoders including, but not limited to,
 AMR-WB, single stream LATM/LOAS, G.722 ADPCM, a native VP8 decoder
 and HE-AACv2. Additionally, many new de/muxers such as WebM in Matroska, Apple
 HTTP Live Streaming, SAP, IEC 61937 (S/PDIF) have been added.
 
 See the Changelog file for a list of significant changes.
 
-Please note that our policy on bug reports has not changed. We still only accept
+Bugreports against FFmpeg git master or the most recent FFmpeg release are
-bug reports against HEAD of the FFmpeg trunk repository. If you are experiencing
+accepted. If you are experiencing issues with any formally released version of
-issues with any formally released version of FFmpeg, please try a current version
+FFmpeg, please try git master to check if the issue still exists. If it does,
-of the development code to check if the issue still exists. If it does, make your
+make your report against the development code following the usual bug reporting
-report against the development code following the usual bug reporting guidelines.
+guidelines.
+
+Note, if you have difficulty building for mingw, try --disable-outdev=sdl
 
 
 API changes
 -----------
 
-Please see the file doc/APIchanges for programmer-centric information. Note that a
+Please see git log of the public headers or the file doc/APIchanges for
-lot of long-time deprecated APIs have been removed. Also, a number of additional
+programmer-centric information. Note that some long-time deprecated APIs have
-APIs have been deprecated and are scheduled for removal in the next release.
+been removed. Also, a number of additional APIs have been deprecated and might
+be removed in the next release.
 
 
 Other notable changes
 ---------------------
 
+- high quality dithering in swscale to fix banding issues
+- ffmpeg is now interactive and various information can be turned on/off while its running
+- resolution changing support in ffmpeg
+- sdl output device
+- optimizations in libavfilter that make it much faster
+- split, buffer, select, lut, negate filters amongth others
+- more than 50 new video filters from mplayers libmpcodecs
 - many ARM NEON optimizations
-- libswscale cleanup started, optimizations should become easier in the future
 - nonfree libfaad support for AAC decoding removed
 - 4:4:4 H.264 decoding
 - 9/10bit H.264 decoding

ffpresets/libvpx-1080p.ffpreset

@@ -0,0 +1,17 @@
+vcodec=libvpx
+g=120
+rc_lookahead=16
+quality=good
+speed=0
+profile=1
+qmax=51
+qmin=11
+slices=4
+vb=2M
+
+#ignored unless using -pass 2
+maxrate=24M
+minrate=100k
+arnr_max_frames=7
+arnr_strength=5
+arnr_type=3

ffpresets/libvpx-1080p50_60.ffpreset

@@ -0,0 +1,17 @@
+vcodec=libvpx
+g=120
+rc_lookahead=25
+quality=good
+speed=0
+profile=1
+qmax=51
+qmin=11
+slices=4
+vb=2M
+
+#ignored unless using -pass 2
+maxrate=24M
+minrate=100k
+arnr_max_frames=7
+arnr_strength=5
+arnr_type=3

ffpresets/libvpx-360p.ffpreset

@@ -0,0 +1,16 @@
+vcodec=libvpx
+g=120
+rc_lookahead=16
+quality=good
+speed=0
+profile=0
+qmax=63
+qmin=0
+vb=768k
+
+#ignored unless using -pass 2
+maxrate=1.5M
+minrate=40k
+arnr_max_frames=7
+arnr_strength=5
+arnr_type=3

ffpresets/libvpx-720p.ffpreset

@@ -0,0 +1,17 @@
+vcodec=libvpx
+g=120
+rc_lookahead=16
+quality=good
+speed=0
+profile=0
+qmax=51
+qmin=11
+slices=4
+vb=2M
+
+#ignored unless using -pass 2
+maxrate=24M
+minrate=100k
+arnr_max_frames=7
+arnr_strength=5
+arnr_type=3

ffpresets/libvpx-720p50_60.ffpreset

@@ -0,0 +1,17 @@
+vcodec=libvpx
+g=120
+rc_lookahead=25
+quality=good
+speed=0
+profile=0
+qmax=51
+qmin=11
+slices=4
+vb=2M
+
+#ignored unless using -pass 2
+maxrate=24M
+minrate=100k
+arnr_max_frames=7
+arnr_strength=5
+arnr_type=3

ffprobe.c

@@ -395,12 +395,11 @@ int main(int argc, char **argv)
     int ret;
 
     av_register_all();
+    init_opts();
 #if CONFIG_AVDEVICE
     avdevice_register_all();
 #endif
 
-    avformat_opts = avformat_alloc_context();
-
     show_banner();
     parse_options(argc, argv, options, opt_input_file);
 

libavcodec/h264.c

@@ -1744,7 +1744,7 @@ static av_always_inline void hl_decode_mb_idct_luma(H264Context *h, int mb_type,
                         h->hpc.pred16x16_add[h->intra16x16_pred_mode](dest_y, block_offset, h->mb + (p*256 << pixel_shift), linesize);
                     }else{
                         for(i=0; i<16; i++){
-                            if(h->non_zero_count_cache[ scan8[i+p*16] ] || dctcoef_get(h->mb, pixel_shift, i*16))
+                            if(h->non_zero_count_cache[ scan8[i+p*16] ] || dctcoef_get(h->mb, pixel_shift, i*16+p*256))
                                 s->dsp.add_pixels4(dest_y + block_offset[i], h->mb + (i*16+p*256 << pixel_shift), linesize);
                         }
                     }

libavcodec/wmadec.c

@@ -828,7 +828,8 @@ static int wma_decode_superframe(AVCodecContext *avctx,
     }
     if (buf_size < s->block_align)
         return AVERROR(EINVAL);
-    buf_size = s->block_align;
+    if(s->block_align)
+        buf_size = s->block_align;
 
     samples = data;
 
@@ -911,9 +912,8 @@ static int wma_decode_superframe(AVCodecContext *avctx,
     }
 
 //av_log(NULL, AV_LOG_ERROR, "%d %d %d %d outbytes:%d eaten:%d\n", s->frame_len_bits, s->block_len_bits, s->frame_len, s->block_len,        (int8_t *)samples - (int8_t *)data, s->block_align);
-
     *data_size = (int8_t *)samples - (int8_t *)data;
-    return s->block_align;
+    return buf_size;
  fail:
     /* when error, we reset the bit reservoir */
     s->last_superframe_len = 0;