[FFmpeg-cvslog] avcodec/dvdsubdec: fix out of bounds accesses

ffmpeg | branch: release/2.3 | wm4 | Mon Jan 5 04:45:26 2015 +0100| [f03888b449faf2888a149cae3b340ea13c6f85fa] | committer: Michael Niedermayer avcodec/dvdsubdec: fix out of bounds accesses The code blindly trusted buffer offsets read from the file in the RLE decoder. Explicitly check the

[FFmpeg-cvslog] avcodec/dvdsubdec: error on bitmaps with size 0

ffmpeg | branch: master | wm4 | Wed Jan 7 23:57:50 2015 +0100| [bcaa9099b3648b47060e1724a97dc98b63c83702] | committer: Michael Niedermayer avcodec/dvdsubdec: error on bitmaps with size 0 Attemtping to decode them could lead to invalid writes with some fuzzed samples. Signed-off-by: Michael

[FFmpeg-cvslog] avcodec/dvdsubdec: fix accessing dangling pointers

ffmpeg | branch: master | wm4 | Thu Jan 8 17:19:17 2015 +0100| [816577716bc6170bccfea3b9e865618b69a4b426] | committer: Michael Niedermayer avcodec/dvdsubdec: fix accessing dangling pointers dvdsub_decode() can call append_to_cached_buf() 2 times, the second time with ctx->buf as argument.

[FFmpeg-cvslog] vp9: fix parser return values in error case

ffmpeg | branch: master | wm4 | Fri Jan 9 02:13:36 2015 +0100| [b88e80589bd11ef935a5e9dab53d4edb00de16e4] | committer: Michael Niedermayer vp9: fix parser return values in error case The parser must always set the out_size and out_data pointers. The API seems to require it, and the common

[FFmpeg-cvslog] avcodec/dvdsubdec: fix out of bounds accesses

ffmpeg | branch: release/2.5 | wm4 | Mon Jan 5 04:45:26 2015 +0100| [e0a12b3dc3a252412e91416ebbbf9449e82e4bd0] | committer: Michael Niedermayer avcodec/dvdsubdec: fix out of bounds accesses The code blindly trusted buffer offsets read from the file in the RLE decoder. Explicitly check the

[FFmpeg-cvslog] avcodec/dvdsubdec: error on bitmaps with size 0

ffmpeg | branch: release/2.5 | wm4 | Wed Jan 7 23:57:50 2015 +0100| [3d71024f8ab35721f674f116ac84d734a1c6d517] | committer: Michael Niedermayer avcodec/dvdsubdec: error on bitmaps with size 0 Attemtping to decode them could lead to invalid writes with some fuzzed samples. Signed-off-by

[FFmpeg-cvslog] avcodec/dvdsubdec: fix accessing dangling pointers

ffmpeg | branch: release/2.5 | wm4 | Thu Jan 8 17:19:17 2015 +0100| [b895e29941f06ec7c96a8a0a5036cc11274fdb51] | committer: Michael Niedermayer avcodec/dvdsubdec: fix accessing dangling pointers dvdsub_decode() can call append_to_cached_buf() 2 times, the second time with ctx->buf

[FFmpeg-cvslog] vp9: fix parser return values in error case

ffmpeg | branch: release/2.5 | wm4 | Fri Jan 9 02:13:36 2015 +0100| [3032291b3a88fb25e3cf0d09a61792768e84a578] | committer: Michael Niedermayer vp9: fix parser return values in error case The parser must always set the out_size and out_data pointers. The API seems to require it, and the

[FFmpeg-cvslog] rm: fix memory leak on init failure

ffmpeg | branch: master | wm4 | Fri Jan 9 21:24:54 2015 +0100| [aab74a38b8ba910a383dd82953061d5c42772ae9] | committer: Michael Niedermayer rm: fix memory leak on init failure AVInputFormat.read_close is not called if AVInputFormat.read_header fails, so this needs to be handled separately

[FFmpeg-cvslog] vp9: avoid infinite loop with broken files

ffmpeg | branch: master | wm4 | Sat Jan 10 18:00:08 2015 +0100| [09b4ad15681be197fff8c57ce7c988a4718d6e03] | committer: Michael Niedermayer vp9: avoid infinite loop with broken files With a certain fuzzed file, the parser will always return 0 consumed bytes, which makes calling code call the

[FFmpeg-cvslog] avcodec/dvdsubdec: fix out of bounds accesses

ffmpeg | branch: release/2.4 | wm4 | Mon Jan 5 04:45:26 2015 +0100| [e2e145db89913e86e9b8573b1b90f001c46dee5e] | committer: Michael Niedermayer avcodec/dvdsubdec: fix out of bounds accesses The code blindly trusted buffer offsets read from the file in the RLE decoder. Explicitly check the

[FFmpeg-cvslog] avcodec/dvdsubdec: error on bitmaps with size 0

ffmpeg | branch: release/2.4 | wm4 | Wed Jan 7 23:57:50 2015 +0100| [2ba1af5791fc405ad648851344fe6dd83d1c92c1] | committer: Michael Niedermayer avcodec/dvdsubdec: error on bitmaps with size 0 Attemtping to decode them could lead to invalid writes with some fuzzed samples. Signed-off-by

[FFmpeg-cvslog] vp9: fix parser return values in error case

ffmpeg | branch: release/2.4 | wm4 | Fri Jan 9 02:13:36 2015 +0100| [1b3332659a77c8167b1d4328a88ff4ae2498147e] | committer: Michael Niedermayer vp9: fix parser return values in error case The parser must always set the out_size and out_data pointers. The API seems to require it, and the

[FFmpeg-cvslog] avcodec/dvdsubdec: fix accessing dangling pointers

ffmpeg | branch: release/2.4 | wm4 | Thu Jan 8 17:19:17 2015 +0100| [3b4e9dddcff61eaf01a9a5b63c1abcc4102a313a] | committer: Michael Niedermayer avcodec/dvdsubdec: fix accessing dangling pointers dvdsub_decode() can call append_to_cached_buf() 2 times, the second time with ctx->buf

[FFmpeg-cvslog] qpeg: avoid pointless invalid memcpy()

ffmpeg | branch: master | wm4 | Tue Jan 13 14:47:47 2015 +0100| [921706691a87c3ea5f5b92afd9b423e5f8c6e9d9] | committer: Michael Niedermayer qpeg: avoid pointless invalid memcpy() If refdata was NULL, the memcpy() ended up copying the same memory block onto itself, which is not only pointless

[FFmpeg-cvslog] avformat/utils: check for malloc failure

ffmpeg | branch: release/2.4 | wm4 | Fri Dec 12 22:22:55 2014 +0100| [b96163f055ee44484ca4a06382221c2abb1e0914] | committer: Michael Niedermayer avformat/utils: check for malloc failure Signed-off-by: Michael Niedermayer (cherry picked from commit a66893ac949864352b36b39e48c4cd72bbd81e54

[FFmpeg-cvslog] h264: unref leaking error recovery picture

ffmpeg | branch: master | wm4 | Mon Jan 19 17:14:44 2015 +0100| [e0e33ebe6bd71fdb211c9af7c7af17d6472c535b] | committer: Michael Niedermayer h264: unref leaking error recovery picture This leaked a frame on each avcodec_flush_buffers() call, if frame threading was enabled. It caused severe

[FFmpeg-cvslog] avcodec/dvdsubdec: fix out of bounds accesses

ffmpeg | branch: release/2.2 | wm4 | Mon Jan 5 04:45:26 2015 +0100| [0d481efb7b81ab2f0491a854a4fd5d8cfb305680] | committer: Michael Niedermayer avcodec/dvdsubdec: fix out of bounds accesses The code blindly trusted buffer offsets read from the file in the RLE decoder. Explicitly check the

[FFmpeg-cvslog] qpeg: avoid pointless invalid memcpy()

ffmpeg | branch: release/2.2 | wm4 | Tue Jan 13 14:47:47 2015 +0100| [b1959b1719c81eb93e06fae4788200d172b7a9fd] | committer: Michael Niedermayer qpeg: avoid pointless invalid memcpy() If refdata was NULL, the memcpy() ended up copying the same memory block onto itself, which is not only

[FFmpeg-cvslog] avcodec/dvdsubdec: fix accessing dangling pointers

ffmpeg | branch: release/2.2 | wm4 | Thu Jan 8 17:19:17 2015 +0100| [a4a87a7efd8b72c0871491bcc85330e14beb8bd6] | committer: Michael Niedermayer avcodec/dvdsubdec: fix accessing dangling pointers dvdsub_decode() can call append_to_cached_buf() 2 times, the second time with ctx->buf

[FFmpeg-cvslog] avcodec/dvdsubdec: error on bitmaps with size 0

ffmpeg | branch: release/2.2 | wm4 | Wed Jan 7 23:57:50 2015 +0100| [b76dc8b5b8f7a637b3232329d574a42af35878ca] | committer: Michael Niedermayer avcodec/dvdsubdec: error on bitmaps with size 0 Attemtping to decode them could lead to invalid writes with some fuzzed samples. Signed-off-by

[FFmpeg-cvslog] vp9: fix parser return values in error case

ffmpeg | branch: release/2.2 | wm4 | Fri Jan 9 02:13:36 2015 +0100| [fc1fed62d9da6f752fbfa4f27fbc1aab7f884c67] | committer: Michael Niedermayer vp9: fix parser return values in error case The parser must always set the out_size and out_data pointers. The API seems to require it, and the

[FFmpeg-cvslog] lavu/frame: fix malloc error path in av_frame_copy_props()

ffmpeg | branch: release/2.2 | wm4 | Mon Dec 15 04:32:58 2014 +0100| [b9510b3274cf1e37079aafdeacf020818908d989] | committer: Michael Niedermayer lavu/frame: fix malloc error path in av_frame_copy_props() The error path frees all side data, but forgets to reset the side data count. This can

[FFmpeg-cvslog] avformat/tta: fix crash with corrupted files

ffmpeg | branch: master | wm4 | Tue Feb 3 14:41:10 2015 +0100| [6a0cd529a35190d9374b0b26504e71857cd67b83] | committer: Michael Niedermayer avformat/tta: fix crash with corrupted files av_add_index_entry() can fail, for example because the parameters are invalid, or because memory allocation

[FFmpeg-cvslog] avformat/mpc8: fix hang with fuzzed file

ffmpeg | branch: master | wm4 | Tue Feb 3 19:04:12 2015 +0100| [56cc024220886927350cfc26ee695062ca7ecaf4] | committer: Michael Niedermayer avformat/mpc8: fix hang with fuzzed file This can lead to an endless loop by seeking back a few bytes after each attempted chunk read. Assuming negative

[FFmpeg-cvslog] avformat/mpc8: fix broken pointer math

ffmpeg | branch: master | wm4 | Tue Feb 3 19:04:11 2015 +0100| [b737a2c52857b214be246ff615c6293730033cfa] | committer: Michael Niedermayer avformat/mpc8: fix broken pointer math This could overflow and crash at least on 32 bit systems. Reviewed-by: Reimar Döffinger Signed-off-by: Michael

[FFmpeg-cvslog] avcodec/dvdsubdec: error on bitmaps with size 0

ffmpeg | branch: release/1.2 | wm4 | Wed Jan 7 23:57:50 2015 +0100| [8a3a3e3154b78c8d90d3b62c38054a959911b79f] | committer: Michael Niedermayer avcodec/dvdsubdec: error on bitmaps with size 0 Attemtping to decode them could lead to invalid writes with some fuzzed samples. Signed-off-by

[FFmpeg-cvslog] avformat/mpc8: fix hang with fuzzed file

ffmpeg | branch: release/1.2 | wm4 | Tue Feb 3 19:04:12 2015 +0100| [bdc4abd694a3234ef76d1f41a06e940d8fc9c0ed] | committer: Michael Niedermayer avformat/mpc8: fix hang with fuzzed file This can lead to an endless loop by seeking back a few bytes after each attempted chunk read. Assuming

[FFmpeg-cvslog] avcodec/dvdsubdec: fix out of bounds accesses

ffmpeg | branch: release/1.2 | wm4 | Mon Jan 5 04:45:26 2015 +0100| [931f4313b2ec2e47bb34b6906ec53df6d3d45f9a] | committer: Michael Niedermayer avcodec/dvdsubdec: fix out of bounds accesses The code blindly trusted buffer offsets read from the file in the RLE decoder. Explicitly check the

[FFmpeg-cvslog] avformat/mpc8: fix broken pointer math

ffmpeg | branch: release/1.2 | wm4 | Tue Feb 3 19:04:11 2015 +0100| [49dd89f9027f3def12e170bb7d986d37812eedba] | committer: Michael Niedermayer avformat/mpc8: fix broken pointer math This could overflow and crash at least on 32 bit systems. Reviewed-by: Reimar Döffinger Signed-off-by

[FFmpeg-cvslog] avformat/tta: fix crash with corrupted files

ffmpeg | branch: release/1.2 | wm4 | Tue Feb 3 14:41:10 2015 +0100| [5d7320be52c37e511dec27127e3274c76f0d9f6a] | committer: Michael Niedermayer avformat/tta: fix crash with corrupted files av_add_index_entry() can fail, for example because the parameters are invalid, or because memory

[FFmpeg-cvslog] vp9: assign PTS to visible instead of invisible frames

ffmpeg | branch: master | wm4 | Thu Feb 5 16:43:27 2015 +0100| [a83b6b077c56341300a271ef4a1aaff19af5235d] | committer: Michael Niedermayer vp9: assign PTS to visible instead of invisible frames All the webm/vp9 files I have seen so far can have packets that contain 1 invisible and 1 visible

[FFmpeg-cvslog] matroska: don't complain about unknown elements

ffmpeg | branch: master | wm4 | Mon Feb 9 20:38:58 2015 +0100| [6938a095cb02dddb295f8340fd12f3c13139b5c8] | committer: Michael Niedermayer matroska: don't complain about unknown elements Matroska is an extensible format - unknown elements must be expected. It shouldn't complain

[FFmpeg-cvslog] avutil: move internal function out of public header

ffmpeg | branch: master | wm4 | Tue Feb 10 13:27:02 2015 +0100| [3e18dc235e9cd41f83643af9470270f434363a0a] | committer: Michael Niedermayer avutil: move internal function out of public header Signed-off-by: Michael Niedermayer > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] avutil: check pixdescs in a different place

ffmpeg | branch: master | wm4 | Tue Feb 10 13:27:03 2015 +0100| [92e3a6fdac73f7e1d69d69717219a7538877d7a0] | committer: Michael Niedermayer avutil: check pixdescs in a different place Doing this check in avutil_version() is not appropriate. Also, this code is by default disabled (--assert

[FFmpeg-cvslog] pixdesc: clarify AV_PIX_FMT_FLAG_ALPHA doxygen

ffmpeg | branch: master | wm4 | Tue Feb 10 12:18:13 2015 +0100| [c247fc63e4dcf49995f5ba14021d0dfea6737b74] | committer: Michael Niedermayer pixdesc: clarify AV_PIX_FMT_FLAG_ALPHA doxygen Also make clear that PAL8 can have alpha. Signed-off-by: Michael Niedermayer > http://git.videolan.

[FFmpeg-cvslog] lavf: remove unused code

ffmpeg | branch: master | wm4 | Fri Feb 6 14:53:39 2015 +0100| [6a5b8ca4329039fad44ad50b6496948f4bfacb4c] | committer: Anton Khirnov lavf: remove unused code Nothing uses it, and it provides no public API. Archeological finds: Commit 101036adb9 added the API. Commit a8dd8dc6e9 made

[FFmpeg-cvslog] lavf: move internal fields from public to internal context

ffmpeg | branch: master | wm4 | Fri Feb 6 14:53:40 2015 +0100| [9deaec782810d098bca11c9332fab2d2f4c5fb78] | committer: Anton Khirnov lavf: move internal fields from public to internal context This is not an API change; the fields were explicitly declared private before. Signed-off-by: Anton

[FFmpeg-cvslog] matroska: redo seekhead handling

ffmpeg | branch: master | wm4 | Mon Feb 9 20:39:00 2015 +0100| [cac2295b21519d5a5320fe0787b5fd2b65eb1f8c] | committer: Michael Niedermayer matroska: redo seekhead handling In particular, this reads chained seekheads. This makes seeking faster in files which have the index indirectly linked

[FFmpeg-cvslog] avformat/matroskadec: fix IGNIDX handling

ffmpeg | branch: master | wm4 | Mon Feb 9 20:39:01 2015 +0100| [7862325f8084a45a59fc629051c6c8e6e33318c4] | committer: Michael Niedermayer avformat/matroskadec: fix IGNIDX handling Just because the user requested the seek index to be ignored, we can't just skip essential headers. At

[FFmpeg-cvslog] qpeg: avoid pointless invalid memcpy()

ffmpeg | branch: release/2.5 | wm4 | Tue Jan 13 14:47:47 2015 +0100| [8a6770a214190550ce3431b3679785ea8fbbe985] | committer: Michael Niedermayer qpeg: avoid pointless invalid memcpy() If refdata was NULL, the memcpy() ended up copying the same memory block onto itself, which is not only

[FFmpeg-cvslog] avformat/utils: check for malloc failure

ffmpeg | branch: release/2.5 | wm4 | Fri Dec 12 22:22:55 2014 +0100| [7caee172049a80e8c048e3d04202d409a06beeca] | committer: Michael Niedermayer avformat/utils: check for malloc failure Signed-off-by: Michael Niedermayer (cherry picked from commit a66893ac949864352b36b39e48c4cd72bbd81e54

[FFmpeg-cvslog] avformat/tta: fix crash with corrupted files

ffmpeg | branch: release/2.5 | wm4 | Tue Feb 3 14:41:10 2015 +0100| [563e542b31e2174693207c1a391b71eac978187a] | committer: Michael Niedermayer avformat/tta: fix crash with corrupted files av_add_index_entry() can fail, for example because the parameters are invalid, or because memory

[FFmpeg-cvslog] avformat/mpc8: fix hang with fuzzed file

ffmpeg | branch: release/2.5 | wm4 | Tue Feb 3 19:04:12 2015 +0100| [193440f5662cc7183798d7eba961f98075361dcd] | committer: Michael Niedermayer avformat/mpc8: fix hang with fuzzed file This can lead to an endless loop by seeking back a few bytes after each attempted chunk read. Assuming

[FFmpeg-cvslog] avformat/mpc8: fix broken pointer math

ffmpeg | branch: release/2.5 | wm4 | Tue Feb 3 19:04:11 2015 +0100| [352d17086fd00e48fa0885113abaedc58a40adc8] | committer: Michael Niedermayer avformat/mpc8: fix broken pointer math This could overflow and crash at least on 32 bit systems. Reviewed-by: Reimar Döffinger Signed-off-by

[FFmpeg-cvslog] avformat/tta: fix crash with corrupted files

ffmpeg | branch: release/2.4 | wm4 | Tue Feb 3 14:41:10 2015 +0100| [e2e835f017d6d21fe0e042b1700ff095fc8c4941] | committer: Michael Niedermayer avformat/tta: fix crash with corrupted files av_add_index_entry() can fail, for example because the parameters are invalid, or because memory

[FFmpeg-cvslog] avformat/mpc8: fix broken pointer math

ffmpeg | branch: release/2.4 | wm4 | Tue Feb 3 19:04:11 2015 +0100| [600c6ebc7d0409f244c8b6e7b424f9c3d60ca493] | committer: Michael Niedermayer avformat/mpc8: fix broken pointer math This could overflow and crash at least on 32 bit systems. Reviewed-by: Reimar Döffinger Signed-off-by

[FFmpeg-cvslog] avformat/mpc8: fix hang with fuzzed file

ffmpeg | branch: release/2.4 | wm4 | Tue Feb 3 19:04:12 2015 +0100| [2515de3b1534ac92762e39d914fcad557860d3d8] | committer: Michael Niedermayer avformat/mpc8: fix hang with fuzzed file This can lead to an endless loop by seeking back a few bytes after each attempted chunk read. Assuming

[FFmpeg-cvslog] qpeg: avoid pointless invalid memcpy()

ffmpeg | branch: release/2.4 | wm4 | Tue Jan 13 14:47:47 2015 +0100| [d705125b949bc49889f52ea8ea4a036a7d8d2e2b] | committer: Michael Niedermayer qpeg: avoid pointless invalid memcpy() If refdata was NULL, the memcpy() ended up copying the same memory block onto itself, which is not only

[FFmpeg-cvslog] hls: fix a null pointer dereference

ffmpeg | branch: master | wm4 | Mon Feb 16 19:31:42 2015 +0100| [4bf3fab2811509a4f9ca932191e9bfa801ceeae2] | committer: Michael Niedermayer hls: fix a null pointer dereference Not allocating the pls->ctx will crash in libavformat/hls.c:1410, where it tries to dereference the field. Sam

[FFmpeg-cvslog] avformat/tta: fix crash with corrupted files

ffmpeg | branch: release/2.2 | wm4 | Tue Feb 3 14:41:10 2015 +0100| [c7dc73a6c3339c525d62498964a707936ff838fc] | committer: Michael Niedermayer avformat/tta: fix crash with corrupted files av_add_index_entry() can fail, for example because the parameters are invalid, or because memory

[FFmpeg-cvslog] avformat/mpc8: fix hang with fuzzed file

ffmpeg | branch: release/2.2 | wm4 | Tue Feb 3 19:04:12 2015 +0100| [3d8c51d699165b63f54af309686e27d311a16a59] | committer: Michael Niedermayer avformat/mpc8: fix hang with fuzzed file This can lead to an endless loop by seeking back a few bytes after each attempted chunk read. Assuming

[FFmpeg-cvslog] avformat/mpc8: fix broken pointer math

ffmpeg | branch: release/2.2 | wm4 | Tue Feb 3 19:04:11 2015 +0100| [79d86b844f084b884363396da4ebb561e7c42a73] | committer: Michael Niedermayer avformat/mpc8: fix broken pointer math This could overflow and crash at least on 32 bit systems. Reviewed-by: Reimar Döffinger Signed-off-by

[FFmpeg-cvslog] matroskadec: verify seekhead IDs

ffmpeg | branch: master | wm4 | Fri Jun 12 13:11:41 2015 +0200| [7e240f95818310ed721321e62130aa1c69f9cbe6] | committer: Michael Niedermayer matroskadec: verify seekhead IDs Some files have SeekHead elements with broken IDs. They mismatch with the ID of the destination element. These files are

[FFmpeg-cvslog] tls_gnutls: fix hang on disconnection

ffmpeg | branch: master | wm4 | Sat Jun 13 23:55:21 2015 +0200| [f419da99ef85d49ab04e7e15b76612f4d054] | committer: Michael Niedermayer tls_gnutls: fix hang on disconnection GNUTLS_SHUT_RDWR means GnuTLS will keep waiting for the server's termination reply. But since we don't sh

[FFmpeg-cvslog] tls_gnutls: fix hang on disconnection

ffmpeg | branch: master | wm4 | Sun Jun 14 17:12:53 2015 +0200| [74ea1167d91ccb2e1f2943efa030f2c278b598be] | committer: Martin Storsjö tls_gnutls: fix hang on disconnection GNUTLS_SHUT_RDWR means GnuTLS will keep waiting for the server's termination reply. But since we don't shutdo

[FFmpeg-cvslog] avio: fix potential crashes when combining ffio_ensure_seekback + crc

ffmpeg | branch: master | wm4 | Wed Jun 17 00:21:02 2015 +0200| [dc87758775e2ce8be84e4fe598e12416e83d2845] | committer: Michael Niedermayer avio: fix potential crashes when combining ffio_ensure_seekback + crc Calling ffio_ensure_seekback() if ffio_init_checksum() has been called on the same

[FFmpeg-cvslog] avio: fix potential crashes when combining ffio_ensure_seekback + crc

ffmpeg | branch: release/2.2 | wm4 | Wed Jun 17 00:21:02 2015 +0200| [539603e8770faa416f9b8383ddc8659e3b463bca] | committer: Michael Niedermayer avio: fix potential crashes when combining ffio_ensure_seekback + crc Calling ffio_ensure_seekback() if ffio_init_checksum() has been called on the

[FFmpeg-cvslog] hevc: make avcodec_decode_video2() fail if get_format() fails

ffmpeg | branch: release/2.2 | wm4 | Thu May 14 18:27:31 2015 +0200| [f06168f7dcd2a7ab67f3369986c465eea9cb2402] | committer: Michael Niedermayer hevc: make avcodec_decode_video2() fail if get_format() fails Personally, I need the decoder to back out if get_format() returns no usable pixel

[FFmpeg-cvslog] id3v2: strip trailing spaces from APIC tag

ffmpeg | branch: master | wm4 | Thu May 7 23:56:37 2015 +0200| [d4007d176313f6fe08c4fd2e484495f55e280c2a] | committer: Michael Niedermayer id3v2: strip trailing spaces from APIC tag The APIC description must be unique, and some ID3v2 tag writers add spaces to write several APIC entries with

[FFmpeg-cvslog] tls_gnutls: fix hang on disconnection

ffmpeg | branch: release/2.7 | wm4 | Sat Jun 13 23:55:21 2015 +0200| [c3c8365dbd3b677207998e5120f7b78850e3532c] | committer: Michael Niedermayer tls_gnutls: fix hang on disconnection GNUTLS_SHUT_RDWR means GnuTLS will keep waiting for the server's termination reply. But since we

[FFmpeg-cvslog] avio: fix potential crashes when combining ffio_ensure_seekback + crc

ffmpeg | branch: release/2.7 | wm4 | Wed Jun 17 00:21:02 2015 +0200| [9473e5a05d2cbdd56a7e3923868451a07a0edfe0] | committer: Michael Niedermayer avio: fix potential crashes when combining ffio_ensure_seekback + crc Calling ffio_ensure_seekback() if ffio_init_checksum() has been called on the

[FFmpeg-cvslog] avformat: clarify what package needs to be compiled with SSL support

ffmpeg | branch: master | wm4 | Thu Jun 18 20:36:19 2015 +0200| [f6c3f1ed6048ebc55f652ed59f7af9acebbf57e5] | committer: Michael Niedermayer avformat: clarify what package needs to be compiled with SSL support Try to reduce user confusion. Signed-off-by: Michael Niedermayer > h

[FFmpeg-cvslog] avformat: clarify what package needs to be compiled with SSL support

ffmpeg | branch: release/2.7 | wm4 | Thu Jun 18 20:36:19 2015 +0200| [93d076b4fde12f11c67dcea86a27cc9770450fd4] | committer: Michael Niedermayer avformat: clarify what package needs to be compiled with SSL support Try to reduce user confusion. Signed-off-by: Michael Niedermayer (cherry

[FFmpeg-cvslog] lavu: add an API function to return the FFmpeg version string

ffmpeg | branch: master | wm4 | Tue Jun 30 23:57:53 2015 +0200| [f91126643a91c2d3f8d8e210c8facaf259951b03] | committer: Michael Niedermayer lavu: add an API function to return the FFmpeg version string This returns something like "N-73264-gb54ac84". This is much more usefu

[FFmpeg-cvslog] lavu: add an API function to return the Libav version string

ffmpeg | branch: master | wm4 | Thu Jul 2 19:04:59 2015 +0200| [1316df7aa98c4784f190d107206d0bb12c590b89] | committer: Janne Grunau lavu: add an API function to return the Libav version string This returns something like "v12_dev0-1332-g333a27c". This is much more useful than the

[FFmpeg-cvslog] avformat: don't crash API users when demuxing mp4

ffmpeg | branch: master | wm4 | Tue Jul 14 21:08:24 2015 +0200| [130a8e0eef2f81e0d853117e417b650c3e16d1b7] | committer: Michael Niedermayer avformat: don't crash API users when demuxing mp4 This code is one big chunk of WTF. Signed-off-by: Michael Niedermayer > http://git.vide

[FFmpeg-cvslog] avio: fix potential crashes when combining ffio_ensure_seekback + crc

ffmpeg | branch: release/2.6 | wm4 | Wed Jun 17 00:21:02 2015 +0200| [c049e979b60542c53894e812d5921a455b7c16a3] | committer: Michael Niedermayer avio: fix potential crashes when combining ffio_ensure_seekback + crc Calling ffio_ensure_seekback() if ffio_init_checksum() has been called on the

[FFmpeg-cvslog] avio: fix potential crashes when combining ffio_ensure_seekback + crc

ffmpeg | branch: release/2.5 | wm4 | Wed Jun 17 00:21:02 2015 +0200| [3903a60d4beeb46cda5d830be2eb2bac76171ce8] | committer: Michael Niedermayer avio: fix potential crashes when combining ffio_ensure_seekback + crc Calling ffio_ensure_seekback() if ffio_init_checksum() has been called on the

[FFmpeg-cvslog] rawdec: fix mjpeg probing

ffmpeg | branch: master | wm4 | Wed Jul 29 22:11:18 2015 +0200| [bf51fcd304d5594a4d8eed2bedf0ef0f68fa65f8] | committer: Michael Niedermayer rawdec: fix mjpeg probing There can be other headers than "Content-Type:" (in this case, a "Content-Length:" header was following

[FFmpeg-cvslog] rawdec: fix mjpeg probing buffer size check

ffmpeg | branch: master | wm4 | Wed Jul 29 22:33:44 2015 +0200| [4c6beaed9210f01290e5a5a4e377f93f145172cc] | committer: Michael Niedermayer rawdec: fix mjpeg probing buffer size check Signed-off-by: Michael Niedermayer > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] rawdec: fix mjpeg probing buffer size check

ffmpeg | branch: release/2.5 | wm4 | Wed Jul 29 22:33:44 2015 +0200| [1eb646ec9f87ed488f52561867e107eaee89e20c] | committer: Michael Niedermayer rawdec: fix mjpeg probing buffer size check Signed-off-by: Michael Niedermayer (cherry picked from commit 4c6beaed9210f01290e5a5a4e377f93f145172cc

[FFmpeg-cvslog] rawdec: fix mjpeg probing

ffmpeg | branch: release/2.5 | wm4 | Wed Jul 29 22:11:18 2015 +0200| [c809cf1c5b1b66fbc14795f3de0dde41c6861356] | committer: Michael Niedermayer rawdec: fix mjpeg probing There can be other headers than "Content-Type:" (in this case, a "Content-Length:" header was followi

[FFmpeg-cvslog] lavc: propagate hwaccel errors

ffmpeg | branch: master | wm4 | Wed Aug 5 19:54:41 2015 +0200| [94c0df79c7e8205b9e0ae560bc4ac831c231abb8] | committer: wm4 lavc: propagate hwaccel errors At least the new videotoolbox decoder does not actually set a frame if end_frame fails. This causes the API to return success and signals

[FFmpeg-cvslog] vc1dec: propagate error codes and return meaningful error codes

ffmpeg | branch: master | wm4 | Thu Aug 6 11:20:25 2015 +0200| [8024002d408face5aac598ddfa69a8ff4c111977] | committer: wm4 vc1dec: propagate error codes and return meaningful error codes > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=8024002d408face5aac598ddfa69a8ff4c111

[FFmpeg-cvslog] mmaldec: hack against buffering problems on broken input

ffmpeg | branch: master | wm4 | Tue Aug 11 11:26:19 2015 +0200| [750f72d7758740d79c2b326a273cbe3af4541f18] | committer: wm4 mmaldec: hack against buffering problems on broken input I can't come up with a nice way to handle this. It's hard to keep the lock-stepped input/output in this

[FFmpeg-cvslog] mmaldec: do not mutate user's AVCodecContext extradata field

ffmpeg | branch: master | wm4 | Tue Aug 11 11:30:25 2015 +0200| [7f116973d59bbadce891320e338f5786ce31c65a] | committer: wm4 mmaldec: do not mutate user's AVCodecContext extradata field > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=7f116973d59bbadce891320e338f5786

[FFmpeg-cvslog] mmaldec: fix problems with flush logic

ffmpeg | branch: master | wm4 | Tue Aug 11 12:11:04 2015 +0200| [67db57ea12bf5a9cb7409a7649b476027cab23a1] | committer: wm4 mmaldec: fix problems with flush logic Don't try to do a blocking wait for MMAL output if we haven't even sent a single real packet, but only flush packets. Ob

[FFmpeg-cvslog] lavc: move vdpau decoders under FF_API_VDPAU.

ffmpeg | branch: master | wm4 | Mon Aug 17 17:05:34 2015 -0400| [a383f226f02b969b72d6b286af21d0fcd38ca98a] | committer: Ronald S. Bultje lavc: move vdpau decoders under FF_API_VDPAU. Signed-off-by: Ronald S. Bultje > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] avio: fix potential crashes when combining ffio_ensure_seekback + crc

ffmpeg | branch: release/2.4 | wm4 | Wed Jun 17 00:21:02 2015 +0200| [9d0aa058c997db60de84d2e05b08dabead9cb6e0] | committer: Michael Niedermayer avio: fix potential crashes when combining ffio_ensure_seekback + crc Calling ffio_ensure_seekback() if ffio_init_checksum() has been called on the

[FFmpeg-cvslog] rawdec: fix mjpeg probing

ffmpeg | branch: release/2.4 | wm4 | Wed Jul 29 22:11:18 2015 +0200| [2d582d142c185fbf7bc9240903423c9bbac7784f] | committer: Michael Niedermayer rawdec: fix mjpeg probing There can be other headers than "Content-Type:" (in this case, a "Content-Length:" header was followi

[FFmpeg-cvslog] rawdec: fix mjpeg probing buffer size check

ffmpeg | branch: release/2.4 | wm4 | Wed Jul 29 22:33:44 2015 +0200| [43956940ea20b8f51ec6d34ced42c63f853072bb] | committer: Michael Niedermayer rawdec: fix mjpeg probing buffer size check Signed-off-by: Michael Niedermayer (cherry picked from commit 4c6beaed9210f01290e5a5a4e377f93f145172cc

[FFmpeg-cvslog] configure: aac encoder requires lpc

ffmpeg | branch: master | wm4 | Tue Aug 25 18:54:39 2015 +0200| [94a4be85a1fcd57fdfd45dc291813d6afef8a87b] | committer: wm4 configure: aac encoder requires lpc Fixes compilation with --disable-encoders --enable-encoder=aac > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] avcodec/vdpau: remove incompatible-libav-abi hack from vdpau_render_state struct

ffmpeg | branch: master | wm4 | Sat Sep 5 16:45:26 2015 -0300| [89b0a9f3fa9d6f32d99d03473ab1cb2110a7c9dd] | committer: James Almer avcodec/vdpau: remove incompatible-libav-abi hack from vdpau_render_state struct > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] lavu: extend size of the AVPixFmtDescriptor.flags field

ffmpeg | branch: master | wm4 | Wed Sep 2 18:59:37 2015 +0200| [b8b5d8274471129f122858bc74ad09284dae6ab7] | committer: Vittorio Giovara lavu: extend size of the AVPixFmtDescriptor.flags field Signed-off-by: Vittorio Giovara > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] oggdec: fix invalid free on error

ffmpeg | branch: master | wm4 | Sat Aug 30 15:39:15 2014 +0200| [b173f5c15572cc82f68128599722e689df4ff137] | committer: Michael Niedermayer oggdec: fix invalid free on error The read_packet callback passes a pointer to a stack-allocated AVPacket. Attempting to free it with av_free() makes no

[FFmpeg-cvslog] avformat/realtextdec: UTF-16 support

ffmpeg | branch: master | wm4 | Tue Sep 2 20:54:16 2014 +0200| [b7f641dc9bff452c0af75700d1127f54dadf4247] | committer: Clément Bœsch avformat/realtextdec: UTF-16 support Also remove ff_smil_extract_next_chunk - this was the last user of it. > http://git.videolan.org/gitweb.cgi/ffmpeg.git

[FFmpeg-cvslog] avformat/assdec: UTF-16 support

ffmpeg | branch: master | wm4 | Tue Sep 2 20:48:45 2014 +0200| [3e8426170ce005c111dfcae7982e18b647b7383f] | committer: Clément Bœsch avformat/assdec: UTF-16 support Use the UTF-16 BOM to detect UTF-16 encoding. Convert the file contents to UTF-8 on the fly using FFTextReader, which acts as

[FFmpeg-cvslog] avformat/srtdec: speed up probing

ffmpeg | branch: master | wm4 | Thu Sep 4 21:16:13 2014 +0200| [c368538667121fb20c5414dfaf0fb273035a0854] | committer: Clément Bœsch avformat/srtdec: speed up probing > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=c368538667121fb20c5414dfaf0fb273035a0854 --- libavfor

[FFmpeg-cvslog] avformat/srtdec: UTF-16 support

ffmpeg | branch: master | wm4 | Tue Sep 2 20:52:07 2014 +0200| [d658ef18e3d1ebe63a64f404ac4646890ecf02c9] | committer: Clément Bœsch avformat/srtdec: UTF-16 support > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=d658ef18e3d1ebe63a64f404ac4646890ecf02c9 --- libavformat/srtde

[FFmpeg-cvslog] avformat/samidec: UTF-16 support

ffmpeg | branch: master | wm4 | Tue Sep 2 20:53:08 2014 +0200| [231a514dd3d246b38d0bf2a400a1759902a6f051] | committer: Clément Bœsch avformat/samidec: UTF-16 support ff_smil_extract_next_chunk() is still used by RealText. > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] oggdec: fix invalid free on error

ffmpeg | branch: release/2.2 | wm4 | Sat Aug 30 15:39:15 2014 +0200| [f1685bd31a1921bb239e77ce852c7a0aa513017b] | committer: Michael Niedermayer oggdec: fix invalid free on error The read_packet callback passes a pointer to a stack-allocated AVPacket. Attempting to free it with av_free

[FFmpeg-cvslog] avformat/mp3dec: fix gapless audio support

ffmpeg | branch: master | wm4 | Sat Sep 20 13:48:05 2014 +0200| [d87fe2687fdc5b1cb9aaec957afadb56d207618f] | committer: Michael Niedermayer avformat/mp3dec: fix gapless audio support The code already had skipping of initial padding, but discarding trailing frame padding was missing. This is

[FFmpeg-cvslog] avformat/mp3dec: avoid early EOF with concatenated gapless mp3s

ffmpeg | branch: master | wm4 | Sun Sep 21 12:51:41 2014 +0200| [6c7f1155bb648eced8e5aa08b1fd490df2f8b325] | committer: Michael Niedermayer avformat/mp3dec: avoid early EOF with concatenated gapless mp3s Consider a file created with something like: cat file1.mp3 file2.mp3 > result.

[FFmpeg-cvslog] avformat: add SUP/PGS subtitle demuxer

ffmpeg | branch: master | wm4 | Wed Sep 24 17:38:58 2014 +0200| [7dd2005e1ac3b86fc999dd40f888b82fb5e47e8d] | committer: Clément Bœsch avformat: add SUP/PGS subtitle demuxer Signed-off-by: Clément Bœsch > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] avformat/aviobuf: fix avio_flush() for read streams

ffmpeg | branch: master | wm4 | Tue Sep 30 18:46:48 2014 +0200| [c8422f04a3256a9abc01588f1b4aa95f5aadc891] | committer: Michael Niedermayer avformat/aviobuf: fix avio_flush() for read streams avio_flush() did nothing useful for read streams. Fix it to behave as expected, and discard the

[FFmpeg-cvslog] avcodec, avutil: allow more control about how samples are skipped

ffmpeg | branch: master | wm4 | Sat Sep 27 16:47:09 2014 +0200| [cdd6f059a65f28ff7a18ccf1194e9554adad1a1b] | committer: Michael Niedermayer avcodec, avutil: allow more control about how samples are skipped Add CODEC_FLAG2_SKIP_MANUAL (exposed as "skip_manual"), which makes the deco

[FFmpeg-cvslog] oggdec: fix invalid free on error

ffmpeg | branch: release/2.3 | wm4 | Sat Aug 30 15:39:15 2014 +0200| [9e43d92d6a16417aa06b62f0375212d528d1a484] | committer: Michael Niedermayer oggdec: fix invalid free on error The read_packet callback passes a pointer to a stack-allocated AVPacket. Attempting to free it with av_free

[FFmpeg-cvslog] mmaldec: support MPEG-4

ffmpeg | branch: master | wm4 | Thu Jan 28 17:24:40 2016 +0100| [7b1b53f3a4569343dee1f0e2493b7cf3aa45731a] | committer: wm4 mmaldec: support MPEG-4 I guess this means part 2 in both cases. Not sure which profiles exactly are actually supported properly. > http://git.videolan.org/gitweb.

[FFmpeg-cvslog] configure: fix mmal build dependencies

ffmpeg | branch: master | wm4 | Thu Jan 28 17:23:23 2016 +0100| [0badf4564a90ffa6f104b275277a6015a1d33905] | committer: wm4 configure: fix mmal build dependencies The mmal decoders do not depend on the software decoders. > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

<    1   2   3   4   5   >