[FFmpeg-devel] [PATCH 6/6] alac: frame multi-threading support

[Paul B Mahol]

On 6/4/13, Michael Niedermayer <michaelni at gmx.at> wrote:
> On Mon, Jun 03, 2013 at 04:47:57PM +0000, Paul B Mahol wrote:
>> Signed-off-by: Paul B Mahol <onemda at gmail.com>
>> ---
>>  libavcodec/alac.c | 13 +++++++++++--
>>  1 file changed, 11 insertions(+), 2 deletions(-)
>
> ./ffmpeg-new -i alac-in-matroska.mka -t 2 -f framecrc -
> 0,          0,          0,     4096,    16384, 0x5ba4d20a
> 0,      36864,      36864,     4096,    16384, 0x05f28187
> 0,      40960,      40960,     4096,    16384, 0x38b2bd46
> 0,      45056,      45056,     4096,    16384, 0xef838394
> 0,      65536,      65536,     4096,    16384, 0xf409eb89
> 0,      69632,      69632,     4096,    16384, 0xcdd66995
> 0,      73728,      73728,     4096,    16384, 0x948e9aca
> 0,      77824,      77824,     4096,    16384, 0x7c0358ff
> [framecrc @ 0x1f65ea0] Non-monotonous DTS in output stream 0:0; previous:
> 77824, current: 32768; changing to 77824. This may result in incorrect
> timestamps in the output file.
> 0,      77824,      77824,     4096,    16384, 0x59d1e746
> 0,      86016,      86016,     4096,    16384, 0x4cfca3d4
>
> vs. with -threads 1
> 0,          0,          0,     4096,    16384, 0x5ba4d20a
> 0,       4096,       4096,     4096,    16384, 0x05f28187
> 0,       8192,       8192,     4096,    16384, 0x38b2bd46
> 0,      12288,      12288,     4096,    16384, 0xef838394
> 0,      16384,      16384,     4096,    16384, 0xf409eb89
> 0,      20480,      20480,     4096,    16384, 0xcdd66995
> 0,      24576,      24576,     4096,    16384, 0x948e9aca
> 0,      28672,      28672,     4096,    16384, 0x7c0358ff
> 0,      32768,      32768,     4096,    16384, 0x59d1e746
> 0,      36864,      36864,     4096,    16384, 0x4cfca3d4
> 0,      40960,      40960,     4096,    16384, 0x30e7d12c
> 0,      45056,      45056,     4096,    16384, 0xf3ed4a7c
> 0,      49152,      49152,     4096,    16384, 0x10b82c3a
> 0,      53248,      53248,     4096,    16384, 0xef42ff63
> 0,      57344,      57344,     4096,    16384, 0x067dd582
> 0,      61440,      61440,     4096,    16384, 0x93d0e8e7
> 0,      65536,      65536,     4096,    16384, 0xf44d786e
> 0,      69632,      69632,     4096,    16384, 0x9fd955ca
> 0,      73728,      73728,     4096,    16384, 0x24ee3023
> 0,      77824,      77824,     4096,    16384, 0x4033d7c4
> 0,      81920,      81920,     4096,    16384, 0x55903ef9
>
> [...]

I couldn't reproduce this. Is this still happen with new 1st patch of queue?

> --
> Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB
>
> I am the wisest man alive, for I know one thing, and that is that I know
> nothing. -- Socrates
>