[FFmpeg-devel] [PATCH] pthread_frame: allow per-field ThreadFrame owners.

[Ronald S. Bultje]

Hi,

On Mon, Apr 3, 2017 at 10:24 AM, Ronald S. Bultje <rsbultje at gmail.com>
wrote:

> This tries to handle cases where separate invocations of decode_frame()
> (each running in separate threads) write to respective fields in the
> same AVFrame->data[]. Having per-field owners makes interaction between
> readers (the referencing thread) and writers (the decoding thread)
> slightly more optimal if both accesses are field-based, since they will
> use the respective producer's thread objects (mutex/cond) instead of
> sharing the thread objects of the first field's producer.
>
> In practice, this fixes the following tsan-warning in fate-h264:
>
> WARNING: ThreadSanitizer: data race (pid=21615)
>   Read of size 4 at 0x7d640000d9fc by thread T2 (mutexes: write M1006):
>     #0 ff_thread_report_progress pthread_frame.c:569
> (ffmpeg:x86_64+0x100f7cf54)
> [..]
>   Previous write of size 4 at 0x7d640000d9fc by main thread (mutexes:
> write M1004):
>     #0 update_context_from_user pthread_frame.c:335
> (ffmpeg:x86_64+0x100f81abb)
> ---
>  libavcodec/h264_slice.c    |  8 +++++---
>  libavcodec/pthread_frame.c | 18 ++++++++++--------
>  libavcodec/thread.h        |  2 +-
>  libavcodec/utils.c         |  7 ++++---
>  4 files changed, 20 insertions(+), 15 deletions(-)
>

Ping.

(I realize this patch is probably very hard to review, but I'd really
appreciate a second set of eyes/brains on my proposed solution, exactly
because it's hard - and also because it touches a very fundamental core
element of frame-mt design.)

Ronald