Re: [FFmpeg-devel] [PATCH v4] vvcdec: add thread executor

[Nuo Mi <nuomi2021@gmail.com>]

On Tue, Aug 22, 2023 at 8:23 PM James Almer <jamrial@gmail.com> wrote:

> On 8/22/2023 9:03 AM, Nuo Mi wrote:
> > On Thu, Aug 17, 2023 at 12:42 AM Rémi Denis-Courmont <remi@remlab.net>
> > wrote:
> >
> >> Le tiistaina 15. elokuuta 2023, 17.50.13 EEST Nuo Mi a écrit :
> >>> The executor design pattern was inroduced by java
> >>> <
> >>
> https://docs.oracle.com/en/java/javase/20/docs/api/java.base/java/util/conc
> >>> urrent/Executor.html> it also adapted by python
> >>> <https://docs.python.org/3/library/concurrent.futures.html>
> >>> Compared to handcrafted thread pool management, it greatly simplifies
> the
> >>> thread code. ---
> >>>   libavutil/Makefile   |   2 +
> >>>   libavutil/executor.c | 201
> +++++++++++++++++++++++++++++++++++++++++++
> >>>   libavutil/executor.h |  67 +++++++++++++++
> >>>   3 files changed, 270 insertions(+)
> >>>   create mode 100644 libavutil/executor.c
> >>>   create mode 100644 libavutil/executor.h
> >>>
> >>> diff --git a/libavutil/Makefile b/libavutil/Makefile
> >>> index 7828c94dc5..4711f8cde8 100644
> >>> --- a/libavutil/Makefile
> >>> +++ b/libavutil/Makefile
> >>> @@ -31,6 +31,7 @@ HEADERS = adler32.h
> >>
> >>>               \ encryption_info.h
> >>
> >>>   \ error.h                                                       \
> >> eval.h
> >>>                                                       \ +
> >>> executor.h                                                    \ fifo.h
> >>
> >>>                                                    \ file.h
> >>
> >>>                                        \ frame.h
> >>
> >>>                            \ @@ -127,6 +128,7 @@ OBJS = adler32.o
> >>
> >>>                                              \ encryption_info.o
> >>
> >>>                                     \ error.o
> >>
> >>>                            \ eval.o
> >>
> >>>                   \ +       executor.o
> >>
> >>>                  \ fifo.o
> >>
> >>>         \ file.o
> >>   \
> >>> file_open.o                                                      \ diff
> >>> --git a/libavutil/executor.c b/libavutil/executor.c
> >>> new file mode 100644
> >>> index 0000000000..6b47ba0bf2
> >>> --- /dev/null
> >>> +++ b/libavutil/executor.c
> >>> @@ -0,0 +1,201 @@
> >>> +/*
> >>> + * Copyright (C) 2023 Nuo Mi
> >>> + *
> >>> + * This file is part of FFmpeg.
> >>> + *
> >>> + * FFmpeg is free software; you can redistribute it and/or
> >>> + * modify it under the terms of the GNU Lesser General Public
> >>> + * License as published by the Free Software Foundation; either
> >>> + * version 2.1 of the License, or (at your option) any later version.
> >>> + *
> >>> + * FFmpeg is distributed in the hope that it will be useful,
> >>> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> >>> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
> >>> + * Lesser General Public License for more details.
> >>> + *
> >>> + * You should have received a copy of the GNU Lesser General Public
> >>> + * License along with FFmpeg; if not, write to the Free Software
> >>> + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
> >> 02110-1301
> >>> USA + */
> >>> +#include "internal.h"
> >>> +#include "mem.h"
> >>> +#include "thread.h"
> >>> +
> >>> +#include "executor.h"
> >>> +
> >>> +#if !HAVE_THREADS
> >>> +#define pthread_create(t, a, s, ar)     0
> >>
> >> Err, this should probably return an error value, e.g. ENOSYS or ENOTSUP.
> >>
> >>> +#define pthread_join(t, r)              do {} while(0)
> >>
> >> ...and this should be an unreachable(), assert(0), abort() or something
> >> along
> >> those lines.
> >>
> >>> +
> >>> +#define pthread_cond_init(c, a)         0
> >>> +#define pthread_cond_broadcast(c)       do {} while(0)
> >>> +#define pthread_cond_signal(c)          do {} while(0)
> >>> +#define pthread_cond_wait(c, m)         do {} while(0)
> >>> +#define pthread_cond_destroy(c)         do {} while(0)
> >>> +
> >>> +#define pthread_mutex_init(m, a)        0
> >>> +#define pthread_mutex_lock(l)           do {} while(0)
> >>> +#define pthread_mutex_unlock(l)         do {} while(0)
> >>> +#define pthread_mutex_destroy(l)        do {} while(0)
> >>> +#endif
> >>> +
> >>> +typedef struct ThreadInfo {
> >>> +    AVExecutor *e;
> >>> +    pthread_t thread;
> >>
> >> Does this even work if !HAVE_THREAD ?
> >
> > Yes. The definition provided by the following include chain:
> >
>  "mem.h"->"avutil.h"->"common.h"->"<stdlib.h>"->"<sys/types.h>"->"<bits/pthreadtypes.h>"
>
> Why would bits/pthreadtypes.h be available if there's no pthreads?
>
I guess it is related to POSIX. If you check <sys/types.h>, you will find
the following code:

/* Now add the thread types.  */
#if defined __USE_POSIX199506 || defined __USE_UNIX98
# include <bits/pthreadtypes.h>
#endif

> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel@ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
>
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".