[FFmpeg-devel] [PATCH 5/5] avformat/libsrt: add snddropdelay parameter for srt

lance.lmwang at gmail.com lance.lmwang at gmail.com
Sun Oct 10 06:00:01 EEST 2021


On Sun, Oct 10, 2021 at 01:22:49AM +0800, "zhilizhao(赵志立)" wrote:
> 
> 
> > On Sep 30, 2021, at 9:14 AM, lance.lmwang at gmail.com wrote:
> > 
> > From: Limin Wang <lance.lmwang at gmail.com>
> > 
> > Signed-off-by: Limin Wang <lance.lmwang at gmail.com>
> > ---
> > doc/protocols.texi   | 10 ++++++++++
> > libavformat/libsrt.c |  7 +++++++
> > 2 files changed, 17 insertions(+)
> > 
> > diff --git a/doc/protocols.texi b/doc/protocols.texi
> > index 726e5f1..9c246f8 100644
> > --- a/doc/protocols.texi
> > +++ b/doc/protocols.texi
> > @@ -1496,6 +1496,16 @@ when the old encryption key is decommissioned. Default is -1.
> > -1 means auto (0x1000 in srt library). The range for
> > this option is integers in the 0 - @code{INT_MAX}.
> > 
> > + at item snddropdelay=@var{microseconds}
> > +The sender's delay before dropping packets. This delay is
> > +added to the default drop delay time interval value.
> > +Keep in mind that the longer the delay, the more probable it
> > +becomes that packets would be retransmitted uselessly because
> > +they will be dropped by the receiver anyway.
> > +
> > +Default is -1 means auto which typically means do not drop
> > +packets on the sender at all.
> 
> There are two issues here:
> 
> Firstly, -1 means auto for our libsrt wrapper. "typically means do not drop
> packets on the sender at all” is incorrect. The default value of libsrt is
> 0 for live mode, -1 for file mode. Live mode is the default and the file mode
> doesn’t matter much for FFmpeg, TBH.
> 
> Secondly, since our wrapper and libsrt defined different sematic for -1, now
> it’s impossible to pass -1 to libsrt.

Yes, the default is misleading. I'll add -2 for auto for wrapper, it'll be unset mode and 
will be use the default value of libsrt. Then -1 can pass to libsrt.c also.

> 
> And the description from libsrt doc
> 
> > Keep in mind that the longer the delay, the more probable it becomes that packets would be
> > retransmitted
> 
> can mislead the user to set a higher value for it, which is not a good idea.
> I don’t know how to improve that.
>  
> > +
> > @item payload_size=@var{bytes}
> > Sets the maximum declared size of a packet transferred
> > during the single call to the sending function in Live
> > diff --git a/libavformat/libsrt.c b/libavformat/libsrt.c
> > index c6308d1..13697d2 100644
> > --- a/libavformat/libsrt.c
> > +++ b/libavformat/libsrt.c
> > @@ -65,6 +65,7 @@ typedef struct SRTContext {
> >     int enforced_encryption;
> >     int kmrefreshrate;
> >     int kmpreannounce;
> > +    int64_t snddropdelay;
> > #endif
> >     int mss;
> >     int ffs;
> > @@ -111,6 +112,7 @@ static const AVOption libsrt_options[] = {
> >     { "enforced_encryption", "Enforces that both connection parties have the same passphrase set",                              OFFSET(enforced_encryption), AV_OPT_TYPE_BOOL,  { .i64 = -1 }, -1, 1,         .flags = D|E },
> >     { "kmrefreshrate",       "The number of packets to be transmitted after which the encryption key is switched to a new key", OFFSET(kmrefreshrate),       AV_OPT_TYPE_INT,   { .i64 = -1 }, -1, INT_MAX,   .flags = D|E },
> >     { "kmpreannounce",       "The interval between when a new encryption key is sent and when switchover occurs",               OFFSET(kmpreannounce),       AV_OPT_TYPE_INT,   { .i64 = -1 }, -1, INT_MAX,   .flags = D|E },
> > +    { "snddropdelay",        "The sender's delay(in microseconds) before dropping packets",                                     OFFSET(snddropdelay),        AV_OPT_TYPE_INT64,   { .i64 = -1 }, -1, INT64_MAX,   .flags = D|E },
> > #endif
> >     { "mss",            "The Maximum Segment Size",                                             OFFSET(mss),              AV_OPT_TYPE_INT,      { .i64 = -1 }, -1, 1500,      .flags = D|E },
> >     { "ffs",            "Flight flag size (window size) (in bytes)",                            OFFSET(ffs),              AV_OPT_TYPE_INT,      { .i64 = -1 }, -1, INT_MAX,   .flags = D|E },
> > @@ -318,6 +320,7 @@ static int libsrt_set_options_pre(URLContext *h, int fd)
> >     int latency = s->latency / 1000;
> >     int rcvlatency = s->rcvlatency / 1000;
> >     int peerlatency = s->peerlatency / 1000;
> > +    int snddropdelay = s->snddropdelay / 1000;
> >     int connect_timeout = s->connect_timeout;
> > 
> >     if ((s->mode == SRT_MODE_RENDEZVOUS && libsrt_setsockopt(h, fd, SRTO_RENDEZVOUS, "SRTO_RENDEZVOUS", &yes, sizeof(yes)) < 0) ||
> > @@ -334,6 +337,7 @@ static int libsrt_set_options_pre(URLContext *h, int fd)
> > #endif
> >         (s->kmrefreshrate >= 0 && libsrt_setsockopt(h, fd, SRTO_KMREFRESHRATE, "SRTO_KMREFRESHRATE", &s->kmrefreshrate, sizeof(s->kmrefreshrate)) < 0) ||
> >         (s->kmpreannounce >= 0 && libsrt_setsockopt(h, fd, SRTO_KMPREANNOUNCE, "SRTO_KMPREANNOUNCE", &s->kmpreannounce, sizeof(s->kmpreannounce)) < 0) ||
> > +        (s->snddropdelay  >= 0 && libsrt_setsockopt(h, fd, SRTO_SNDDROPDELAY,  "SRTO_SNDDROPDELAY",  &snddropdelay, sizeof(snddropdelay)) < 0) ||
> > #endif
> >         (s->mss >= 0 && libsrt_setsockopt(h, fd, SRTO_MSS, "SRTO_MSS", &s->mss, sizeof(s->mss)) < 0) ||
> >         (s->ffs >= 0 && libsrt_setsockopt(h, fd, SRTO_FC, "SRTO_FC", &s->ffs, sizeof(s->ffs)) < 0) ||
> > @@ -549,6 +553,9 @@ static int libsrt_open(URLContext *h, const char *uri, int flags)
> >         if (av_find_info_tag(buf, sizeof(buf), "kmpreannounce", p)) {
> >             s->kmpreannounce = strtol(buf, NULL, 10);
> >         }
> > +        if (av_find_info_tag(buf, sizeof(buf), "snddropdelay", p)) {
> > +            s->snddropdelay = strtoll(buf, NULL, 10);
> > +        }
> > #endif
> >         if (av_find_info_tag(buf, sizeof(buf), "mss", p)) {
> >             s->mss = strtol(buf, NULL, 10);
> > -- 
> > 1.8.3.1
> > 
> > _______________________________________________
> > ffmpeg-devel mailing list
> > ffmpeg-devel at ffmpeg.org
> > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
> > 
> > To unsubscribe, visit link above, or email
> > ffmpeg-devel-request at ffmpeg.org with subject "unsubscribe".
> > 
> 

-- 
Thanks,
Limin Wang


More information about the ffmpeg-devel mailing list