1
0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2024-12-07 11:13:41 +02:00
Commit Graph

154 Commits

Author SHA1 Message Date
James Almer
092cbbfa74 avformat/tcp: re-add checks for setsockopt return values
Originally added in 0ed0af595b and lost
by mistake in ef71ef5f30.

Signed-off-by: James Almer <jamrial@gmail.com>
2018-09-11 15:48:55 -03:00
James Almer
ef71ef5f30 Merge commit '8c76bfacf663ff71cee5264a74d0f9c86addd325'
* commit '8c76bfacf663ff71cee5264a74d0f9c86addd325':
  tcp: Use ff_connect_parallel for RFC 8305 style connecting

Merged-by: James Almer <jamrial@gmail.com>
2018-09-11 13:54:59 -03:00
Martin Storsjö
8c76bfacf6 tcp: Use ff_connect_parallel for RFC 8305 style connecting
Signed-off-by: Martin Storsjö <martin@martin.st>
2018-08-31 12:26:52 +03:00
Jun Zhao
0ed0af595b lavf/tcp: check return value of setsockopt.
when setsockopt fail, use ff_log_net_error to dump the string
describing for error number.

Signed-off-by: Jun Zhao <mypopydev@gmail.com>
2018-08-18 16:16:11 +08:00
Jun Zhao
790cf9518a lavf/tcp: add option to setting Maximum Segment Size
This can change the the MSS value announced to the other end in
the initial TCP packet, it's can be used when failed Path MTU
discovery.

Signed-off-by: Jun Zhao <mypopydev@gmail.com>
2018-07-26 19:03:24 +08:00
Stephan Holljes
3717512282 lavf/tcp.c: Free allocated client URLContext in case of error.
Signed-off-by: Stephan Holljes <klaxa1337@googlemail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2018-04-19 03:15:32 +02:00
wm4
0e1f771d22 tcp: properly return EOF
There is no POSIX error code for EOF - recv() signals EOF by simply
returning 0. But libavformat recently changed its conventions and
requires an explicit AVERROR_EOF, or it might get into an endless retry
loop, consuming 100% CPU while doing nothing.
2017-12-31 16:14:23 +01:00
Carl Eugen Hoyos
c3b5ea7530 lavf/tcp: Fix the type of the optlen argument to getsockopt().
Fixes a warning on aix:
libavformat/tcp.c:283:58: warning: passing argument 5 of 'getsockopt' from incompatible pointer type
2017-11-18 17:40:18 +01:00
Aman Gupta
80bb81a8f3 avformat/tcp: add option to enable TCP_NODELAY
This can reduce latency and increase throughput, particularly on high
latency networks.

Signed-off-by: Aman Gupta <aman@tmm1.net>
Reviewed-by: Jeyapal, Karthick <kjeyapal@akamai.com>
2017-11-17 10:48:55 -08:00
Simon Thelen
54b6bef6e1 libavformat/tcp: fix return code for tcp_accept
ff_accept can return AVERROR(ETIMEDOUT) and errno will be 0 (or
undefined), return ret instead and return ff_neterror() in
ff_poll_interrupt instead of AVERROR(errno) to parse WSAGetLastError on
Windows.

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-05-10 14:00:20 +02:00
Joel Cunningham
8c8e5d5286 HTTP: improve performance by reducing forward seeks
This commit optimizes HTTP performance by reducing forward seeks, instead
favoring a read-ahead and discard on the current connection (referred to
as a short seek) for seeks that are within a TCP window's worth of data.
This improves performance because with TCP flow control, a window's worth
of data will be in the local socket buffer already or in-flight from the
sender once congestion control on the sender is fully utilizing the window.

Note: this approach doesn't attempt to differentiate from a newly opened
connection which may not be fully utilizing the window due to congestion
control vs one that is. The receiver can't get at this information, so we
assume worst case; that full window is in use (we did advertise it after all)
and that data could be in-flight

The previous behavior of closing the connection, then opening a new
with a new HTTP range value results in a massive amounts of discarded
and re-sent data when large TCP windows are used.  This has been observed
on MacOS/iOS which starts with an initial window of 256KB and grows up to
1MB depending on the bandwidth-product delay.

When seeking within a window's worth of data and we close the connection,
then open a new one within the same window's worth of data, we discard
from the current offset till the end of the window.  Then on the new
connection the server ends up re-sending the previous data from new
offset till the end of old window.

Example (assumes full window utilization):

TCP window size: 64KB
Position: 32KB
Forward seek position: 40KB

      *                      (Next window)
32KB |--------------| 96KB |---------------| 160KB
        *
  40KB |---------------| 104KB

Re-sent amount: 96KB - 40KB = 56KB

For a real world test example, I have MP4 file of ~25MB, which ffplay
only reads ~16MB and performs 177 seeks. With current ffmpeg, this results
in 177 HTTP GETs and ~73MB worth of TCP data communication.  With this
patch, ffmpeg issues 4 HTTP GETs and 3 seeks for a total of ~22MB of TCP data
communication.

To support this feature, the short seek logic in avio_seek() has been
extended to call a function to get the short seek threshold value.  This
callback has been plumbed to the URLProtocol structure, which now has
infrastructure in HTTP and TCP to get the underlying receiver window size
via SO_RCVBUF.  If the underlying URL and protocol don't support returning
a short seek threshold, the default s->short_seek_threshold is used

This feature has been tested on Windows 7 and MacOS/iOS.  Windows support
is slightly complicated by the fact that when TCP window auto-tuning is
enabled, SO_RCVBUF doesn't report the real window size, but it does if
SO_RCVBUF was manually set (disabling auto-tuning). So we can only use
this optimization on Windows in the later case

Signed-off-by: Joel Cunningham <joel.cunningham@me.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-02-14 17:00:30 +01:00
Joel Cunningham
f3778108d3 tcp: set socket buffer sizes before listen/connect/accept
From e24d95c0e06a878d401ee34fd6742fcaddeeb95f Mon Sep 17 00:00:00 2001
From: Joel Cunningham <joel.cunningham@me.com>
Date: Mon, 9 Jan 2017 13:37:51 -0600
Subject: [PATCH] tcp: set socket buffer sizes before listen/connect/accept

Attempting to set SO_RCVBUF and SO_SNDBUF on TCP sockets after connection
establishment is incorrect and some stacks ignore the set call on the socket at
this point.  This has been observed on MacOS/iOS.  Windows 7 has some peculiar
behavior where setting SO_RCVBUF after applies only if the buffer is increasing
from the default while decreases are ignored.  This is possibly how the incorrect
usage has gone unnoticed

Unix Network Programming Vol. 1: The Sockets Networking API (3rd edition, seciton 7.5):

"When setting the size of the TCP socket receive buffer, the ordering of the
function calls is important.  This is because of TCP's window scale option,
which is exchanged with the peer on SYN segments when the connection is
established. For a client, this means the SO_RCVBUF socket option must be
set before calling connect.  For a server, this means the socket option must
be set for the listening socket before calling listen.  Setting this option
for the connected socket will have no effect whatsoever on the possible window
scale option because accept does not return with the connected socket until
TCP's three-way handshake is complete.  This is why the option must be set on
the listening socket. (The sizes of the socket buffers are always inherited from
the listening socket by the newly created connected socket)"

Signed-off-by: Joel Cunningham <joel.cunningham@me.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-01-26 20:19:18 +01:00
Michael Niedermayer
c6bc823eea avformat/tcp: Put struct sockaddr_in6 under #if
Fixes: error: dereferencing pointer to incomplete type

Tested-by: Dave Yeo <daveryeo@telus.net>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2016-11-06 21:29:12 +01:00
liu jc
c0f6eff6a7 avformat/tcp: workaround for IOS9 getaddrinfo in IPv6 only network use hardcode IPv4 address can not resolve port number.
Signed-off-by: liujingchao <jcliu@outlook.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2016-11-05 18:48:00 +01:00
Martin Storsjö
136c3438bb tcp: Use rw_timeout for setting the connect/listen timeouts
Apply the default value for timeout in code instead of via the
avoption, to allow distinguishing the default value from the user
not setting anything at all.

Signed-off-by: Martin Storsjö <martin@martin.st>
2016-03-24 10:34:24 +02:00
Derek Buitenhuis
9c75148e6e Merge commit '2758cdedfb7ac61f8b5e4861f99218b6fd43491d'
This commit also disables the async fate test, because it
used internal APIs in a non-kosher way, which no longer
exists.

* commit '2758cdedfb7ac61f8b5e4861f99218b6fd43491d':
  lavf: reorganize URLProtocols

Merged-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
2016-02-29 16:51:10 +00:00
Anton Khirnov
2758cdedfb lavf: reorganize URLProtocols
Instead of a linked list constructed at av_register_all(), store them
in a constant array of pointers.

Since no registration is necessary now, this removes some global state
from lavf. This will also allow the urlprotocol layer caller to limit
the available protocols in a simple and flexible way in the following
commits.
2016-02-22 11:30:58 +01:00
Perette Barella
84110f4f77 libavformat/tcp.c : add send_buffer_size and recv_buffer_size options
adds two new options that may be set via the dictionary:

- send_buffer_size
- recv_buffer_size

When present, setsockopt() is used with SO_SNDBUF and SO_RCVBUF to set
socket buffer sizes.  I chose to make send and receive independent
because buffering requirements are often asymmetric.

Errors in setting the buffer size mean the socket will use its
default, so they are ignored.

There is no sanity checking on values, as the kernel/socket layers
already impose reasonable limits if asked for something crazy.

Rationale for enlarging receive buffers is to reduce susceptibility
to intermittent network delays/congestion.  I added setting the send
buffer for symmetry.

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2016-01-22 15:36:30 +01:00
Alexander S. Drozdov
6a8d58d693 avformat/tcp: TCP Protocol: fix descriptor leak on listen and interrupt
If we try to listen on TCP port and ff_listen() fails on
interrupt callback socket (bind) descriptor overwrites and
does not closed at all.

As a result, we can't rebind to the same port.

Reviewed-by: Stephan Holljes <klaxa1337@googlemail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2015-09-11 21:44:30 +02:00
Stephan Holljes
75235a2550 lavf/tcp: increase range for listen and call the underlying socket operations accordingly
Signed-off-by: Stephan Holljes <klaxa1337@googlemail.com>
2015-08-01 00:58:31 +02:00
Stephan Holljes
21198155a7 lavf/tcp: add tcp_accept
Signed-off-by: Stephan Holljes <klaxa1337@googlemail.com>
2015-08-01 00:58:31 +02:00
Michael Niedermayer
71288bf80f Merge commit '27852f2f1dec3749ea79883b70484c841169f747'
* commit '27852f2f1dec3749ea79883b70484c841169f747':
  libavformat: Handle error return from ff_listen_bind

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2015-04-09 20:58:11 +02:00
Anders Nystrom
27852f2f1d libavformat: Handle error return from ff_listen_bind
Handle error return from ff_listen_bind without leaking file descriptors.

Signed-off-by: Anders Nystrom <anders.nystrom@southpole.se>
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
2015-04-09 14:32:37 +02:00
Michael Niedermayer
c5c4e5839a Merge commit 'e14f98c62fdf8744b07419314095d1b3248cce75'
* commit 'e14f98c62fdf8744b07419314095d1b3248cce75':
  tcp: Clarify the units for the timeout avoptions

Conflicts:
	libavformat/tcp.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2015-02-28 23:20:39 +01:00
Martin Storsjö
e14f98c62f tcp: Clarify the units for the timeout avoptions
Signed-off-by: Martin Storsjö <martin@martin.st>
2015-02-28 22:07:24 +02:00
Carl Eugen Hoyos
1f4bce894a lavf/tcp: Clarify that the -timeout option takes microseconds. 2014-11-12 13:52:56 +01:00
Michael Niedermayer
ae17a0c258 Merge commit '1d4579e38ecad578c86516fad2837a273d11b320'
* commit '1d4579e38ecad578c86516fad2837a273d11b320':
  tcp: Add AVOption support

Conflicts:
	libavformat/tcp.c

See: 2e009c6042
Merged-by: Michael Niedermayer <michaelni@gmx.at>
2014-10-11 00:39:01 +02:00
Luca Barbato
1d4579e38e tcp: Add AVOption support 2014-10-10 16:29:07 +02:00
Michael Niedermayer
08a110ca87 Merge commit '6ee1cb5740e7490151db7dcec7e20ceaf8a2fe1f'
* commit '6ee1cb5740e7490151db7dcec7e20ceaf8a2fe1f':
  libavformat: use MSG_NOSIGNAL when applicable

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2014-08-25 13:27:21 +02:00
Rémi Denis-Courmont
6ee1cb5740 libavformat: use MSG_NOSIGNAL when applicable
If the remote end of a connection oriented socket hangs up, generating
an EPIPE error is preferable over an unhandled SIGPIPE signal.

Signed-off-by: Martin Storsjö <martin@martin.st>
2014-08-25 10:49:14 +03:00
Stefano Sabatini
1b4da43ce0 doc/protocols/tcp,lavf/tcp: apply minor fixes to TCP protocol documentation 2013-12-25 18:42:26 +01:00
Stefano Sabatini
702d9a23b3 lavf/tcp: honor listen option value in a tag
Consistent with what is done in udp.c, allow to explicitly disable listen
with listen=0.
2013-12-25 18:42:20 +01:00
Martin Storsjö
a6a4596e12 tcp: Explicitly convert a pointer to a boolean integer
This fixes warnings about making integers from pointers without
a cast, and avoids the theoretical case where the lower 32 bits of
the pointer would all be zero where the implicit cast wouldn't give
the right result.

Signed-off-by: Martin Storsjö <martin@martin.st>
2013-09-09 23:57:04 +03:00
Michael Niedermayer
22fbc7f8be avformat/tcp: fix pointer to int warning
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-08-25 19:44:40 +02:00
Michael Niedermayer
89efaabc99 Merge commit 'abe5268c3328bf0e8fcfb7dc6e231b8920177c3a'
* commit 'abe5268c3328bf0e8fcfb7dc6e231b8920177c3a':
  tcp: Use a different log message and level if there's more addresses to try

Conflicts:
	libavformat/tcp.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-08-06 17:53:01 +02:00
Michael Niedermayer
2539767206 Merge commit 'bb9378251a167ef0116f263912e57f715c1e02ac'
* commit 'bb9378251a167ef0116f263912e57f715c1e02ac':
  network: Use SOCK_CLOEXEC when available

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-08-06 15:40:05 +02:00
Martin Storsjö
abe5268c33 tcp: Use a different log message and level if there's more addresses to try
This lowers the level of warnings printed if trying to connect
to a host name that provides both v6 and v4 addresses but the
service only is available on the v4 address (often occurring for
'localhost', with servers that aren't v6-aware).

Signed-off-by: Martin Storsjö <martin@martin.st>
2013-08-06 10:39:39 +03:00
Luca Barbato
bb9378251a network: Use SOCK_CLOEXEC when available 2013-08-05 16:13:36 +02:00
Michael Niedermayer
68b7b534be tcp: Use a default timeout of 5 sec for opening a connection but not for receiving packets
This should be closer to how tcp behaved longer ago and should
fix the issue with idle connections timing out.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-06 01:33:19 +02:00
Michael Niedermayer
390aabb2a1 tcp: Fix the default timeout
Fixes Ticket2694

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-06 00:17:31 +02:00
Michael Niedermayer
82070b01b8 Merge commit '9835abb6d63fb07613994ae90e72fef758149408'
* commit '9835abb6d63fb07613994ae90e72fef758149408':
  network: uniform ff_listen_bind and ff_listen_connect

Conflicts:
	libavformat/network.c
	libavformat/tcp.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-06-05 11:56:56 +02:00
Luca Barbato
9835abb6d6 network: uniform ff_listen_bind and ff_listen_connect
Document the functions and have both use a millisecond timeout and
check for interrupt.
2013-06-04 14:56:52 +02:00
Michael Niedermayer
54ddbb477b Merge remote-tracking branch 'qatar/master'
* qatar/master:
  network: factor out connect-listening code

Conflicts:
	libavformat/network.h
	libavformat/tcp.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-06-02 11:02:02 +02:00
Michael Niedermayer
4d4f5911d3 Merge commit '28306e6d620c109ddd672f7243adfbc2bbb3b18f'
* commit '28306e6d620c109ddd672f7243adfbc2bbb3b18f':
  network: factor out bind-listening code
  use my full first name instead of short one in copyrights

Conflicts:
	libavformat/tcp.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-06-02 10:30:35 +02:00
Luca Barbato
f849a77e67 network: factor out connect-listening code
Introduce ff_listen_connect, to be shared with the other
non-tcp network protocols.
2013-06-01 15:29:53 +02:00
Luca Barbato
28306e6d62 network: factor out bind-listening code
Introduce ff_listen_bind, to be shared with the other non-tcp
network protocols.
2013-06-01 15:29:53 +02:00
Michael Niedermayer
9cf788eca8 avformat/avdevice: add missing time.h includes
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-03-12 15:28:22 +01:00
Andrey Utkin
2e009c6042 tcp: accept params through avio_open2() options
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2012-10-09 17:24:55 +02:00
Andrey Utkin
949acefc11 tcp: use timeout option consistently
Default _connection_ timeout value 5 seconds is left as i/o timeout also

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2012-10-09 17:18:43 +02:00
Michael Niedermayer
f7d4e26c6a Merge remote-tracking branch 'qatar/master'
* qatar/master:
  rtmp: Add a new option 'rtmp_pageurl'
  doc: Update the description of the rtmp_tcurl option
  rtmp: Make the description of the rtmp_tcurl option more generic
  libfdk-aacenc: add LATM/LOAS encapsulation support
  sctp: add port missing error message
  tcp: add port missing error message
  avfilter: Fix printf format string conversion specifier

Conflicts:
	libavcodec/version.h
	libavfilter/avfilter.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2012-07-25 22:00:06 +02:00