]>
git.sesse.net Git - vlc/log
Rémi Denis-Courmont [Sun, 2 Mar 2014 15:04:35 +0000 (17:04 +0200)]
wasapi: audio capture client module (fixes #7205)
David Fuhrmann [Sun, 22 Mar 2015 18:02:28 +0000 (19:02 +0100)]
puzzle: fix error check, do not leak pictures on error
close #14006
Rémi Denis-Courmont [Sun, 22 Mar 2015 17:58:21 +0000 (19:58 +0200)]
Real RTSP: discard unsupported legacy username and password syntax
This is better than leaking them to the DNS resolver. This also avoids
confusing error messages if live555 is missing or failing.
Francois Cartegnie [Sun, 22 Mar 2015 15:18:20 +0000 (16:18 +0100)]
mux: mp4: handle box/bo failed alloc/realloc
refs #14193
Francois Cartegnie [Sat, 21 Mar 2015 18:29:35 +0000 (19:29 +0100)]
mux: avi: handle failed reallocs
refs #14193
Francois Cartegnie [Sat, 21 Mar 2015 18:27:53 +0000 (19:27 +0100)]
mux: avi: don't try to delete failed stream
Francois Cartegnie [Sat, 21 Mar 2015 17:48:06 +0000 (18:48 +0100)]
mux: avi: fix leak on format failure
Francois Cartegnie [Sat, 21 Mar 2015 14:08:16 +0000 (15:08 +0100)]
vlc_bits: use vlc_common and guards against failed realloc
refs #14193
Rémi Denis-Courmont [Sat, 21 Mar 2015 20:19:13 +0000 (22:19 +0200)]
decoder: reduce lock scope and cosmetic
p_owner->b_paused can only be written by the owner, so the owner thread
does not need to lock to read it. (However, locking is necessary to
write it from the owner thread, or to read it from the decoder thread.)
Rémi Denis-Courmont [Sat, 21 Mar 2015 20:02:42 +0000 (22:02 +0200)]
aout: remove aout_DecIsEmpty()
Rémi Denis-Courmont [Sat, 21 Mar 2015 19:59:58 +0000 (21:59 +0200)]
decoder: drain the audio output properly
Rémi Denis-Courmont [Sat, 21 Mar 2015 19:56:50 +0000 (21:56 +0200)]
aout: add wait parameter to aout_DecFlush()
Rémi Denis-Courmont [Sat, 21 Mar 2015 19:58:53 +0000 (21:58 +0200)]
decoder: cosmetic changes
Rémi Denis-Courmont [Sat, 21 Mar 2015 18:38:12 +0000 (20:38 +0200)]
decoder: inline DecoderSignalWait()
Rémi Denis-Courmont [Sat, 21 Mar 2015 18:43:33 +0000 (20:43 +0200)]
decoder: do not mix and match condidition variable and mutex pairings
Rémi Denis-Courmont [Sat, 21 Mar 2015 18:16:35 +0000 (20:16 +0200)]
decoder: do not wait for buffering when there is no data
This removes small hacks surrounding this corner case.
Rémi Denis-Courmont [Sat, 21 Mar 2015 18:06:26 +0000 (20:06 +0200)]
decoder: remove BLOCK_FLAG_CORE_EOS
Do not rely on allocating a block for draining the decoder and output.
Rémi Denis-Courmont [Sat, 21 Mar 2015 18:23:01 +0000 (20:23 +0200)]
fifo: fix FIFO bytes size computation
Rémi Denis-Courmont [Sat, 21 Mar 2015 17:47:18 +0000 (19:47 +0200)]
decoder: remove unnecessary special case
All audio decoders support the *pp_block == NULL case, and return NULL.
This enables audio decoders to drain cleanly.
Rémi Denis-Courmont [Sat, 21 Mar 2015 17:29:55 +0000 (19:29 +0200)]
decoder: simplify audio path
DecoderPlayAudio() has no effects with a NULL block.
David Fuhrmann [Sat, 21 Mar 2015 17:30:21 +0000 (18:30 +0100)]
macosx: fix main menu initialization order on startup
MainMenus p_intf can be already accessed before
applicationWillFinishLaunching was called.
This should fix quite some crashes reported by the users.
Rémi Denis-Courmont [Sat, 21 Mar 2015 17:09:45 +0000 (19:09 +0200)]
es_out: use input_DecoderDrain()
Rémi Denis-Courmont [Sat, 21 Mar 2015 17:09:31 +0000 (19:09 +0200)]
decoder: add input_DecoderDrain()
David Fuhrmann [Sat, 21 Mar 2015 17:04:09 +0000 (18:04 +0100)]
macosx: fix wrong ref counting in media info
Rémi Denis-Courmont [Sat, 21 Mar 2015 17:00:57 +0000 (19:00 +0200)]
decoder: factor some common code into DecoderUpdateFormatLocked()
Rémi Denis-Courmont [Sat, 21 Mar 2015 16:25:03 +0000 (18:25 +0200)]
decoder: reorder to avoid forward declation, no functional changes
Rémi Denis-Courmont [Sat, 21 Mar 2015 16:10:38 +0000 (18:10 +0200)]
decoder: fix data race in sout
Rémi Denis-Courmont [Sat, 21 Mar 2015 15:48:07 +0000 (17:48 +0200)]
decoder: merge b_exit into b_flushing
There is nothing special about exiting at this level. The decoder
should discard its buffers and return to reading the input FIFO at the
earliest (where it gets canceled). This is the same as flushing.
Rémi Denis-Courmont [Sat, 21 Mar 2015 16:03:33 +0000 (18:03 +0200)]
decoder: fix data race in input_DecoderChangePause()
See also
0fde3beaa85528e555bac9c1b327ba5a32da67e9 .
Rémi Denis-Courmont [Sat, 21 Mar 2015 15:59:20 +0000 (17:59 +0200)]
decoder: fix data race in input_DecoderFrameNext()
See also
0fde3beaa85528e555bac9c1b327ba5a32da67e9 .
Rémi Denis-Courmont [Sat, 21 Mar 2015 15:58:27 +0000 (17:58 +0200)]
es_out: do not start buffering when already buferring
This fixes an assertion failure.
Regression from
621400186a49b282197267e588a030b425829a64 .
Rémi Denis-Courmont [Sat, 21 Mar 2015 15:57:46 +0000 (17:57 +0200)]
decoder: add input_DecoderFlush()
David Fuhrmann [Sat, 21 Mar 2015 16:31:20 +0000 (17:31 +0100)]
macosx: Update progress dialog on the main thread, make check thread safe
This should fix some crashes as reported by the users.
Rémi Denis-Courmont [Sat, 21 Mar 2015 12:15:22 +0000 (14:15 +0200)]
vlc_plugin: fix non-LGPL plugins meta infos
Rémi Denis-Courmont [Sat, 21 Mar 2015 11:42:56 +0000 (13:42 +0200)]
vlc_plugin: allow overriding module meta-infos
Steve Lhomme [Thu, 19 Mar 2015 12:08:12 +0000 (12:08 +0000)]
directsound: fix TimeGet returning a positive value on error
Fixes #14186
a positive HRESULT means it succeeded.
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
Carlos Fenollosa [Wed, 18 Mar 2015 17:43:23 +0000 (18:43 +0100)]
Added forward and rewind (30s) buttons to mobile HTTP remote control
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
Thomas Guillem [Wed, 18 Mar 2015 09:16:44 +0000 (09:16 +0000)]
decoder: fix race in spu_new_buffer
There is a race when starting video with subtitles at a given position (via
--start-time).
If all decoders are flushed early, the video decoder may not have created the
Vout via the decoder_NewPicture call. In that case, spu_new_buffer will be
blocking for 6 seconds. Indeed, spu_new_buffer can block for maximum 6 seconds
when it's waiting for a Vout.
To solve this race, abort spu_new_buffer if decoder is flushing.
How to reproduce the race condition:
./vlc --start-time 3600 <video_with_subtitles>
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
Eugen Geist [Thu, 19 Mar 2015 16:57:05 +0000 (16:57 +0000)]
Update bookmark list before showing dialog.
Fixes #2100.
Override toggleVisible of BookmarkDialog to call update before using toggleVisible of QVLCFrame
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
Steve Lhomme [Fri, 20 Mar 2015 08:07:05 +0000 (08:07 +0000)]
libass: use the visible area dimensions, not the input dimensions
Fixes incorrect offset when playing http://streams.videolan.org/misc/test-files-ccc/vsfilter_general_TestCase.mkv
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
Miha Sokolov [Fri, 20 Mar 2015 09:38:51 +0000 (09:38 +0000)]
fix teletext framing code in DVB PES packets ignored
Check the teletext framing code in PES buffer for each of the received lines
and only copy those with correct framing code (p_block->p_buffer[3]) to the
p_sliced buffer that is later forwarded to ZVBI vbi_decode. Invalid lines
will not reach vbi_decode anymore.
When also packets with erroneous framing code are sent to vbi_decode (often
0x00 with some noise), in most cases those are decoded as packet 1/2,
causing the second text line on the teletext page to be overwritten with
spaces. So we need to avoid sending such packets with invalid framing code
to vbi_decode.
Close #14191
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
Jean-Baptiste Kempf [Fri, 20 Mar 2015 14:47:04 +0000 (15:47 +0100)]
Win32: don't call setsockopt if compiling for Windows 8.0
Ilkka Ollakka [Thu, 19 Mar 2015 06:52:30 +0000 (08:52 +0200)]
udp: adjust receive buffer for windows 7 and earlier
Try to detect runtime what windows version it is running to see if
receive buffers should be increased or leave to OS handling.
Fixes #14200
Rémi Denis-Courmont [Thu, 19 Mar 2015 21:51:21 +0000 (23:51 +0200)]
ALSA: request large enough buffers first (refs #10422)
With the current input clock and decoder buffering code, getting large
buffers is essential. So favor that over having short audio periods.
(Ideally the audio period should probably be an half or a third of the
PTS delay, rather than hard-coded to 40 ms, but that is a separate
issue).
Francois Cartegnie [Thu, 19 Mar 2015 17:57:44 +0000 (18:57 +0100)]
vlc_bits.h: drop len field
Francois Cartegnie [Thu, 19 Mar 2015 18:00:36 +0000 (19:00 +0100)]
mux: mp4: vlc_bits regression (fix #14192)
commit
6c9bc13fca989feca53a1ccdd609d6c6995a3879
Rémi Denis-Courmont [Thu, 19 Feb 2015 22:11:23 +0000 (00:11 +0200)]
playlist: remove activity callback in favor of input-current
Rémi Denis-Courmont [Thu, 19 Mar 2015 18:10:12 +0000 (20:10 +0200)]
playlist: move input-current, fix race
See also
a6b259b733a0ba0c064377cf07b5bd0c33584b17 .
Rémi Denis-Courmont [Tue, 17 Mar 2015 18:08:02 +0000 (20:08 +0200)]
block: remove block_FifoPace()
Rémi Denis-Courmont [Wed, 18 Mar 2015 23:04:50 +0000 (01:04 +0200)]
decoder: avoid block_FifoPace(), fix races
Rémi Denis-Courmont [Wed, 18 Mar 2015 22:39:29 +0000 (00:39 +0200)]
decoder: remove unused parameter
DecoderSignalWait() had no observable effects if has_data was false. It
could wake up the condition variable, but no thread would wait on that
event.
Rémi Denis-Courmont [Tue, 17 Mar 2015 18:07:48 +0000 (20:07 +0200)]
block: remove block_FifoWake()
Rémi Denis-Courmont [Tue, 17 Mar 2015 18:19:04 +0000 (20:19 +0200)]
decoder: avoid block_FifoWake()
Rémi Denis-Courmont [Tue, 17 Mar 2015 17:49:42 +0000 (19:49 +0200)]
udp: use the new FIFO functions
Rémi Denis-Courmont [Tue, 17 Mar 2015 17:33:31 +0000 (19:33 +0200)]
block: rewrite block_Fifo(Empty,Get,Put) with the new functions
Rémi Denis-Courmont [Tue, 24 Feb 2015 21:07:36 +0000 (23:07 +0200)]
block: add low-level functions for block FIFOs
In some cases, the thread(s) consuming a FIFO needs to wake up in other
circumstances than the FIFO being non-empty. For that purpose, this new
set of functions is vastly more flexible than block_FifoWake().
Rémi Denis-Courmont [Thu, 19 Mar 2015 17:05:49 +0000 (19:05 +0200)]
block: split block_t and block_fifo_t code
Rémi Denis-Courmont [Tue, 17 Mar 2015 20:15:09 +0000 (22:15 +0200)]
decoder: remove ARM-specific limitation
There are ARM servers and or HD/UHD-capable STBs. The distinction no
longer makes sense.
Rémi Denis-Courmont [Tue, 17 Mar 2015 19:38:31 +0000 (21:38 +0200)]
es_out: remove tautology
es->p_dec is obviously non-NULL when the loop starts, and the loop
cannot change the value.
Rémi Denis-Courmont [Wed, 18 Mar 2015 21:29:30 +0000 (23:29 +0200)]
decoder: remove tautology, add asserts to waiting code
Rémi Denis-Courmont [Tue, 17 Mar 2015 20:41:46 +0000 (22:41 +0200)]
decoder: remove invalid call, fix races
Calling input_DecoderStopWait() from the decoder thread makes no sense
as it signals a condition variable that only the decoder thread itself
would ever wait on.
More importantly, the call breaks assumptions made by the input thread.
For instance, input_DecoderDecode() assumes p_owner->b_waiting is only
changed by the decoder owner (i.e. input) thread.
Francois Cartegnie [Thu, 19 Mar 2015 14:07:26 +0000 (15:07 +0100)]
vlc_bits: add bo_set_64_be
Francois Cartegnie [Thu, 19 Mar 2015 14:07:05 +0000 (15:07 +0100)]
vlc_bits: fix bo_set_32be
Francois Cartegnie [Wed, 18 Mar 2015 17:11:49 +0000 (18:11 +0100)]
demux: ts: fix IOD signedness/warning
Francois Cartegnie [Wed, 18 Mar 2015 16:52:28 +0000 (17:52 +0100)]
demux: ts: fix opus signedness/warning
Francois Cartegnie [Wed, 18 Mar 2015 16:51:01 +0000 (17:51 +0100)]
opus: fix duration signedness
Francois Cartegnie [Wed, 18 Mar 2015 16:45:37 +0000 (17:45 +0100)]
demux: ts: fix signedness/warnings
Steve Lhomme [Thu, 19 Mar 2015 13:12:37 +0000 (13:12 +0000)]
MKV: factorize the es_out_Send() and es_out_Control() calls
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
Steve Lhomme [Wed, 18 Mar 2015 10:05:55 +0000 (10:05 +0000)]
DirectX/GDI: use COINIT_MULTITHREADED instead of COINIT_APARTMENTTHREADED
The renderer may be updated in the DMO thread which uses COINIT_MULTITHREADED
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
Steve Lhomme [Wed, 18 Mar 2015 11:53:53 +0000 (11:53 +0000)]
MKV: packetized codec can use VLC_TS_INVALID as i_pts for extra Block frames
Fixes WMA 9.1 Lossless playback through avcodec
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
Steve Lhomme [Thu, 19 Mar 2015 10:06:30 +0000 (10:06 +0000)]
MKV: fix COOK/ATRAC3 seeking
Fixes #14210
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
Steve Lhomme [Wed, 18 Mar 2015 14:35:46 +0000 (14:35 +0000)]
MKV: fix ATRAC3 playback
Fixes #14205
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
Steve Lhomme [Wed, 18 Mar 2015 11:48:42 +0000 (11:48 +0000)]
MKV: A_MS/ACM is not packetized
Just like in AVI
See modules/demux/avi/avi.c l461
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
Steve Lhomme [Wed, 18 Mar 2015 09:57:55 +0000 (09:57 +0000)]
DMO: use COINIT_MULTITHREADED instead of COINIT_APARTMENTTHREADED
Fixes #14202
All the other CoInitializeEx() calls and now using multithreading, except for
this code which is called in UpdateRects()
See modules/video_output/msw/common.c l286
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
Francois Cartegnie [Tue, 17 Mar 2015 13:10:14 +0000 (14:10 +0100)]
mux: mp4: handle both short and long NALU startcodes (fix #14185)
Francois Cartegnie [Tue, 17 Mar 2015 19:13:30 +0000 (20:13 +0100)]
Francois Cartegnie [Wed, 18 Mar 2015 11:56:15 +0000 (12:56 +0100)]
httplive: leaks on playlist reload/merge (fix #14171)
Francois Cartegnie [Tue, 17 Mar 2015 21:38:56 +0000 (22:38 +0100)]
demux: ts: gather data for uncreated es
should go to pre es queue otherwise first frames/gop are dropped
Gina Dobrescu [Tue, 17 Mar 2015 21:27:35 +0000 (21:27 +0000)]
Interface: HTTP server: set proper align when expanding menu buttons
Fixes #13855
The small menu buttons on the left doesn't move downwards anymore when
the menu is expanded
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
Petri Hintukainen [Tue, 17 Mar 2015 18:51:10 +0000 (18:51 +0000)]
sftp: implement directory listing
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
Steve Lhomme [Wed, 18 Mar 2015 06:26:55 +0000 (06:26 +0000)]
MKV: use the more accurate i_pcr for GET_TIME and GET_POSITION
similar to what avformat does
I forgot to re-submit that one after the other commits went through.
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
Rémi Denis-Courmont [Tue, 17 Mar 2015 17:16:44 +0000 (19:16 +0200)]
block: simplify block_FifoRelease
Since the FIFO is being destroyed, there is no need to lock and to
update the FIFO internal state.
Rémi Denis-Courmont [Tue, 17 Mar 2015 17:41:04 +0000 (19:41 +0200)]
udp: avoid alloc/free on I/O error
Rémi Denis-Courmont [Tue, 17 Mar 2015 18:45:51 +0000 (20:45 +0200)]
decoder: fix data race in input_DecoderIsEmpty()
p_dec->fmt_out is owned by the decoder plugin, and can only safely be
accessed by the core from within decoder callbacks, notably the format
update callbacks.
Outside that context, p_owner->fmt has to be used. It contains a copy
of p_dec->fmt_out at the last format update.
Rémi Denis-Courmont [Tue, 17 Mar 2015 17:00:47 +0000 (19:00 +0200)]
decoder: remove redundant call to block_FifoEmpty()
block_FifoRelease() takes care of it.
Steve Lhomme [Tue, 17 Mar 2015 13:41:16 +0000 (13:41 +0000)]
Revert "MKV: Always set an i_pts in a lace otherwise it disturbs seeking performance"
This reverts commit
6a16055a3331a4022d12ba29ab002a3257e4c89c .
Fixes #14194
Vorbis doesn't work properly with forced bogus PTS.
The seek issue in #13575 is solved with the DirectSound fixes from #14186
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
Petri Hintukainen [Tue, 17 Mar 2015 07:01:23 +0000 (07:01 +0000)]
do not install dsm plugin twice
On ma, 2015-03-16 at 13:59 +0300, Rémi Denis-Courmont wrote:
> Le 2015-03-16 13:24, Petri Hintukainen a écrit :
> > I tried to enable dsm access plugin in Linux (to get directory
> > browsing
> > support). But "make install" failed with error:
> >
> > libtool: install: error: cannot install `libdsm_plugin.la' to a
> > directory not ending in /usr/lib/vlc/plugins/services_discovery
> >
> > It looks like the module is installed twice: first to access/, then
> > to
> > services_discovery/. Installing it only to services_discovery/ seems
> > to
> > work.
>
> The (rest of the) Makefile rules are in access/ though...
Removing it from services_discovery/Makefile.am in commit
http://git.videolan.org/gitweb.cgi/vlc.git/?a=commit;h=
42350148049e6a91db1f6876d1e0765640da0f7d
didn't fix the problem with make install.
With
../configure --prefix=/usr --enable-dsm --disable-smb
make && sudo make install
I still get
libtool: install: error: cannot install `libdsm_plugin.la' to a
directory not ending in /usr/lib/vlc/plugins/services_discovery
Alternative patch attached.
(or maybe I have broken libtool ... ?)
>From
43e5502bb44f8611eb1d61cc8fa6c1c711dce32b Mon Sep 17 00:00:00 2001
From: Petri Hintukainen <phintuka@gmail.com>
Date: Tue, 17 Mar 2015 09:52:47 +0200
Subject: [PATCH] dsm: fix runtime library search path
Fixes "libtool: install: error: cannot install `libdsm_plugin.la'
to a directory not ending in /usr/lib/vlc/plugins/services_discovery"
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
Steve Lhomme [Mon, 16 Mar 2015 13:13:09 +0000 (13:13 +0000)]
MKV: fix a memory leak
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
Jean-Baptiste Kempf [Tue, 17 Mar 2015 17:03:17 +0000 (18:03 +0100)]
Qt: do not emit signals from the MIM constructor
Should close #14162
Jean-Baptiste Kempf [Tue, 17 Mar 2015 16:23:54 +0000 (17:23 +0100)]
Qt: MIM: code cosmetics
Jean-Baptiste Kempf [Tue, 17 Mar 2015 16:21:58 +0000 (17:21 +0100)]
Qt: MIM initialize p_input through constructor parameter
Francois Cartegnie [Tue, 17 Mar 2015 15:57:10 +0000 (16:57 +0100)]
mux: mp4: fix NALU size endianness
Francois Cartegnie [Tue, 17 Mar 2015 15:03:44 +0000 (16:03 +0100)]
vlc_block: fix NULL dereference in block_ChainGather
Francois Cartegnie [Tue, 17 Mar 2015 14:59:41 +0000 (15:59 +0100)]
demux: mp4: don't NULL dereference on failed realloc
Francois Cartegnie [Tue, 17 Mar 2015 00:48:01 +0000 (01:48 +0100)]
demux: ts: set fmt id on all es and add collision avoidance for extra es
Hugo Beauzée-Luyssen [Tue, 17 Mar 2015 09:50:51 +0000 (10:50 +0100)]
direct3d11: Fix mingw build
Francois Cartegnie [Mon, 16 Mar 2015 23:38:46 +0000 (00:38 +0100)]
demux: ts: fix potential segfault
introduced by
1dec820c43290fe19ac8b291d70353943275b8fc
Francois Cartegnie [Mon, 16 Mar 2015 19:48:30 +0000 (20:48 +0100)]
demux: ts: fix uninitialized fields with teletext es
Francois Cartegnie [Mon, 16 Mar 2015 19:10:38 +0000 (20:10 +0100)]
demux: ts: update probing
fixes invalid length/time with arib
and handles shared pid
Jean-Baptiste Kempf [Mon, 16 Mar 2015 18:11:18 +0000 (19:11 +0100)]
MFT: GUID_AC3 should be defined in Mingw-w64 4.0
Of course, it's technically a GCC bug, because selectany is broken, but
there is no simple fix for that.