Nicholas Nethercote
7df92b2298
Bug 1203005 - Remove all traces of -Wdeclaration-after-statement. r=glandium.
...
Even though we compile C code as C99, we used to need
-Wdeclaration-after-statement because MSVC didn't allow declarations after
statements.
However, Visual Studio 2013 added support, so we can now merrily mix
declarations and statements everywhere. Hooray.
2015-09-09 04:57:44 -07:00
Matthew Gregan
c949250be2
Bug 1199794 - Add NULL checks to avoid crashing during media playback if the audio device removed. r=padenot
2015-08-28 19:13:00 -04:00
Nicholas Nethercote
69d088e45f
Bug 1198334 (part 1) - Replace the opt-in FAIL_ON_WARNINGS with the opt-out ALLOW_COMPILER_WARNINGS. r=glandium.
...
The patch removes 455 occurrences of FAIL_ON_WARNINGS from moz.build files, and
adds 78 instances of ALLOW_COMPILER_WARNINGS. About half of those 78 are in
code we control and which should be removable with a little effort.
2015-08-27 20:44:53 -07:00
Matthew Gregan
5c2236fa18
Bug 1197049 - Retain stream volume across device changes. r=padenot
2015-08-21 14:46:22 +12:00
Matthew Gregan
0a16b8f06b
Bug 1136360 - Report actual number of frames written to ReleaseBuffer. r=jesup
2015-08-21 18:09:34 +12:00
Matthew Gregan
b8e7d35ed9
Bug 1136360 - Account for stream and device latency in stream position calculation. r=jesup
2015-08-21 13:53:21 +12:00
Matthew Gregan
42e93061e3
Backed out 2 changesets (bug 1136360)
...
Backed out changeset 004f6db9c2dd (bug 1136360)
Backed out changeset 9b568fb8d447 (bug 1136360)
2015-08-24 13:25:57 +12:00
Matthew Gregan
e640422013
Bug 1136360 - Report actual number of frames written to ReleaseBuffer. r=jesup
2015-08-21 18:09:34 +12:00
Matthew Gregan
1da5a97706
Bug 1136360 - Account for stream and device latency in stream position calculation. r=jesup
2015-08-21 13:53:21 +12:00
Matthew Gregan
a4a6d5cc9a
Bug 1195058 - Handle pa_context_new returning NULL in OOM situations. r=padenot
2015-08-18 12:34:15 +12:00
Matthew Gregan
f0a620f81a
Bug 1195061 - Correctly restore audio hardware mix format after attempting to use an alternative format. r=padenot
2015-08-18 10:59:14 +12:00
JW Wang
dd02890934
Bug 1189866 - revert code that accidentally got into bug 1187817 patch 5. r=me.
2015-08-01 13:58:39 +08:00
JW Wang
0af35fe560
Bug 1187817. Part 5 - assert some code in the audio thread and don't enter the monitor. r=kinetik.
2015-07-31 13:26:06 +08:00
JW Wang
9afacdd82c
Bug 1189197 - enqueue a silent frame to kick off the buffer queue callbacks. r=kinetik.
2015-07-31 13:25:59 +08:00
Matthew Gregan
b0a98f661f
Bug 1103824 - Back out changeset for causing regression reported in bug 1176499.
2015-06-23 17:02:51 -07:00
Matthew Gregan
4a86a9bc10
Bug 1103824 - Ensure first initialization of IAudioClient happens on STA thread. r=padenot
2015-05-14 18:33:23 +12:00
Ted Mielczarek
fe4092720a
bug 1171072 - make libcubeb build on iOS. r=kinetik
2015-02-06 14:50:40 -05:00
Benjamin Chen
e2e98f6828
Bug 1165963 - Fix regression by bug 1104643: Detect "over compensation" and reset the anchor. r=padenot
2015-06-02 17:27:09 +08:00
Benjamin Chen
05d3e3d64c
Bug 1104643 - Interpolate the audio clock by system clock if the qudio clock is not going forward. r=padenot
2015-05-05 16:00:14 +08:00
David Major
ed6ca74f4f
Bug 1157835: Remove the MSVC_ENABLE_PGO flag from the build system. r=glandium
2015-04-27 19:59:27 -04:00
Chris Pearce
4aa83eaea9
Bug 1136360 - Backout 3920b67e97a3 to fix A/V sync regressions (bug 1148299 & bug 1157886). r=backout
2015-04-28 11:37:33 +12:00
Jacek Caban
df6dbaa8a7
Bug 1155624 - follow up: MSVC requires including <algorithm> for std::max. CLOSED TREE
2015-04-20 18:41:12 +02:00
Jacek Caban
a33ef4a63c
Bug 1155624 - Use std::max instead of max macro in cubeb_wasapi.cpp. r=kinetik
2015-04-20 18:18:56 +02:00
Paul Adenot
8c5829b980
Bug 979705 - Fix typo.
2015-04-20 17:05:26 +02:00
Paul17041993
2af31c5b7f
Bug 979705 - Re-write the upmixing functions in the WASAPI cubeb backend to fix mono audio and surround compatibility. r=padenot
2015-04-20 15:58:27 +02:00
Carsten "Tomcat" Book
9292d28542
Backed out changeset c511ccad7676 (bug 979705) for bustage on a CLOSED TREE
2015-04-20 16:44:41 +02:00
Paul17041993
510aeb0311
Bug 979705 - Re-write the upmixing functions in the WASAPI cubeb backend to fix mono audio and surround compatibility. r=padenot
2015-04-20 15:58:27 +02:00
Paul Adenot
8c600ca84e
Bug 1136360 - Take into account the output device latency in the clock, and be more robust about rounding error accumulation, in cubeb_wasapi.cpp. r=kinetik
2015-03-16 18:12:38 +01:00
Matthew Gregan
4fc21b3855
Bug 1144199 - Require multiple consecutive timeouts to accumulate before triggering timeout error handling in libcubeb's WASAPI backend; this avoids spurious timeout errors triggered by system sleep/wake cycles. r=padenot
2015-04-11 13:24:52 +12:00
Matthew Gregan
9cfaa9f421
Bug 1134977 - Release IAudioStreamVolume when closing WASAPI stream. Refixes bug 1109802. r=padenot
2015-04-02 10:08:07 +13:00
Matthew Gregan
7c58f24efc
Bug 1143968 - Allow a cubeb_stream in error to be stopped without triggering a fatal assert. r=padenot
2015-03-18 15:23:19 +13:00
Alexandre Ratchov
157ac815f9
Bug 1144087 - "Audio glitches on OpenBSD caused by rounding errors." r=kinetik
2015-03-18 12:19:00 +01:00
Matthew Gregan
cc687d3c79
Bug 1142746 - Make unexpected SL_PLAYEVENT_HEADATMARKER notification non-fatal. r=brsun
2015-03-17 13:30:40 +13:00
Ryan VanderMeulen
52711c3d87
Backed out changeset 7a53ee0cc3ae (bug 1136360) for Windows mochitest crashes.
2015-03-16 15:04:51 -04:00
Paul Adenot
2d95627b11
Bug 1136360 - Take into account the output device latency in the clock, and be more robust about rounding error accumulation, in cubeb_wasapi.cpp. r=kinetik
2015-03-16 18:12:38 +01:00
Matthew Gregan
08abd3d433
Bug 1136107 - Handle reconfiguring audio device if it went away while the stream was stopped. r=padenot
2015-03-02 18:07:43 +13:00
Matthew Gregan
f2307b8011
Bug 1135878 - Simply post-error cleanup logic in WASAPI cubeb backend. r=padenot
2015-02-24 13:42:33 +13:00
Matthew Gregan
1b8685913f
Bug 1133386 - Introduce an XASSERT() macro to libcubeb rather than (ab)using assert(). r=padenot f=dmajor
2015-02-20 13:42:14 +13:00
Paul Adenot
6ebc006367
Bug 1124411 - Add timeout when calling WaitForMultipleObjects to wallpaper over a deadlock. r=kinetik
2015-02-19 19:35:07 +13:00
Matthew Gregan
11b8535276
Bug 1134102 - Move cubeb's stream reconfiguration to render thread. r=padenot
2015-02-18 18:43:07 +13:00
Paul Adenot
c0397fa4b7
Bug 1133190 - Don't use auto_unlock in paths where lock is being destroyed. r=kinetik
2015-02-19 19:35:06 +13:00
Paul Adenot
389746ec78
Bug 1132034 - Properly round the number of frame for the accumulating clock in WASAPI. r=kinetik
2015-02-19 19:35:06 +13:00
Matthew Gregan
092fdd407d
Bug 1134078 - Don't restart stopped cubeb streams when handling device change notifications. r=padenot
2015-02-18 16:06:55 +13:00
Matthew Gregan
2bf339f2a9
Bug 1132257 - Update cubeb from upstream. r=padenot
2015-02-17 11:52:59 +13:00
Matthew Gregan
9d80ee47b6
Bug 1131788 - cubeb: Unable to use InterlockedAdd64 on MSVC2010 (and mingw). r=padenot
2015-02-10 17:45:00 +13:00
David Major
1ccaf82726
Bug 1131871 - auto_com should only uninitialize when successful. r=padenot
2015-02-12 10:33:45 +13:00
Paul Adenot
318ed9a8bd
Bug 1131768 - Unlock before tearing down the stream in case of error, to avoid recursive locking. r=kinetik
2015-02-11 00:12:09 +01:00
Paul Adenot
3830f5aad3
Bug 1127213 - Fix various issues with the device change notification in the WASAPI cubeb backend. r=kinetik
...
This patch does the following:
- Introduces an owned_critical_section object to be able to assert that a
current thread owns a critical section
- Change the auto_lock to use the above, add auto_unlock (basically like the
Gecko AutoEnter/AutoExit things)
- Fix an issue during audio output device switch where the clock would use the
old sample rate. Apparently I did not notice this because my headset and the
sound card on this laptop have the same rate
- Check that we could acquire a device_enumerator in the ctor before
deallocating in the dtor, as that can happen if a ton of streams are running at
once (I had this issue running the full mochitest suite)
- Stop getting another device_enumator in unregister_notification_client, fixing a leak
- Assert that setup_wasapi_stream and close_wasapi_stream are called with the lock held, this was the cause of the crash for this bug
- Make close_wasapi_stream clear out its state to make sure setup_wasapi_stream
and close_wasapi_stream are called in the right order (especially, not two
setup_wasapi_stream without close in between, since that would leak stuff)
- In wasapi_stream_destroy, unregister the notification client before destroying
the CRITICAL_SECTION (this was the cause of a crash I duped against this bug)
2015-02-09 14:42:43 +01:00
Carsten "Tomcat" Book
8da798492b
Backed out changeset 2f46afa97421 (bug 1127213) for another bustage on a CLOSED TREE
2015-02-09 16:51:14 +01:00
Paul Adenot
3800ae52f4
Bug 1123768 - Backout bug 1108455 to avoid truncating the end of audio streams on Vista+. r=kinetik
2015-02-09 14:43:03 +01:00
