OSDN Git Service
Jean-Baptiste Queru [Mon, 8 Jul 2013 18:25:40 +0000 (11:25 -0700)]
am
06197925: am
267363ee: Merge "Prevent Media Server crash while the AwesomePlayer getBitrate return true with mBitrate equals to zero."
* commit '
061979253985e59c4fcb82ede232717736e633b1':
Prevent Media Server crash while the AwesomePlayer getBitrate return true with mBitrate equals to zero.
Jean-Baptiste Queru [Mon, 8 Jul 2013 18:25:34 +0000 (11:25 -0700)]
am
7266bc80: am
ca7f1609: Merge "Prevent double disconnects in RTSPSource"
* commit '
7266bc80850dc327e9f18c27b81613f959399630':
Prevent double disconnects in RTSPSource
Jean-Baptiste Queru [Mon, 8 Jul 2013 18:21:48 +0000 (11:21 -0700)]
am
267363ee: Merge "Prevent Media Server crash while the AwesomePlayer getBitrate return true with mBitrate equals to zero."
* commit '
267363ee2836ff9bc18d7fdd93aaedf19de546ad':
Prevent Media Server crash while the AwesomePlayer getBitrate return true with mBitrate equals to zero.
Jean-Baptiste Queru [Mon, 8 Jul 2013 18:21:45 +0000 (11:21 -0700)]
am
ca7f1609: Merge "Prevent double disconnects in RTSPSource"
* commit '
ca7f160921a4a1a5da32ebe8fab1067bd8998a1f':
Prevent double disconnects in RTSPSource
Jean-Baptiste Queru [Mon, 8 Jul 2013 18:17:43 +0000 (18:17 +0000)]
Merge "Prevent Media Server crash while the AwesomePlayer getBitrate return true with mBitrate equals to zero."
Jean-Baptiste Queru [Mon, 8 Jul 2013 18:17:02 +0000 (18:17 +0000)]
Merge "Prevent double disconnects in RTSPSource"
Jeff Tinker [Sun, 7 Jul 2013 19:26:16 +0000 (12:26 -0700)]
am
eced745a: Merge "Part of fix for TEE resource leaks when app crashes" into jb-mr2-dev
* commit '
eced745af7d2d27a8c45e04eb071d4fb30cd3e48':
Part of fix for TEE resource leaks when app crashes
Jeff Tinker [Sun, 7 Jul 2013 19:21:55 +0000 (19:21 +0000)]
Merge "Part of fix for TEE resource leaks when app crashes" into jb-mr2-dev
Jeff Tinker [Wed, 3 Jul 2013 22:38:58 +0000 (15:38 -0700)]
Part of fix for TEE resource leaks when app crashes
bug:
9672606
Change-Id: I98cf65260d5e43d34e3897452f6f26ed2c89294e
Eric Laurent [Wed, 3 Jul 2013 15:16:46 +0000 (08:16 -0700)]
am
f59a4b39: fix scheduling policy service death detection
* commit '
f59a4b393f4844c5bbc8d6212364bdddea33d232':
fix scheduling policy service death detection
hovanchen [Tue, 2 Jul 2013 07:07:44 +0000 (15:07 +0800)]
Prevent Media Server crash while the AwesomePlayer getBitrate return true with mBitrate equals to zero.
[Cause]
- getBitrate(int64_t *bitrate) will return true if mBitrate = 0.
- Then, in getCachedDuration_l(int64_t *durationUs, bool *eos),
we might execute the following function when
getBitrate(&bitrate) = true and bitrate = 0.
*durationUs = cachedDataRemaining * 8000000ll / bitrate;
- Mediaserver will be crashed when divided by zero.
- The mediaserver crash often occur when DLNA IOP tests on Golden DMS
with AAC_ADTS_320 audio.
[Solution]
- Prevent the divide function executing when bitrate is zero.
Change-Id: I4439d92cee5faec95df2109e9186c33b3fff6c66
Eric Laurent [Tue, 2 Jul 2013 18:15:41 +0000 (11:15 -0700)]
fix scheduling policy service death detection
Check status of transactions to scheduling policy service
and re-acquire a binder interface in case of DEAD_OBJECT.
Bug:
8875559.
Change-Id: I1e00bd44e2d4723b3ec95d5c31d9652ba08e238a
The Android Open Source Project [Mon, 1 Jul 2013 14:42:56 +0000 (07:42 -0700)]
Reconcile with jb-mr2-release jb-mr2-zeroday-release - do not merge
Change-Id: Idf0f0cdf02e6e94603ff003b6baf869b4414e1cf
Fredrik Rosin [Wed, 6 Mar 2013 12:42:53 +0000 (13:42 +0100)]
Prevent double disconnects in RTSPSource
Prevent a null pointer from occuring by checking that we are not
already disconnected before performing a disconnect, in case
multiple disconnects have been issued.
Change-Id: Ib88800cd246086fd57f77d10748d41b1aefe2084
The Android Automerger [Sun, 30 Jun 2013 12:55:22 +0000 (05:55 -0700)]
merge in jb-mr2-zeroday-release history after reset to jb-mr2-dev
Andreas Huber [Fri, 28 Jun 2013 22:31:37 +0000 (15:31 -0700)]
am
47a1fe04: am
a9aa6baa: Merge "Set u.ext_data to null after it has been freed"
* commit '
47a1fe04637793f48d808261b1fac47acc3de1e1':
Set u.ext_data to null after it has been freed
Andreas Huber [Fri, 28 Jun 2013 22:29:01 +0000 (15:29 -0700)]
am
a9aa6baa: Merge "Set u.ext_data to null after it has been freed"
* commit '
a9aa6baa5d86fc1a5bd37fc636d766d9e3ad2507':
Set u.ext_data to null after it has been freed
Andreas Huber [Fri, 28 Jun 2013 22:26:42 +0000 (22:26 +0000)]
Merge "Set u.ext_data to null after it has been freed"
Andreas Huber [Wed, 29 May 2013 21:20:52 +0000 (14:20 -0700)]
Update HTTP proxy configuration for all media playback inside stagefright.
Change-Id: Ie0dd00045aba668d8b49da73224e7a7c9c04f69b
related-to-bug:
8873723
(cherry picked from commit
2704965b8a1ff3b7450ff58ccecf86d8ec688c40)
Glenn Kasten [Fri, 28 Jun 2013 16:33:15 +0000 (09:33 -0700)]
am
83790fcc: am
bdf33c98: Merge "Fix compilation warning: incorrect format specifier for type \'ssize_t\'."
* commit '
83790fcce4f5cda1e7c89f1cbed452426502f227':
Fix compilation warning: incorrect format specifier for type 'ssize_t'.
Glenn Kasten [Fri, 28 Jun 2013 16:30:58 +0000 (09:30 -0700)]
am
bdf33c98: Merge "Fix compilation warning: incorrect format specifier for type \'ssize_t\'."
* commit '
bdf33c98912090176183d3e9b9a9217dbbdca6da':
Fix compilation warning: incorrect format specifier for type 'ssize_t'.
Glenn Kasten [Fri, 28 Jun 2013 16:28:48 +0000 (16:28 +0000)]
Merge "Fix compilation warning: incorrect format specifier for type 'ssize_t'."
Eino-Ville Talvala [Fri, 28 Jun 2013 14:07:29 +0000 (07:07 -0700)]
am
a691ff3c: Camera2/3: Don\'t allow recording and callbacks to coexist.
* commit '
a691ff3c03e38e148bbefed35ebb15e552a12613':
Camera2/3: Don't allow recording and callbacks to coexist.
Christer Fletcher [Tue, 18 Jun 2013 06:55:09 +0000 (08:55 +0200)]
Set u.ext_data to null after it has been freed
SEGV_MAPPER crash could happen in MetaData::typed_data::freeStorage.
The faulty address could be both random pointers and
deadbaad. It
seems like free could be called more than once on u.ext_data. Set
u.ext_data to null after it has been freed as there already is a
null check before trying to free it.
Change-Id: I710c49f9c231bd8852159914c0b2add58a685f0e
Eino-Ville Talvala [Thu, 27 Jun 2013 01:23:23 +0000 (18:23 -0700)]
Camera2/3: Don't allow recording and callbacks to coexist.
- Tear down conflicting streams when necessary.
- Shut down callbacks if recording starts
- Do not allow callbacks to start if recording is active
Per the current camera API, recording and preview callbacks cannot be
active simultaneously. However, the framework did not explicitly
disallow this, and in fact left the streams configured once they were
created, even if switching between the two operational modes.
In addition, no guards existed for trying to enable both recording and
callbacks at the same time.
Bug:
9423825
Change-Id: I7d6e6114c2e14fcfb5299b4c72ad557895cbf4b8
Scott Fan [Sun, 28 Apr 2013 02:13:54 +0000 (10:13 +0800)]
Fix compilation warning: incorrect format specifier for type 'ssize_t'.
warning: format '%ld' expects argument of type 'long int',
but argument 7 has type 'ssize_t {aka int}' [-Wformat]
Change-Id: I823835a26320a3abb82f8dff9fbe82e5132e7556
Signed-off-by: Scott Fan <fancp2007@gmail.com>
The Android Open Source Project [Tue, 25 Jun 2013 23:12:03 +0000 (16:12 -0700)]
Reconcile with jb-mr2-zeroday-release - do not merge
Change-Id: I4de46f3bfcfb80867bbeb52a76d383f5870e74f4
Jean-Baptiste Queru [Tue, 25 Jun 2013 17:40:38 +0000 (10:40 -0700)]
am
3a4ff0b8: am
d74a9ee8: Merge "libstagefright: fix the duration "00:00" for some httplive link"
* commit '
3a4ff0b8bf464c9c286d63326bea42609527051e':
libstagefright: fix the duration "00:00" for some httplive link
Jean-Baptiste Queru [Tue, 25 Jun 2013 17:38:44 +0000 (10:38 -0700)]
am
d74a9ee8: Merge "libstagefright: fix the duration "00:00" for some httplive link"
* commit '
d74a9ee83fd6e033fd2260ca3777390d7e5f817a':
libstagefright: fix the duration "00:00" for some httplive link
Jean-Baptiste Queru [Tue, 25 Jun 2013 17:37:28 +0000 (17:37 +0000)]
Merge "libstagefright: fix the duration "00:00" for some httplive link"
The Android Automerger [Mon, 24 Jun 2013 21:55:06 +0000 (14:55 -0700)]
merge in jb-mr2-zeroday-release history after reset to jb-mr2-dev
The Android Automerger [Mon, 24 Jun 2013 21:41:40 +0000 (14:41 -0700)]
Merge remote-tracking branch 'origin/jb-mr2-zeroday-release' into HEAD
The Android Open Source Project [Fri, 21 Jun 2013 18:56:55 +0000 (11:56 -0700)]
Reconcile with jb-mr2-zeroday-release - do not merge
Change-Id: Id1c9edbb5b7b398fb863cbbd79e673ab2c10df9a
The Android Automerger [Fri, 21 Jun 2013 18:35:26 +0000 (11:35 -0700)]
merge in jb-mr2-zeroday-release history after reset to jb-mr2-dev
Jean-Michel Trivi [Fri, 21 Jun 2013 05:23:58 +0000 (22:23 -0700)]
am
209bbbcf: Apply EQ and bass boost first in effect chain
* commit '
209bbbcf4190231f9dede758cbe77d109919f9f1':
Apply EQ and bass boost first in effect chain
Jean-Michel Trivi [Thu, 20 Jun 2013 23:49:25 +0000 (16:49 -0700)]
Apply EQ and bass boost first in effect chain
EQ and bass boost can improperly interact with other effects
(e.g. virtualizer) outside of the bundle. Apply them first
in an effect chain, which is equivalent to using them to affect
the content of a track, rather than how a track is presented,
when additional effects are cascaded behind EQ and BB.
Bug
8589316
Change-Id: Ifb34e46bc1969f613cdc90ee2a79af677e2fad92
Glenn Kasten [Thu, 20 Jun 2013 22:16:00 +0000 (15:16 -0700)]
am
f3e21c30: am
60f3f102: Merge "Prevent AudioCommands being freed before read"
* commit '
f3e21c301b8fef49eecf9c716373fcbfbaa47670':
Prevent AudioCommands being freed before read
Glenn Kasten [Thu, 20 Jun 2013 22:13:19 +0000 (15:13 -0700)]
am
60f3f102: Merge "Prevent AudioCommands being freed before read"
* commit '
60f3f102ba2973cc31e5c7ea00d95bbf19968e8b':
Prevent AudioCommands being freed before read
Glenn Kasten [Thu, 20 Jun 2013 22:11:25 +0000 (22:11 +0000)]
Merge "Prevent AudioCommands being freed before read"
Christer Fletcher [Fri, 18 Jan 2013 14:27:03 +0000 (15:27 +0100)]
Prevent AudioCommands being freed before read
When AudioCommandThread::threadLoop process AudioCommands it was
possible for it to delete a command where the posting thread still
hadn't read that status from it.
If a second command signaled the thread loop to continue after it
had inserted a new command while the thread loop was actually waiting
for the first command to report that it had read the status the
thread loop would continue and delete the first command.
Changed the wait condition when waiting for the calling thread to
read status to use command->mCond instead of mWaitWorkCV. This way
it's guaranteed that the signal to continue comes from the correct
thread.
Change-Id: Ia69b48cb4fdfaf8b4c83b56a197fb9f2058a92d1
Eino-Ville Talvala [Wed, 19 Jun 2013 21:23:22 +0000 (14:23 -0700)]
am
e6478de7: Camera2/3: Avoid shutdown race in callback processor.
* commit '
e6478de72a8ae1d3abb3e8c20898977c45354a4f':
Camera2/3: Avoid shutdown race in callback processor.
Eino-Ville Talvala [Wed, 19 Jun 2013 15:42:17 +0000 (08:42 -0700)]
Camera2/3: Avoid shutdown race in callback processor.
It's possible, during shutdown, for callback processor's heap to be
destroyed when it's about to send the callback back to the user.
Properly copy the heap reference to a local variable before unlocking
the mutex.
Bug:
9485959
Change-Id: I301347b77145f19c7ac721b9127dc74f122acce2
Eino-Ville Talvala [Mon, 17 Jun 2013 23:23:56 +0000 (16:23 -0700)]
am
241b5279: Camera3: Power: Properly shut down unused thread.
* commit '
241b52798809d8db3d369af05ace1f73f723f29b':
Camera3: Power: Properly shut down unused thread.
Eino-Ville Talvala [Mon, 17 Jun 2013 20:29:16 +0000 (13:29 -0700)]
Camera3: Power: Properly shut down unused thread.
Bug:
9466221
Change-Id: I9ed0740561b2e04954eebc1b8aec89cfab40a3a0
The Android Open Source Project [Sun, 16 Jun 2013 14:44:00 +0000 (07:44 -0700)]
Reconcile with jb-mr2-zeroday-release - do not merge
Change-Id: I2ab98a5fccea49e0f4e6d22fddca485ad50b63b8
The Android Automerger [Sun, 16 Jun 2013 12:55:26 +0000 (05:55 -0700)]
merge in jb-mr2-zeroday-release history after reset to jb-mr2-dev
Eino-Ville Talvala [Fri, 14 Jun 2013 15:35:48 +0000 (08:35 -0700)]
am
26bc908b: Camera2/3: Make sure to wait for idle after still capture.
* commit '
26bc908b3dafb5d80f092a22c5c71116526eae06':
Camera2/3: Make sure to wait for idle after still capture.
Eino-Ville Talvala [Fri, 14 Jun 2013 15:35:48 +0000 (08:35 -0700)]
am
214a17fd: Camera3: Always close HAL device even in face of errors.
* commit '
214a17fd37ef85fc841d3157b1e9096e1aa1b42f':
Camera3: Always close HAL device even in face of errors.
Glenn Kasten [Fri, 14 Jun 2013 12:48:38 +0000 (05:48 -0700)]
Glenn Kasten [Fri, 14 Jun 2013 12:43:03 +0000 (05:43 -0700)]
Glenn Kasten [Fri, 14 Jun 2013 12:39:16 +0000 (12:39 +0000)]
Merge "Fix typo."
Eino-Ville Talvala [Fri, 14 Jun 2013 01:04:28 +0000 (18:04 -0700)]
Camera2/3: Make sure to wait for idle after still capture.
While this wait should be immediate, it's important to guarantee
that preview is stopped when takePicture completes, and that the
device knows itself to be idle.
Bug:
9413025
Change-Id: I1c151ba920e226107ee048918187364990f444c0
Eino-Ville Talvala [Thu, 13 Jun 2013 19:20:02 +0000 (12:20 -0700)]
Camera3: Always close HAL device even in face of errors.
To improve robustness, make sure the HAL device instance is always
closed, even if it cannot be put into the IDLE state.
Bug:
9414167
Change-Id: I8e7035ff0afe7214951f9824e561b340cb37592e
Eino-Ville Talvala [Thu, 13 Jun 2013 17:34:07 +0000 (10:34 -0700)]
am
1191454a: Merge "Camera3: Skipping no-ops good, not updating state machine bad." into jb-mr2-dev
* commit '
1191454a3a960a1972b0f36e625796c6c1b2c9ca':
Camera3: Skipping no-ops good, not updating state machine bad.
Eino-Ville Talvala [Thu, 13 Jun 2013 17:34:07 +0000 (10:34 -0700)]
am
c4a8d2dd: Merge "Camera2/3: Create JPEG stream unconditionally." into jb-mr2-dev
* commit '
c4a8d2ddfff6d91e493692251733f20ece3e713d':
Camera2/3: Create JPEG stream unconditionally.
Eino-Ville Talvala [Thu, 13 Jun 2013 17:33:20 +0000 (17:33 +0000)]
Merge "Camera3: Skipping no-ops good, not updating state machine bad." into jb-mr2-dev
Eino-Ville Talvala [Thu, 13 Jun 2013 17:33:07 +0000 (17:33 +0000)]
Merge "Camera2/3: Create JPEG stream unconditionally." into jb-mr2-dev
The Android Open Source Project [Thu, 13 Jun 2013 15:22:42 +0000 (08:22 -0700)]
Reconcile with jb-mr2-zeroday-release - do not merge
Change-Id: Id9419bf573d550f1ebebcc96fe615e1aaf8f36ef
The Android Automerger [Thu, 13 Jun 2013 12:55:17 +0000 (05:55 -0700)]
merge in jb-mr2-zeroday-release history after reset to jb-mr2-dev
The Android Open Source Project [Thu, 13 Jun 2013 01:04:12 +0000 (18:04 -0700)]
Reconcile with jb-mr2-release - do not merge
Change-Id: I024b04bf1e47cc7c10209e83f640cc6f12e03274
Eino-Ville Talvala [Thu, 13 Jun 2013 00:06:41 +0000 (17:06 -0700)]
Camera3: Skipping no-ops good, not updating state machine bad.
If a HAL-level stream reconfigure is not needed, still need to
transition camera device to ACTIVE state in configureStreams.
Otherwise, shutdown functions will not wait for device to drain
properly.
Bug:
9413025
Change-Id: I0b15792097c7e5771015f2a362735d5f88b8c6ea
Eric Laurent [Tue, 11 Jun 2013 21:38:48 +0000 (14:38 -0700)]
audioflinger: fix effects on direct output threads
PlaybackThread::addTrack_l() uses the assumption that
effects are attached to a track only if the track accumulation
buffer is different from the mixer thread output buffer.
This is not true for direct output threads where only one track is active
an only one buffer is needed.
This assumption is an optimization to avoid checking for effect chains with
the same session ID each time a track is processed. The optimization
is not key if only one track is attached to the thread which is the case for
direct outputs.
Current code fails to increment the active track count in the effect chain
on direct output threads when a track is started thus making the effect
framework clear the mix buffer and produce silence each time the mixer runs.
The fix consists in removing the optimization described above.
Bug:
9324989.
Change-Id: Id7a6337450ed90d326299c2ce9fc02f4b9e2fa6f
Lajos Molnar [Wed, 12 Jun 2013 16:29:43 +0000 (09:29 -0700)]
am
0662f7b6: Merge "stagefright: synchronously change nativewindow crop" into jb-mr2-dev
* commit '
0662f7b6735df0a140f6512285d05f8f9a38bb98':
stagefright: synchronously change nativewindow crop
Lajos Molnar [Wed, 12 Jun 2013 16:27:40 +0000 (16:27 +0000)]
Merge "stagefright: synchronously change nativewindow crop" into jb-mr2-dev
Eric Laurent [Wed, 12 Jun 2013 15:46:03 +0000 (08:46 -0700)]
am
7be48189: Merge "audioflinger: fix effects on direct output threads" into jb-mr2-dev
* commit '
7be48189c0aefb82825a7689d9e1005175570779':
audioflinger: fix effects on direct output threads
Eric Laurent [Wed, 12 Jun 2013 15:44:42 +0000 (15:44 +0000)]
Merge "audioflinger: fix effects on direct output threads" into jb-mr2-dev
The Android Open Source Project [Wed, 12 Jun 2013 14:23:29 +0000 (07:23 -0700)]
Reconcile with jb-mr2-zeroday-release - do not merge
Change-Id: I87a7986866714bfd922259cf961d6d5f5b752362
The Android Automerger [Wed, 12 Jun 2013 12:55:52 +0000 (05:55 -0700)]
merge in jb-mr2-zeroday-release history after reset to jb-mr2-dev
Lajos Molnar [Wed, 12 Jun 2013 00:56:24 +0000 (17:56 -0700)]
stagefright: synchronously change nativewindow crop
Only change the nativewindow crop before queuing frames
with the new crop.
Change-Id: Icd8a84c82e66a8be3d6bc1675c777b1de8dc39b9
Signed-off-by: Lajos Molnar <lajos@google.com>
Bug:
9395753
Eino-Ville Talvala [Wed, 12 Jun 2013 01:21:05 +0000 (18:21 -0700)]
am
ea26c777: Camera3: Skip no-op stream configurations.
* commit '
ea26c7772f4721657db409068d4bed194ae49c94':
Camera3: Skip no-op stream configurations.
Eino-Ville Talvala [Tue, 11 Jun 2013 23:04:06 +0000 (16:04 -0700)]
Camera3: Skip no-op stream configurations.
If configuring the same set of streams more than once, don't actually
call into the HAL for the second and subsequent configure calls, since
they're no-ops.
This can speed up camera operation substantially if the
HAL implementation does not detect no-ops on its own and does a full
shutdown/restart on each configure call.
Bug:
9392513
Change-Id: I23baf4acbae2304735899adcf8e17565fa94d31d
Eric Laurent [Tue, 11 Jun 2013 21:38:48 +0000 (14:38 -0700)]
audioflinger: fix effects on direct output threads
PlaybackThread::addTrack_l() uses the assumption that
effects are attached to a track only if the track accumulation
buffer is different from the mixer thread output buffer.
This is not true for direct output threads where only one track is active
an only one buffer is needed.
This assumption is an optimization to avoid checking for effect chains with
the same session ID each time a track is processed. The optimization
is not key if only one track is attached to the thread which is the case for
direct outputs.
Current code fails to increment the active track count in the effect chain
on direct output threads when a track is started thus making the effect
framework clear the mix buffer and produce silence each time the mixer runs.
The fix consists in removing the optimization described above.
Bug:
9324989.
Change-Id: Id7a6337450ed90d326299c2ce9fc02f4b9e2fa6f
Eino-Ville Talvala [Mon, 10 Jun 2013 22:12:01 +0000 (15:12 -0700)]
Camera2/3: Create JPEG stream unconditionally.
Instead of creating the JPEG output stream only at first use,
create it at time of preview start. This is important for reducing
the first-capture latency on HAL3 devices, and for ZSL on them.
Bug:
9339858
Change-Id: I4296f706cfd151c47ef315149e87d01fe554aaa4
The Android Open Source Project [Mon, 10 Jun 2013 15:13:42 +0000 (08:13 -0700)]
Reconcile with jb-mr2-release jb-mr2-zeroday-release - do not merge
Change-Id: I9864de7c2785357cea0c4b669607fee76e9f09e8
The Android Automerger [Sun, 9 Jun 2013 12:55:29 +0000 (05:55 -0700)]
merge in jb-mr2-zeroday-release history after reset to jb-mr2-dev
Zhijun He [Thu, 6 Jun 2013 23:52:02 +0000 (16:52 -0700)]
camera2/3: Fix deadlock take picture while preview callback
Acquiring mInputMutex before SharedParameters mutex in preview callback thread
causes circular locking dependency between callback thread and capture callback
thread.
Enforce the right lock ordering to break the dead lock.
Bug:
9323319
Change-Id: Iea2e025e4d9e29debcf74297c19930075017e179
Zhijun He [Fri, 7 Jun 2013 16:59:47 +0000 (09:59 -0700)]
am
ba696bbf: Merge "camera2/3: Fix deadlock take picture while preview callback" into jb-mr2-dev
* commit '
ba696bbff88d5627beaa0be95be78ba30138983d':
camera2/3: Fix deadlock take picture while preview callback
Zhijun He [Fri, 7 Jun 2013 16:58:12 +0000 (16:58 +0000)]
Merge "camera2/3: Fix deadlock take picture while preview callback" into jb-mr2-dev
Eric Laurent [Fri, 7 Jun 2013 15:57:00 +0000 (08:57 -0700)]
am
18b46679: Merge "AudioFlinger: do not cache audio device when 0." into jb-mr2-dev
* commit '
18b466790aa0a55ff4282ed8b5e550e58e0444b2':
AudioFlinger: do not cache audio device when 0.
Eric Laurent [Fri, 7 Jun 2013 15:52:59 +0000 (15:52 +0000)]
Merge "AudioFlinger: do not cache audio device when 0." into jb-mr2-dev
Glenn Kasten [Fri, 7 Jun 2013 14:09:39 +0000 (07:09 -0700)]
am
911a184b: am
132bc5bd: Merge "Noise will be heard if audio sample rate not matched with audio track"
* commit '
911a184bb5f5177a00bb4a8c80453f22e67b449c':
Noise will be heard if audio sample rate not matched with audio track
Glenn Kasten [Fri, 7 Jun 2013 14:06:15 +0000 (07:06 -0700)]
am
132bc5bd: Merge "Noise will be heard if audio sample rate not matched with audio track"
* commit '
132bc5bd1c68aa807a20e1655f009f74433be588':
Noise will be heard if audio sample rate not matched with audio track
Glenn Kasten [Fri, 7 Jun 2013 14:04:45 +0000 (14:04 +0000)]
Merge "Noise will be heard if audio sample rate not matched with audio track"
Ming Zhou [Wed, 27 Feb 2013 09:12:37 +0000 (17:12 +0800)]
Noise will be heard if audio sample rate not matched with audio track
When audio sample rate which set to audio track is not the same with
the actual pcm data, noise will be heard. Fix the bug when write 8 bit
pcm samples.
Change-Id: Idcb0d7b0e9aaa250dd22b758c8337e23d1706049
Signed-off-by: Ming Zhou <b42586@freescale.com>
Signed-off-by: guoyin.chen <guoyin.chen@freescale.com>
Eric Laurent [Fri, 7 Jun 2013 01:29:01 +0000 (18:29 -0700)]
AudioFlinger: do not cache audio device when 0.
AudioFlinger should not cache the audio device
passed to the audio HAL when it is AUDIO_DEVICE_NONE
but keep previous valid selection instead.
Bug:
9323399.
Change-Id: I6f9480e55a21be4115453e6a5eebc5cf2536c476
Zhijun He [Thu, 6 Jun 2013 23:52:02 +0000 (16:52 -0700)]
camera2/3: Fix deadlock take picture while preview callback
Acquiring mInputMutex before SharedParameters mutex in preview callback thread
causes circular locking dependency between callback thread and capture callback
thread.
Enforce the right lock ordering to break the dead lock.
Bug:
9323319
Change-Id: Iea2e025e4d9e29debcf74297c19930075017e179
Zhijun He [Thu, 6 Jun 2013 20:51:22 +0000 (13:51 -0700)]
camera3: Fix crash in set error state.
When camera open fails in camera3device initialize call, setErrorStateLockedV()
tries to access RequestThread object that is not created yet.
Bug:
9319518
Change-Id: Id5c6c0150ecaecb198045d5c6eb1b2ceebf5a727
Zhijun He [Thu, 6 Jun 2013 03:00:09 +0000 (20:00 -0700)]
camera2/3: Add protection to callback processing call
Callback stream in callbackprocessor could be delleted while process or
discard callback are ongoing, which leads CpuConsumer related call crash.
Bug
9303911
Change-Id: Ib6137e512208e5603468eba166640729b744e8f9
Zhijun He [Thu, 6 Jun 2013 21:18:06 +0000 (14:18 -0700)]
am
b05eeaed: camera3: Fix crash in set error state.
* commit '
b05eeaedacaff92b6e5ac89f99b0fccdf7643f09':
camera3: Fix crash in set error state.
Zhijun He [Thu, 6 Jun 2013 20:51:22 +0000 (13:51 -0700)]
camera3: Fix crash in set error state.
When camera open fails in camera3device initialize call, setErrorStateLockedV()
tries to access RequestThread object that is not created yet.
Bug:
9319518
Change-Id: Id5c6c0150ecaecb198045d5c6eb1b2ceebf5a727
Zhijun He [Thu, 6 Jun 2013 18:09:54 +0000 (11:09 -0700)]
am
62a8f67f: camera2/3: Add protection to callback processing call
* commit '
62a8f67f33faf19d0ff815672dd19bb71dd0d650':
camera2/3: Add protection to callback processing call
Zhijun He [Thu, 6 Jun 2013 03:00:09 +0000 (20:00 -0700)]
camera2/3: Add protection to callback processing call
Callback stream in callbackprocessor could be delleted while process or
discard callback are ongoing, which leads CpuConsumer related call crash.
Bug
9303911
Change-Id: Ib6137e512208e5603468eba166640729b744e8f9
The Android Open Source Project [Thu, 6 Jun 2013 17:37:16 +0000 (10:37 -0700)]
Reconcile with jb-mr2-release jb-mr2-zeroday-release - do not merge
Change-Id: I0aa9bb7e725a8352da3eae512c2a62df4f12e3b3
The Android Automerger [Thu, 6 Jun 2013 12:55:27 +0000 (05:55 -0700)]
merge in jb-mr2-zeroday-release history after reset to jb-mr2-dev
Masaki Muranaka [Thu, 6 Jun 2013 09:36:34 +0000 (09:36 +0000)]
Fix typo.
Change-Id: Ia72ed2bef54751dc2d6c3379a58de95a0e9ebfdd
Eino-Ville Talvala [Tue, 4 Jun 2013 23:34:35 +0000 (16:34 -0700)]
Camera2/3: Move recording frame processing to its own thread.
Due to new deadlock opportunities from HAL3 integration, recording
frame processing needs to move out of the new frame handler into its
own thread in StreamingProcessor.
Bug:
9282913
Change-Id: I8d2926f2de9d0d395eef3c328c65caebbdbd2827
Lajos Molnar [Fri, 26 Apr 2013 23:11:45 +0000 (16:11 -0700)]
stagefright: Fix port-reconfiguration & output-buffer-filled race condition
Remove the invalid assumption that when a port-reconfiguration event
is received, buffers cannot be downstream (waiting to be rendered).
Luckily, these buffers are properly handled (freed) after they are
sent to be rendered.
Change-Id: I1df39c1ffc2bfb96f8b7b4ee5be07cae654f956f
Signed-off-by: Lajos Molnar <lajos@google.com>
Bug:
8736466
Eino-Ville Talvala [Wed, 5 Jun 2013 18:43:49 +0000 (11:43 -0700)]
am
254446a8: Camera2/3: Move recording frame processing to its own thread.
* commit '
254446a80cb13d5beb606bfc3e60a17473869478':
Camera2/3: Move recording frame processing to its own thread.
The Android Automerger [Wed, 5 Jun 2013 17:11:30 +0000 (10:11 -0700)]
merge in jb-mr2-zeroday-release history after reset to jb-mr2-dev
Eino-Ville Talvala [Tue, 4 Jun 2013 23:34:35 +0000 (16:34 -0700)]
Camera2/3: Move recording frame processing to its own thread.
Due to new deadlock opportunities from HAL3 integration, recording
frame processing needs to move out of the new frame handler into its
own thread in StreamingProcessor.
Bug:
9282913
Change-Id: I8d2926f2de9d0d395eef3c328c65caebbdbd2827