Commit Graph

2689 Commits

Author SHA1 Message Date
Michael Lentine
2bb7270783 am 15b2ecf5: am dcc959be: Merge "Adding eglInitialize and eglTerminate image creation and deletion." into lmp-mr1-dev
* commit '15b2ecf599414c86cb198f3ffc126b50f88b5aa0':
  Adding eglInitialize and eglTerminate image creation and deletion.
2014-10-21 18:43:47 +00:00
Michael Lentine
15b2ecf599 am dcc959be: Merge "Adding eglInitialize and eglTerminate image creation and deletion." into lmp-mr1-dev
* commit 'dcc959beb77099d319a94e3985da2f4ea4064e7b':
  Adding eglInitialize and eglTerminate image creation and deletion.
2014-10-21 18:35:24 +00:00
Michael Lentine
dcc959beb7 Merge "Adding eglInitialize and eglTerminate image creation and deletion." into lmp-mr1-dev 2014-10-21 18:20:45 +00:00
Jesse Hall
937052fb1c am a069e853: am 649bafb7: Merge "bufferqueue: workaround: allow NULL fence with queueBuffer (DO NOT MERGE)" into lmp-dev
* commit 'a069e8532ba272ac5b935e332e426a3dacf827fc':
  bufferqueue: workaround: allow NULL fence with queueBuffer (DO NOT MERGE)
2014-10-20 14:31:38 +00:00
Jesse Hall
a069e8532b am 649bafb7: Merge "bufferqueue: workaround: allow NULL fence with queueBuffer (DO NOT MERGE)" into lmp-dev
* commit '649bafb7ba2c3204295bc3451ff7af7e28e714ed':
  bufferqueue: workaround: allow NULL fence with queueBuffer (DO NOT MERGE)
2014-10-20 14:17:18 +00:00
Jesse Hall
5b0cbcf946 bufferqueue: workaround: allow NULL fence with queueBuffer (DO NOT MERGE)
On one device there is a bug, not yet root-caused, that causes fence
fds to not make it across binder from producer to consumer in the
IGraphicBufferProducer::queueBuffer call. Rather than returning an
error, which the producer typically treats as a fatal error, this
change allows the buffer to be queued with no fence. This avoids an
application crash at the risk of (likely single-frame) visible
corruption.

Bug: 17946343
Change-Id: I9ca89f94098c455e1e90f5f58d5336c936b04a9c
2014-10-19 04:57:10 +00:00
Dan Albert
0e978be998 am 3235d8e4: am c52e54c4: Merge "Some more test cleanup."
* commit '3235d8e477e1a0ad72d22acfff7d9966f882daa6':
  Some more test cleanup.
2014-10-17 17:57:51 +00:00
Dan Albert
3235d8e477 am c52e54c4: Merge "Some more test cleanup."
* commit 'c52e54c47184704613d446eb15d051ceb43d82b3':
  Some more test cleanup.
2014-10-17 17:53:10 +00:00
Dan Albert
ed7d193ae9 Some more test cleanup.
Needed to build successfully when libc++ is the default STL.

Also fix a missing include. Not sure how this was getting by before.
Probably some transitive include.

Change-Id: Ie96b79e96b4e154d786fbf2d5ae5cf5892608837
2014-10-16 20:46:05 -07:00
Dan Albert
fad2e3abf0 am 27b780b5: am 9cc84950: Merge "Don\'t check if this == NULL."
* commit '27b780b517eb18582f978da64febb31ecfa8e4bf':
  Don't check if this == NULL.
2014-10-15 22:57:13 +00:00
Dan Albert
27b780b517 am 9cc84950: Merge "Don\'t check if this == NULL."
* commit '9cc849500df0bba42d3ed19b90c38ff1dffd2262':
  Don't check if this == NULL.
2014-10-15 22:52:51 +00:00
Dan Albert
9cc849500d Merge "Don't check if this == NULL." 2014-10-15 22:45:07 +00:00
Narayan Kamath
25c5d725f0 am 328f9c25: am 16410d45: am 9ea09757: Fix broken error check in Parcel::readBlob
* commit '328f9c259ba9d3586a2496d887a96b610d112b7b':
  Fix broken error check in Parcel::readBlob
2014-10-10 18:42:19 +00:00
Narayan Kamath
328f9c259b am 16410d45: am 9ea09757: Fix broken error check in Parcel::readBlob
* commit '16410d458fdf9915dc8cb82b1bae3949a8100124':
  Fix broken error check in Parcel::readBlob
2014-10-10 17:00:18 +00:00
Narayan Kamath
16410d458f am 9ea09757: Fix broken error check in Parcel::readBlob
* commit '9ea09757fa63ad2a9bea9c846b51c1210d69c79f':
  Fix broken error check in Parcel::readBlob
2014-10-09 19:17:44 +00:00
Michael Lentine
78be65e7bf Adding eglInitialize and eglTerminate image creation and deletion.
Previously it was possible to have the driver's eglTerminate called beofre
eglDestroyImageKHR in GLConsumer. This was because we didn't increment the
refcount for the lifetime of the image. This could lead to a crash or a deadlock
when multiple threads called terminate and destroy simultaneously.

Bug: 17700483
Change-Id: I7010d0f1b3db875332e95630b5e098a5564ba755
2014-10-09 10:26:37 -07:00
Jesse Hall
3dbf96cf06 am 20fca75f: am b7702c5c: Merge "BufferQueueProducer: Throttle EGL frame production." into lmp-mr1-dev
* commit '20fca75fbd9141ebae6c9a28f5c09a837eb9caad':
  BufferQueueProducer: Throttle EGL frame production.
2014-10-08 17:04:38 +00:00
Jesse Hall
20fca75fbd am b7702c5c: Merge "BufferQueueProducer: Throttle EGL frame production." into lmp-mr1-dev
* commit 'b7702c5ce7f010fa5f19646fd79f28ff04e7014b':
  BufferQueueProducer: Throttle EGL frame production.
2014-10-08 16:59:38 +00:00
Jesse Hall
b7702c5ce7 Merge "BufferQueueProducer: Throttle EGL frame production." into lmp-mr1-dev 2014-10-08 16:50:13 +00:00
Narayan Kamath
9ea09757fa Fix broken error check in Parcel::readBlob
mmap returns MAP_FAILED (which is -1) and not NULL on
failure.

Diagnosed by cferris.

bug: 17909809

Change-Id: I609788ebf94742ef88af002d2d3f3bc9b9e520ac
2014-10-08 17:37:38 +01:00
Jesse Hall
f570be2ef9 am f3b5cea6: am 366998d8: am 78639c22: Merge "Parcel: extra validation/debug code for writeDupFileDescriptor" into lmp-dev
* commit 'f3b5cea662da350684b9fd7a8f486c42e57bee6d':
  Parcel: extra validation/debug code for writeDupFileDescriptor
2014-10-07 00:40:24 +00:00
Jesse Hall
f3b5cea662 am 366998d8: am 78639c22: Merge "Parcel: extra validation/debug code for writeDupFileDescriptor" into lmp-dev
* commit '366998d87109578eb8c33d2e94d61fe7f189beb8':
  Parcel: extra validation/debug code for writeDupFileDescriptor
2014-10-07 00:34:57 +00:00
Jesse Hall
78639c22ac Merge "Parcel: extra validation/debug code for writeDupFileDescriptor" into lmp-dev 2014-10-06 22:36:54 +00:00
Jesse Hall
ccf851f108 Parcel: extra validation/debug code for writeDupFileDescriptor
Temporary extra debug validation for b/17477219: a Parcel recipient is
getting a positive but invalid fd unexpectedly. Trying to track down
where it's coming from.

Debug code for bug: 17477219
Change-Id: Idb1e71621025a3928c7adc88fd44790e1abd2a01
2014-10-06 15:23:56 -07:00
Aravind Akella
5b8d3eed87 am ae29ee0a: am 93176114: am a7eb4b74: Merge "Fix sockfd leakage in SensorService." into lmp-dev
* commit 'ae29ee0a9eea291316248a90799ff28f35defeb6':
  Fix sockfd leakage in SensorService.
2014-10-03 23:57:35 +00:00
Aravind Akella
ae29ee0a9e am 93176114: am a7eb4b74: Merge "Fix sockfd leakage in SensorService." into lmp-dev
* commit '93176114cf34c9874c2db70fd7665e9f95dda9d0':
  Fix sockfd leakage in SensorService.
2014-10-03 22:39:19 +00:00
Eric Penner
99a0afbaee BufferQueueProducer: Throttle EGL frame production.
Throttling was previously controlled by a combination of the
driver and the number of buffers in the queue. This patch makes
a more consistent trade-off, which allows two GPU frames pending
but not three. More buffering could improve throughput in the
case of varying frame times, but this also increases latency.

Bug: 17502897
Change-Id: I4ee68019ca94c635294c5959931a555a6c4ef2df
2014-10-03 14:57:09 -07:00
Aravind Akella
a7eb4b74f7 Merge "Fix sockfd leakage in SensorService." into lmp-dev 2014-10-03 21:45:37 +00:00
Aravind Akella
8a96955c8e Fix sockfd leakage in SensorService.
i) Call removeFd() only if the fd in the BitTube has been
previously added to the Looper. Use a flag to determine whether the fd
has been previously added or not.
ii) Increment mPendingFlushEventsToSend after holding a connectionLock.
iii) Store the number of acks that are pending in SensorEventQueue
 and send them all at once.

Bug: 17472228
Change-Id: I1ec834fea1112a9cfbd9cddd2198438793698502
2014-10-02 18:59:56 -07:00
Jesse Hall
6051dc2fcc am 8619132f: am 74efd6e6: am f05ef4e5: Merge "Surface: cancel the dequeued buffer when requestBuffer fails" into lmp-dev
* commit '8619132f167afab3775315acdf6e7097fff41a91':
  Surface: cancel the dequeued buffer when requestBuffer fails
2014-10-02 23:46:41 +00:00
Jesse Hall
8619132f16 am 74efd6e6: am f05ef4e5: Merge "Surface: cancel the dequeued buffer when requestBuffer fails" into lmp-dev
* commit '74efd6e6bec8a49f66090a5c0c18b65ba8a6fcad':
  Surface: cancel the dequeued buffer when requestBuffer fails
2014-10-02 23:43:31 +00:00
Jesse Hall
f05ef4e542 Merge "Surface: cancel the dequeued buffer when requestBuffer fails" into lmp-dev 2014-10-02 23:11:09 +00:00
Michael Lentine
08c356c68a am dea55870: am 36462d9d: Merge "Add more logging for dup(fd) failure" into lmp-dev
* commit 'dea558703a737d7e017a4f7991ea7b0367916654':
  Add more logging for dup(fd) failure
2014-10-02 21:34:52 +00:00
Michael Lentine
dea558703a am 36462d9d: Merge "Add more logging for dup(fd) failure" into lmp-dev
* commit '36462d9de00105ae46ee3dbb0f644fa6af599f97':
  Add more logging for dup(fd) failure
2014-10-02 19:44:43 +00:00
Jesse Hall
9f5a1b6abd Surface: cancel the dequeued buffer when requestBuffer fails
Partial fix for bug: 17477219
Change-Id: Ibf5a9e26e02c4be8854925a77a70f5c9c7dcf6f2
2014-10-02 18:16:19 +00:00
Michael Lentine
36273c991b Add more logging for dup(fd) failure
Bug: 17477219
Change-Id: Ide0ae16d777c9af783023c705c18a93c00999147
2014-10-02 18:13:02 +00:00
Dianne Hackborn
7a29f687db am e5a59e37: am 5ee2c9d2: Work on issue #17656716: Unhandled exception in Window Manager
* commit 'e5a59e37f2a160f71459b1888ed2d268f9faa10e':
  Work on issue #17656716: Unhandled exception in Window Manager
2014-10-02 02:00:11 +00:00
Dianne Hackborn
e5a59e37f2 am 5ee2c9d2: Work on issue #17656716: Unhandled exception in Window Manager
* commit '5ee2c9d20c0fbecf6b4a482eb8d8bfdf85d0d424':
  Work on issue #17656716: Unhandled exception in Window Manager
2014-10-02 00:57:48 +00:00
Dianne Hackborn
5ee2c9d20c Work on issue #17656716: Unhandled exception in Window Manager
Try to clean up the code paths coming in and out of binder IPCs to
plug any places where we could disrupt the gather flag of a thread,
causing it to keep gathering stack crawls (which is the thing that
is causing our strict mode data to become so large).

We now take care of saving and restoring this state in the core
IPC code path, not at the Java layer.

Change-Id: I73d564778da127bdce00f304225930e7f2318293
2014-09-30 11:30:03 -07:00
Dan Albert
5193aea69a am 56b3f887: am 6ee97e74: Merge "Fix a null dereference."
* commit '56b3f887a2a859eaac9f3c3eb2fa18eb5bf396da':
  Fix a null dereference.
2014-09-23 20:55:50 +00:00
Dan Albert
56b3f887a2 am 6ee97e74: Merge "Fix a null dereference."
* commit '6ee97e74d2c972dec2aa6a2f231b718eae54898f':
  Fix a null dereference.
2014-09-23 17:54:39 +00:00
Dan Albert
c78a0c1733 Fix a null dereference.
Change-Id: I7224c3368ff414620062bc08b80a6a94d55d7931
2014-09-23 09:51:51 -07:00
Lajos Molnar
147372fdf9 resolved conflicts for merge of 67d8bd66 to lmp-mr1-dev-plus-aosp
Change-Id: I5e7e6fdf0d2afe6c3e4a0ed6ef96afb4713a72bc
2014-09-19 06:23:46 -07:00
Lajos Molnar
e7bda06398 resolved conflicts for merge of 67d8bd66 to lmp-dev-plus-aosp
Change-Id: Idbe39c6de8fe31ac57a9238c18b4c20d66dc73ad
2014-09-18 20:05:27 -07:00
Lajos Molnar
67d8bd66aa surfaceflinger: add getDisplayStats() method
This is used by media service to schedule video frames at the
proper time, based on precise vsync timings.

Bug: 14659809
Change-Id: I1a90603f3dc09dca9aa4f90a3aa845fab56e0a5e
2014-09-18 17:03:20 -07:00
Dan Albert
1474f8864f Don't check if this == NULL.
Entering a method with this == NULL is undefined behavior. Clang whines
about this.

Change-Id: Ibde628395ca10dfef0d2f59e81280576f104b83c
2014-09-16 17:52:06 -07:00
Dan Albert
94ba882df2 am 4355c0c9: am e1e3d278: Merge "Clean cruft out of makefiles."
* commit '4355c0c98644c37f0511257403bbfb71b342e2df':
  Clean cruft out of makefiles.
2014-09-12 19:28:17 +00:00
Hans Boehm
d898618914 am e03abcd6: am 4a7001ca: Merge "Work around C11 const atomic restrictions."
* commit 'e03abcd61b127b10391b8257c35fbe66053e417a':
  Work around C11 const atomic restrictions.
2014-09-12 19:28:16 +00:00
Aravind Akella
f627c969e1 am 3d0daa57: am 8493b79e: SensorService fixes
* commit '3d0daa57f5f180418fdecbf06f6a5a41e566f6d9':
  SensorService fixes
2014-09-12 19:20:58 +00:00
Dan Albert
4355c0c986 am e1e3d278: Merge "Clean cruft out of makefiles."
* commit 'e1e3d278549eb6389eb0b5fa3f13a4ac079a55a2':
  Clean cruft out of makefiles.
2014-09-12 16:22:03 +00:00