Commit Graph

2035 Commits

Author SHA1 Message Date
Jamie Gennis
1df8c34585 libgui: disallow NULL Fence pointers
This change eliminates the uses of a NULL sp<Fence> indicating that no waiting
is required.  Instead we use a non-NULL but invalid Fence object for which the
wait methods will return immediately.

Bug: 7892871
Change-Id: I5360aebe3090422ef6920d56c99fc4eedc642e48
2013-02-12 18:15:38 -08:00
Nick Kralevich
c4fd05b172 Revert "try to fix win_sdk build."
A better change was checked into system/core. See commit
941daef629bd571032851edf7ae1dce24266640e

This reverts commit fa99d30ec7.
2013-02-02 18:09:15 -08:00
Nick Kralevich
fa99d30ec7 try to fix win_sdk build.
Change-Id: I55e78e339874a6d6a3381c2550556b65c7ec1ca0
2013-02-01 16:09:23 -08:00
Jesse Hall
7c36cd2315 Fix memory leak when getting signal time for unsignaled fences
Bug: 8000983
Change-Id: Ie88ed23fc5e8c95bfade364e090e90b7f070fa8e
2013-01-14 16:33:29 -08:00
Andy McFadden
466a192d20 Minor tweaks.
Change-Id: I9f82f574112141b6aa074c42303056cd52d5c478
2013-01-08 11:25:51 -08:00
Andy McFadden
2adaf04fab Rename ISurfaceTexture and SurfaceTexture
The C++ class names don't match what the classes do, so rename
ISurfaceTexture to IGraphicBufferProducer, and SurfaceTexture to
GLConsumer.

Bug 7736700

Change-Id: Ia03e468888025b5cae3c0ee1995434515dbea387
2012-12-18 13:10:48 -08:00
Marco Nelissen
cce6230b3c Merge "Add NULL check to CallStack::toString" 2012-12-17 13:42:05 -08:00
Marco Nelissen
32130635f6 Add NULL check to CallStack::toString
CallStack::toString() has a 0 default argument, which ends up getting
passed to strlen(), resulting in a crash.

Change-Id: If706aff8c400983670f49cdbb66e11191ac76e0e
2012-12-17 10:28:20 -08:00
Jamie Gennis
c911ea5318 stop using a deprecated SurfaceTextureClient ctor
Change-Id: I8c8282a0debd551db290dd6849faf272a88c704c
2012-12-12 12:38:36 -08:00
Jamie Gennis
3c1ada9822 Merge "SurfaceTextureClient: remove a deprecated ctor" 2012-12-12 12:16:21 -08:00
Jamie Gennis
4def860030 SurfaceTextureClient: remove a deprecated ctor
Change-Id: Ibf2746fdca26851677cd90199c60eab80f6d0c73
2012-12-12 12:13:54 -08:00
Jamie Gennis
c6f04666c2 Merge "SurfaceFlinger: Move GraphicBufferAlloc to libgui" 2012-12-12 10:56:42 -08:00
Jamie Gennis
392edd88cb SurfaceFlinger: Move GraphicBufferAlloc to libgui
This change moves the GraphicBufferAlloc class from SurfaceFlinger to libgui.

Change-Id: Idf31d2004efa2651b60590733f73c4a7b831e8a9
2012-12-11 17:54:29 -08:00
Andy McFadden
97eba8904c Avoid unnecessary texture bind
In SurfaceFlingerConsumer, check to see if native fence sync is
enabled.  If so, defer the texture binding step to Layer::onDraw.

Change-Id: I7d4034a31c0143207eea2509dfa13ef3820f9b8c
2012-12-11 16:26:51 -08:00
Andy McFadden
bf974abe92 Refactor SurfaceTexture a bit.
Rearranges updateTexImage() so that the SurfaceFlinger-specific
behavior is in a new SurfaceFlingerConsumer subclass.

SurfaceTexture behavior should not be altered.  Instead of
acquire-bind-release we now do acquire-release-bind, but since
it's all done with the lock held there shouldn't be any
externally-visible change.

Change-Id: Ia566e4727945e2cfb9359fc6d2a8f8af64d7b7b7
2012-12-11 09:42:51 -08:00
Jamie Gennis
a436799668 am 50f8d285: am d4e70620: am 72c3f7d8: Revert "ConsumerBase: free buffers outside the lock"
* commit '50f8d2856b6644059b31684687842bf9c1f2e814':
  Revert "ConsumerBase: free buffers outside the lock"
2012-12-07 11:09:12 -08:00
Jamie Gennis
50f8d2856b am d4e70620: am 72c3f7d8: Revert "ConsumerBase: free buffers outside the lock"
* commit 'd4e706208c99ff1716fb0009ec867ff29ff04a7c':
  Revert "ConsumerBase: free buffers outside the lock"
2012-12-07 11:07:03 -08:00
Jamie Gennis
7df16644c4 am 2d659286: am 2e59d2c3: DO NOT MERGE GraphicBufferAllocator: make frees async
* commit '2d659286b59dc7ff7378fe967f1c425e92155995':
  DO NOT MERGE GraphicBufferAllocator: make frees async
2012-12-07 10:59:29 -08:00
Jamie Gennis
72c3f7d881 Revert "ConsumerBase: free buffers outside the lock"
This reverts commit b21a4e3b5f.
2012-12-07 10:33:07 -08:00
Jamie Gennis
2e59d2c3fd DO NOT MERGE GraphicBufferAllocator: make frees async
This change makes GraphicBufferAllocator::free queue a job to another thread to
perform the actual free operation.  This prevents potentially slow free
operations from blocking rendering.

Bug: 7675940
Change-Id: Id61099d66bb4c3949d04184e0d7f192ac18076b4
2012-12-07 10:32:13 -08:00
Jamie Gennis
b5fe96b813 am 05989772: am 3ed2736c: am b21a4e3b: ConsumerBase: free buffers outside the lock
* commit '05989772d5b46cd5328e88d546f04deef39cc3c8':
  ConsumerBase: free buffers outside the lock
2012-12-06 23:24:25 -08:00
Jamie Gennis
05989772d5 am 3ed2736c: am b21a4e3b: ConsumerBase: free buffers outside the lock
* commit '3ed2736c10efb2f18062591e308036837d9725a4':
  ConsumerBase: free buffers outside the lock
2012-12-06 19:00:57 -08:00
Jamie Gennis
b21a4e3b5f ConsumerBase: free buffers outside the lock
This change makes ConsumerBase::onBuffersReleased hold a reference to all its
gralloc buffers until after the mutex is unlocked.  This prevents slow
gralloc::free calls from causing lock contention with rendering threads.

Bug: 7675940
Change-Id: I0ec805d1b612afeeecfffec03f982371d27d93be
2012-12-06 18:17:35 -08:00
Chris Craik
0a73f822b1 Add LinearAllocator
Moving from external/webkit/Source/WebCore/platform/graphics/android/utils/

Change-Id: If91830aa9b207dbc8692b2ca7c4a0b76778addd5
2012-12-05 17:34:54 -08:00
Alex Ray
653078f652 utils: clarify scoped tracing functionality
ScopedTrace objects were being used in place of ATRACE_NAME because of a
misunderstanding of it's function.  Cleared up documentation for usage.

Also explicitly use global namespace for sysprop callback.

Change-Id: I7c248b486b614ccdb841659ca0dcfc644fda670a
2012-11-30 19:54:00 -08:00
Alex Ray
b7e06477f0 utils: Use cutils tracing functionality.
Tracing functionality has moved to cutils.

Change-Id: Ie78ccc1d59dd5178f5058fbc3858a37f9adce552
2012-11-29 13:33:14 -08:00
Jamie Gennis
c849b6f44c Merge "SurfaceFlinger: refactor frame time tracking" 2012-11-20 12:20:52 -08:00
Mathias Agopian
b171fe9f47 am 594a8773: am 764c197c: fix typo that broke all the builds
* commit '594a8773c1ef5632a8eb2ef61a86c319460d4248':
  fix typo that broke all the builds
2012-11-20 07:13:12 -08:00
Mathias Agopian
73d5662e4c fix typo that broke all the builds
cherry pick into master since auto-merger is blocked

Bug: 7584338
Change-Id: Ie7d7c238de1fd224b3b0bae9669a8dcb2f700a79
2012-11-19 22:13:49 -08:00
Jamie Gennis
82dbc7429f SurfaceFlinger: refactor frame time tracking
This change moves the frame time history tracking code out of Layer and into a
new class called FrameTracker.  It also changes the tracking to use signal
timestamps from fences when available for more accurate results.

Change-Id: I323c5f075c58bf86ce363b52af885ad0f6365f2b
2012-11-19 17:45:09 -08:00
Mathias Agopian
764c197c6f fix typo that broke all the builds
Bug: 7584338
Change-Id: Ieb8c27a544ac583af9aa1e0376e33a673d2d9673
2012-11-19 16:50:24 -08:00
Mathias Agopian
bd3e837ec7 am c208cdd4: am 1d3612f1: am 2a8c49eb: fix an out-of-bounds memory access
* commit 'c208cdd402d767228a34074d589c6368cc656b7d':
  fix an out-of-bounds memory access
2012-11-19 15:17:27 -08:00
Mathias Agopian
c208cdd402 am 1d3612f1: am 2a8c49eb: fix an out-of-bounds memory access
* commit '1d3612f11cb0a56062059dad4643657750c23aab':
  fix an out-of-bounds memory access
2012-11-19 15:14:08 -08:00
Mathias Agopian
2a8c49eb5d fix an out-of-bounds memory access
in this particular case, this OOB is always harmless
(and that's why it didn't get fixed from MR1), however,
it interfers with valgrind debugging.

Change-Id: Ic977e03287e59c4b124a89146c9023bd0cb540a8
2012-11-15 17:19:48 -08:00
Andy McFadden
2bdd8fd65a am 41f7736a: am 434e21bf: am 3fdeb48e: Reduce emulator logspam
* commit '41f7736ad1e72df8796416a82ce3cede6193d1eb':
  Reduce emulator logspam
2012-11-09 22:49:18 -08:00
Andy McFadden
41f7736ad1 am 434e21bf: am 3fdeb48e: Reduce emulator logspam
* commit '434e21bfd6b14f8d99b9bcdb829e7252edd574f3':
  Reduce emulator logspam
2012-11-09 22:45:13 -08:00
Andy McFadden
3fdeb48e3c Reduce emulator logspam
The emulator doesn't support systrace, but we should point that out
at most once per process.

Bug 7436352

Change-Id: I06b2c1ea0df6c02c11cd2496423c337f8d7c62a1
2012-11-08 16:40:11 -08:00
Igor Murashkin
74284b9467 Merge "Change ConsumerBase's FrameAvailableListener to be a weak pointer" 2012-11-05 13:47:37 -08:00
Igor Murashkin
a4a3149a36 Change ConsumerBase's FrameAvailableListener to be a weak pointer
This prevents strong reference cycles when the listener implementation also
holds a strong pointer to the ConsumerBase

Bug: 7425644
Change-Id: I1514b13a32b18d421c902dddebec0765a989c55c
2012-11-05 13:39:02 -08:00
Jeff Brown
f3f7db6346 Allow global transactions to nest.
This change fixes a number of small glitches that can occur when
multiple components in the same process are updating surfaces.
One would expect that updates to disjoint sets of surfaces would
not collide but this is not the case.  The first component to
close the global transaction causes all pending updates to
be applied, including those that another component might not
have finished setting up if it also had an open transaction
at the same time.

Change-Id: I99345958581abbe0e1e325a5bcba37e8941a313a
2012-11-02 14:02:52 -07:00
Raph Levien
3bc7b27571 Merge "Add an LRU cache plus hashing primitives" 2012-10-29 08:59:54 -07:00
Raph Levien
8185e47822 Add an LRU cache plus hashing primitives
This patch adds a hashtable-based LRU cache. This should be
significantly higher performance than the GenerationCache it is intended
to replace. It is a large part of the fix for bug 7271109
TextLayoutCache low-level performance issues.

We added a new method to BasicHashtable to detect when rehashing is
needed, because the internal linked list pointers would get invalidated
by that rehashing.

Also, the hash_type specialized to pointers had a small flaw.

Change-Id: I950c2083f96519777b851dbe157100e0a334caec
2012-10-26 16:09:22 -07:00
Dave Burke
0491524f83 am 296cd1cb: am 04075569: Revert "Revert "put back the unused virtuals in Vector<>""
* commit '296cd1cb9ac7b7ea201662860347755d0349a7de':
  Revert "Revert "put back the unused virtuals in Vector<>""
2012-10-25 11:48:40 -07:00
Dave Burke
296cd1cb9a am 04075569: Revert "Revert "put back the unused virtuals in Vector<>""
* commit '04075569b565c949a6db8b4e8a9f10d281de4d9c':
  Revert "Revert "put back the unused virtuals in Vector<>""
2012-10-25 11:45:41 -07:00
Dave Burke
04075569b5 Revert "Revert "put back the unused virtuals in Vector<>""
This reverts commit 225c66a48c

Change-Id: If31a04b81052cbc7dd7bf237c07107c33066d03d
2012-10-25 11:40:51 -07:00
Mike Lockwood
126d215667 More changes to fix PDK build:
move screencap back to frameworks/base
move libdiskusage over from frameworks/base

Change-Id: I8415716634a4cb5b0fae15ef15ff21ee0a08906a
2012-10-24 12:30:41 -07:00
Jamie Gennis
3b6055a718 am ba7dc2db: Merge "SurfaceFlinger: add support for secure displays" into jb-mr1-dev
* commit 'ba7dc2db6e93a2407c8c328f2838591b7b760658':
  SurfaceFlinger: add support for secure displays
2012-10-22 13:54:50 -07:00
Jamie Gennis
dd3cb84cfb SurfaceFlinger: add support for secure displays
This change adds support for displays that are not allowed to display surfaces
with the eSecure flag set.  All non-virtual displays are considered secure,
while virtual displays have their secure-ness specified at creation time.

Bug: 7368436
Change-Id: I81ad535d2d1e5a7ff78269017e85b111f0098500
2012-10-22 13:41:21 -07:00
Jamie Gennis
9097e14443 am 7f79a2bd: Merge "BufferQueue: alloc without holding the lock" into jb-mr1-dev
* commit '7f79a2bd5c92bc626890dc16ffb8cd2de934e5fb':
  BufferQueue: alloc without holding the lock
2012-10-17 09:37:40 -07:00
Jamie Gennis
7f79a2bd5c Merge "BufferQueue: alloc without holding the lock" into jb-mr1-dev 2012-10-17 09:32:43 -07:00