OSDN Git Service
Romain Guy [Mon, 11 Mar 2013 21:38:26 +0000 (21:38 +0000)]
Merge "Add Thread::isRunning and Condition::signal(WakeUpType)" into jb-mr2-dev
Romain Guy [Mon, 11 Mar 2013 21:34:56 +0000 (14:34 -0700)]
Add Thread::isRunning and Condition::signal(WakeUpType)
The signal() method is useful to choose whether to wake up one or
all threads.
Change-Id: I062ab6d3ddd306a9fb735549ea140e2a76eed75a
Jean-Baptiste Queru [Mon, 11 Mar 2013 20:44:33 +0000 (13:44 -0700)]
am
1d6e52d1: am
50d36d58: am
75bcd737: am
4ddcb758: (-s ours) am
9ef024ec: (-s ours) Reconcile with jb-mr1-release - do not merge
* commit '
1d6e52d17e2df913e10925c5e3e652d3421ff320':
Jean-Baptiste Queru [Mon, 11 Mar 2013 20:13:37 +0000 (13:13 -0700)]
am
50d36d58: am
75bcd737: am
4ddcb758: (-s ours) am
9ef024ec: (-s ours) Reconcile with jb-mr1-release - do not merge
* commit '
50d36d58369cdf48bc6fcdba96b36c846be8587a':
Jean-Baptiste Queru [Mon, 11 Mar 2013 20:00:44 +0000 (13:00 -0700)]
am
75bcd737: am
4ddcb758: (-s ours) am
9ef024ec: (-s ours) Reconcile with jb-mr1-release - do not merge
* commit '
75bcd737770f85114673c4411b8e2ecc0dd32bdf':
Jean-Baptiste Queru [Mon, 11 Mar 2013 19:41:50 +0000 (12:41 -0700)]
am
4ddcb758: (-s ours) am
9ef024ec: (-s ours) Reconcile with jb-mr1-release - do not merge
* commit '
4ddcb75861cfb2835fcc20f2071222d2f55450ba':
jp abgrall [Mon, 11 Mar 2013 19:28:06 +0000 (12:28 -0700)]
am
ba8abb51: Merge "Ensure dumpstate children die"
* commit '
ba8abb517d1eddebeda62bf907fa446e3c3b81f6':
Ensure dumpstate children die
Jean-Baptiste Queru [Mon, 11 Mar 2013 19:23:09 +0000 (12:23 -0700)]
am
9ef024ec: (-s ours) Reconcile with jb-mr1-release - do not merge
* commit '
9ef024ec8ae0318c89cc65dc0591aa40695393fc':
Jean-Baptiste Queru [Mon, 11 Mar 2013 19:19:30 +0000 (12:19 -0700)]
Reconcile with jb-mr1-release - do not merge
jp abgrall [Mon, 11 Mar 2013 18:09:35 +0000 (18:09 +0000)]
Merge "Ensure dumpstate children die"
John Michelau [Thu, 7 Mar 2013 21:35:35 +0000 (15:35 -0600)]
Ensure dumpstate children die
Use prctl(PR_SET_PDEATHSIG, SIGKILL) in each forked child to ensure
it dies when dumpstate dies.
This is important for two cases:
- dumpstate runs a timer for each child process. On expiration, it
sends SIGTERM to kill the process. Sometimes SIGTERM isn't enough
to kill a hung process, so the child lives on.
- When dumpstate is killed by the user before completing, outstanding
children continue to run and generate output.
Change-Id: I96e0dc918c26d56c9fee53611980ba2edd238712
Mathias Agopian [Fri, 8 Mar 2013 21:54:46 +0000 (13:54 -0800)]
make sure to call compositionComplete after taking a screenshot
older drivers which are doing implicit synchronization need this
or they could deadlock.
Bug:
8341885
Change-Id: Icd980a6be16071678d6151e34725b3c1c547d7ee
Mathias Agopian [Fri, 8 Mar 2013 21:18:52 +0000 (13:18 -0800)]
add two libgui tests
we check that the order in which we destroy GLConsumer wrt.
releasing the corresponding EGLSurface via eglMake(Un)Current
doesn't leak a buffer.
On at least 2 devices this test doesn't pass.
Change-Id: I63ab83951b4b0a977f38571158f948cbd9dc7cec
Mathias Agopian [Fri, 8 Mar 2013 20:28:42 +0000 (20:28 +0000)]
Merge changes Ibb3b6ff0,I2341e20c into jb-mr2-dev
* changes:
rename binder services main thread to Binder_*
limit number of extra binder threads in SF to 4
Jesse Hall [Thu, 7 Mar 2013 23:43:56 +0000 (23:43 +0000)]
Merge "When disconnecting a surface, cancel don't queue the buffer" into jb-mr2-dev
Jesse Hall [Thu, 7 Mar 2013 23:41:45 +0000 (23:41 +0000)]
Merge "Defer destroying surfaces until not current" into jb-mr2-dev
Mathias Agopian [Thu, 7 Mar 2013 23:34:28 +0000 (15:34 -0800)]
rename binder services main thread to Binder_*
When a binder service's main thread joins the thread pool
it retains its name (whatever the exec name was), which is
very confusing in systrace.
we now rename that thread just like its friends in the
thread pool.
Change-Id: Ibb3b6ff07304b247cfc6fb1694e72350c579513e
Mathias Agopian [Thu, 7 Mar 2013 23:32:47 +0000 (15:32 -0800)]
limit number of extra binder threads in SF to 4
it should be enough and it makes debugging easier.
Bug:
6609290
Change-Id: I2341e20cfcfde184fe1ed9b925215f8fa3481ced
Jesse Hall [Thu, 7 Mar 2013 23:14:18 +0000 (15:14 -0800)]
When disconnecting a surface, cancel don't queue the buffer
This isn't really right either, but avoids having an extra buffer that
the consumer has to drain which it might not be expecting.
To be correct, disconnecting a surface from a context should retain
the current buffer and continue using it when reconnected. The buffer
should only be canceled when the surface is destroyed. That will wait
for a later change.
Bug:
8320762
Change-Id: I5efa39c741193ca4f5612ea9de001ccbb683b345
Jesse Hall [Thu, 7 Mar 2013 17:56:26 +0000 (09:56 -0800)]
Defer destroying surfaces until not current
Bug:
8320762
Change-Id: I1320cf87923bcc5b795a86a13193363a49e29653
Jean-Baptiste Queru [Thu, 7 Mar 2013 19:35:19 +0000 (11:35 -0800)]
resolved conflicts for merge of
bfb2aa46 to jb-mr2-dev
Jean-Baptiste Queru [Thu, 7 Mar 2013 19:08:08 +0000 (19:08 +0000)]
Merge "Tweaks for forward compatibility"
Mathias Agopian [Thu, 7 Mar 2013 06:37:33 +0000 (22:37 -0800)]
another attempt to fix the build
Change-Id: Ie0245db97170cca9a2cd8783b0fd512daf0d30a6
Mathias Agopian [Thu, 7 Mar 2013 05:34:39 +0000 (21:34 -0800)]
attempt to fix build
Change-Id: Ifb186d2f157ad02f593bf9b681aab53575f183ce
Mathias Agopian [Thu, 7 Mar 2013 04:52:24 +0000 (04:52 +0000)]
Merge changes I7f696e42,I51c6df3c into jb-mr2-dev
* changes:
Make GraphicBuffer a RefBase
Fix RefBase debugging. O_CREAT must specify the mode.
Mathias Agopian [Thu, 7 Mar 2013 04:42:56 +0000 (20:42 -0800)]
workaround to fix screenshot leak on N4
Bug:
8322020
Change-Id: Ie60af0eb431866b8d64b2674ae7bd8b5ee05f5d6
Mathias Agopian [Thu, 7 Mar 2013 02:50:52 +0000 (18:50 -0800)]
Make GraphicBuffer a RefBase
- this gives us access to RefBase's refcounting debugging
- it doesn't cost much because GraphicBuffer already has a vtable
Change-Id: I7f696e421fea14b14bfaeb83880689b83e96af4d
Mathias Agopian [Thu, 7 Mar 2013 01:51:15 +0000 (17:51 -0800)]
Fix RefBase debugging. O_CREAT must specify the mode.
Change-Id: I51c6df3cfd59b20ca73c3edee86bc2f74dbde1b1
Mathias Agopian [Thu, 7 Mar 2013 01:47:34 +0000 (01:47 +0000)]
Merge changes I66511c08,Ia051949f,Ic7451365,I5b571a4c into jb-mr2-dev
* changes:
Get rid of LayerBase.
Make LayerDim a regular Layer instead of a LayerBase
fold LayerBaseClient into LayerBase
Remove support for ScreenshotLayer
Jean-Baptiste Queru [Wed, 6 Mar 2013 23:04:42 +0000 (15:04 -0800)]
Tweaks for forward compatibility
Change-Id: I136fcbbc14072ee5f5281d09445f28d083ed3ce1
Mathias Agopian [Wed, 6 Mar 2013 01:47:11 +0000 (17:47 -0800)]
Get rid of LayerBase.
The functionality of LayerBase and Layer is folded
into Layer. There wasn't a need for this abstraction
anymore.
Change-Id: I66511c08cc3d89009ba4deabf47e26cd4cfeaefb
Mathias Agopian [Tue, 5 Mar 2013 23:50:58 +0000 (15:50 -0800)]
Make LayerDim a regular Layer instead of a LayerBase
this is in preparation to get rid of LayerBase entirely
Change-Id: Ia051949fc5205fd87371331145356ee11598a597
Mathias Agopian [Tue, 5 Mar 2013 23:14:58 +0000 (15:14 -0800)]
fold LayerBaseClient into LayerBase
Change-Id: Ic745136522df59c42f0885fd969e75ea55d09f01
Mathias Agopian [Tue, 5 Mar 2013 22:51:39 +0000 (14:51 -0800)]
Remove support for ScreenshotLayer
Change-Id: I5b571a4cf3faa77d2c4aca916fa4bd00a1065bb9
Jeff Sharkey [Wed, 6 Mar 2013 01:57:26 +0000 (17:57 -0800)]
Return NO_MEMORY when glReadPixels() fails.
Change-Id: Ic66134ef457e8442ae9135e9ec50e3d02932a253
Igor Murashkin [Tue, 5 Mar 2013 21:58:00 +0000 (21:58 +0000)]
Merge "CpuConsumer_test: Extend test with new formats RGBA8888 and optional Y8/Y16" into jb-mr2-dev
Igor Murashkin [Wed, 6 Feb 2013 01:54:32 +0000 (17:54 -0800)]
CpuConsumer_test: Extend test with new formats RGBA8888 and optional Y8/Y16
Change-Id: I0a0f6ce73516883bf50c749a1759abbdb3748e08
Andy McFadden [Sat, 2 Mar 2013 00:25:32 +0000 (16:25 -0800)]
Add eglPresentationTimeANDROID
Added EGL extension to set a timestamp on a surface.
Also, fix JNI encoding of "long" in glgen.
Bug
8191230
Change-Id: I38b7334bade3f8ff02bffe600bb74469ef22c164
Mathias Agopian [Tue, 5 Mar 2013 02:30:13 +0000 (02:30 +0000)]
Merge "rework screenshot API and implementation" into jb-mr2-dev
Jesse Hall [Tue, 5 Mar 2013 01:04:53 +0000 (01:04 +0000)]
Merge "Init displays to null layer stack" into jb-mr2-dev
Mathias Agopian [Mon, 4 Mar 2013 22:12:23 +0000 (22:12 +0000)]
Merge changes I0e90b3f3,Ib7769bde,I4c25f34f,I1ec6400a into jb-mr2-dev
* changes:
implement display projection clipping in h/w composer
refactor the crop region for hwc is calculated/set
apply the projection's viewport to the visibleregion passed to hwc
set correct crop rectangle in LayerBase::setCrop
Jesse Hall [Sun, 3 Mar 2013 00:25:32 +0000 (16:25 -0800)]
Fix reading NULL Surfaces from Parcels
Writing a NULL Surface was being read as a non-NULL Surface with NULL
mGraphicBufferProducer. Before the SurfaceTextureClient -> Surface
refactoring, you'd get a NULL Surface, and some code relies on that.
Bug:
8291161
Change-Id: I477bfe8882693e53a5f604a3d2c9e3cfe24473b4
Mathias Agopian [Fri, 1 Mar 2013 21:42:57 +0000 (13:42 -0800)]
rework screenshot API and implementation
- SurfaceFlinger now supports to take a screenshot
directly into an IGraphicBufferProducer
- reimplement the IMemoryHeap screenshot on top
of the above
- reimplement LayerScreenshot such that its
BufferQueue is directly used as the destination
of the screenshot. LayerScreenshot is now a thin
wrapper around Layer
Bug:
6940974
Change-Id: I69a2096b44b91acbb99eba16f83a9c78d94e0d10
Dave Burke [Fri, 1 Mar 2013 20:39:03 +0000 (20:39 +0000)]
Revert "Change SurfaceControl setPosition to take floats"
Temporary, to fix weekend build, until we get Nvidia code drop.
This reverts commit
9a867a8798fa6ea21f6341db31e38ea64fde6c83
DO NOT MERGE
Change-Id: I7b5dbc4db46ef3d97dc8598057d5487d6971178b
Nick Kralevich [Fri, 1 Mar 2013 18:05:56 +0000 (18:05 +0000)]
am
22641851: Merge "dumpstate: use correct header file."
* commit '
22641851c48fee6e79d01adcfab01c0b625cfb3c':
dumpstate: use correct header file.
Nick Kralevich [Fri, 1 Mar 2013 17:42:32 +0000 (17:42 +0000)]
Merge "installd: use correct header file." into jb-mr2-dev
Nick Kralevich [Fri, 1 Mar 2013 16:39:57 +0000 (16:39 +0000)]
Merge "dumpstate: use correct header file."
Jesse Hall [Wed, 20 Feb 2013 00:13:35 +0000 (16:13 -0800)]
Init displays to null layer stack
When a display is added, initialize it to use an empty layer stack, so
if it is somehow visible it will show black. It will be assigned the
real layer stack -- along with a projection and other properties -- by
window manager soon. Normally a display remains blanked until window
manager has decided what to show on it, but for HDMI connected at boot
that isn't currently the case.
Bug:
7258935
Change-Id: Ic9bb25f7a9b8d9d3772b097ab1d6fa03bc8780a1
Mathias Agopian [Fri, 1 Mar 2013 01:12:07 +0000 (17:12 -0800)]
implement display projection clipping in h/w composer
- cropping to the projection's "viewport" is "simply"
accomplished by intersecting it with the window crop
expressed in layerstack space.
Bug:
7149437
Change-Id: I0e90b3f37945292314b5d78a8f134935967e8053
Mathias Agopian [Thu, 28 Feb 2013 06:03:19 +0000 (22:03 -0800)]
refactor the crop region for hwc is calculated/set
- the crop region is now always calculated and set
in LayerBase::setGeometry which uses new virtuals to
access the "content" crop and transform (which are
provided by the Layer subclass)
Change-Id: Ib7769bdec0917dd248f926600c14ddf9ea84897a
Mathias Agopian [Wed, 27 Feb 2013 00:54:05 +0000 (16:54 -0800)]
apply the projection's viewport to the visibleregion passed to hwc
each desplay's projection's viewport essentially clips each layer,
so this should be reflected in the visibleregion passed to
h/w composer.
DisplayDevice getViewport and getFrame are now guaranteed to return
valid Rects.
Change-Id: I4c25f34fb26af10179eb26d429ca6c384c671e91
Mathias Agopian [Wed, 27 Feb 2013 00:37:53 +0000 (16:37 -0800)]
set correct crop rectangle in LayerBase::setCrop
The crop always had left=top=0, because the crop position
and the layer's transform were merged together in
computeBounds() (which really used to compute the
bounds in screen space, which we usually call the
"frame" elsewhere in the code)
Note: in practice this crop value is not used by
hwc, because it's overridden in Layer::setGeometry(), which
is why this bug was never apparent.
Change-Id: I1ec6400a8fc8314408e4252708f43ea98c2fe64e
Nick Kralevich [Fri, 1 Mar 2013 00:59:13 +0000 (16:59 -0800)]
installd: use correct header file.
This is a port of https://android-review.googlesource.com/52540 ,
which doesn't apply cleanly to master.
Change-Id: I2824bb7afd4e343b5dc8bed9f24002ec1dcbf07f
Nick Kralevich [Fri, 1 Mar 2013 00:46:22 +0000 (16:46 -0800)]
dumpstate: use correct header file.
Change-Id: I006d7a219ce6309259f271beff6011cc266f1a61
Eino-Ville Talvala [Thu, 28 Feb 2013 19:01:32 +0000 (11:01 -0800)]
CpuConsumer: Add optional asynchronous mode
Bug:
8290146
Bug:
8291751
Change-Id: I9c8ac4bff38b0411e987a204e540d018dba6d0b4
Eino-Ville Talvala [Thu, 28 Feb 2013 22:08:34 +0000 (14:08 -0800)]
CpuConsumer: Don't unlock buffers on producer disconnect
Bug:
8291751
Change-Id: I062a3d34b41183d07fb6b9109cdb6bf0c0c75672
Jesse Hall [Thu, 28 Feb 2013 00:39:25 +0000 (16:39 -0800)]
Initialize DisplayData fences to NO_FENCE, not NULL
Also fix another place that was checking for NULL fence rather than
Fence::isValid().
Bug
8283950
Change-Id: Ie06db327eb416828d8dac139171d96d4470b2e35
Chris Craik [Tue, 26 Feb 2013 03:12:47 +0000 (19:12 -0800)]
Add createTJunctionFreeRegion
T-junction free regions are useful for rendering regions with various
geometric transformations, and the Region's span-ordered, sorted rect
list supports T-junction free storage without modification.
This approach creates a T-junction free region by splitting each
rectangle that is part of a vertical T-junction. This approach is two
pass (up and down) so that divisions can trickle up/down to other
adjacent spans.
Change-Id: Ifcf5e6fe0034c96b00ef09a4433b2b0fce8f4300
Igor Murashkin [Wed, 27 Feb 2013 00:43:20 +0000 (00:43 +0000)]
Merge "GraphicBuffer.h: Remove redundant friend class declaration causing warnings" into jb-mr2-dev
Igor Murashkin [Tue, 26 Feb 2013 22:23:11 +0000 (14:23 -0800)]
GraphicBuffer.h: Remove redundant friend class declaration causing warnings
Change-Id: I8a4b0e0b4d0a1cf472de820553c31f2d42044543
Mathias Agopian [Tue, 26 Feb 2013 00:53:28 +0000 (00:53 +0000)]
Merge "Added a test that checks the error behavior of ConsumerBase::abandon" into jb-mr2-dev
Mathias Agopian [Tue, 26 Feb 2013 00:53:13 +0000 (00:53 +0000)]
Merge "add some missing parameter validation in GLES java bindings" into jb-mr2-dev
Mathias Agopian [Sat, 23 Feb 2013 03:34:49 +0000 (19:34 -0800)]
Added a test that checks the error behavior of ConsumerBase::abandon
We check that calling eglSwapBuffers() on an abandonned BufferQueue
return EGL_BAD_SURFACE -- this is to ensure consistancy between
drivers.
Change-Id: Ibb548e0cf767ceee69f2fc4a85811d15a6522277
Mathias Agopian [Sat, 23 Feb 2013 03:34:06 +0000 (19:34 -0800)]
add some missing parameter validation in GLES java bindings
- added support for comments in checks.spec
- added most missing checks
- added and commented with // special-cased functions
- added and commented with # functions that are still missing validation checks
- moved glGet* to a special case and updated all the "pnames" from the khronos spec
- changed ifcheck to default to 1 value. this allows us to simplify the checks.spec file
and handle unknown pnames automatically (they'll be validated against 1 value, if
it happens to need more, the call will go through but the validation will not
happen).
- refactored the cpp headers in to a common header + GLES version specific
header
Bug:
7402895
Change-Id: Ib5c68ca0ca416407b4cfa36e3a21901b2d6263ab
Jamie Gennis [Mon, 25 Feb 2013 21:37:54 +0000 (13:37 -0800)]
SurfaceFlinger: fix a couple NULL fence checks
This change replaces checks for a NULL fence pointer with calls to
Fence::isValid. There should no longer be NULL fences.
Change-Id: If17c9c132fcb1801531bf7588f8ba53476c57dad
Jesse Hall [Sun, 24 Feb 2013 00:18:55 +0000 (00:18 +0000)]
Merge changes Ic344ef63,If9bb02be,Ic8f94634
* changes:
Symlink /system/lib/libGLESv3.so -> libGLESv2.so
Add ES3 support to libGLESv2 and tracing tools
Import OpenGL ES 3.0 headers from Khronos SVN
Jesse Hall [Sat, 23 Feb 2013 04:56:31 +0000 (04:56 +0000)]
Merge changes I45836219,I70a4075e
* changes:
Regenerate API/trace files for constness change
Import constness fix from gl3.h into gl2.h.
Mathias Agopian [Sat, 23 Feb 2013 04:14:19 +0000 (04:14 +0000)]
Merge "sort the check file in alphabetical order"
Mathias Agopian [Sat, 23 Feb 2013 04:13:57 +0000 (04:13 +0000)]
Merge "update glgen to take into account recent changes"
Mathias Agopian [Sat, 23 Feb 2013 01:48:39 +0000 (17:48 -0800)]
sort the check file in alphabetical order
this simplifies a lot comparing it to the spec files.
Change-Id: I4b9b5e7469338a626100472d360ea26a902017bc
Mathias Agopian [Sat, 23 Feb 2013 04:04:22 +0000 (20:04 -0800)]
update glgen to take into account recent changes
- it looks like UserId has been renamed to UserHandle in the
framework.
- also for some reason I don't understand glgen didn't seem
to be up to date with respect to the generated bindings in the tree.
It's like the bindings were generated with a more recent version of
glgen (maybe it was never checked in). So we fix that here.
Change-Id: Ie49522ebf67fcab9213246b4d93500e37a3cbc05
Dmitry Shmidt [Fri, 22 Feb 2013 19:27:58 +0000 (11:27 -0800)]
Remove obsolte 'su' from 'ping' call (to work on user build)
Change-Id: I2d391bbd39fc591eef33b9c9e4b4cdbda29e41e7
Signed-off-by: Dmitry Shmidt <dimitrysh@google.com>
Michael Wright [Fri, 22 Feb 2013 06:41:15 +0000 (22:41 -0800)]
Change SurfaceControl setPosition to take floats
Bug:
8153162
Change-Id: If3c4b7a3e124ace25999eb9a569dc9a806a1fbc2
Mathias Agopian [Thu, 21 Feb 2013 23:23:21 +0000 (23:23 +0000)]
Merge "get rid of Surface::getISurfaceTexture()"
Ed Heyl [Wed, 20 Feb 2013 21:47:29 +0000 (13:47 -0800)]
Merge remote-tracking branch 'goog/jb-mr2-release'
Mathias Agopian [Wed, 20 Feb 2013 02:24:40 +0000 (18:24 -0800)]
get rid of Surface::getISurfaceTexture()
this was there just for legacy reasons.
Change-Id: I0a48d6e04d7efa65033f9f10c91bba557f16ae8b
Eino-Ville Talvala [Tue, 19 Feb 2013 20:54:57 +0000 (12:54 -0800)]
Fixes to native app ops services
Bug:
8181262
Change-Id: I3d3f20453f6c6f2756c988363c5b5453ef309cb6
The Android Automerger [Tue, 19 Feb 2013 17:51:24 +0000 (09:51 -0800)]
merge in jb-mr2-release history after reset to master
Mathias Agopian [Sat, 16 Feb 2013 00:34:04 +0000 (16:34 -0800)]
fix build: SurfaceTextureClient to Surface rename
Change-Id: I0041e6f49bc637bc52161c83f0bd44101246f480
Mathias Agopian [Fri, 15 Feb 2013 22:59:09 +0000 (14:59 -0800)]
fix build: SurfaceTextureClient to Surface rename
Change-Id: I54d68241f1080eef640aefe71f33912ad79345ef
Mathias Agopian [Fri, 15 Feb 2013 22:48:52 +0000 (14:48 -0800)]
fix build: SurfaceTextureClient to Surface rename
Change-Id: Ibcf98337940f9195bcff7b5e8bddb6e912e65163
Mathias Agopian [Fri, 15 Feb 2013 01:11:02 +0000 (17:11 -0800)]
Refactoring: Rename SurfaceTextureClient to Surface
Change-Id: Ibed34175ae273608393aaa5f0a7df207dc40d709
Mathias Agopian [Thu, 14 Feb 2013 20:28:24 +0000 (20:28 +0000)]
Merge "remove access ANDROID_VIEW_SURFACE_JNI_ID"
Mathias Agopian [Thu, 14 Feb 2013 20:24:37 +0000 (20:24 +0000)]
Merge "get rid of Surface identity and token"
Elliott Hughes [Thu, 14 Feb 2013 19:42:06 +0000 (11:42 -0800)]
am
45b0e435: Merge "Simplify OpenGL TLS access on ARM."
# Via Elliott Hughes (1) and Gerrit Code Review (1)
* commit '
45b0e4350d3404bd194edc569e36ecca017c3d76':
Simplify OpenGL TLS access on ARM.
Elliott Hughes [Thu, 14 Feb 2013 18:31:20 +0000 (18:31 +0000)]
Merge "Simplify OpenGL TLS access on ARM."
The Android Automerger [Thu, 14 Feb 2013 13:55:53 +0000 (05:55 -0800)]
merge in jb-mr2-release history after reset to master
Elliott Hughes [Thu, 14 Feb 2013 01:30:54 +0000 (17:30 -0800)]
Simplify OpenGL TLS access on ARM.
bionic now assumes the TLS register is available, so OpenGL can too.
Change-Id: If2b56a4c08de9f887759b78f70022026a181dc47
Jesse Hall [Sat, 9 Feb 2013 00:13:50 +0000 (16:13 -0800)]
Symlink /system/lib/libGLESv3.so -> libGLESv2.so
The platform has a unified GLES2/GLES3 wrapper in libGLESv2.so, which
is what bundled binaries should link against. The NDK will have
separate GLES2 and GLES3 libraries so that applications will get a
link error if they accidentally use GLES3 entry points in a GLES2
application. This symlink allows GLES3 NDK apps to load using the
unified wrapper.
Change-Id: Ic344ef63c334047fccebb55b7cb2fb23a4c2d61b
Alex Ray [Wed, 13 Feb 2013 23:29:59 +0000 (15:29 -0800)]
atrace: add HAL tag
Change-Id: Idfd4f2075e8cd82833ef06d4487423c62b59fe11
Mathias Agopian [Wed, 13 Feb 2013 23:26:48 +0000 (15:26 -0800)]
remove access ANDROID_VIEW_SURFACE_JNI_ID
Change-Id: I0ef474ed9be2cd26fa520ef871ffcf21a25e4994
Mathias Agopian [Tue, 12 Feb 2013 00:40:36 +0000 (16:40 -0800)]
get rid of Surface identity and token
we use the IBinder instead.
Change-Id: I4aa0b58869ba43f19980013620051e5a261b062d
Jamie Gennis [Wed, 13 Feb 2013 23:22:04 +0000 (23:22 +0000)]
Merge changes I50831d4e,I5360aebe
* changes:
flatland: remove an unneeded #include
libgui: disallow NULL Fence pointers
The Android Automerger [Wed, 13 Feb 2013 13:55:55 +0000 (05:55 -0800)]
merge in jb-mr2-release history after reset to master
Jamie Gennis [Wed, 9 Jan 2013 02:05:17 +0000 (18:05 -0800)]
flatland: remove an unneeded #include
Change-Id: I50831d4efd543664ff7df7aaef35d842aebf1bd6
Jamie Gennis [Thu, 20 Dec 2012 22:05:45 +0000 (14:05 -0800)]
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
Jesse Hall [Fri, 8 Feb 2013 19:13:46 +0000 (11:13 -0800)]
Add ES3 support to libGLESv2 and tracing tools
Since ES3 is backwards compatible with ES2, a new wrapper isn't
necessary, and the Khronos implementation guidelines recommend
supporting both versions with the same library.
Change-Id: If9bb02be60ce01cc5fe25d1f40c4e7f37244ebf6
Jesse Hall [Fri, 8 Feb 2013 18:35:43 +0000 (10:35 -0800)]
Import OpenGL ES 3.0 headers from Khronos SVN
Add MODULE_LICENSE_MIT: The GLES headers use an MIT license or a
slight variant of it (the "SGI Free Software License B version 2.0",
http://oss.sgi.com/projects/FreeB/).
Change-Id: Ic8f946341cdd0d579f51b6f7a3ee109cd31b1691
Dianne Hackborn [Tue, 12 Feb 2013 23:54:27 +0000 (23:54 +0000)]
Merge "Implement issue #
6646859: 4K!!!! 4K!!!! 4K!!!!"
Dianne Hackborn [Tue, 12 Feb 2013 23:36:31 +0000 (15:36 -0800)]
Implement issue #
6646859: 4K!!!! 4K!!!! 4K!!!!
Change-Id: I1bb630c58d0d41446b726de1b9d2c3875fcbbceb