Commit Graph

337 Commits

Author SHA1 Message Date
Jesse Hall
2f4b68d21c SurfaceFlinger: fix layer removal race condition
Layer::lockPageFlip() and layer::onRemove() could be called on
different threads and race such that lockPageFlip() successfully
called mSurfaceTexture->updateTexImage() but then gets NULL back from
mSurfaceTexture->getCurrentBuffer(), leading to a crash.

This change moves Layer::onRemove() calls to
SurfaceFlinger::commitTransaction() so they happen after the Layer is
done being drawn from and only happen on the main surfaceflinger
thread.

Change-Id: I4b550caadff4cc1878d7c3bca6129193fb0c713e
2011-12-02 10:03:25 -08:00
Mathias Agopian
363c4a9fa0 Merge "add a way to access the version string of the h/w implementation of EGL" into ics-mr1 2011-11-30 13:52:21 -08:00
Mathias Agopian
bc2d79ed7a add a way to access the version string of the h/w implementation of EGL
we use a hidden egl extension. the version string is printed
in SF's dumpsys log.

Change-Id: I123eb4bde6de462bb2404c67b74d6d6219a48d6a
2011-11-29 18:00:35 -08:00
Jamie Gennis
86edf4f647 SurfaceTexture: add EGL_KHR_fence_sync option
This change adds a compile-time option for SurfaceTexture to use the
EGL_KHR_fence_sync extension to synchronize access to Gralloc buffers.

Bug: 5122031
Change-Id: I7e973a358631fff5308acf377581b811911fe790
2011-11-29 17:09:01 -08:00
Mathias Agopian
b8bb31dc10 temporarily hardcode Nexus S refresh rate
Change-Id: Id8465f679b96636041b4b891d0b358c387d456c3
2011-11-23 14:51:09 -08:00
Mathias Agopian
465494486c remove hack that hardcoded the refresh rate
Change-Id: Ibdd40b337c547e80ef8541856015f1380c8d58ed
2011-11-16 00:08:08 -08:00
Mathias Agopian
4b9511c161 rework a bit how we manage EGL extensions
- don't advertise extensions that are not supported
  by any implementation

- remove EGL_ANDROID_swap_rectangle which is not
  implemented by anybody and confuses people

- add some comments about mandatory extensions

Bug: 5428001
Change-Id: Id8dc48116ac1d1eb79ec9ef55d03e29d4257c1f3
2011-11-14 19:05:45 -08:00
Dave Burke
2f78cf1122 Back to 59fps
Change-Id: I8d89e5e27f1abc5a2e36cb832eb4db2ca83c2cc6
2011-11-09 17:54:37 -08:00
Mathias Agopian
386aa98c4f Fix an issue where we could wait for a non-existing transaction
This fixes the issue:
"Call not sent" dialog takes too long to dismiss after hitting OK"

Note: the system would recover after a 5 second timeout.

Bug: 5534520
Change-Id: Ifa37e594b50581f498479a5858672441b3d7dd87
2011-11-07 21:58:03 -08:00
Mathias Agopian
a5f9676ecc Merge "Fix rotation displays frame N-1 briefly while rotating" into ics-mr1 2011-11-07 21:33:53 -08:00
Mathias Agopian
385977f6d6 workaround to set the WM refresh rate to 48 Hz for tuna devices only
this is TEMPORARY until we can do this properly.

Bug: 5572464
Change-Id: I6537706d636a83a4a50e2900d6b829dd89b6f245
2011-11-04 18:46:11 -07:00
Mathias Agopian
4a9ac37fe2 Fix rotation displays frame N-1 briefly while rotating
The ScreenShot layer is now created hidden. The screenshot itself
is aquired during the transaction when the layer is made visible.
This guarantees the screenshot and the layer happen atomically
with respect to screen updates.

Bug: 5534521
Change-Id: Ida23e1f13d5716ec83b78a15712e0646d6cf8729
2011-11-04 15:15:32 -07:00
Mathias Agopian
62f7114719 fix Corrupted graphics while playing You Tube on orientation change
Bug: 5432124
Change-Id: If948b9797b1ec6fff80ca5ea94508abcaced9f31
2011-10-26 15:11:59 -07:00
Xavier Ducrohet
7f6aec0185 Merge "Fix software GL renderer." 2011-10-21 16:22:15 -07:00
Xavier Ducrohet
4c4163b53e Fix software GL renderer.
Change-Id: I07ab2709fa694e41aa3fcc5b9dfc809f55853ab4
2011-10-21 16:18:48 -07:00
Mathias Agopian
51726c381e am aa938c8d: Merge "mDirtyRegion is single threaded, but could be accessed from a hwc thread" into ics-mr0
* commit 'aa938c8d9c0e71c9b556657cb33794210ce6ebf8':
  mDirtyRegion is single threaded, but could be accessed from a hwc thread
2011-10-21 16:00:00 -07:00
Mathias Agopian
0dfb7b73a4 mDirtyRegion is single threaded, but could be accessed from a hwc thread
We now have mInvalidateRegion which holds the region to invalidate, it
can be set from any thread as long as mInvalidateLock is held. We use
fine-grained locking here because mInvalidateRegion can be set from anywhere,
in particular frmo HWC callbacks.

Bug: 5466774
Change-Id: Iafca20aa3f5b25a87755e65bde7b769aa8f997bc
2011-10-21 15:41:01 -07:00
Mathias Agopian
7ec662c07a am 16bece04: Merge "added dpi and refresh rate info in SF\'s dumpsys" into ics-mr0
* commit '16bece04bdbac9af5228436f70267ec5763315de':
  added dpi and refresh rate info in SF's dumpsys
2011-10-20 17:48:02 -07:00
Mathias Agopian
d5e4ef9e43 added dpi and refresh rate info in SF's dumpsys
Change-Id: I2327248eb1993689367e3daeaccba74c172dfceb
2011-10-20 17:22:38 -07:00
Dave Burke
818fb24357 am 70ac412b: Merge "Add a LayerScreenshot" into ics-mr0
* commit '70ac412b2fe7be2507189a9fdfb30c43b36d56ac':
  Add a LayerScreenshot
2011-10-18 23:00:08 -07:00
Mathias Agopian
675370ab9a am f7613743: Merge "Don\'t call hwc set() if there is nothing new to do" into ics-mr0
* commit 'f7613743f29f38cbbaf4ae09996c380845a17daf':
  Don't call hwc set() if there is nothing new to do
2011-10-18 20:24:23 -07:00
Mathias Agopian
5ba8f366bd am 23bf2505: Merge "Make sure set GL state properly" into ics-mr0
* commit '23bf250522b7dff9acd14c5900b66b1df9c76745':
  Make sure set GL state properly
2011-10-18 20:24:21 -07:00
Mathias Agopian
118d0245ee Add a LayerScreenshot
A LayerScreenshot is a special type of layer that contains a screenshot of
the screen acquired when its created. It works just like LayerDim.

Make sure to call compositionComplete() after rendering into a FBO.

Bug: 5446982, 5467587, 5466259
Change-Id: I5d8a1b4c327f9973d950cd4f4c0bca7f62825cd4
2011-10-18 20:21:47 -07:00
Mathias Agopian
3a3cad30c4 Don't call hwc set() if there is nothing new to do
there was situations where SF's main loop would run (as if there was
an invalidate), but the dirty region was empty (so no new buffers
were retired). In this case we return early and don't swap, which
would cause drawing artifacts.

Bug: 5476838
Change-Id: Id3b7bf4b7aabec7919c50d9278eb2165973a4c3d
2011-10-18 17:39:09 -07:00
Mathias Agopian
c492e67810 Make sure set GL state properly
when taking a screenshot, in particular, we could end up
with stale GL state when drawing LayerDim which resulted
in incortect rendering.

Bug: 5467587
Change-Id: Id9fbed2843481d31063620f3662b364c7e3ac781
2011-10-18 15:32:57 -07:00
Mathias Agopian
cdec8f01aa am 840b8a67: Revert "Add a LayerScreenshot"
* commit '840b8a678537519c27ddf2f818494eaa20a135d4':
  Revert "Add a LayerScreenshot"
2011-10-17 11:32:19 -07:00
Mathias Agopian
5bacca6238 am 4fb6416e: Merge "we need to guarantee that h/w comp set() is called when a buffer has been retired" into ics-mr0
* commit '4fb6416e3a21031a88921a784ae62b13d8a1a39f':
  we need to guarantee that h/w comp set() is called when a buffer has been retired
2011-10-17 11:32:16 -07:00
Elliott Hughes
090cb44b94 Merge branch 'master' of ssh://android-git:29418/platform/frameworks/base 2011-10-17 11:32:08 -07:00
Mathias Agopian
e9800c8311 Revert "Add a LayerScreenshot"
This reverts commit d6809f40cf61203573ec5dbc437f695cd132cc18.
2011-10-16 23:54:25 -07:00
Mathias Agopian
a44b041639 we need to guarantee that h/w comp set() is called when a buffer has been retired
removed a test that could violate this guarantee. note that we
have no proof tha this ever happened, but consequences could be
hard lock-ups.

the code here was intended to track the region to update for displays
that can do partial update. the logic discarded the update entirely
if that region was empty. instead we just redraw the whole thing
(note that we should never be there with an empty region and retired
buffers though).

Bug: 5466259

Change-Id: I91ccab3b1a599e729e438eb833939e2236da6854
2011-10-16 18:59:38 -07:00
Jamie Gennis
a402c4c991 SurfaceFlinger: Remove display freezing code
This change removes the dead code from SurfaceFlinger that resulted from
disabling support for freezing the display.

Change-Id: I4e5ff00c94b4c7a79af2f65c9850c135210068ed
2011-10-14 16:44:08 -07:00
Jamie Gennis
23c2c5d1c6 SurfaceFlinger: add some layer update tests
This change adds two tests for SurfaceFlinger's behavior when updating
the position and size of a layer.

Change-Id: Id5e3ca1d7e629a3bd2c2d28275d80c7f9256d6da
2011-10-14 16:11:51 -07:00
Jamie Gennis
28378392fd SurfaceFlinger: make sync transactions explicit
This change enables a layer or orientation update transaction sent to
SurfaceFlinger to explicitly request a synchronous transaction.

Change-Id: I97cbba610c13679849f66114b216fa6dbf12f2a9
2011-10-14 16:11:51 -07:00
Mathias Agopian
f171ab6da9 Add a LayerScreenshot
A LayerScreenshot is a special type of layer that contains a screenshot of
the screen acquired when its created. It works just like LayerDim.

Bug: 5446982
Change-Id: I7814aff2380e7e146937f2b641907be2a30c76cc
2011-10-14 14:32:48 -07:00
Jamie Gennis
b8d69a55f1 SurfaceFlinger: update orientation via transactions
This change merges the ISurfaceComposer::setOrientation functionality
into ISurfaceComposer::setTransactionState.  It enables the window
manager to atomically update both the display orientation and the
position and size of the windows in a single transaction with
SurfaceFlinger.

Bug: 5439574
Change-Id: I18a8ccc564d7d760ef8afb2d015ccdb7a7963900
2011-10-11 13:34:01 -07:00
Mathias Agopian
e64206915d Merge "fix an issue where the screen could stay off" 2011-10-10 22:21:57 -07:00
Mathias Agopian
f7cdd053fe fix an issue where the screen could stay off
this would happen when toggling on/off/on very fast, the screen
could stay black (while the panel is on).

Bug: 5429724
Change-Id: Ic8aa6aff066e6267923c0d47ef65e314e7bb6d41
2011-10-10 22:18:55 -07:00
Mathias Agopian
1dea97d662 Merge "Fix screen off animation when in landscape" 2011-10-10 19:05:23 -07:00
Mathias Agopian
a9040d0eef Fix screen off animation when in landscape
Change-Id: I4bc5b12d7a64a4bf8b9a851594be4d60b790d1ed
2011-10-10 19:02:07 -07:00
Mathias Agopian
8cb9d95002 Merge "enable EGL_IMG_context_priority for the omap4 platform" 2011-10-10 12:38:16 -07:00
Jamie Gennis
9575f60722 SurfaceFlinger: screenshots w/ protected buffers
This change modifies SurfaceFlinger's screenshot behavior when a layer
with a protected buffer is visible.  The previous behavior was to simply
fail the screenshot.  The new behavior is to render the screenshot using
a placeholder texture where the protected buffer would have been.

Change-Id: I5e50cb2f3b31b2ea81cfe291c9b4a42e9ee71874
2011-10-07 17:53:37 -07:00
Mathias Agopian
57bf9e7100 enable EGL_IMG_context_priority for the omap4 platform
Bug: 5311015
Change-Id: Icaad3d1c963d382ac0a4b9121fe20d3408846c2b
2011-10-07 15:42:53 -07:00
Mathias Agopian
e2c2f9213f Force a repaint when hwc invalidate hook is called
without this prepare() would be called but not set() since the
dirty region is empty.

Change-Id: I038acfbdad4c16015357ccde4d1949391d6f989d
2011-10-05 15:00:22 -07:00
Mathias Agopian
aebac5f34a improve hwc log output
we now print the handle of each buffer

Change-Id: I70884ae458a9dcf10cfc403258d48eaa4618b7f4
2011-09-29 18:07:08 -07:00
Jamie Gennis
ed984d9406 Merge changes Ib302d79e,Ib4ee085f
* changes:
  SurfaceFlinger: set layer names on SurfaceTextures
  SurfaceTexture: add name support
2011-09-27 14:24:09 -07:00
Jamie Gennis
e790db9980 am da42e0fa: Merge "SurfaceFlinger: fix setting default buffer size" into ics-factoryrom
* commit 'da42e0faaa95c796dc49f3b38023562dab6584a7':
  SurfaceFlinger: fix setting default buffer size
2011-09-27 12:32:17 -07:00
Jamie Gennis
1b5f4e61ff Merge "SurfaceFlinger: fix setting default buffer size" into ics-factoryrom 2011-09-27 12:28:59 -07:00
Jamie Gennis
9a375927a5 am 3780d380: Merge "SurfaceFlinger: fix the transform hint" into ics-factoryrom
* commit '3780d38010f50ccce508871058d54c1921936e0c':
  SurfaceFlinger: fix the transform hint
2011-09-26 19:44:41 -07:00
Jamie Gennis
2a0d5b6084 SurfaceFlinger: fix setting default buffer size
This change makes SurfaceFlinger update a SurfaceTextureLayer's default
buffer size in response to a window resize even if the Layer is in fixed
size mode.

Change-Id: I3871c8f19a045d03904d1284390cad7659be71ff
Bug: 5366426
2011-09-26 18:38:52 -07:00
Mathias Agopian
4801404dcb don't clear the framebuffer when the framebuffer is not used
when areas of the FB are undefined (transparent windows on top of
nothing), we clear those areas before composition.
however, it makes no sense to do this when the FB is not in use
(case where hwc handles all layers)

Bug: 5360529
Change-Id: I761a7d3664228ebbee0944aca8b0cc278fc94c9c
2011-09-26 16:14:05 -07:00