Commit Graph

48966 Commits

Author SHA1 Message Date
Lajos Molnar
dde86f7b18 media/hardware/HardwareAPI.h: Added PrepareForAdaptivePlaybackParams
This is the param struct used by stagefright to notify codec that
the playback may include resolution changes, so that it can prepare
accordingly to avoid port reconfiguration.

Change-Id: I64e53b7ebf74c4f9d07cc9fe0cd65be55f666afe
Signed-off-by: Lajos Molnar <lajos@google.com>
Related-to-bug: 7093648
2013-08-12 12:56:08 -07:00
Mathias Agopian
96a7dc08f2 Merge "make sure we have a context when creating the 1st surface" into klp-dev 2013-08-09 04:41:19 +00:00
Mathias Agopian
9870c9b66c make GLConsumer's debug texture static
this is so that we don't burn a gralloc handle and a bit
of memory per GLConsumer.

Change-Id: I30c3e7ec45d7f1b5426aca8e06c7716870877706
2013-08-08 17:46:48 -07:00
Mathias Agopian
a6bb107434 make sure we have a context when creating the 1st surface
also add an option to dump the stack trace when
calling a GL function without a context.

Change-Id: I57b72bb8c322ac4253c3077bf150621bd9863b69
2013-08-08 16:38:06 -07:00
Jesse Hall
6c913be9ca Add ISurfaceComposer::destroyDisplay
Bug: 10191053
Change-Id: Ia89286f95421344a60ffedaaca5825c4e3cd7f9e
2013-08-08 13:12:29 -07:00
Mathias Agopian
74020ca133 Merge "don't hardcode debug texture dimensions" 2013-08-07 20:36:07 +00:00
Mathias Agopian
45263e2475 don't hardcode debug texture dimensions
Change-Id: I356333801d246a8cafb3e977a77dc7325e8b822d
2013-08-07 13:35:20 -07:00
Jean-Baptiste Queru
0de07adb02 am 046b72fc: Merge "fix a possible deadlock when removing a layer and destroying a client"
* commit '046b72fc5d93aae3b47b88b9f6683512d01d75d0':
  fix a possible deadlock when removing a layer and destroying a client
2013-08-07 13:18:51 -07:00
Todd Poynor
b98240cc6c batteryservice: add battery current now and charge counter data
Change-Id: I4b3ea25342d4a03a8e38bfee2ef3b1ce54f82582
2013-08-07 18:30:13 +00:00
Jesse Hall
43c2718b3f Merge "Add EGL_KHR_get_all_proc_addresses to extension string" 2013-08-07 15:23:05 +00:00
Jesse Hall
ff866e7642 Merge "Set the swap interval of virtual display surfaces to 0" 2013-08-07 15:23:00 +00:00
Jesse Hall
f460f55c84 Set the swap interval of virtual display surfaces to 0
This prevents slow/malicious virtual display consumers from
backpressuring SurfaceFlinger and preventing it from updating other
displays.

Bug: 10193714
Change-Id: I3e877d97202628d2d9abea24b66576f38299c14d
2013-08-06 19:32:02 -07:00
Mathias Agopian
5f51ed9499 fix tests more
Change-Id: I05c3dd6a22440844f8bd071f4c2b620b2bd8c720
2013-08-06 16:47:04 -07:00
Mathias Agopian
3dce3360e7 fix tests
Change-Id: I0f4d613481eda2c44a2d631a81d75de7e52e7a62
2013-08-06 16:42:05 -07:00
Mathias Agopian
3ca76f416b remove support for glReadPixels screenshot path
this was only needed on some chipset we're not
supporting in KLP.

Change-Id: I2e8fc07f135030bd61d2e241a8e426f1a4316425
2013-08-06 16:07:33 -07:00
Jesse Hall
21558daf69 Add EGL_KHR_get_all_proc_addresses to extension string
We've had the extended behavior since Android 4.3, but the extension
hadn't been ratified by Khronos yet so we couldn't add it to the
extension string.

Bug: 9681677
Change-Id: I78842316a3ab7a3f66a0ac1a4742d837de664c4c
2013-08-06 15:33:46 -07:00
Mathias Agopian
126b9c8e95 Fix build. 3rd attempt
Change-Id: I4317b34fd07890ea7cf9db51be42d72ef21aca89
2013-08-06 15:19:15 -07:00
Mathias Agopian
db89edc94b All consumers now take an IGraphicBufferConsumer instead of a BufferQueue
this means they only have access to the consumer end of
the interface. we had a lot of code that assumed consumers
where holding a BufferQueue (i.e.: both ends), so most of
this change is untangling in fix that

Bug: 9265647
Change-Id: Ic2e2596ee14c7535f51bf26d9a897a0fc036d22c
2013-08-06 20:05:36 +00:00
Jean-Baptiste Queru
046b72fc5d Merge "fix a possible deadlock when removing a layer and destroying a client" 2013-08-06 18:37:12 +00:00
Mathias Agopian
e5886d9178 fix a possible deadlock when removing a layer and destroying a client
generally the last reference to a Layer is released in commitTransaction()
with mStateLock held. Layer itself only holds weak references to Client,
however, ~Layer() briefly promotes this weak reference -- during that time
the all other strong references to that Client go away, ~Layer is left with
the last one... then hell breaks loose as ~Client is called, which in turn
needs to acquire mStateLock.

We fix this by holding a temporary copy of the drawing state during
the transaction so that the side-effects of copying the current
state into the drawing state are seen only after mStateLock has
been released.

Bug: 9106453
Change-Id: Ic5348ac12283500ead87286a37565e8da35f1db2
2013-08-05 14:28:46 -07:00
Eino-Ville Talvala
687821c05b Merge "BufferQueue: Support query of consumer usage bits" 2013-08-05 21:10:45 +00:00
Andy McFadden
d9ba9707b0 Merge "Roll back recent change" 2013-08-02 22:34:13 +00:00
Andy McFadden
4b49e08385 Roll back recent change
Disable dropping of frames based on timestamp.  Resume auto-
generating timestamps in Surface.

Bug 10151804

Change-Id: I15de26158e1d7ef22a5b150e685a126dc48ae2b4
2013-08-02 15:31:45 -07:00
Mathias Agopian
c1c39f8807 Merge "Binderize the consumer side of BufferQueue" 2013-08-02 20:35:26 +00:00
Andy McFadden
86cd09d388 Merge "Drop frames based on presentation timestamp" 2013-08-02 16:53:11 +00:00
Mathias Agopian
a4e19521ac Binderize the consumer side of BufferQueue
While currently untested, this should allow to move the
BuffereQueue in the consumer process and have everything
work as usual.

Bug: 9265647

Change-Id: I9ca8f099f7c65b9a27b7e7a3643b46d1b58eacfc
2013-08-01 17:20:08 -07:00
Andy McFadden
14fab7dd79 Drop frames based on presentation timestamp
If there are two or more buffers pending that are ready for
immediate presentation, drop all but the last one.

Any code that didn't explicitly specify timestamps for buffers
was using the default value (auto-generated "now").  As a result,
surfaceflinger would drop frames whenever more than one buffer
was queued.  We now use zero as the auto-generated timestamp,
and we don't set the timestamp in eglBeginFrame().

Change-Id: I187f42d33de227cd3411ff0dcd3b9ce1961457eb
2013-08-01 15:58:08 -07:00
Mathias Agopian
ba93b3f8e4 log an error when dequeueBuffer gets a NULL Fence
Bug: 9858745
Change-Id: If35d6a7d770bf0a42d55bc55880a9ddb721fcfd0
2013-08-01 15:48:40 -07:00
Mathias Agopian
87f9b87796 remove unused code form GraphicBuffer
Change-Id: I0dd67f1f93f654331e8f17b55730931dbc695ca7
2013-07-31 20:10:10 -07:00
Alex Ray
b304edb0b1 remove libutils
libutils is being moved from frameworks/native/ to system/core/ in order
to facilitate native C++ platform (non-frameworks) code.

Change-Id: I44089fb960591a40b8a9c30faabb10459d107d71
2013-07-31 13:48:57 -07:00
Brian Carlstrom
8776c34ca2 Fix BlobCache_test
Change-Id: I5b20aeb4dcff80f1e18e1165b8461d20034dc1ac
2013-07-31 01:08:36 -07:00
Mathias Agopian
e142428a9c Make Flattenable not virtual
Fallout from the Flattenable change, update all its uses.

Additionnaly, fix/tighten size checks when (un)flatten()ing
things.

Removed the assumption by some flattenables (e.g.: Fence)
that the size passed to them would be exact (it can
and will be larger in some cases)

The code in Parcel is a bit complicated so that we don't
have to expose the full implementation (and also to
keep the code smallish).

Change-Id: I0bf1c8aca2a3128491b4f45510bc46667e566dde
2013-07-30 21:19:13 -07:00
Mathias Agopian
1d76781b7a Make Flattenable not virtual (libutils)
Making an object Flattenable doesn't force it to
become virtual anymore. For instance, Fence and GraphicBuffer
are now non-virtual classes.

Also change Flatennable protocol a bit so that it updates
its parameters (pointers, sizes) to make it easier
to implement a flattenable in terms of other flattenables.

Change-Id: Ie81dc7637180b3c2cfcbaf644f8987ca804eb891
2013-07-30 21:19:02 -07:00
Jamie Gennis
7925b6bdb8 Merge "EGL: fix eglTerminate blob cache behavior" 2013-07-30 22:37:58 +00:00
Jamie Gennis
5539e219de EGL: fix eglTerminate blob cache behavior
Bug: 9610573
Change-Id: I258e7e0d82ab76fbacf0b9c25d4f92fca6df04ac
2013-07-30 15:12:16 -07:00
Eino-Ville Talvala
f7c6087bcc BufferQueue: Support query of consumer usage bits
Bug: 9592202
Change-Id: I61a17758200e61adaca695b426db3df0f6637d4c
2013-07-30 21:38:34 +00:00
Wink Saville
f97b6e218a am 9f476fd0: Merge "Add a symbol to represent MNC=0"
* commit '9f476fd08079701d1ad0f7c591667b6e673ed38e':
  Add a symbol to represent MNC=0
2013-07-30 12:28:02 -07:00
Wink Saville
9f476fd080 Merge "Add a symbol to represent MNC=0" 2013-07-30 19:19:41 +00:00
Mike J. Chen
aaff4ef717 SurfaceFlinger: Set the result of binderized screencapture
A recent change to screencapture to have the call to
IGraphicBufferProducer happen on the incoming binder
thread didn't set the result so the result was always
returned as NO_ERROR.

This made screencap fail on some devices (e.g. Wolfie)
which relies on some kind of fallback mechanism to
generate the screencap but the fallback mechanism
doesn't get triggered because the error isn't returned.

Bug: 9989385

Change-Id: I2aee91ea1034869fcbb0f49b9a0087c3cff43bbe
Signed-off-by: Mike J. Chen <mjchen@google.com>
2013-07-30 10:24:22 -07:00
Mathias Agopian
ad678e18b6 single buffer mode for BufferQueue
Bug: 9891035
Change-Id: Id1ab5f911a6dc4c1d8235e65775b3d3635231ad4
2013-07-26 18:45:02 -07:00
Mathias Agopian
bf5b849ec7 fix build
Change-Id: I04df8ec96510b16e84054089f7c02609d8fe2b0c
2013-07-26 17:26:16 -07:00
Mathias Agopian
c2414bb098 get rid of HAL pixelformats 5551 and 4444
Change-Id: I39409103144c1ba09078c4ddc323a58987fb09ea
2013-07-26 15:43:33 -07:00
Mathias Agopian
dd533712f8 make transparent pixels transparents again
apparently boolean logic in hard...

Change-Id: Ie70401fe12e0fa00b0b8c7cc475c1438cdbbf531
2013-07-26 15:31:39 -07:00
Johan Redestig
8df483c1fd Add a symbol to represent MNC=0
Bug: 7170488
Change-Id: Icfd39fd0c739216e89446252ea0e7ceba6f002c6
2013-07-26 18:39:09 +00:00
Mathias Agopian
5773d3f5b2 get rid of PixelFormatInfo and simplify things
Change-Id: I025a362cc12d5b9b794fac14be500e25aab65396
2013-07-25 19:24:31 -07:00
Mathias Agopian
1fe854f678 get rid of PIXEL_FORMAT_A_8
Change-Id: I4e409a69c01b95b48f20112084dd927e2d043407
2013-07-25 17:25:05 -07:00
Mathias Agopian
9642dba110 Merge "Fix a race in BufferQueue" 2013-07-25 02:17:52 +00:00
Todd Poynor
87bf0d97be add libbatteryservice
IBatteryPropertiesListener is used by clients such as BatteryService to
receive notifications of changed battery/power status (from healthd).

IBatteryPropertiesRegistrar manages registrations of
IBatteryPropertiesListener clients.

Add BatteryService native header with defines from BatteryManager, and
class BatteryProperties to pass battery/power status around.

Change-Id: I44e2736e503b586794097b7b9d31d07fefd9b15a
2013-07-25 01:57:32 +00:00
Todd Poynor
a7b0f04aaf defaultServiceManager wait for service manager ready
If the service manager is not ready when we attempt to create a local
proxy, fail and retry.

Change-Id: I7d7300bc07cd70608793479aa3da282d066da7f6
2013-07-24 16:18:12 -07:00
Jamie Gennis
1d4f79c482 Merge "surfaceflinger: add frame duration logging" 2013-07-24 20:47:37 +00:00