OSDN Git Service

android-x86/frameworks-native.git
10 years agoAdd support for mmc trace events
Ken Sumrall [Wed, 3 Jul 2013 19:32:03 +0000 (12:32 -0700)]
Add support for mmc trace events

Change-Id: I1d9f2a85e50f8e30d418fb97c01ce7b0fbb680bd

10 years agoam 8b931916: (-s ours) am f4f2bd21: Tweak for forward compatibility
Jean-Baptiste Queru [Tue, 2 Jul 2013 19:41:22 +0000 (12:41 -0700)]
am 8b931916: (-s ours) am f4f2bd21: Tweak for forward compatibility

* commit '8b9319162d4f574fc45f88e6f9e28e62a4b5da78':
  Tweak for forward compatibility

10 years agoam f4f2bd21: Tweak for forward compatibility
Jean-Baptiste Queru [Tue, 2 Jul 2013 19:33:15 +0000 (12:33 -0700)]
am f4f2bd21: Tweak for forward compatibility

* commit 'f4f2bd21d5b57f1598172ddbc44817c10785bdb3':
  Tweak for forward compatibility

10 years agoTweak for forward compatibility
Jean-Baptiste Queru [Tue, 2 Jul 2013 17:46:02 +0000 (10:46 -0700)]
Tweak for forward compatibility

Change-Id: I6be78b607789c469f4ffad849a493d78e2539450

10 years agodo not build libinput in PDK
keunyoung [Tue, 2 Jul 2013 17:45:21 +0000 (10:45 -0700)]
do not build libinput in PDK

- skia is not available in PDK, and it causes build break.

bug: 9528927
Change-Id: I89663e4a4bd52858506e3b52fbf8ca85d9992363

10 years agoMerge "Move input library code from frameworks/base."
Jeff Brown [Tue, 2 Jul 2013 03:20:11 +0000 (03:20 +0000)]
Merge "Move input library code from frameworks/base."

10 years agoMove input library code from frameworks/base.
Jeff Brown [Tue, 2 Jul 2013 02:10:31 +0000 (19:10 -0700)]
Move input library code from frameworks/base.

Change-Id: I4983db61b53e28479fc90d9211fafff68f7f49a6

10 years agoimprove GLES jumptables
Mathias Agopian [Sat, 15 Jun 2013 02:08:36 +0000 (19:08 -0700)]
improve GLES jumptables

in the common case this saves one instructions per jump
(which will help with the i-cache).

this change also gets rid of the "use slow tls" option,
which was useless. So at least now architectures that don't have
assembly bindings will perform much better.

Change-Id: I31be6c06ad2136b50ef3a1ac14682d7812ad40d2

10 years agoMerge "SurfaceFlinger: fix releaseBuffer in updateTexImage"
Lajos Molnar [Fri, 28 Jun 2013 00:42:23 +0000 (00:42 +0000)]
Merge "SurfaceFlinger: fix releaseBuffer in updateTexImage"

10 years agoSurfaceFlinger: fix releaseBuffer in updateTexImage
Lajos Molnar [Thu, 27 Jun 2013 18:51:25 +0000 (11:51 -0700)]
SurfaceFlinger: fix releaseBuffer in updateTexImage

Allow releasing a buffer acquired in the past.

Change-Id: I3f27f1caec109c1a47286fc4d2235c6e1afe9022
Signed-off-by: Lajos Molnar <lajos@google.com>
Bug: 9516405

10 years agoAdd checkin dumps to bug reports.
Dianne Hackborn [Thu, 27 Jun 2013 01:59:09 +0000 (18:59 -0700)]
Add checkin dumps to bug reports.

This is a duplicate of the associated human-readable dumps
of these services, but much better for automated parsing.

Change-Id: I902b768b03aeac1472f03f96913f9b8ada5dba9c

10 years agoMerge "Small clarifications to EGL_ANDROID_presentation_time"
Jesse Hall [Wed, 26 Jun 2013 14:51:25 +0000 (14:51 +0000)]
Merge "Small clarifications to EGL_ANDROID_presentation_time"

10 years agoSmall clarifications to EGL_ANDROID_presentation_time
Jesse Hall [Wed, 26 Jun 2013 14:50:32 +0000 (07:50 -0700)]
Small clarifications to EGL_ANDROID_presentation_time

Change-Id: I4365a981776ab0a5719c268258b87ff1418f9aeb

10 years agoam db8c267e: am 3c0425cd: am 0ecf0b8d: Merge "Additional parameter validation for...
Jean-Baptiste Queru [Wed, 26 Jun 2013 00:02:09 +0000 (17:02 -0700)]
am db8c267e: am 3c0425cd: am 0ecf0b8d: Merge "Additional parameter validation for EGL functions"

* commit 'db8c267ec26997771eb718c29d3dacd4e738051b':
  Additional parameter validation for EGL functions

10 years agoam 3c0425cd: am 0ecf0b8d: Merge "Additional parameter validation for EGL functions"
Jean-Baptiste Queru [Tue, 25 Jun 2013 22:37:44 +0000 (15:37 -0700)]
am 3c0425cd: am 0ecf0b8d: Merge "Additional parameter validation for EGL functions"

* commit '3c0425cd95a7918d1dbfecf32aef3b238bf236a5':
  Additional parameter validation for EGL functions

10 years agoam 0ecf0b8d: Merge "Additional parameter validation for EGL functions"
Jean-Baptiste Queru [Tue, 25 Jun 2013 22:35:05 +0000 (15:35 -0700)]
am 0ecf0b8d: Merge "Additional parameter validation for EGL functions"

* commit '0ecf0b8db74682222bf095a235e28edefff871a5':
  Additional parameter validation for EGL functions

10 years agoMerge "Additional parameter validation for EGL functions"
Jean-Baptiste Queru [Tue, 25 Jun 2013 22:31:13 +0000 (22:31 +0000)]
Merge "Additional parameter validation for EGL functions"

10 years agoMerge "Add video extension OMX_VIDEO_CodingVP9 to support VP9 decoder."
Lajos Molnar [Tue, 25 Jun 2013 22:07:39 +0000 (22:07 +0000)]
Merge "Add video extension OMX_VIDEO_CodingVP9 to support VP9 decoder."

10 years agoFix includes in HardwareAPI.h to be absolute and not relative
Andreas Huber [Tue, 25 Jun 2013 17:08:42 +0000 (10:08 -0700)]
Fix includes in HardwareAPI.h to be absolute and not relative

(but using system header syntax)

Change-Id: I504c5c5330ddd1e3b3d2fb4760b0235a4fda84d8

11 years agoAdd video extension OMX_VIDEO_CodingVP9 to support VP9 decoder.
hkuang [Mon, 24 Jun 2013 18:06:44 +0000 (11:06 -0700)]
Add video extension OMX_VIDEO_CodingVP9 to support VP9 decoder.

Change OMX_Video_CodingVPX to OMX_VIDEO_CodingVP8 in order
to be consistant with official OMX_VideoExt.h and avoid the
confusion with new extension OMX_VIDEO_CodingVP9.

This change needs to be merged with changes in frameworks/av to
avoid build break due to header change.

Change-Id: Iaa2dbafa76dc24ad3cf52a9bf3dee312e3c64bfa

11 years agoMerge "media/hardware/HardwareAPI.h: Added metadata struct used by video decoder"
Lajos Molnar [Mon, 24 Jun 2013 17:47:16 +0000 (17:47 +0000)]
Merge "media/hardware/HardwareAPI.h: Added metadata struct used by video decoder"

11 years agoFix wrapper equals()
Andy McFadden [Fri, 21 Jun 2013 23:02:11 +0000 (16:02 -0700)]
Fix wrapper equals()

Make it faster and correcter.

Also, fixed the "gen" script to update the static/egl/*.java files
as part of doing an update.  Altered the "git" lines to change
directories before invoking git -- necessary because we're now
split between frameworks/base and frameworks/native.

Bug 9204146

Change-Id: Ie60504b1932349e6f2c9d3868dee6a645ef2b36e

11 years agomedia/hardware/HardwareAPI.h: Added metadata struct used by video decoder
Lajos Molnar [Fri, 3 May 2013 20:20:53 +0000 (13:20 -0700)]
media/hardware/HardwareAPI.h: Added metadata struct used by video decoder

This is the metadata layout when used by the video decoder.

Change-Id: I3497305ad23b3fc4e1e8eaaea7b0890bc1ffd5ad
Signed-off-by: Lajos Molnar <lajos@google.com>
Bug: 7093648

11 years agoMoving dex2oat installd patches from frameworks/base
Brian Carlstrom [Fri, 22 Mar 2013 01:20:22 +0000 (18:20 -0700)]
Moving dex2oat installd patches from frameworks/base

(cherry picked from commit 3fa3a652ba9a7c5b23b6f2bc7c1851d4e23e430f)

Change-Id: I2f9bb66e1b0463bef41805f2d220160880121813

11 years agoMerge "Add a note about emulated sdcard in rawbu usage dump"
Amith Yamasani [Fri, 14 Jun 2013 18:00:01 +0000 (18:00 +0000)]
Merge "Add a note about emulated sdcard in rawbu usage dump"

11 years agoAdd a note about emulated sdcard in rawbu usage dump
Amith Yamasani [Fri, 14 Jun 2013 17:13:53 +0000 (10:13 -0700)]
Add a note about emulated sdcard in rawbu usage dump

Change-Id: I95af9175235a18716d63a863b482a323392e4fa2

11 years agoadd dumpsys -l to list available services
keunyoung [Thu, 13 Jun 2013 22:08:51 +0000 (15:08 -0700)]
add dumpsys -l to list available services

- for people with poor memory like me :)

Change-Id: I60160ca333ac4864bb8328f164cc2e9eba3660fc

11 years agoam 03c06bc9: (-s ours) Reconcile with jb-mr2-release - do not merge
The Android Open Source Project [Thu, 13 Jun 2013 01:06:09 +0000 (18:06 -0700)]
am 03c06bc9: (-s ours) Reconcile with jb-mr2-release - do not merge

* commit '03c06bc9f44626eae1af9901aaddfb4e6e911c6a':
  Prevent IPCThreadState::joinThreadPool from spinning on unexpected error

11 years agoReconcile with jb-mr2-release - do not merge
The Android Open Source Project [Thu, 13 Jun 2013 01:04:25 +0000 (18:04 -0700)]
Reconcile with jb-mr2-release - do not merge

Change-Id: Ib28265f84583e4c1c8860f359145f38737817e99

11 years agoPrevent IPCThreadState::joinThreadPool from spinning on unexpected error
Jeff Tinker [Tue, 11 Jun 2013 18:30:21 +0000 (11:30 -0700)]
Prevent IPCThreadState::joinThreadPool from spinning on unexpected error

Adds logging to help determine what is happening to the /dev/binder
fd and dump the process state when it happens.

bug: 8912673
Change-Id: I2aa0c66fc499e91e0bf9ee4ae20404bec35adc82

11 years agoam cd9482d9: (-s ours) Reconcile with jb-mr2-zeroday-release - do not merge
The Android Open Source Project [Wed, 12 Jun 2013 14:26:00 +0000 (07:26 -0700)]
am cd9482d9: (-s ours) Reconcile with jb-mr2-zeroday-release - do not merge

* commit 'cd9482d961d1c14b6dea7279f8524e29908b6a27':

11 years agoReconcile with jb-mr2-zeroday-release - do not merge
The Android Open Source Project [Wed, 12 Jun 2013 14:23:36 +0000 (07:23 -0700)]
Reconcile with jb-mr2-zeroday-release - do not merge

Change-Id: I02e004028811a3efad8c010ed19e216bb87814d7

11 years agomerge in jb-mr2-zeroday-release history after reset to jb-mr2-dev
The Android Automerger [Wed, 12 Jun 2013 12:56:00 +0000 (05:56 -0700)]
merge in jb-mr2-zeroday-release history after reset to jb-mr2-dev

11 years agoam de836890: am ef07386e: Prevent IPCThreadState::joinThreadPool from spinning on...
Jeff Tinker [Wed, 12 Jun 2013 03:11:43 +0000 (20:11 -0700)]
am de836890: am ef07386e: Prevent IPCThreadState::joinThreadPool from spinning on unexpected error

* commit 'de836890abc7b5381e285833052375cbb85730cb':
  Prevent IPCThreadState::joinThreadPool from spinning on unexpected error

11 years agoam ef07386e: Prevent IPCThreadState::joinThreadPool from spinning on unexpected error
Jeff Tinker [Wed, 12 Jun 2013 03:09:27 +0000 (20:09 -0700)]
am ef07386e: Prevent IPCThreadState::joinThreadPool from spinning on unexpected error

* commit 'ef07386e2fca73680214ececc3c9c0ecbb0f6d88':
  Prevent IPCThreadState::joinThreadPool from spinning on unexpected error

11 years agoPrevent IPCThreadState::joinThreadPool from spinning on unexpected error
Jeff Tinker [Tue, 11 Jun 2013 18:30:21 +0000 (11:30 -0700)]
Prevent IPCThreadState::joinThreadPool from spinning on unexpected error

Adds logging to help determine what is happening to the /dev/binder
fd and dump the process state when it happens.

bug: 8912673
Change-Id: I2aa0c66fc499e91e0bf9ee4ae20404bec35adc82

11 years agoCpuConsumer: Add set buffer size and format functions.
Zhijun He [Thu, 6 Jun 2013 18:59:21 +0000 (11:59 -0700)]
CpuConsumer: Add set buffer size and format functions.

Add setDefaultBufferSize() and setDefaultBufferFormat(). ImageReader JNI need
them.

Bug: 9254294
Change-Id: I7d2464d43b0ca73fbb834ed22cecbfbb30eef60c

11 years agoclean-up SurfaceFlinger a bit
Mathias Agopian [Wed, 5 Jun 2013 23:59:15 +0000 (16:59 -0700)]
clean-up SurfaceFlinger a bit

- most methods on Layer didn't need to be virtual
- more consistency in naming drawing/current state

Change-Id: Ieb7b4951b40fc673b807994ed21ae4aea7281068

11 years agoMerge "Basic binary event-log helpers"
Mathias Agopian [Thu, 6 Jun 2013 00:01:15 +0000 (00:01 +0000)]
Merge "Basic binary event-log helpers"

11 years agofix a possible deadlock when removing a layer and destroying a client
Mathias Agopian [Wed, 5 Jun 2013 21:30:54 +0000 (14:30 -0700)]
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

11 years agoBasic binary event-log helpers
Mathias Agopian [Wed, 5 Jun 2013 04:50:31 +0000 (21:50 -0700)]
Basic binary event-log helpers

EventLog can be used from SurfaceFlinger to log jank
statistics events to the binary event log.

Change-Id: If0fbbe7d7111f10957697166d37fd0c3418109bb

11 years agoCpuConsumer: Remove CpuConsumer JNI ID.
Zhijun He [Mon, 3 Jun 2013 15:10:16 +0000 (08:10 -0700)]
CpuConsumer: Remove CpuConsumer JNI ID.

Not used anywhere, and it is pretty confusing to have it here.

Bug: 9254294
Change-Id: I12b761532918b546a0f56c74ccf81b2c119ccd66

11 years agoMerge "Added bitwise-or and bitwise-and to BitSet"
Michael Wright [Fri, 31 May 2013 03:31:49 +0000 (03:31 +0000)]
Merge "Added bitwise-or and bitwise-and to BitSet"

11 years agoMerge "free EGL resources associated to a thread when it terminates"
Mathias Agopian [Fri, 31 May 2013 02:01:30 +0000 (02:01 +0000)]
Merge "free EGL resources associated to a thread when it terminates"

11 years agoAdded bitwise-or and bitwise-and to BitSet
Michael Wright [Tue, 21 May 2013 21:11:34 +0000 (14:11 -0700)]
Added bitwise-or and bitwise-and to BitSet

Change-Id: I9bbf41f9d2d4a2593b0e6d7d8be7e283f985bade

11 years agofree EGL resources associated to a thread when it terminates
Mathias Agopian [Thu, 30 May 2013 23:07:36 +0000 (16:07 -0700)]
free EGL resources associated to a thread when it terminates

destroyed but current-to-a-thread resources are only destroyed
when they're made not-current; however, we were not destroying
those when the thread itself terminated, causing these resources
to be leaked forever.

we now install a tls-key destructor that takes care of this
by calling eglReleaseThread upon thread termination.

Bug: 9209776
Change-Id: I88298a34e3a27488eb81eab76717715569c7d57c

11 years agoMerge "Bring back overloading to print String8/16."
Jeff Sharkey [Thu, 30 May 2013 21:33:36 +0000 (21:33 +0000)]
Merge "Bring back overloading to print String8/16."

11 years agoMerge changes Id5584bc1,I8ca826d0
Mathias Agopian [Thu, 30 May 2013 21:30:17 +0000 (21:30 +0000)]
Merge changes Id5584bc1,I8ca826d0

* changes:
  revert parts of dc5b63e40, which made gyro drift estimation unstable
  use gyro timestamp directly in fusion

11 years agorevert parts of dc5b63e40, which made gyro drift estimation unstable
Mathias Agopian [Thu, 30 May 2013 21:26:36 +0000 (14:26 -0700)]
revert parts of dc5b63e40, which made gyro drift estimation unstable

    initialize the system covariance matrix with non-zero
    values for the gyro-bias part. this improves the initial
    bias estimation speed significantly.

the initial covariance matrix should be small because the drift
changes slowly. the real problem is that we're not starting with
a good estimate of the drift, which this algorithm relies on.

so with this revert, it'll take a while for the drift to be estimated
but it won't be unstable.

Change-Id: Id5584bc114a2390d507643b2451b2650c1b90721

11 years agoam 255405eb: am 86b2ecdd: Merge "fix mem leak on error handling"
Jeff Brown [Thu, 30 May 2013 21:23:58 +0000 (14:23 -0700)]
am 255405eb: am 86b2ecdd: Merge "fix mem leak on error handling"

* commit '255405eb19e6a3fa7f7bb4de9b07beacdf65ece9':
  fix mem leak on error handling

11 years agoam 75765d3a: am 1c45dc66: Merge "Fix AMotionEvent_getHistorical* to take const AInput...
Andrew Hsieh [Thu, 30 May 2013 21:23:56 +0000 (14:23 -0700)]
am 75765d3a: am 1c45dc66: Merge "Fix AMotionEvent_getHistorical* to take const AInputEvent* motion_event"

* commit '75765d3af58ad7ed97c0d646c936294ba861848f':
  Fix AMotionEvent_getHistorical* to take const AInputEvent* motion_event

11 years agouse gyro timestamp directly in fusion
Mathias Agopian [Thu, 30 May 2013 21:18:23 +0000 (14:18 -0700)]
use gyro timestamp directly in fusion

we used to estimate the gyro rate and deduce the period from that
but it turns out this is causing problems.

Bug: 5192288
Change-Id: I8ca826d0e11e488587bcaa1720de99e92b82f191

11 years agoBring back overloading to print String8/16.
Jeff Sharkey [Thu, 30 May 2013 20:53:39 +0000 (13:53 -0700)]
Bring back overloading to print String8/16.

Change-Id: Iecab20db2b8574d5f1e6fdefdc2c8d4ce6c37121

11 years agoam 86b2ecdd: Merge "fix mem leak on error handling"
Jeff Brown [Thu, 30 May 2013 02:01:17 +0000 (19:01 -0700)]
am 86b2ecdd: Merge "fix mem leak on error handling"

* commit '86b2ecdd311fca2e1e4a3504bf96bb23c5630492':
  fix mem leak on error handling

11 years agoMerge "fix mem leak on error handling"
Jeff Brown [Thu, 30 May 2013 01:58:45 +0000 (01:58 +0000)]
Merge "fix mem leak on error handling"

11 years agoam 1c45dc66: Merge "Fix AMotionEvent_getHistorical* to take const AInputEvent* motion...
Andrew Hsieh [Thu, 30 May 2013 01:33:33 +0000 (18:33 -0700)]
am 1c45dc66: Merge "Fix AMotionEvent_getHistorical* to take const AInputEvent* motion_event"

* commit '1c45dc66c77f0ddb246a55d780206a183740ff7f':
  Fix AMotionEvent_getHistorical* to take const AInputEvent* motion_event

11 years agoMerge "Fix AMotionEvent_getHistorical* to take const AInputEvent* motion_event"
Andrew Hsieh [Thu, 30 May 2013 01:30:48 +0000 (01:30 +0000)]
Merge "Fix AMotionEvent_getHistorical* to take const AInputEvent* motion_event"

11 years agofix mem leak on error handling
Hyejin Kim [Sat, 9 Mar 2013 02:28:54 +0000 (11:28 +0900)]
fix mem leak on error handling

Change-Id: I40d8e371c46184aa701fa676246201a9d8f6e27d

11 years agoam 4b2b0a74: am 74bd799e: Merge "Exit dumpstate on SIGPIPE to avoid cascading child...
jp abgrall [Tue, 28 May 2013 22:52:27 +0000 (15:52 -0700)]
am 4b2b0a74: am 74bd799e: Merge "Exit dumpstate on SIGPIPE to avoid cascading child crashes"

* commit '4b2b0a7494c872b7d545a9abe2d08ca0d0325073':
  Exit dumpstate on SIGPIPE to avoid cascading child crashes

11 years agoam 31ab96fc: am 135e4dbd: Merge "utils: Allow non android namespaces to use ALOGD_IF_...
Jeff Brown [Tue, 28 May 2013 22:52:26 +0000 (15:52 -0700)]
am 31ab96fc: am 135e4dbd: Merge "utils: Allow non android namespaces to use ALOGD_IF_SLOW"

* commit '31ab96fcc1810e044f8bdb35d3bfbbba2fa6bafd':
  utils: Allow non android namespaces to use ALOGD_IF_SLOW

11 years agoam 74bd799e: Merge "Exit dumpstate on SIGPIPE to avoid cascading child crashes"
jp abgrall [Tue, 28 May 2013 22:46:14 +0000 (15:46 -0700)]
am 74bd799e: Merge "Exit dumpstate on SIGPIPE to avoid cascading child crashes"

* commit '74bd799ec20d2420459fa382533b1cc39ca9a54c':
  Exit dumpstate on SIGPIPE to avoid cascading child crashes

11 years agoam 135e4dbd: Merge "utils: Allow non android namespaces to use ALOGD_IF_SLOW"
Jeff Brown [Tue, 28 May 2013 22:46:03 +0000 (15:46 -0700)]
am 135e4dbd: Merge "utils: Allow non android namespaces to use ALOGD_IF_SLOW"

* commit '135e4dbd65d49730fbd6f830de7b303e01d20d5f':
  utils: Allow non android namespaces to use ALOGD_IF_SLOW

11 years agoMerge "Exit dumpstate on SIGPIPE to avoid cascading child crashes"
jp abgrall [Tue, 28 May 2013 21:18:59 +0000 (21:18 +0000)]
Merge "Exit dumpstate on SIGPIPE to avoid cascading child crashes"

11 years agoMerge "utils: Allow non android namespaces to use ALOGD_IF_SLOW"
Jeff Brown [Tue, 28 May 2013 20:20:14 +0000 (20:20 +0000)]
Merge "utils: Allow non android namespaces to use ALOGD_IF_SLOW"

11 years agoExit dumpstate on SIGPIPE to avoid cascading child crashes
John Michelau [Mon, 6 May 2013 21:42:02 +0000 (16:42 -0500)]
Exit dumpstate on SIGPIPE to avoid cascading child crashes

When dumpstate ignores SIGPIPE it can lead to a cascade of tombstones /
coredumps since many of its children don't handle or ignore it.  It's
best to just exit dumpstate once the pipe is broken.

Change-Id: Ic0c57ecf4171f0c0a07837e51c41cb1876e1350c

11 years agoFix AMotionEvent_getHistorical* to take const AInputEvent* motion_event
Andrew Hsieh [Mon, 27 May 2013 04:26:04 +0000 (12:26 +0800)]
Fix AMotionEvent_getHistorical* to take const AInputEvent* motion_event

See b.android.com/55873

Change-Id: I92c549e1a595662d4df8fa0e47675ccb2b8814fd

11 years agoMerge "Prevent opaque windows from making framebuffer translucent"
Jesse Hall [Fri, 24 May 2013 20:54:18 +0000 (20:54 +0000)]
Merge "Prevent opaque windows from making framebuffer translucent"

11 years agoMerge "Add RenderScript category for atrace."
Tim Murray [Fri, 24 May 2013 20:36:16 +0000 (20:36 +0000)]
Merge "Add RenderScript category for atrace."

11 years agoPrevent opaque windows from making framebuffer translucent
Jesse Hall [Tue, 21 May 2013 22:36:55 +0000 (15:36 -0700)]
Prevent opaque windows from making framebuffer translucent

To keep the code readable now that we have four different texenv
configurations, this change separates the decisions about what
configuration to use from the GL calls to set up the configuration.

Bug: 8963244
Change-Id: Ia07a306a7809ba8f93493d0160ccbd509e948581

11 years agoBufferQueue: remove freeAllBuffersExceptHeadLocked()
Lajos Molnar [Mon, 6 May 2013 23:23:07 +0000 (16:23 -0700)]
BufferQueue: remove freeAllBuffersExceptHeadLocked()

Now that we are having separate buffer-instances for the buffer-
queue, we can free all buffers; we don't have to keep the head
alive.

Change-Id: I023e9161a2501d99333f8868ce438afa914ec50f
Signed-off-by: Lajos Molnar <lajos@google.com>
Related-to-bug: 7093648

11 years agoBufferQueue: track buffer-queue by instance vs. by reference
Lajos Molnar [Fri, 3 May 2013 21:50:50 +0000 (14:50 -0700)]
BufferQueue: track buffer-queue by instance vs. by reference

Instead of representing the buffer-queue as a vector of buffer
indices, represent them as a vector of BufferItems (copies).
This allows modifying the buffer slots independent of the queued
buffers.

As part of this change, BufferSlot properties that are only
been relevant in the buffer-queue have been removed.

Also, invalid scalingMode in queueBuffer now returns an error.

ConsumerBase has also changed to allow reuse of the same
buffer slots by different buffers.

Change-Id: If2a698fa142b67c69ad41b8eaca6e127eb3ef75b
Signed-off-by: Lajos Molnar <lajos@google.com>
Related-to-bug: 7093648

11 years agoAdd RenderScript category for atrace.
Tim Murray [Thu, 23 May 2013 21:39:42 +0000 (14:39 -0700)]
Add RenderScript category for atrace.

Change-Id: I069a301329f4a579c7b2716fa52c3f4a0ae0062b

11 years agoMake trace macros slightly more robust
Tim Murray [Thu, 23 May 2013 20:44:08 +0000 (13:44 -0700)]
Make trace macros slightly more robust

Change-Id: I9544eb7b27fc1a971cabadd8d5b4b4b80678febf

11 years agoPower manager now takes calling package name arg.
Dianne Hackborn [Mon, 20 May 2013 18:24:31 +0000 (11:24 -0700)]
Power manager now takes calling package name arg.

Change-Id: Iae57b8eb7abe4e66f0d2645db8cdf57acf7dcd3e

11 years agoam f0e52b64: (-s ours) Reconcile with jb-mr2-release - do not merge
The Android Open Source Project [Mon, 20 May 2013 15:22:44 +0000 (08:22 -0700)]
am f0e52b64: (-s ours) Reconcile with jb-mr2-release - do not merge

* commit 'f0e52b648b435634d8cbcf52cac86c8dac9e4a63':

11 years agoReconcile with jb-mr2-release - do not merge
The Android Open Source Project [Mon, 20 May 2013 15:20:26 +0000 (08:20 -0700)]
Reconcile with jb-mr2-release - do not merge

Change-Id: I400727c668eff57b5760595787d21dd021464587

11 years agomerge in jb-mr2-release history after reset to jb-mr2-dev
The Android Automerger [Sun, 19 May 2013 12:55:29 +0000 (05:55 -0700)]
merge in jb-mr2-release history after reset to jb-mr2-dev

11 years agoMerge "wifi-display: add support for metadata mode on encoder output"
Chong Zhang [Sat, 18 May 2013 00:02:10 +0000 (00:02 +0000)]
Merge "wifi-display: add support for metadata mode on encoder output"

11 years agoMerge "Remove unnecessary header file."
Nick Kralevich [Fri, 17 May 2013 21:58:50 +0000 (21:58 +0000)]
Merge "Remove unnecessary header file."

11 years agoam 2cad801d: (-s ours) am 23dacde4: [DO NOT MERGE] fix a bug where surfaceflinger...
Mathias Agopian [Fri, 17 May 2013 20:55:28 +0000 (13:55 -0700)]
am 2cad801d: (-s ours) am 23dacde4: [DO NOT MERGE] fix a bug where surfaceflinger and system_server could deadlock

* commit '2cad801d0d05190cf4ab7edc7ecffc5862ee18dd':
  [DO NOT MERGE] fix a bug where surfaceflinger and system_server could deadlock

11 years agoam 23dacde4: [DO NOT MERGE] fix a bug where surfaceflinger and system_server could...
Mathias Agopian [Fri, 17 May 2013 20:53:31 +0000 (13:53 -0700)]
am 23dacde4: [DO NOT MERGE] fix a bug where surfaceflinger and system_server could deadlock

* commit '23dacde4b624921df4f2c9646d20ca7acae357b6':
  [DO NOT MERGE] fix a bug where surfaceflinger and system_server could deadlock

11 years ago[DO NOT MERGE] fix a bug where surfaceflinger and system_server could deadlock
Mathias Agopian [Thu, 16 May 2013 02:14:52 +0000 (19:14 -0700)]
[DO NOT MERGE] fix a bug where surfaceflinger and system_server could deadlock

because surfaceflinger handles screenshot in a different
thread from the binder thread that requested it and because
the IGraphicBufferProducer is a synchronous interface
calling back into the system server; it is possible for
the latter to run out of binder threads (b/c it holds
a lock while calling into SF).

The solution is to make sure all calls on IGraphicBufferProducer
happen on the incoming binder thread. We achieve this by creating
a IGBP wrapper which is given to the screenshot code.

Bug: 8734824
Change-Id: Ife2441c7322e51ecfb20e0df03dacf6bce49578e

11 years agoRemove unnecessary header file.
Nick Kralevich [Fri, 17 May 2013 18:06:39 +0000 (11:06 -0700)]
Remove unnecessary header file.

In b2cac4ff37ed2a3521efb307b807b12f649634b8, the last
reference to property_get was removed, but the include
file remained. Remove it.

This is causing me unrelated problems.

Change-Id: I87bbfbfa5871f6da4a1b00037a225f67bbd75881

11 years agoam e6eef6c1: am ea74d3b7: make the warning timout of Fence::waitForever() implicit...
Mathias Agopian [Fri, 17 May 2013 17:18:45 +0000 (10:18 -0700)]
am e6eef6c1: am ea74d3b7: make the warning timout of Fence::waitForever() implicit and longer

* commit 'e6eef6c136b007f3a7e1e19884875b855c0c5f95':
  make the warning timout of Fence::waitForever() implicit and longer

11 years agoam ea74d3b7: make the warning timout of Fence::waitForever() implicit and longer
Mathias Agopian [Fri, 17 May 2013 17:16:51 +0000 (10:16 -0700)]
am ea74d3b7: make the warning timout of Fence::waitForever() implicit and longer

* commit 'ea74d3b78d607cde17790a7bb83e6f68ffd34cfd':
  make the warning timout of Fence::waitForever() implicit and longer

11 years agoMerge "various fixes to the sensorservice"
Mathias Agopian [Fri, 17 May 2013 01:49:42 +0000 (01:49 +0000)]
Merge "various fixes to the sensorservice"

11 years agovarious fixes to the sensorservice
Mathias Agopian [Sat, 11 May 2013 02:32:39 +0000 (19:32 -0700)]
various fixes to the sensorservice

1) "google" sensors are now reporting AOSP as the vendor string
2) don't expose the system's sensor fusion if the HAL provides it
3) use uncalibrated gyro if availble for the system's sensor fusion

Change-Id: I25140436cdb29d55e39fd6fbbf8c44a410a83d5c

11 years agomake the warning timout of Fence::waitForever() implicit and longer
Mathias Agopian [Fri, 17 May 2013 01:03:22 +0000 (18:03 -0700)]
make the warning timout of Fence::waitForever() implicit and longer

- timeout is now 3 seconds instead of 1
- simplifies the API a bit
- allows us to change/tweak this timeout globaly

Bug: 8988871

Change-Id: I8d3c6ec43a372f602fb3f29856710339f86c0ec9

11 years agowifi-display: add support for metadata mode on encoder output
Chong Zhang [Thu, 16 May 2013 20:11:27 +0000 (13:11 -0700)]
wifi-display: add support for metadata mode on encoder output

Bug: 8968123

Change-Id: I3d2b567d4908f5f49c0694589920babe1f6752cd

11 years agoam 92cf3553: (-s ours) Reconcile with jb-mr2-release - do not merge
The Android Open Source Project [Thu, 16 May 2013 12:48:22 +0000 (05:48 -0700)]
am 92cf3553: (-s ours) Reconcile with jb-mr2-release - do not merge

* commit '92cf355308ffb28f8a460d69915ca5955016a632':

11 years agoReconcile with jb-mr2-release - do not merge
The Android Open Source Project [Thu, 16 May 2013 12:43:41 +0000 (05:43 -0700)]
Reconcile with jb-mr2-release - do not merge

Change-Id: Ic457c583534afce9b1697b3b34221a729e2bc5b2

11 years agoam 28fc78bf: am e8fed71d: Merge "Abort surface composition if hw surface is invalid"
Jesse Hall [Tue, 14 May 2013 06:31:54 +0000 (23:31 -0700)]
am 28fc78bf: am e8fed71d: Merge "Abort surface composition if hw surface is invalid"

* commit '28fc78bf5fbb6288a27f3f25565a960a1873b0ef':
  Abort surface composition if hw surface is invalid

11 years agoam e8fed71d: Merge "Abort surface composition if hw surface is invalid"
Jesse Hall [Tue, 14 May 2013 06:28:45 +0000 (23:28 -0700)]
am e8fed71d: Merge "Abort surface composition if hw surface is invalid"

* commit 'e8fed71da2d3e03ef933d308520bab5b860f05aa':
  Abort surface composition if hw surface is invalid

11 years agoMerge "Abort surface composition if hw surface is invalid"
Jesse Hall [Tue, 14 May 2013 00:24:28 +0000 (00:24 +0000)]
Merge "Abort surface composition if hw surface is invalid"

11 years agoAbort surface composition if hw surface is invalid
Michael Chock [Mon, 4 Mar 2013 23:15:46 +0000 (15:15 -0800)]
Abort surface composition if hw surface is invalid

If an invalid display device is detected during surface composition
(e.g., a simulated secondary display is removed), abort the
composition.

Change-Id: Ia6afb2e287882d8ae0614eb25463d3f85b687adf

11 years agoam 7a486528: am f0dcff89: Merge "gl_code.cpp: Remove a DOS style newline"
Jesse Hall [Mon, 13 May 2013 20:08:54 +0000 (13:08 -0700)]
am 7a486528: am f0dcff89: Merge "gl_code.cpp: Remove a DOS style newline"

* commit '7a4865287460a5f0ff26e22be5f0271735d32ac7':
  gl_code.cpp: Remove a DOS style newline

11 years agoam f0dcff89: Merge "gl_code.cpp: Remove a DOS style newline"
Jesse Hall [Mon, 13 May 2013 20:05:36 +0000 (13:05 -0700)]
am f0dcff89: Merge "gl_code.cpp: Remove a DOS style newline"

* commit 'f0dcff89acb998179757ca66b5c6a2c585a980e9':
  gl_code.cpp: Remove a DOS style newline

11 years agoMerge "gl_code.cpp: Remove a DOS style newline"
Jesse Hall [Mon, 13 May 2013 18:53:10 +0000 (18:53 +0000)]
Merge "gl_code.cpp: Remove a DOS style newline"

11 years agomerge in jb-mr2-release history after reset to jb-mr2-dev
The Android Automerger [Mon, 13 May 2013 12:55:40 +0000 (05:55 -0700)]
merge in jb-mr2-release history after reset to jb-mr2-dev

11 years agotake the "transparent region" into account for blending
Mathias Agopian [Sat, 11 May 2013 01:01:12 +0000 (18:01 -0700)]
take the "transparent region" into account for blending

until now it was only used to discard a layer entirely.
we're now reducing the size of the layer if it is still
visible, if possible.

this works for instance when a surfaceView is used and
only the menu bar is displayed over it.

Change-Id: I3f5527c5cd1e69ecc968272c8948f1513ada8c55