| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
| |
* Revert change that broke HW keyboards
* Add workaround var for devices with broken kernels
Change-Id: I5bcda9c0464f75cc080c224382eaecbf48613bf0
|
|\ |
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This reverts commit 5a63233c9ca4f7aa2e57c776fb275e5c3d78d2ff
and b1b664a242ea311c7a96d7e614cc95799aa0b8b3
Patch Set 2: Do not change default LIGHT_DECREASE setting
Change-Id: I17616d17c86a2954d5c530c356b0541af5664ad5
|
| |
| |
| |
| |
| |
| | |
rotation
Change-Id: I42694931cb1a754aa6c7652c0a96304ef1f88492
|
|/
|
|
|
| |
Change-Id: I533cadf0363a9de278730c29c97e342e26bf8726
Signed-off-by: Cuong <cuongbui@gmail.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
in addition to converting ABS_MT_TOUCH_MAJOR to ABS_MT_PRESSURE, it also converts ABS_MT_WIDTH_MAJOR to ABS_MT_TOUCH_MAJOR.
also add 2 workarounds for some touchscreen driver which
1. has BTN_TOUCH feature but it doesn't send BTN_TOUCH event
2. sends multi-touch event for not-in-use pointer
reference: http://source.android.com/tech/input/touch-devices.html
Change-Id: I7b7f8802e45bf9d9f8de00373559a30ea839a178
|
|
|
|
|
|
| |
Needed on (at least) crespo, aries and galaxys2
Change-Id: Iaff5379775b9a19bd3cd8598e885e885cc0e8492
|
|\ |
|
| |
| |
| |
| |
| |
| |
| |
| | |
by commit 4aed78b5056560f499e5953f659fa90a06ecc38a lidOpen was reverted.
fix to make lidOpen true when sw == 0.
and fix getStateLocked()/getState() to get UP/DOWN state correctly.
Change-Id: I22a185478be860f1c88cf56d7ef3d0b3a8617459
|
|/
|
|
|
|
|
|
|
| |
Whenever we are allocating a new buffer, we delete the reference to
the previous buffer at that buffer slot. The newly allocated Graphic
Buffer is also added at the correct buffer slot in the
GraphicBufferAllocator.
Change-Id: Ie91e61e39b18759893c773ebda9ce7d7654d4144
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
- Don't read or set "persist.service.adb.enable", it is deprecated
and is read and cleared by UsbDeviceManager every boot. Trying
to read it will always yield "" because it was already cleared
by UsbDeviceManager (thus disabling adb debug every boot). Trying
to set it doesn't have any effect because adb is now started by
"sys.usb.config".
- In UsbDeviceManager, don't call AdbSettingsObserver when enabling
or disabling adb over network because that function is only for
enabling/disabling adb.
Change-Id: Ib6c1a655277c7f6d7395595ccfe3bac29c0582fe
|
|\
| |
| |
| |
| |
| | |
* changes:
Hold a reference to the allocated buffers in GraphicBufferAlloc
opengl: remove unused adreno flag from the makefile
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Store the allocated GraphicBuffers in a vector list. This list is
cleared only when SurfaceTextures requests that the buffers be freed.
Premature deletion of GraphicBuffers and therefore deallocation of
gralloc handles are prevented.
This change partially reverts:
commit 5f05f99aaedaba18c426fac287bcb18d56dbe881
Author: Mathias Agopian <mathias@google.com>
Date: Fri Apr 8 19:10:43 2011 -0700
Fix a GraphicBuffer leak in SurfaceTexture
This leak was intentional, it was there to deal with the fact that
some gralloc implementations don't track buffer handles with
file-descriptors so buffers needed to stay alive until there were
registered, which is not guaranteed by binder transactions.
Change-Id: Ia732806bf8d8d15b66944569e8aa2120641d9812
CRs-fixed: 327655
|
|\ \
| | |
| | |
| | | |
and undocked modes. It does not make sence to not allow screen to dim in either instance. Better options are welcome." into ics
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
and undocked modes. It does not make sence to not allow screen to dim
in either instance.
Better options are welcome.
Change-Id: I501ec96fce64b9629a198ab713858eabee602600
|
|\ \ \
| |_|/
|/| |
| | | |
into ics
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
The feature can be used/tested without the Settings part :
setprop service.adb.tcp.port 5555
Note: This ADB setting is not persistent (for security purpose) and
require init.rc implementation event like this : http://bit.ly/AdbTcpIP
Change-Id: Idc1bbbc0301e575870d10f47cb27dedcb4f16dba
|
|\ \ \
| | |/
| |/|
| | | |
to decrease when changing to dimmer location when phone is not connected to a dock. Previouos setting had reversed." into ics
|
| |/
| |
| |
| |
| |
| |
| | |
This allows for backlight to decrease when changing to dimmer location
when phone is not connected to a dock. Previouos setting had reversed.
Change-Id: I3f9e44dec8e3deb9ecb2aeb0f775383bbd7b3d57
|
|/
|
|
| |
Change-Id: Id14f70f7e21b689c08cadef69f69da711c7dc566
|
|\
| |
| |
| | |
(lost in merge)" into ics
|
| |
| |
| |
| |
| |
| |
| | |
Same as I4b501139b660255fc935d6a12fd6ecf44d06ec6e, which was lost
in the merge
Change-Id: I0208731a35b81e2df8f6c45615fe16caed66b558
|
|\ \ |
|
| | |
| | |
| | |
| | |
| | |
| | | |
Patchset 2: Clean code
Change-Id: Ib1fc59833d066dee9c4c5f02a9bf1a912463cb1e
|
|\ \ \ |
|
| |/ /
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
(Broke in I9c88704484f4a23b3d643597f35b55d74cf36b56)
- Do not mHandler.removeCallbacks(mAutoBrightnessTask) every time
a value is received from the sensor because that will ensure that
the task ever get ran! (We schedule it 2 seconds into the future
but receives a value from the sensor every second.)
- Replace the algorithm for brightness change (non-filtering) with
the original algorithm before the patch. (Not sure why the patch
changed it.)
Change-Id: I71f6f23c53974fbb763d46ea4023623d2bbeb250
|
|\ \ \
| |/ /
|/| | |
|
| | |
| | |
| | |
| | |
| | |
| | | |
Rebased for ICS 4.0.3
Change-Id: I029d8eeca4de6d92499856feab2761a8bfa9a526
|
| | |
| | |
| | |
| | | |
Change-Id: I7be5345707e00b20ad96ebd3958bbf7f9fb5a22f
|
|\ \ \ |
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
This is required in order to make the trackball on devices that has it to
work properly, like it does before ICS. In addition to this patch, you just
need an .idc file that matches the device name for your the trackball. You can
find it by either looking in a kernel log or a logcat.
For example, a device with a Crucial optical navigation pad, will be called
curcial-oj.idc
The .idc file should contain the following:
cursor.mode = navigation
cursor.orientationAware = 1
Place the file in /system/usr/idc folder.
Enable in BoardConfig.mk with BOARD_USE_LEGACY_TRACKPAD := true
Change-Id: I57e9912b9e4d6242c980c99f4fd28b0d50f19bb8
|
|\ \ \ \
| |/ / /
|/| | | |
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Call libQcomUI's implementation of clearRegion instead of glClear. This
implementation is specifically designed to optimize the drawWormhole for non GPU
composition cases, in order to improve performace. If the qcom version
of this API returns an error, we fallback to SurfaceFlinger's implementation
of drawWormhole. The clearRegion API is only called if there are no layers
being drawn by the GPU.
Change-Id: I1a50cf5ba2a88ffcadf13ff78df16f5a30dfff3a
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
layer count
This change considers both gpu and c2d compositions while calculating FB layer count
Change-Id: I57fe19650030bb49f8bf1afe88fe26c6731b91da
|
| | | |
| | | |
| | | |
| | | | |
Change-Id: I7ea6fc824b8038a2b84e53cc8057b17cd6bcd074
|
| | | |
| | | |
| | | |
| | | | |
Change-Id: Ieaa1db0fe80b8c90cdcb24c4b415199232c06d67
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
set string com.qc.hdmi_out=true to enable java hdmi service. set false by default
set TARGET_QCOM_HDMI_OUT true in BoardConfig to enable qcom hdmi bits
Change-Id: I9286bd01085657251d83b4649d0ef86433947765
|
| | | |
| | | |
| | | |
| | | | |
Change-Id: I339eed86d842621013be1d3672552bdec5a1b167
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
If the HWC sets the HWC_SKIP_COMPOSITION, SurfaceFlinger should skip
the composition step.
Change-Id: I8d9f6bc61b3788c73d4ea950f1e9a4d78277ae9f
|
| | | |
| | | |
| | | |
| | | | |
Change-Id: Iacd8de7a9395ee3f68e2a5677f254920f2edfc2f
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
This change contains following commits.
commit 27889d5e0c7951d9cc454c0cdc84bb6c4e4d93b4
Author: Arun Kumar K.R <c_akkr@quicinc.com>
Date: Wed Jun 15 15:12:33 2011 -0700
frameworks/base: Remove unnecessary ioctls from HDMI Daemon
- Remove FBIO_BLANK_POWERDOWN as the device will be powered down when the
ref count reaches zero in framebuffer
- Remove MSMFB_OVERLAY_PLAY_ENABLE as it will be depricated
(cherry picked from commit 9dfcc8c1c5b6a4695ef807d11b67cae25ff8fcfd)
Change-Id: I28cbf52edd3c244f1f04be3df5bc4c5a091fde13
commit 98774ca7b39995a169022289274cdc4e99dc4848
Author: Arun Kumar K.R <c_akkr@quicinc.com>
Date: Fri May 27 14:12:18 2011 -0700
frameworks/base: Set hdmiON property before informing SF
In the HDMI Daemon mirroring is enabled set the hdmiON system
property before informing surface flinger to start mirroring.
Change-Id: Ib897cfe9f9e113553cb0272e3d0fc04437b3577b
commit d6c4ac202205cfdbacd49951bbdfe42cb9ced971
Author: Arun Kumar K.R <c_akkr@quicinc.com>
Date: Mon May 9 20:15:44 2011 -0700
frameworks/base: Changes for HDMI HDCP Compliance
- For HDCP compliance, broadcast HDMIONEvent after we receive
the audio_on event from the daemon based on HDCP
- In HDMI Daemon check if the HDCP keys are present based on that
broadcast for Audio.
- Remove redundant calls for PUT_VSCREENINFO and handle this in
setResolution.
Change-Id: I7fa23ac1a384fcc32687b287dbb828445d4d81a4
commit 798421991797c249e9eb554f91d49e0d898327fa
Author: Arun Kumar K.R <c_akkr@quicinc.com>
Date: Thu Apr 7 18:55:44 2011 -0700
HDMIDaemon: call FBIO_PUTVSCREENINFO ioctl for valid mode
Analog TV does not have the EDID information.
For analog TV no need to call the ioctl, as it defaults to
its original resolution
Change-Id: I18daab570167004be76d47cd58413408a20b0987
CRs-fixed: 281752
commit c512a0c47116394de431b6957213ac2cabcfc091
Author: Arun Kumar K.R <c_akkr@quicinc.com>
Date: Wed Mar 16 13:12:53 2011 -0700
Hdmi Daemon: Inform SF about HDMI mirroring when HDMI is enabled
When HDMI is connected, inform SF about mirroring only once.
Avoid multiple enableHDMI calls to SF when cable is connected.
Change-Id: I00d72571d2796e204401f57a87d9e580a3c8e994
CRs-fixed: 279095
commit 92de304205eb65e5bfd6976eef6396e908df4829
Author: Kinjal Bhavsar <kbhavsar@codeaurora.org>
Date: Mon Mar 28 17:43:22 2011 -0700
frameworks/base: Use sanitized kernel headers
Add the exported sanitized kernel headers to include path in hdmid
and surfaceflinger.
As hdmid uses some bionic specific structs, explicitly add path of
kernel header files it uses.
Change-Id: Id0daed4a425976fad3116de2d2054c9c386770fe
commit bd484e4de3031b393e2b4e1c91a17c2b9bc4acac
Author: Arun Kumar K.R <c_akkr@quicinc.com>
Date: Mon Mar 7 19:37:56 2011 -0800
frameworks/base: HPD enable on sysfs
Fixes the issue where HDMI user setting is ON, Cable is
connected and on reboot, we don't see HDMI output.
Enables/Disables HPD by writing to sysfs node from Daemon
Removes mirroring calls from HDMIService.
Change-Id: Ide8ac1d60d9b04be2dcb79323f42750ef8f045c3
commit d2076706b4a1636c3e67131498824048b43046cb
Author: Saurabh Shah <saurshah@codeaurora.org>
Date: Thu Oct 28 10:26:44 2010 -0700
frameworks/base/:Interface for action-safe width, height
Interface and implementation for setting action-safe
width and height from userspace.
commit 5446eab60c8c88417375607dea5ccc246cdd9bd0
Author: Arun Kumar K.R <akkr@codeaurora.org>
Date: Mon Nov 8 18:29:49 2010 -0800
frameworks/base: Enable/Disable HDMI UI mirroring based on HPD
commit 24d813ec5e79ae92e7efc2f4286bb4425dec1053
Author: Arun Kumar K.R <akkr@codeaurora.org>
Date: Thu Sep 23 17:06:26 2010 -0700
frameworks\base: Support for switching resolution in HDMI daemon
Use property enable.hdmi.edid to enable the dynamic resolution
switch Set hdmi.yRes to required resolution.
commit 60208c205645f7da7d148655d638d36bc7cca121
Author: Arun Kumar K.R <akkr@codeaurora.org>
Date: Tue Aug 31 19:21:53 2010 -0700
frameworks/base: SF to control HDMI output
Change SF to control HDMI output.
HDMI Daemon will call SF to control HDMI out.
HPD based on fbopen/fbclose, which is controlled by User settings.
commit b180d502bc53bb55bc168048d7b691f33899e4db
Author: Urs Muff <umuff@codeaurora.org>
Date: Wed Aug 4 14:14:00 2010 -0600
frameworks/base: hdmid: HPD / EDID integration
Adding support for dynamic resolution switching and HDP and EDID
commit a192fa8f272c113996a65bca4a68fa328f2fb4f4
Author: Omprakash Dhyade <odhyade@codeaurora.org>
Date: Wed Aug 18 13:39:49 2010 -0700
frameworks/base: Call open only user says HDMI ON
HDMI Clock remains ON when fb1 is opened from HDMI Daemon
so call open fb1 only when user sets HDMI ON.
commit db05d0df83e568e31b273e1f667539a02156fcbb
Author: Omprakash Dhyade <odhyade@codeaurora.org>
Date: Thu Jul 15 18:18:28 2010 -0700
frameworks/base: Add HPD support
Add HPD support to HDMI daemon.
commit 25f7beb6a82aff11f0cc2ed9850e40278820c32c
Author: Omprakash Dhyade <odhyade@codeaurora.org>
Date: Tue Jun 1 16:56:22 2010 -0700
frameworks/base: Add HDMIService for 7x30 target
Add HDMI deamon which communicates with HDMI driver.
Add HDMI service to communicate with HDMI deamon
and send intent to applications when HDMI connected
and disconnected.
Change-Id: I416ccb87dd356a08299c3180275aa53a4fe7e412
Conflicts:
libs/surfaceflinger_client/ISurfaceComposer.cpp
Typo fix
Change-Id: I70e22fd74eab7b9ee2d3f33e97b8256c03cce2c1
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Add support for drawing buffers which are not supported by the GPU.
These buffers are drawn using the overlay or C2D. During rotation,
when the draw operation is performed, the GPU clears the clip region
for this buffer.
Change-Id: I8738dcffbfb31db330872ec2c295181b540e03fb
|
| | |/
| |/|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
C2D composition can handle fade cases. Don't skip layers in fade
case if C2D composition is used. Send the alpha value of the layer
to the HWC
Change-Id: I2751bcb175679b063617102cecda874bc94ad5cf
CRs-fixed: 288257
|
| |/
|/| |
|
|/
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Moving average filter for light sensor, support for custom levels, battery stats fix and dim fix.
Optimization: Don't compute the sum of all samples in filter window every run. Instead, keep track of the sum, subtract phased out sample, add new sample and divide new sum by number of valid samples. Massive CPU cycle savings.
Thanks to Jonathan (zenpoy on gmail dot com) for pointing me in the right direction!
Fix: The code for stopping the filter if the sensor has been stable the entire filter window time was only enabled if debug logging was enabled. Oops. CPU cycle savings in steady ambient light.
Tweak light sensor filter
1. Only allow decrease if user permitted it
2. Only perform filter reset on sensor increase and current average is below 1500 ("indoors")
3. Filter reset in 2. means pushing the filter half way towards new sample
4. Filter reset is disabled by default
This change requires commits from CMParts:
5f4026702b67b5d23c9a992f408e3861c90566db
f9238a01e4a4a64cb414dbdada36c97ae2a37db9
0a4b490035c333b3d77e6e59fcc0f7d16d78ebda
5036cc8217daf59aa9f2800bc3fa40e4dd4b3117
b43caaab217848668241681613ce8d54ceda69e9
cc0682a4cc43d0058db8b5195d4d329336f2d3fd
Change-Id: I9c88704484f4a23b3d643597f35b55d74cf36b56
|
|\
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
https://android.googlesource.com/platform/frameworks/base into aosp
Conflicts:
core/res/res/values/strings.xml
policy/src/com/android/internal/policy/impl/GlobalActions.java
services/java/com/android/server/WifiService.java
services/sensorservice/SensorDevice.cpp
telephony/java/com/android/internal/telephony/RIL.java
Change-Id: I89ec56d5ff282a2eb879ca40fb6d74ebcf752837
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
We need to work more like before in determining whether the menu
key is needed -- in some cases look back in the window list to
determine this if we don't know the value from the current window.
This requires adding a new private flag indicating whether the
compat menu state is known for a window, which is set by
PhoneWindow as part of its existing process of computing the flag
for its own windows.
Now we can have a new API on WindowState to determine the value
of this flag for a window, which if needed walks back in the window list
to find a window the value is known for (or stops at what the policy
has determined is the top full-screen window, so we stop like we used
to at things like the lock screen or the bottom of an application).
Change-Id: I829de6d629b5af8bcb422cb85249ee4041c7205e
|
| |
| |
| |
| |
| | |
Bug: 5750339
Change-Id: I4b739dbd076b270a5a2b2e2a1e0f6e445d8c124e
|
| |\ |
|
| | |
| | |
| | |
| | |
| | | |
Bug: 5714296
Change-Id: Idba41b9f3907747a84387ea4f3e930b3a185a06e
|