Overdrive enablement bit sits at 6th bit of control register 0.
Current shifting overwrites overdrive high bit.
Change-Id: I4cdd435da86a234ccf0465462be7626162cc8a9d
Signed-off-by: Chun Zhang <chunz@codeaurora.org>
Provide a callback machanism to all the modules registered
for HDMI cable notifications. The registered modules will be
notified everytime the cable is connected or disconnected.
The callback will be scheduled on global kernel workqueue.
Change-Id: I74630748448d8479b6e0cda6ccd148cbeae48a51
Signed-off-by: Ajay Singh Parmar <aparmar@codeaurora.org>
ssd2080m requires panel reset after LP11 timing starts,
add extra 50ms delay after reset to ensure the safety.
Change-Id: I5278443a5555dcc59498a52984c9523134b35d50
Signed-off-by: Ray Zhang <rayz@codeaurora.org>
Driver needs dummy functions defined for cases when CONFIG_PM
is not enabled.
Change-Id: I4e9f0885eb9f7a45589b2362e468335a3ca57cba
Signed-off-by: Amy Maloche <amaloche@codeaurora.org>
Commit "6e3630b5f28c60c403af8c6b3c87679d3d98d6af: usb: dwc3-msm: Support
external charging detection " has added check not to start charger
detection in case charger detection is done by external circuit. when otg
driver calls charger_detection API to set charger type to invalid as part
of VBUS disconnect handling, this check prevents to do that. This results
next charger detection not happening after DCP charger disconnection.
Hence fix the issue by moving down the check in the charger_detection API.
Change-Id: I5c614a208f22e4cd9f334c80dae6831bcff64ada
Signed-off-by: Vijayavardhan Vennapusa <vvreddy@codeaurora.org>
Refactor PM callbacks and remove redundant variable.
Add support for configuration of clockcontrol event.
Now the clock's state may be set by pm calls (as before)
or by a transfer's begining and ending.
Runtime-PM is still used in autosuspend configuration for
expensive NoC voting/unvoting since that may go to remote
processor.
CRs-Fixed: 592435
Change-Id: I54f7c67f768336f66765a04006b4c8e4734ac277
Signed-off-by: Gilad Avidov <gavidov@codeaurora.org>
Signed-off-by: Sagar Dharia <sdharia@codeaurora.org>
Signed-off-by: Sana Venkat Raju <c_vsana@codeaurora.org>
Workqueue might have pending connect/disconnect works when
composite driver calls qdss unbind as part of composition
switch or remove. As qdss_unbind frees up device references
e.g. ep->driver_data, flush pending works from unbind to avoid
any access to them later from work_fn.
Change-Id: I2ea3988ccebb6fc34772b30965ab8a1cb74d5464
Signed-off-by: Manu Gautam <mgautam@codeaurora.org>
NUMTOUCH field in T9 object of Atmel's config data is changed
to 0x05 from 0x0A. Config version information is also updated to
29.5.1 from 29.4.0
CRs-Fixed: 598507
Change-Id: I561198c502dcd22f9157d61b158176e0ce42a24c
Signed-off-by: Himanshu Aggarwal <haggarwa@codeaurora.org>
The side buttons bitmap was expanded to allow sending button
events for tool type buttons (stylus, rubber) and also for a new
button which signals that the hovering icon should be displayed.
The event source field was removed as it was previously used to
set the tool type, which is now done explicitly via the buttons
bitmap.
Change-Id: Id08921c392eeec8928ca0c730988f8c83be2ba8f
Signed-off-by: Lior Barenboim <liorb@codeaurora.org>
Allow mma8x5x accelerometer work on interrupt mode and issue
interrupt to wake up application processor when data is ready.
Change-Id: Iadecfa9aeaa8e225098d049659802dcd7f976f48
Signed-off-by: Bingzhe Cai <bingzhec@codeaurora.org>
When the charger is in the chg_pause state, the ibat regulation loop can
cause the batfet to oscillate due to VCP. This occurs because of an
hardware issue with the IBAT_MAX stepper when the charger is paused.
When charging is paused, the buck accuracy degrades and can cause VPH to
dip below vbat, triggering VCP. The batfet toggling causes faulty IBAT
readings by the BMS.
Fix this by disabling the ibat regulation loop when the battery goes
into the hot or cold state. This will fix the batfet toggling issue as
well as the faulty ibat readings.
Change-Id: I74bab3e6fd0c1be0e83100672262fc504eee3396
Signed-off-by: Xiaozhe Shi <xiaozhes@codeaurora.org>
Previously, the BMS was directing the IADC driver to calibrate for trim
every SoC cycle assuming that the batfet is open. This is incorrect. The
trim calibration should only happen when the batfet is closed. Fix this
by calling the calibration routing with the batfet closed option set.
Additionally, the batfet status was not being updated if the IADC
revision was old. This is not correct. Fix this by moving the revision
check so that the batfet status is always saved but does not apply the
work around if the revision is old.
Change-Id: I4ac9ff5f7baa5f1be23c98fd86d2d34ba743c004
Signed-off-by: Xiaozhe Shi <xiaozhes@codeaurora.org>
The kickoff code relies on the fact that write back arguments are zero
initialized to check for available buffer. If it's not zero initialized
it may pick dirty buffer address from stack.
Change-Id: I7309f8fe508640833a182a8ecdb2d3be1797fab7
Signed-off-by: Adrian Salido-Moreno <adrianm@codeaurora.org>
When a buffer has been properly processed, client may choose to use sync
fence to acknowledge when buffer is complete. In such cases dequeue is
not necessary, allow this use case by allowing client to re-queue
writeback buffers as long as the buffer has been properly processed
already.
Change-Id: Ibac1eb9710477734e1ee2daf8136085d4439b0d4
Signed-off-by: Adrian Salido-Moreno <adrianm@codeaurora.org>
Provide an API for drivers to read from miscellaneous registers
in the MSM TLMM region.
Change-Id: I100ab4ed4643e234073fc4c67a254052480dfc7d
Signed-off-by: Rohit Vaswani <rvaswani@codeaurora.org>
