This change is added to keep sensor code alinged for porting.

Change-Id: I2f671f26a16883acfc463cd02663b6c5d8ec790f
Signed-off-by: Raja Mallik <rmallik@codeaurora.org>

Add cam-res-mgr entry in cci node for multiple
flash handle

Change-Id: Ic77e95561a3f2e4a0602051b6e3047a1bf4580d7
Signed-off-by: Shankar Ravi <rshankar@codeaurora.org>

Migrate the flash changes from mainline.

Change-Id: Ia8401e2d3e658666916beb6cbccdb76aaf8ccb3e
Signed-off-by: Shankar Ravi <rshankar@codeaurora.org>
Signed-off-by: Raja Mallik <rmallik@codeaurora.org>

Changes added for camera driver from msm-4.14 'commit <efd1feba>
("msm: camera: ife: clean up HW manager sequence")' till 'commit
<f4ae227b> ("msm: camera: core: Avoid concurrency between
stop and pfdump")'

  commit 'f4ae227b':
  msm: camera: core: Avoid concurrency between stop and pfdump
  msm: camera: isp: Switching between rdi and pix rup for fs2 state machine
  msm: camera: icp: Increase number of supported contexts
  msm: camera: isp: Dump registers based on debugfs
  msm: camera: Listen to error interrupts once CSID is enabled
  msm: camera: cci: Add serialization of requests from diff i2c freq modes
  msm: camera: sync: Update error message in sync driver
  msm: camera: reqmgr: Move trace event post applying
  msm: camera: reqmgr: Correct the check for lower pd devices
  msm: camera: reqmgr: Reset sync link initial sync req
  msm: camera: icp: Add fw error message in kmd logging
  msm: camera: sync: Report error if getting sync obj reference fails
  msm: camera: icp: Update frame process command
  msm: camera: sensor: Add csi phy reset sequence
  msm: camera: core: Update the log level

Retain existing ION, CMA alloc, SMMU, for camera modules for msm-4.9.

Change-Id: I5fc3c798372ac824ded2c093f7da59da5816d015
Signed-off-by: Raja Mallik <rmallik@codeaurora.org>

msm: cam: isp: Add state machine for FS2
msm: camera: isp: Improve logging by adding context info
msm: camera: icp: Reduce delay between subsequent HFI polls
msm: camera: cdm: Avoid presence of stale tags in BL_FIFO
msm: camera: lrme: correct fetch engine width size for P010 format
msm: camera: fd: Add mutex protection to frame done processing
msm: camera: crm: Fix use-after-free for link
msm: camera: reqmgr: Modify frame sync logic
msm: camera: reqmgr: Fix frame sync issue on SAT mode
msm: camera: reqmgr: Validate num links in link control
msm: camera: reqmgr: Validate num links in link control
msm: camera: reqmgr: Fix frame skip condition
msm: camera: reqmgr: Deliver matching frames for master-slave sync
msm: camera: reject request id earlier than last flush request
msm: camera: core: Cleanup the logic for shutdown
msm: camera: Add stride info as part of pf dump
msm: camera: Enable fetch engine to support FS2 feature for IFE sdm845
msm: camera: isp: Change epoch to 2/3rd frame
msm: camera: memmgr: Allow alloc flags with 0

Changes added from camera from msm-4.14 'commit <a50bf345>
("msm: camera: flash: Moving flash off operation in stop dev ioctl")'
till 'commit <efd1feba> ("msm: camera: ife: clean up HW manager
sequence")'

Retain existing ION, CMA alloc, SMMU, for camera modules for msm-4.9.
Retain cpas hw tcsr specific changes for sdm670.

Change-Id: I053a47b2662fb2b545f694633afac784f641d14f
Signed-off-by: Suresh Vankadara <svankada@codeaurora.org>
Signed-off-by: Alok Pandey <akumarpa@codeaurora.org>
Signed-off-by: Raja Mallik <rmallik@codeaurora.org>

Change-Id: Id11b8b13a1eaf4965aaa3fe3b376139690bccd86

When there is a SYS_ERROR the instance object is getting
freed even before notifying the client causing the kernel
NULL pointer dereferece. (When the max number of instances
reached latest instance will be freed)

Change-Id: I5c93401d650228755516e362ceb8c16a356f2653
Signed-off-by: Mahesh Voorugonda <mvooru@codeaurora.org>

Adding code changes to validate buffer size.
While calling ipa_read verifying the kernel buffer
size in range or not.

Change-Id: Idc608c2cf0587a00f19ece38a4eb646f7fde68e3
Signed-off-by: Praveen Kurapati <pkurapat@codeaurora.org>

Changes added for camera ISP, ICP, JPEG, fd, cpas, cdm, core from msm-4.14
'commit <a50bf345> ("msm: camera: flash: Moving flash off operation
in stop dev ioctl")' till 'commit <efd1feba> ("msm: camera: ife:
clean up HW manager sequence")'

Retain existing ION, CMA alloc, SMMU, for camera modules for msm-4.9.

Change-Id: Ia93df8a17445918d1cf5374fc1f3dfa4666f4c43
Signed-off-by: Raja Mallik <rmallik@codeaurora.org>

Changes added for camera ISP, ICP, JPEG, fd, cpas, cdm, core from msm-4.14
'commit <a50bf345> ("msm: camera: flash: Moving flash off operation
in stop dev ioctl")' till 'commit <efd1feba> ("msm: camera: ife:
clean up HW manager sequence")'

Retain existing ION, CMA alloc, SMMU, for camera modules for msm-4.9.

Change-Id: Ia93df8a17445918d1cf5374fc1f3dfa4666f4c43
Signed-off-by: Raja Mallik <rmallik@codeaurora.org>

xlog debug config on perf is not needed

Change-Id: Iaae11f8c0aadde07e54a3a21710e8828c3a94683
Signed-off-by: Jayaprakash <jmadiset@codeaurora.org>

Change-Id: I0f91cdd5bd321581dcad22f6f1fcb4eb39ffea86

Update the permissions of vibrator state and add set state
function for the userspace to access the vibrator state.

Change-Id: I71b3a8668330b9709b041ea4b3f54766030406ae
Signed-off-by: Kavya Nunna <knunna@codeaurora.org>

When there is a SYS_ERROR the instance object is getting
freed even before notifying the client causing the kernel
NULL pointer dereferece. To avoid this concurrency between
video close session and H/W error, unlocking the instance
lock after notifying the client.

Change-Id: I37027ffbf54143b89b93387e7cd5115b1bcfb71f
Signed-off-by: Mahesh Voorugonda <mvooru@codeaurora.org>

QM215 uses gpio based flash and torch
Add gpio flash driver for qm215 qrd.

Change-Id: Ib26c80141207030ed0d2a9f14c95622ce260a9b6
Signed-off-by: Naresh.B <cnbomme@codeaurora.org>
Signed-off-by: VijayaKumar T M <vtmuni@codeaurora.org>

This reverts commit e94591d0.

This cleanup broke the parsing of procfs mount parameters.

Bug: 79705088
Change-Id: I445c816cec4ea97b2f6f4abf516109f17b0f214a
Signed-off-by: Alistair Strachan <astrachan@google.com>
Git-commit: c396a018
Git-repo: https://android.googlesource.com/kernel/common/


Signed-off-by: Sachin Grover <sgrover@codeaurora.org>

Since message received from spi cannot be trusted there is possibility
of out-of-bound read if received read_id is not in range of fifo.
The patch validate rx_fifo_read index of edge info for remote side.

Change-Id: I3d3fa749935f477e5f98f986adc24e6e6a682d4d
Signed-off-by: Hardik Arya <harya@codeaurora.org>

Avoid calling icnss_fw_crashed two times in case of
back to back root pd shutdown.

Change-Id: I84a6bd9945a27219ddca2ffe2b0929451974a053
Signed-off-by: Anurag Chouhan <achouhan@codeaurora.org>

In fast_smmu_alloc size_t variable size is type casted to int
variable count, this variable count can get truncated and can
result in memory corruption during unmap, make count as size_t
and also add a check for count as sg_alloc_table_from_pages
accepts unsigned int value for count.

Change-Id: I4780a554c5c062fd9dd229e5cc0ac804b1ba31d8
Signed-off-by: Vijayanand Jitta <vjitta@codeaurora.org>

If code is merged upstream first it is possible that the same author
may end up attempting to merge the same code into a downstream kernel.
Since code coming from upstream is always a backport and may involve
modifications it is appropriate for the author to add another
Signed-off-by line to the bottom of the commit log but this will
cause checkpatch to complain about duplicate signatures.  By making
the duplicate signature warning a different error class
(DUPLICATE_SIGN_OFF instead of BAD_SIGN_OFF) this will let us ignore
this error with --ignore DUPLICATE_SIGN_OFF without losing the benefits
of the other signature checks.

Change-Id: Ic0dedbadcc1b319061db324f88740e5320bfff09
Signed-off-by: Jordan Crouse <jcrouse@codeaurora.org>

Currently, the perf's hotplug management is split between the
core and the PMU driver. That is, the PMU driver is responsible
for starting and stopping the perf-event counter in the cases
where the CPU is coming up and going down, respectively. However,
this approach is not scalable for the growing number of PMU drivers.
Moreover, due to partial cleanup at the perf-core, currently the
perf-events are prone to race conditions, for example between the
CPU hotplug teardown callback (perf_event_exit_cpu()) and the hrtimer
handler (perf_mux_hrtimer_handler()) that's used for multiplexing
the events.

Hence, the approach here is to achieve the hotplug management
at the perf-core level. The idea is to detach the event from the
context (perf_remove_from_context()), when the CPU is about to come
down and re-attach it back (perf_install_in_context()), when the CPU
comes back online.

The approach involves removing the logic for maintaining zombie
events (PERF_EVENT_STATE_ZOMBIE) and let the dormant list itself
carry the events whose CPUs are offline.

Change-Id: I39503d1f4b0b3937920b666a4f04875162fad3b3
Signed-off-by: Raghavendra Rao Ananta <rananta@codeaurora.org>
Signed-off-by: Swetha Chikkaboraiah <schikk@codeaurora.org>