commit 5b127e3b introduced accesses to pm_qos requests in workqueues
and hence some race conditions leading sometimes to kernel panics.
To be thread-safe, initialize the request once and for all and only
update its value to block or not s0i1.

Change-Id: I6516c60496974fd6d98421b794f4f6fb0718b1ce
Tracked-On: https://jira01.devtools.intel.com/browse/AW-6335


Signed-off-by: Simon Dubray <simonx.dubray@intel.com>
Reviewed-on: https://android.intel.com/597693


Reviewed-by: jenkins_ndg <jenkins_ndg@intel.com>
Reviewed-by: Korpershoek, MattijsX <mattijsx.korpershoek@intel.com>
Reviewed-by: Whitfield, MichaelX <michaelx.whitfield@intel.com>
Reviewed-by: Rouis, KhalifaX <khalifax.rouis@intel.com>
Reviewed-by: Jacquet, CyrilX <cyrilx.jacquet@intel.com>
Reviewed-by: Ledentec, AlexandreX <alexandrex.ledentec@intel.com>
Reviewed-by: Zaghdoud, WalidX <walidx.zaghdoud@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

Required to add revision of the git tree in kernel version.

Change-Id: Ib94d9eea64cd1f95ae415c204dac7e0035f3351f
Tracked-On: https://jira01.devtools.intel.com/browse/AW-6329


Signed-off-by: Victor Tasayco Loarte <victorx.tasayco.loarte@intel.com>
Reviewed-on: https://android.intel.com/597308


Reviewed-by: jenkins_ndg <jenkins_ndg@intel.com>
Reviewed-by: Korpershoek, MattijsX <mattijsx.korpershoek@intel.com>
Reviewed-by: Bertolin, PierreX <pierrex.bertolin@intel.com>

power_island_get/put() calls must be always balanced.
Make sure that error cases are properly handled by all subsystems.

Change-Id: Ic5c9b39d733fcac5d0d311267b3b4803878b53d6
Tracked-On: https://jira01.devtools.intel.com/browse/AW-6243


Signed-off-by: Simon Dubray <simonx.dubray@intel.com>
Reviewed-on: https://android.intel.com/594944


Reviewed-by: jenkins_ndg <jenkins_ndg@intel.com>
Reviewed-by: Ledentec, AlexandreX <alexandrex.ledentec@intel.com>
Reviewed-by: Korpershoek, MattijsX <mattijsx.korpershoek@intel.com>
Reviewed-by: Zaghdoud, WalidX <walidx.zaghdoud@intel.com>
Reviewed-by: Louis, FabienX <fabienx.louis@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

In order to recover after a dbi hang, enter_dsr_locked() must complete
even if the fifo is not yet empty.

Change-Id: Idad78d08ba94d28506f8f2d00ed8ca5a81cdef2e
Tracked-On: https://jira01.devtools.intel.com/browse/AW-6262


Signed-off-by: Simon Dubray <simonx.dubray@intel.com>
Reviewed-on: https://android.intel.com/595500


Reviewed-by: jenkins_ndg <jenkins_ndg@intel.com>
Reviewed-by: Zaghdoud, WalidX <walidx.zaghdoud@intel.com>
Reviewed-by: Wang, WenshengX <wenshengx.wang@intel.com>
Reviewed-by: Liu, WeiX W <weix.w.liu@intel.com>
Reviewed-by: Lachaud, EtienneX <etiennex.lachaud@intel.com>
Reviewed-by: Ledentec, AlexandreX <alexandrex.ledentec@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

To match rules related to uid/gid for syn recv packets
we need to get the full socket from request_sock struct.

Bug: 63917742
Change-Id: I03acb2251319fd800d0e36a6dde30fc1fbb7d1b0
Tracked-On: https://jira01.devtools.intel.com/browse/AW-5883


Signed-off-by: Simon Dubray <simonx.dubray@intel.com>
Reviewed-on: https://android.intel.com/595180


Reviewed-by: jenkins_ndg <jenkins_ndg@intel.com>
Reviewed-by: Korpershoek, MattijsX <mattijsx.korpershoek@intel.com>
Reviewed-by: Nassiet, GaelleX <gaellex.nassiet@intel.com>
Reviewed-by: Riviere, NathanX <nathanx.riviere@intel.com>
Reviewed-by: Louis, FabienX <fabienx.louis@intel.com>
Tested-by: Nassiet, GaelleX <gaellex.nassiet@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

vndbinder is needed as binder device in the Gold transition as recommended
by Google .In fact, in Android O, Binders will be seperated for several pieces
of the system (hw, vendor..)

Add vndbinder in the kernel config

Change-Id: Ifca98c3a88db61d229506994c7e2bf74ffffcee2
Tracked-On: https://jira01.devtools.intel.com/browse/AW-6268


Signed-off-by: Zairi Wajdi <wajdix.zairi@intel.com>
Signed-off-by: Maher KHADRAOUI <maherx.khadraoui@intel.com>
Reviewed-on: https://android.intel.com/595523


Reviewed-by: Korpershoek, MattijsX <mattijsx.korpershoek@intel.com>
Reviewed-by: Zaghdoud, WalidX <walidx.zaghdoud@intel.com>
Reviewed-by: Louis, FabienX <fabienx.louis@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>
Reviewed-by: Ledentec, AlexandreX <alexandrex.ledentec@intel.com>

Updates the battery charging status on interrupts and
sysfs access.

Change-Id: I1429d1d08be9c06f541c096460c6b65564744f19
Tracked-On: https://jira01.devtools.intel.com/browse/AW-6151


Signed-off-by: Hubert CHAUMETTE <hubertx.chaumette@intel.com>
Signed-off-by: Marouane <marouanex.saadani@intel.com>
Reviewed-on: https://android.intel.com/553873


Reviewed-by: jenkins_ndg <jenkins_ndg@intel.com>
Reviewed-by: Ghaddab, RiadhX <riadhx.ghaddab@intel.com>
Reviewed-by: Ferrari, AlainX <alainx.ferrari@intel.com>
Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

The reported capacity is set to 100% only after finishing the full charge.

Change-Id: I41343017877dbff1eba0abec58bf646b3bc36459
Tracked-On: https://jira01.devtools.intel.com/browse/AW-6151


Signed-off-by: Marouane <marouanex.saadani@intel.com>
Reviewed-on: https://android.intel.com/593252


Reviewed-by: jenkins_ndg <jenkins_ndg@intel.com>
Reviewed-by: Ghaddab, RiadhX <riadhx.ghaddab@intel.com>
Reviewed-by: Ferrari, AlainX <alainx.ferrari@intel.com>
Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

Change-Id: I30111f1e37a0e4b955f639bb4a2a0e6511068a87
Tracked-On: https://jira01.devtools.intel.com/browse/AW-6191


Signed-off-by: Hubert CHAUMETTE <hubertx.chaumette@intel.com>
Reviewed-on: https://android.intel.com/572396


Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Korpershoek, MattijsX <mattijsx.korpershoek@intel.com>
Reviewed-by: Ghaddab, RiadhX <riadhx.ghaddab@intel.com>
Reviewed-by: jenkins_ndg <jenkins_ndg@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

Add mutex to ensure updating the taos_device_lux global table through
taos_luxtable_store() doesn't race with read access by other functions.

The previous method was inefficient: disabling the chip requires
superfluous i2c transactions likely to take longer than the memcpy().

Change-Id: If749fccf83e7bc0e8db3073b828a19fd4265b9f4
Tracked-On: https://jira01.devtools.intel.com/browse/AW-6191


Signed-off-by: Hubert CHAUMETTE <hubertx.chaumette@intel.com>
Reviewed-on: https://android.intel.com/572395


Reviewed-by: jenkins_ndg <jenkins_ndg@intel.com>
Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Ghaddab, RiadhX <riadhx.ghaddab@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>
Reviewed-by: Korpershoek, MattijsX <mattijsx.korpershoek@intel.com>

Ensure that polling is really cancelled when disabling the sensor and
never runs on resume if the timer has expired during suspend.
delayed_work is equivalent and simpler than the timer/wq combination.

Change-Id: I5f5dffcfc06b95fc218fc21f264c3ed88b3e4d2d
Tracked-On: https://jira01.devtools.intel.com/browse/AW-6191


Signed-off-by: Hubert CHAUMETTE <hubertx.chaumette@intel.com>
Reviewed-on: https://android.intel.com/572394


Reviewed-by: jenkins_ndg <jenkins_ndg@intel.com>
Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Korpershoek, MattijsX <mattijsx.korpershoek@intel.com>
Reviewed-by: Ghaddab, RiadhX <riadhx.ghaddab@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

Ensure the bus is awake before accessing it.

Change-Id: I16cb9f5f4e6218be31bacd8ef2b495b4896a71f6
Tracked-On: https://jira01.devtools.intel.com/browse/AW-6191


Signed-off-by: Hubert CHAUMETTE <hubertx.chaumette@intel.com>
Reviewed-on: https://android.intel.com/572391


Reviewed-by: jenkins_ndg <jenkins_ndg@intel.com>
Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Korpershoek, MattijsX <mattijsx.korpershoek@intel.com>
Reviewed-by: Ghaddab, RiadhX <riadhx.ghaddab@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

- On resume the ADC was disabled regardless of whether it
was enabled before going to suspend or not. As a consequence
the LUX values are sometimes not available after a resume
- This patch restore the ADC status on resume

Change-Id: I3bc5b7997526830f2f35e4db6391e916693527a5
Tracked-On: https://jira01.devtools.intel.com/browse/AW-6191


Signed-off-by: Riadh GHADDAB <riadhx.ghaddab@intel.com>
Reviewed-on: https://android.intel.com/594058


Reviewed-by: Chaumette, HubertX <hubertx.chaumette@intel.com>
Tested-by: Chaumette, HubertX <hubertx.chaumette@intel.com>
Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: jenkins_ndg <jenkins_ndg@intel.com>
Reviewed-by: Korpershoek, MattijsX <mattijsx.korpershoek@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

Commit 939c7a4f ("tracing: Introduce saved_cmdlines_size file")
introduced ability to change saved cmdlines size. This resized saved
command lines but missed resizing tgid mapping as well.

Another issue is that when the resize happens, it removes saved command
lines and reallocates new memory for it. This introduced a race
condition when reading the global savecmd as this can be freed in the
middle of accessing it causing a use after free access. Fix this by
implementing locking.

Signed-off-by: Adrian Salido <salidoa@google.com>
Bug: 36007735
Tracked-On: https://jira01.devtools.intel.com/browse/AW-6222


Change-Id: I334791ac35f8bcbd34362ed112aa624275a46947
Signed-off-by: Victor Tasayco Loarte <victorx.tasayco.loarte@intel.com>
Reviewed-on: https://android.intel.com/594734


Reviewed-by: Louis, FabienX <fabienx.louis@intel.com>
Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Ledentec, AlexandreX <alexandrex.ledentec@intel.com>
Reviewed-by: Zairi, WajdiX <wajdix.zairi@intel.com>
Reviewed-by: Akue, LoicX <loicx.akue@intel.com>

Make sure segno and blkoff read from raw image are valid.

(url https://sourceforge.net/p/linux-f2fs/mailman/message/35835945

)

Signed-off-by: Jin Qian <jinqian@google.com>
Bug: 36588520
Change-Id: Iba66ab97d3d0870ea48b5ef192d9075f225a934a
Tracked-On: https://jira01.devtools.intel.com/browse/AW-6222


Signed-off-by: Victor Tasayco Loarte <victorx.tasayco.loarte@intel.com>
Reviewed-on: https://android.intel.com/594733


Reviewed-by: Louis, FabienX <fabienx.louis@intel.com>
Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Ledentec, AlexandreX <alexandrex.ledentec@intel.com>
Reviewed-by: Zairi, WajdiX <wajdix.zairi@intel.com>
Reviewed-by: Akue, LoicX <loicx.akue@intel.com>

F2FS uses 4 bytes to represent block address. As a result, supported
size of disk is 16 TB and it equals to 16 * 1024 * 1024 / 2 segments.

Change-Id: I4b7d0024fcc530a1d6a553c26fe5145a789ef39c
Tracked-On: https://jira01.devtools.intel.com/browse/AW-6222


Signed-off-by: Jin Qian <jinqian@google.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Signed-off-by: Victor Tasayco Loarte <victorx.tasayco.loarte@intel.com>
Reviewed-on: https://android.intel.com/594732


Reviewed-by: Louis, FabienX <fabienx.louis@intel.com>
Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Ledentec, AlexandreX <alexandrex.ledentec@intel.com>
Reviewed-by: Zairi, WajdiX <wajdix.zairi@intel.com>
Reviewed-by: Akue, LoicX <loicx.akue@intel.com>

f2fs currently only supports 4KB block size and 2MB segment size.
Sanity check log_blocks_per_seg == 9, i.e. 2MB/4KB = (1 << 9)

Partially
(cherry-picked from commit 9a59b62fd88196844cee5fff851bee2cfd7afb6e)

f2fs: do more integrity verification for superblock

Do more sanity check for superblock during ->mount.

Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>

Signed-off-by: Jin Qian <jinqian@google.com>
Bug: 36817013
Change-Id: I0be52e54fba82083068337ceb9f7ad985a87319f
Tracked-On: https://jira01.devtools.intel.com/browse/AW-6222


Signed-off-by: Victor Tasayco Loarte <victorx.tasayco.loarte@intel.com>
Reviewed-on: https://android.intel.com/594731


Reviewed-by: Louis, FabienX <fabienx.louis@intel.com>
Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Zairi, WajdiX <wajdix.zairi@intel.com>
Reviewed-by: Akue, LoicX <loicx.akue@intel.com>

The handling of the might_cancel queueing is not properly protected, so
parallel operations on the file descriptor can race with each other and
lead to list corruptions or use after free.

Protect the context for these operations with a seperate lock.

The wait queue lock cannot be reused for this because that would create a
lock inversion scenario vs. the cancel lock. Replacing might_cancel with an
atomic (atomic_t or atomic bit) does not help either because it still can
race vs. the actual list operation.

Change-Id: I5ccfd9bff148278d50589582450755946cb85f0a
Reported-by: Dmitry Vyukov <dvyukov@google.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: "linux-fsdevel@vger.kernel.org"
Cc: syzkaller <syzkaller@googlegroups.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: linux-fsdevel@vger.kernel.org
Link: http://lkml.kernel.org/r/alpine.DEB.2.20.1701311521430.3457@nanos


Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tracked-On: https://jira01.devtools.intel.com/browse/AW-6222


Signed-off-by: Victor Tasayco Loarte <victorx.tasayco.loarte@intel.com>
Reviewed-on: https://android.intel.com/594730


Reviewed-by: Louis, FabienX <fabienx.louis@intel.com>
Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Ledentec, AlexandreX <alexandrex.ledentec@intel.com>
Reviewed-by: Zairi, WajdiX <wajdix.zairi@intel.com>
Reviewed-by: Akue, LoicX <loicx.akue@intel.com>

Set the safeguard threshold value "volt_min_lim" to 3.2v, since we
have changed the fuel guage characterization table.
This reverts commit 49e53970.

Change-Id: Ia9065996600b3e7d2741c91f05cd3f6afab7d0be
Tracked-On: https://jira01.devtools.intel.com/browse/AW-6112


Signed-off-by: Marouane <marouanex.saadani@intel.com>
Reviewed-on: https://android.intel.com/592210


Reviewed-by: Chaumette, HubertX <hubertx.chaumette@intel.com>
Reviewed-by: Ferrari, AlainX <alainx.ferrari@intel.com>
Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Korpershoek, MattijsX <mattijsx.korpershoek@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

On resume, gfx driver needs to put a pm_qos preventing s0ix until
display resume, else there are some scu hangs when changing the power
state of other devices. This patch forces a fake display activity 250ms
after resume to allow s0ix and improve power consumption.

Change-Id: Ie0143d0bcd9176db8b307b0b43c4ac05c66c27f4
Tracked-On: https://jira01.devtools.intel.com/browse/AW-6176


Signed-off-by: Simon Dubray <simonx.dubray@intel.com>
Reviewed-on: https://android.intel.com/593703


Reviewed-by: Nassiet, GaelleX <gaellex.nassiet@intel.com>
Reviewed-by: jenkins_ndg <jenkins_ndg@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>
Reviewed-by: Korpershoek, MattijsX <mattijsx.korpershoek@intel.com>

If there is an alarm firing in the next 2s, alarmtimer makes suspend
abort with a pm_wakeup_event until the next alarm. Add a new workqueue
to detect that such scenario will happen and call pm_wakeup_event at
this time to be more power friendly.

Change-Id: I443589c346bd30387fc4994a9e2a26a534d4f9d5
Tracked-On: https://jira01.devtools.intel.com/browse/AW-6148


Signed-off-by: Simon Dubray <simonx.dubray@intel.com>
Reviewed-on: https://android.intel.com/592013


Reviewed-by: Chaumette, HubertX <hubertx.chaumette@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

Reducing delays during mmc init allows to save 50ms on resume time.

Change-Id: I187c923b825c05d48fbf4f2d1a5c6fe3629dfba7
Tracked-On: https://jira01.devtools.intel.com/browse/AW-6147


Signed-off-by: Simon Dubray <simonx.dubray@intel.com>
Reviewed-on: https://android.intel.com/591905


Reviewed-by: Nassiet, GaelleX <gaellex.nassiet@intel.com>
Reviewed-by: Lachaud, EtienneX <etiennex.lachaud@intel.com>
Tested-by: Lachaud, EtienneX <etiennex.lachaud@intel.com>
Reviewed-by: Chaumette, HubertX <hubertx.chaumette@intel.com>
Reviewed-by: Ferrari, AlainX <alainx.ferrari@intel.com>
Reviewed-by: Korpershoek, MattijsX <mattijsx.korpershoek@intel.com>
Reviewed-by: Pierre, AlexandreX <alexandrex.pierre@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

config RCU_FAST_NO_HZ helps to make rcu less agressive by increasing
the grace period to several ticks instead of one (default is four).
Setting this parameter will have no visible performance impact but
save some power in most active use cases.

Change-Id: Iadac95d6435909afab54b4b9b0c1db107b35fe30
Tracked-On: https://jira01.devtools.intel.com/browse/AW-6165


Signed-off-by: Simon Dubray <simonx.dubray@intel.com>
Reviewed-on: https://android.intel.com/593407


Reviewed-by: Korpershoek, MattijsX <mattijsx.korpershoek@intel.com>
Reviewed-by: Pierre, AlexandreX <alexandrex.pierre@intel.com>
Reviewed-by: Lachaud, EtienneX <etiennex.lachaud@intel.com>
Tested-by: Lachaud, EtienneX <etiennex.lachaud@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

Force DMA RX to be processed and DMA channel to be reprogrammed
in case of OHRM HSU driver suspend failing on DMA RX busy.

The device suspend failing was causing platform to never suspend and
then draining power.

Change-Id: I109ff44fb0a5a69a499fb9df4125b7131859267c
Tracked-On: https://jira01.devtools.intel.com/browse/AW-5242


Signed-off-by: Celine Laurencin <celinex.laurencin@intel.com>
Reviewed-on: https://android.intel.com/590358


Reviewed-by: Jacquet, CyrilX <cyrilx.jacquet@intel.com>
Reviewed-by: Miloni, MarcX <marcx.miloni@intel.com>
Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Korpershoek, MattijsX <mattijsx.korpershoek@intel.com>
Reviewed-by: Zaghdoud, WalidX <walidx.zaghdoud@intel.com>
Reviewed-by: Ben Alaya, AymenX <aymenx.ben.alaya@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

As our devices are closed, mmc insertion/removal is useless.

Change-Id: I796ca594dfee1357f282100edeb637e82ef71210
Tracked-On: https://jira01.devtools.intel.com/browse/AW-6147


Signed-off-by: Simon Dubray <simonx.dubray@intel.com>
Reviewed-on: https://android.intel.com/591456


Reviewed-by: Nassiet, GaelleX <gaellex.nassiet@intel.com>
Reviewed-by: Lachaud, EtienneX <etiennex.lachaud@intel.com>
Tested-by: Lachaud, EtienneX <etiennex.lachaud@intel.com>
Reviewed-by: Chaumette, HubertX <hubertx.chaumette@intel.com>
Reviewed-by: Ferrari, AlainX <alainx.ferrari@intel.com>
Reviewed-by: Korpershoek, MattijsX <mattijsx.korpershoek@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

Move platform-specific lux algorithms to platform data and add a new
one

Change-Id: I9254b431e08d1584954fcbc4019942596e217709
Tracked-On: https://jira01.devtools.intel.com/browse/AW-5823


Signed-off-by: Hubert Chaumette <hubertx.chaumette@intel.com>
Reviewed-on: https://android.intel.com/589196


Reviewed-by: Ghaddab, RiadhX <riadhx.ghaddab@intel.com>
Reviewed-by: Korpershoek, MattijsX <mattijsx.korpershoek@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

Restore ALS parameters on resume. Fixes sensor latency after resume.
Also move the initialization delay after powering up rather than before
enabling to avoid unneeded delays on disabled/enabled transitions.

Change-Id: I1fd01bfb5c0841f51730006257a6ea6d51ff588a
Tracked-On: https://jira01.devtools.intel.com/browse/AW-6028


Signed-off-by: Hubert Chaumette <hubertx.chaumette@intel.com>
Reviewed-on: https://android.intel.com/591152


Reviewed-by: Nassiet, GaelleX <gaellex.nassiet@intel.com>
Reviewed-by: Ghaddab, RiadhX <riadhx.ghaddab@intel.com>
Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Korpershoek, MattijsX <mattijsx.korpershoek@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

This patch initially merged for power saving has potentially a bad
impact by increasing the dbi_hang issue occurrence on all platforms.

This reverts commit 78bd997a.

Change-Id: I5e57c0f2526cc45456f5e3d57f577e108921c1a7
Tracked-On: https://jira01.devtools.intel.com/browse/AW-6085


Signed-off-by: Alexandre Ledentec <alexandrex.ledentec@intel.com>
Reviewed-on: https://android.intel.com/591672


Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Zaghdoud, WalidX <walidx.zaghdoud@intel.com>
Reviewed-by: Akue, LoicX <loicx.akue@intel.com>

When calculating po->tp_hdrlen + po->tp_reserve the result can overflow.

Fix by checking that tp_reserve <= INT_MAX on assign.

Change-Id: I22f6c56ae02c435f9d5413effad43f32d9c18cdb
Tracked-On: https://jira01.devtools.intel.com/browse/AW-5928


Signed-off-by: Loic Akue <loicx.akue@intel.com>
Reviewed-on: https://android.intel.com/590711


Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

When calculating rb->frames_per_block * req->tp_block_nr the result
can overflow.

Add a check that tp_block_size * tp_block_nr <= UINT_MAX.

Since frames_per_block <= tp_block_size, the expression would
never overflow.

Change-Id: I71b9547a7f832ceda2349108e2129905cb6d1121
Tracked-On: https://jira01.devtools.intel.com/browse/AW-5928


Signed-off-by: Loic Akue <loicx.akue@intel.com>
Reviewed-on: https://android.intel.com/590710


Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

Subtracting tp_sizeof_priv from tp_block_size and casting to int
to check whether one is less then the other doesn't always work
(both of them are unsigned ints).

Compare them as is instead.

Also cast tp_sizeof_priv to u64 before using BLK_PLUS_PRIV, as
it can overflow inside BLK_PLUS_PRIV otherwise.

Change-Id: Ia02b8748d4af204e75aabb1e5031f563bc92a44e
Tracked-On: https://jira01.devtools.intel.com/browse/AW-5928


Signed-off-by: Loic Akue <loicx.akue@intel.com>
Reviewed-on: https://android.intel.com/590709


Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

In the current DCCP implementation an skb for a DCCP_PKT_REQUEST packet
is forcibly freed via __kfree_skb in dccp_rcv_state_process if
dccp_v6_conn_request successfully returns.

However, if IPV6_RECVPKTINFO is set on a socket, the address of the skb
is saved to ireq->pktopts and the ref count for skb is incremented in
dccp_v6_conn_request, so skb is still in use. Nevertheless, it gets freed
in dccp_rcv_state_process.

Fix by calling consume_skb instead of doing goto discard and therefore
calling __kfree_skb.

Similar fixes for TCP:

fb7e2399 [TCP]: skb is unexpectedly freed.
0aea76d35c9651d55bbaf746e7914e5f9ae5a25d tcp: SYN packets are now
simply consumed

Change-Id: I572812d1789144cb363b84a12d733049e0d58e96
Tracked-On: https://jira01.devtools.intel.com/browse/AW-5928


Signed-off-by: Loic Akue <loicx.akue@intel.com>
Reviewed-on: https://android.intel.com/590708


Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

Andrey Konovalov got crashes in __ip_options_echo() when a NULL skb->dst
is accessed.

ipv4_pktinfo_prepare() should not drop the dst if (evil) IP options
are present.

We could refine the test to the presence of ts_needtime or srr,
but IP options are not often used, so let's be conservative.

Thanks to syzkaller team for finding this bug.

Change-Id: I76d78d2ec831ff447334e8c71efb72736305c8f9
Fixes: d826eb14 ("ipv4: PKTINFO doesnt need dst reference")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Andrey Konovalov <andreyknvl@google.com>
Tracked-On:  https://jira01.devtools.intel.com/browse/AW-5928


Signed-off-by: Loic Akue <loicx.akue@intel.com>
Reviewed-on: https://android.intel.com/590707


Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

This reverts commit f0ce0eee6b71bc310153edb87e66e6b25e12fece.

Bug: 34951864
Bug: 36468447
Change-Id: I9667f6a1b5ddd0ecc358bd9f2fb4ce8e62185486
Tracked-On: https://jira01.devtools.intel.com/browse/AW-5928


Signed-off-by: Loic Akue <loicx.akue@intel.com>
Reviewed-on: https://android.intel.com/590701


Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

DSR count had been increased from 2 to 15 three years ago in order to make
it less power-agressive and avoid shutting down the display for nothing.
On our platforms, we need to reduce it in order to increase the s0ix residency
in use cases with a few fps.

A previous patch already decrease this DSR count from 15 to 4.
Now let's revert completely the change made three years ago by setting it to 2

It allows to sleep as much as possible between 2 frames and avoid any
side-effect on dozing needs by requesting unexpected VSYNC reSync.

Change-Id: I595b836698e2f769cfaaabb496d6f440dbef1b0d
Tracked-On: https://jira01.devtools.intel.com/browse/AW-4084


Signed-off-by: Alexandre Ledentec <alexandrex.ledentec@intel.com>
Reviewed-on: https://android.intel.com/571897


Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Tested-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

Make atomic calls to power_island_get() to avoid getting twice an island.

Change-Id: Ia775039100ab1a9897d8ecb9cc8df7c3ca11b1d5
Tracked-On: https://jira01.devtools.intel.com/browse/AW-5804


Signed-off-by: Simon Dubray <simonx.dubray@intel.com>
Reviewed-on: https://android.intel.com/590348


Reviewed-by: Ledentec, AlexandreX <alexandrex.ledentec@intel.com>
Tested-by: Ledentec, AlexandreX <alexandrex.ledentec@intel.com>
Reviewed-by: Gong, Sophia <sophia.gong@intel.com>
Reviewed-by: Lachaud, EtienneX <etiennex.lachaud@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

CONFIG_UID_CPUTIME allows per UID based cpu time statistics
exported to /proc/uid_cputime

Notes:
- This must be enabled according to Android requirements document:
  https://source.android.com/compatibility/android-cdd#84-power-consumption-accounting

- This config is part of android/configs/android-base.cfg

Change-Id: Id91c5c237da8c22f14317ba36500d4afbe67971e
Tracked-On: https://jira01.devtools.intel.com/browse/AW-5992


Signed-off-by: Mattijs Korpershoek <mattijsx.korpershoek@intel.com>
Reviewed-on: https://android.intel.com/590567


Reviewed-by: Samoun, JacquesX <jacquesx.samoun@intel.com>
Reviewed-by: Louis, FabienX <fabienx.louis@intel.com>
Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>
Reviewed-by: Ledentec, AlexandreX <alexandrex.ledentec@intel.com>

 this allows to set platform specific thresholds, which is necessary as
 SPL has a different threshold as RBY.

Change-Id: I72dcd3d141a3e3a5564f052ddd1ea470d06de6f5
Tracked-On: https://jira01.devtools.intel.com/browse/AW-5914


Signed-off-by: Jean-Yves Bottraud <jean-yvesx.bottraud@intel.com>
Reviewed-on: https://android.intel.com/590314


Reviewed-by: Chaumette, HubertX <hubertx.chaumette@intel.com>
Reviewed-by: Korpershoek, MattijsX <mattijsx.korpershoek@intel.com>
Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

Change-Id: I768f04006a62a701232549aa812ae05dbd09cbfc
Tracked-On: https://jira01.devtools.intel.com/browse/AW-5943


Signed-off-by: Louis, FabienX <fabienx.louis@intel.com>
Reviewed-on: https://android.intel.com/589850


Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Korpershoek, MattijsX <mattijsx.korpershoek@intel.com>
Reviewed-by: Ledentec, AlexandreX <alexandrex.ledentec@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>

This patch introduces a new binder_fd_array object,
that allows us to support one or more file descriptors
embedded in a buffer that is scatter-gathered.

Change-Id: I647a53cf0d905c7be0dfd9333806982def68dd74
Tracked-On: https://jira01.devtools.intel.com/browse/AW-5943


Signed-off-by: Martijn Coenen <maco@google.com>
Reviewed-on: https://android.intel.com/589830


Reviewed-by: Louis, FabienX <fabienx.louis@intel.com>
Tested-by: Louis, FabienX <fabienx.louis@intel.com>
Reviewed-by: Dubray, SimonX <simonx.dubray@intel.com>
Reviewed-by: Korpershoek, MattijsX <mattijsx.korpershoek@intel.com>
Reviewed-by: Ledentec, AlexandreX <alexandrex.ledentec@intel.com>
Reviewed-by: Tasayco Loarte, VictorX <victorx.tasayco.loarte@intel.com>