CHROMIUM: md: bootcache check metadata and data lengths

bootcache now verifies that the lengths for metadata and data
make sense and fit with in the partition.

BUG=chromium-os:35379
TEST=ran bootcachetest on stumpy, arm

Change-Id: I6beee3a5b1aa723aaf08c673a004bc4924afc42d
Signed-off-by: Paul Taysom <taysom@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/35578
Reviewed-by: Mandeep Singh Baines <msb@chromium.org>

CHROMIUM: mm: Fix calculation of dirtyable memory

The system uses global_dirtyable_memory() to calculate
number of dirtyable pages/pages that can be allocated
to the page cache.  A bug causes an underflow thus making
the page count look like a big unsigned number.  This in turn
confuses the dirty writeback throttling to aggressively write
back pages as they become dirty (usually 1 page at a time).

Fix is to ensure there is no underflow while doing the math.

Signed-off-by: Sonny Rao <sonnyrao@chromium.org>
Signed-off-by: Puneet Kumar <puneetster@chromium.org>
BUG=chrome-os-partner:16011
TEST=Manual; boot kernel, powerwash, login with testaccount and
make sure no jank occurs on sync of applications

Change-Id: I614e7c3156e014f0f28a4ef9bdd8cb8a2cd07b2a
Reviewed-on: https://gerrit.chromium.org/gerrit/37612
Reviewed-by: Kees Cook <keescook@chromium.org>
Reviewed-by: Olof Johansson <olofj@chromium.org>
Commit-Ready: Puneet Kumar <puneetster@chromium.org>
Reviewed-by: Puneet Kumar <puneetster@chromium.org>
Tested-by: Puneet Kumar <puneetster@chromium.org>

samsung: pm-check: Make checksum algorithm more configurable and add sum

There are good reasons for using different checksums in pm-check.  Add
'sum' and generalize mechanism for switching checksum algorithms.

The 'sum' algorithm is now the default.

BUG=chrome-os-partner:15655
TEST=Put code in to do a bogus corruption of address
0x71735644 to 0xBADC0DE1.  Then:
1. Do a suspend/resume and see sum algorithm (0xffffffff + 0xbadc0de1):
    s3c_pm_check: Restore CRC error at 71734000 (badc0de0 vs ffffffff)
    bitfix_recover_chunk: Attempting recovery at 71734000
    ...fixed 0x71735644 from 0xbadc0de1 to 0x00000000
2. Switch to xor:
    echo -n xor > /sys/module/pm_check/parameters/pm_check_checksum_type
3. Do suspend/resume again and see xor algorithm (0xffffffff ^ 0xbadc0de1)
    s3c_pm_check: Restore CRC error at 71734000 (4523f21e vs ffffffff)
    bitfix_recover_chunk: Attempting recovery at 71734000
    ...fixed 0x71735644 from 0xbadc0de1 to 0x00000000
4. Switch to CRC:
    echo -n crc > /sys/module/pm_check/parameters/pm_check_checksum_type
5. Do suspend/resume again and see crc algorithm
    s3c_pm_check: Restore CRC error at 71734000 (57243536 vs 270b666b)
    bitfix_recover_chunk: Attempting recovery at 71734000
    ...fixed 0x71735644 from 0xbadc0de1 to 0x00000000
TEST=Time by doing this for sum:
    echo Y > /sys/module/pm_check/parameters/pm_check_print_timings
    suspend_stress_test --noerrors_fatal --backup_rtc \
        --suspend_min 15 --suspend_max 15 -c2
    dmesg | grep 'memory scan'
...and saw this, which is same as xor:
    s3c_pm_check: Suspend memory scan took 4519985 usecs
    s3c_pm_check: Resume memory scan took 1181126 usecs
    s3c_pm_check: Suspend memory scan took 4596896 usecs
    s3c_pm_check: Resume memory scan took 1184638 usecs
...note that the above includes bitfix, which is why suspend is 4.5
seconds.

Change-Id: Ifd20c5d3f6da5069634360057120b63e232f8a77
Signed-off-by: Doug Anderson <dianders@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/37592
Reviewed-by: Jon Kliegman <kliegs@chromium.org>

CHROMIUM: config: exynos5: Turn on bitfix

Enable bitfix code.  This code is safe to enable always--it will
autodetect when it's needed.

BUG=chrome-os-partner:15655
TEST=suspend_stress_test

Change-Id: I9fc03f05c1e894dcf196e2e2a256ae87f8dc4384
Signed-off-by: Doug Anderson <dianders@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/37066
Reviewed-by: Jon Kliegman <kliegs@chromium.org>

samsung: snow: bitfix: Recover from bit errors across sleep/wake

Some Snow boards have an issue in read-only firmware that can cause
bit errors across a suspend/resume cycle.  Given the pattern of the
corruption we can actually recover from nearly all bit errors given a
little bit of perparation time.

A full description of the problem and solution can be found in
bitfix-snow.c.

This change recovers from most cases of corruption but can still
be made more robust and optimized.

At the moment this code appears to add about 3.5 seconds to
suspend time and ~nothing to resume time.  That's on top of the
~1 second for pm-check.

BUG=chrome-os-partner:15655
TEST=Run the following script:
  cd /var/log
  touch sst.txt
  echo '------------' >> sst.txt
  date >> sst.txt
  while true; do
    suspend_stress_test --noerrors_fatal --backup_rtc \
      --suspend_min 15 --suspend_max 15 | tee -a sst.txt
    dmesg | \
      grep "\(CRC error\|bitfix\|...fixed\|[rR]ecover\)" \
      >> sst.txt
    dmesg -C
  done
See messages like:
  [  229.959279] s3c_pm_check: Restore CRC error at 86856000 (6a5fe07c vs 6acbe07c)
  [  229.959279] bitfix_recover_chunk: Attempting recovery at 86856000
  [  229.959279] ...fixed 0x86857140 from 0x76b47894 to 0x76b07894
  [  229.959279] ...fixed 0x8685751c from 0x00100051 to 0x00000051
  [  229.959279] ...fixed 0x86857d54 from 0x77773604 to 0x77f73604
  [  229.959279] s3c_pm_check: Recovered
  [  229.959279] s3c_pm_check: Restore CRC error at 86858000 (4e502299 vs 4e5b2299)
  [  229.959279] bitfix_recover_chunk: Attempting recovery at 86858000
  [  229.959279] ...fixed 0x86858358 from 0x76f8a3d8 to 0x76f9a3d8
  [  229.959279] ...fixed 0x86858f2c from 0x00080002 to 0x00000002
  [  229.959279] ...fixed 0x86859b64 from 0x00021a66 to 0x00001a66
  [  229.959279] s3c_pm_check: Recovered
TEST=Run bitfix-test.sh from I26b3938fe5c481968dc019f188f46d211dd6a801
TEST=Time by doing:
  echo Y > /sys/module/pm_check/parameters/pm_check_print_timings
  suspend_stress_test --noerrors_fatal --backup_rtc \
      --suspend_min 15 --suspend_max 15 -c3
  dmesg | grep 'memory scan'
See:
  [ 1433.174264] s3c_pm_check: Suspend memory scan took 4434401 usecs
  [ 1433.174264] s3c_pm_check: Resume memory scan took 1122626 usecs
  [ 1445.639303] s3c_pm_check: Suspend memory scan took 4460913 usecs
  [ 1445.639303] s3c_pm_check: Resume memory scan took 1122503 usecs
  [ 1458.219387] s3c_pm_check: Suspend memory scan took 4455245 usecs
  [ 1458.219387] s3c_pm_check: Resume memory scan took 1122534 usecs
TEST=Boot with old bios and see bitfix get enabled by doing:
  dmesg | grep bitfix_reserve
...and see:
  [    0.000000] bitfix_reserve: Detected firmware that needs bitfix
TEST=Run with legacy BIOS (doesn't set RO string) and enable manually
  echo 'Y' > /sys/module/bitfix_snow/parameters/bitfix_enabled
  echo 'Y' > /sys/module/pm_check/parameters/pm_check_enabled
...see bits get fixed using tests above.

Change-Id: I9c7ac2f85b8d9398c93f486f9401daee1526f571
Signed-off-by: Doug Anderson <dianders@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/36905
Reviewed-by: Sam Leffler <sleffler@chromium.org>
Reviewed-by: Jon Kliegman <kliegs@chromium.org>

PM / Sleep: Mark oops_id as suspend_volatile

Without this if we have a WARN_ON hit in suspend/resume code we'll
detect this as a CRC error.

BUG=chrome-os-partner:15655
TEST=Try to measure timings (which prints out a warning).
  echo Y > /sys/module/pm_check/parameters/pm_check_print_timings
  suspend_stress_test --noerrors_fatal --backup_rtc \
      --suspend_min 15 --suspend_max 15 -c1
Before this CL you'll see a CRC error:
  s3c_pm_check: Restore CRC error at 407bc000 (08e3f07b vs 08e3f079)
  bitfix_recover_chunk: Attempting recovery at 407bc000
  ...fixed 0x407bdeb0 from 0x76d4a47b to 0x76d4a479
After this CL there will be no CRC error.

Change-Id: I276bd69156bea6bd5b4c32274898e7c4ded03eb5
Signed-off-by: Doug Anderson <dianders@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/37591
Reviewed-by: Jon Kliegman <kliegs@chromium.org>

CHROMIUM: drm/i915: Workaround disappearing AVI Infoframe on resume

The AVI Infoframe disappears on SNB after suspend/resume. This patch
works around what seems to be a hardware issue by doing a dummy register
write on DPMS calls.

BUG=chrome-os-partner:16054
TEST=with the HDMI analyzer: suspend/resume parrot, verify that
TEST=the AVI infoframe is still here

Change-Id: I7cb880408db3cd279d09cb7a53f6787c14e03027
Reviewed-on: https://gerrit.chromium.org/gerrit/37596
Tested-by: Stéphane Marchesin <marcheu@chromium.org>
Reviewed-by: Stuart Abercrombie <sabercrombie@chromium.org>
Commit-Ready: Stéphane Marchesin <marcheu@chromium.org>

CHROMIUM: Input: cyapa - Send uevent when device detection fails

Fire an uevent when device detection fails so that userspace can
take some actions (e.g. gathering relevant logs).

Signed-off-by: Yufeng Shen <miletus@chromium.org>
BUG=chromium-os:35169
TEST=No consistent way to repro the device detection failure. So
     no test for now.

Change-Id: Ie2951f6dc69d6afaf9b3e37b97d7383f0eb53df8
Reviewed-on: https://gerrit.chromium.org/gerrit/37147
Reviewed-by: Benson Leung <bleung@chromium.org>
Commit-Ready: Yufeng Shen <miletus@chromium.org>
Tested-by: Yufeng Shen <miletus@chromium.org>

CHROMIUM: Input: psmouse - retry getid command in psmouse_probe()

As the synaptics device may not respond to the first command in psmouse_probe
when a machine is booting up, the patch gives a second probe if the first
one fails.

Signed-off-by: Chung-yih Wang <cywang@chromium.org>
BUG=chrome-os-partner:14638
TEST=on device; do the following steps:

1. make sure there is no other USB devices plugged into the machine.
2. mount -o rw,remount /
3. create a script file test-synaptics-present.conf under /etc/init with following.

cat > /etc/init/test-synaptics-present.conf <<EOF
start on starting system-services
script
  DEV=/sys/devices/platform/i8042/serio4/serio5/input/input9
  sleep 15
  if [ -e \$DEV ]; then
    logger 'Synaptics Trackpad found at \$DEV'
    shutdown -r now
  else
    logger 'Synaptics Trackpad NOT found at \$DEV'
  fi
end script

4. reboot, then the machine should keep rebooting every 15+ seconds.

Change-Id: Ieb81acf75bb79ac9f34f9b01fbe3d8f4b6a367b2
Reviewed-on: https://gerrit.chromium.org/gerrit/34405
Reviewed-by: Daniel Kurtz <djkurtz@chromium.org>
Commit-Ready: Chung-yih Wang <cywang@chromium.org>
Tested-by: Chung-yih Wang <cywang@chromium.org>

PM / Sleep: arm: mmu: Mark fixed ptes as volatile during suspend

Make the page table entries used by kmap_atomic as volatile so we ignore
changes made to them during the memory checksum operations on suspend/resume.

BUG=chrome-os-partner:15914
TEST=Saw location 0xaf7fe400 generate false CRC errors before patch
  After patch changes to that page are ignored.
  Confirmed with additional debug printing that the supression is added

Change-Id: I6766099aa0e86c479b26ee18d69b537493e84dfc
Signed-off-by: Jonathan Kliegman <kliegs@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/37463
Reviewed-by: Michael Spang <spang@chromium.org>
Reviewed-by: Doug Anderson <dianders@chromium.org>

samsung: pm-check: Cleanup chunk exclude code to use suspend_volatile

Recent changes mark memory that may change across suspend/resume
as suspend_volatile.  We can now use this in pm-check, so do so.
That means we can remove some extra globals from printk.c.

This change also cleans up handling of sleep_save_sp:
* Don't exclude *sleep_save_sp.  That was a physical address and is
  also already handled by current_thread_info().
* Don't exclude sleep_save_sp.  Instead: we zero it out before out
  code runs.  This is a transitory value and zeroing it is fine.  Now
  we won't exclude chunks of memory stored near sleep_save_sp.

...and a few other minor fixes:
* Skips regions in s3c_pm_makecheck() too.
* Avoids goto in s3c_pm_runcheck() loop.

BUG=chrome-os-partner:15914
TEST=suspend_stress_test

Change-Id: I5eadc7631c3f31f91138170c5e37430da7e3520b
Signed-off-by: Doug Anderson <dianders@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/37359
Reviewed-by: Jon Kliegman <kliegs@chromium.org>

CHROMIUM: tpm: always run ContinueSelftest at resume

In a previous patch we moved ContinueSelfTest out of the resume path
to speed it up, and we skip the command if enough time has passed
between the resume and the first following TPM command, on the
assumption that the TPM starts and runs the self test to completion
at resume time.  (This applies only to TPMs that are powered off
between suspend and resume.)

The ST Micro TPM behaves differently and needs ContinueSelfTest
to be called at resume, before the next TPM command can be
successfully executed.  We could make the behavior of the
resume callback depend on the TPM model.  However, ContinueSelftest
completes quickly on all TPM models currently used in Chromebooks,
and it's not on any critical path.  So we prefer to make the
code simpler for everybody.

BUG=chrome-os-partner:15694
TEST=tested manually

Change-Id: I62e12db3506947e757ea251a9f8915ffae6ac106
Signed-off-by: Luigi Semenzato <semenzato@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/37468
Reviewed-by: Darren Krahn <dkrahn@chromium.org>
Tested-by: Darren Krahn <dkrahn@chromium.org>
Reviewed-by: Mandeep Singh Baines <msb@chromium.org>

PM / Sleep: ramoops: Mark ramoops buffer as suspend volatile

If we don't have no_console_suspend and we have PSTORE_CONSOLE
defined then we might end up writing to the ramoops buffer during
suspend/resume.  Mark that memory as suspend volatile.

BUG=chrome-os-partner:15914
TEST=suspend_stress_test

Change-Id: I8e64c68938c3151ac4e96ed69d45c1ec40799134
Signed-off-by: Doug Anderson <dianders@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/37387
Reviewed-by: Jon Kliegman <kliegs@chromium.org>
Reviewed-by: Michael Spang <spang@chromium.org>

PM / Sleep: printk: Mark some printk globals as suspend volatile

This tags global variables in printk that are known to be touched
during suspend/resume.

BUG=chrome-os-partner:15914
TEST=suspend_stress_test

Change-Id: I393536bb4fb1b985b43fa1a3bec253ca81cb99b0
Signed-off-by: Doug Anderson <dianders@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/37386
Reviewed-by: Jon Kliegman <kliegs@chromium.org>
Reviewed-by: Michael Spang <spang@chromium.org>

PM / Sleep: Add the concept of suspend_volatile

In order to catch suspend/resume problems we would like to run a CRC
over most of memory before and after sleep and compare the results.
To do this we need to know which memory regions the suspend/resume
code touches and avoid running the CRC over those regions.

We will consider those regions to be "suspend volatile".  The CRC code
will skip chunks of memory tagged as suspend volatile to avoid falsely
reporting CRC errors.

We add the ability to mark global variables as suspend volatile
(keeping them in a special section) and also the ability to register
dynamic chunks of memory as volatile.

BUG=chrome-os-partner:15914
TEST=suspend_stress_test

Change-Id: I7964f5886d6b62fbca8d82986db20eae83ad0918
Signed-off-by: Doug Anderson <dianders@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/37385
Reviewed-by: Michael Spang <spang@chromium.org>
Reviewed-by: Jon Kliegman <kliegs@chromium.org>

addr_overlap: Add functions for checking address range overlap

These functions are simple but nice to have in one place.  Code
is based on in_region() from arch/arm/plat-samsung/pm-check.c.
There is similar code in memblock in memblock_addrs_overlap().

BUG=chrome-os-partner:15914
TEST=Use in future CLs.

Change-Id: I4faec4fb977215e0858bec8f0c0b988edb619457
Signed-off-by: Doug Anderson <dianders@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/37467
Reviewed-by: Jon Kliegman <kliegs@chromium.org>
Reviewed-by: Michael Spang <spang@chromium.org>

CHROMIUM: r8169.c: Fix WOL setting after resume from S3.

After resume from S3, register Config3 comes up as 0x60 (enable WOL),
regardless of previous state. Fix is to save previous state of Config3
and related WOL register bits and restore them upon resume.

BUG=chrome-os-partner:12129
TEST="ethtool eth0" after S3 resume and check that WOL is disabled. Also
verify that network card still functions as normal after resume.

Signed-off-by: Shawn Nematbakhsh <shawnn@chromium.org>
Change-Id: I2495b71029ccc30f406aea60cfd843b3cc6c711e
Reviewed-on: https://gerrit.chromium.org/gerrit/37143
Commit-Ready: Shawn Nematbakhsh <shawnn@google.com>
Reviewed-by: Shawn Nematbakhsh <shawnn@google.com>
Tested-by: Shawn Nematbakhsh <shawnn@google.com>

CHROMIUM: Input: cyapa - Remove capability array

It's not actually used for anything. It's just making our cyapa
structure 14 bytes larger.

Signed-off-by: Benson Leung <bleung@chromium.org>
BUG=chromium-os:21047
TEST=builds clean

Change-Id: I676fb04ce49172fba5a751a24219bd7c3f451d17
Reviewed-on: https://gerrit.chromium.org/gerrit/36410
Tested-by: Benson Leung <bleung@chromium.org>
Reviewed-by: Daniel Kurtz <djkurtz@chromium.org>
Commit-Ready: Benson Leung <bleung@chromium.org>

CHROMIUM: Input: cyapa - Set IRQF_ONESHOT for threaded irq

Starting with commit 1c6c69525b ("genirq: Reject bogus threaded irq requests")
threaded IRQs without a primary handler need to be requested with
IRQF_ONESHOT to make sure the irq isn't unmasked before the thread
is finished.

Signed-off-by: Benson Leung <bleung@chromium.org>
BUG=chromium-os:36125,chromium-os:21047
TEST=verify that the trackpad works, generates events when you move
a finger over it (ie, interrupts are working).

Change-Id: I2ff6294a09509da548624ca1229c5206c758cb7a
Reviewed-on: https://gerrit.chromium.org/gerrit/37493
Tested-by: Benson Leung <bleung@chromium.org>
Reviewed-by: Daniel Kurtz <djkurtz@chromium.org>
Commit-Ready: Benson Leung <bleung@chromium.org>

CHROMIUM: Input: cyapa - Clean up button capabilities

We don't need three boolean variables representing left, middle, right.
Lets just save to u8 btn_capability, and mask for the relevant bits
when we need to report left, middle, or right.

Signed-off-by: Benson Leung <bleung@chromium.org>
BUG=chromium-os:21047
TEST=builds clean
Test on butterfly system. Using a test image.
1. run evtest, select Cypress APA Trackpad (cyapa)
2. left click
3. Verify left click generates BTN_LEFT event
4. right click
5. Verify right click generates BTN_RIGHT event

Change-Id: I8814a63a2b45b9edd54a97a2677e7837d3b312d9
Reviewed-on: https://gerrit.chromium.org/gerrit/36450
Reviewed-by: Daniel Kurtz <djkurtz@chromium.org>
Commit-Ready: Benson Leung <bleung@chromium.org>
Tested-by: Benson Leung <bleung@chromium.org>

BACKPORT: r8169: enable internal ASPM and clock request settings

The following chips need to enable internal settings to let ASPM
and clock request work.

RTL8111E-VL, RTL8111F, RTL8411, RTL8111G
RTL8105, RTL8402, RTL8106

Signed-off-by: Hayes Wang <hayeswang@realtek.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
(cherry picked from commit d64ec841517a25f6d468bde9f67e5b4cffdc67c7)

Resolve conflicts as follows:

In the original CL, changes are made to many chip-specific configuration
functions, such as rtl_hw_start_8168e_2, rtl_hw_start_8402, etc. Our
source is missing some of these configuration functions and others do
not integrate cleanly. Therefore, only apply the changes to the hw_start
functions that integrate cleanly:

rtl_hw_start_8168e_2
rtl_hw_start_8105e_1

TEST=Confirm basic functionality on 8168e and 8105e cards.
BUG=chrome-os-partner:15923

Change-Id: Ib790aeb102920e127c43f74a9643e89afdffc251
Reviewed-on: https://gerrit.chromium.org/gerrit/37418
Reviewed-by: Sameer Nanda <snanda@chromium.org>
Commit-Ready: Shawn Nematbakhsh <shawnn@google.com>
Tested-by: Shawn Nematbakhsh <shawnn@google.com>

UPSTREAM: mwifiex: process RX packets in SDIO IRQ thread directly

ksdioirqd has higher priority than kworker. Process RX packets
in SDIO IRQ thread (ksdioirqd/mmcX) directly instead of deferring
the work to kworker to avoid the extra latency.
This improves TCP throughput 15~20% on an ARM platform with SDIO
2.0 controller.

BUG=chrome-os-partner:15867
TEST=TCP throughput improvements (TX 17%, RX 18%) are measured in
shieldroom with '*_CPU_GOVERNOR' set to 'performance' in
/etc/laptop-mode/conf.d/cpufreq.conf

Change-Id: I2de0020540d4b24e3084be21efa880daf186d43a
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Reviewed-on: https://gerrit.chromium.org/gerrit/37182
Reviewed-by: mukesh agrawal <quiche@chromium.org>
Reviewed-by: Sam Leffler <sleffler@chromium.org>

CHROMIUM: ramoops: Avoid calling time functions when timekeeping disabled

Calling the time functions when timekeeping is disabled will cause a
WARN_ON() to happen.  This appears to deadlock things.  Better to just
not call the function when we shouldn't.

BUG=chrome-os-partner:15655
TEST=suspend_stress_test
...ran with no serial console (AKA "console=" on kernel cmdline).

Change-Id: I1fa3430ea279c4bc79c3fb5e82228b380fecb60a
Signed-off-by: Doug Anderson <dianders@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/37364
Reviewed-by: Sameer Nanda <snanda@chromium.org>
Reviewed-by: Jon Kliegman <kliegs@chromium.org>
Reviewed-by: Kees Cook <keescook@chromium.org>

BACKPORT: module: add flags arg to sys_finit_module()

Thanks to Michael Kerrisk for keeping us honest.  These flags are actually
useful for eliminating the only case where kmod has to mangle a module's
internals: for overriding module versioning.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Lucas De Marchi <lucas.demarchi@profusion.mobi>
Acked-by: Kees Cook <keescook@chromium.org>
(backported from Rusty's modules-next tree)

BUG=None, keeping us in sync with upstream for this API
TEST=daisy build, manual testing
CQ-DEPEND=I5b5a9bee3f9e73eb076f01e8aacee51eff095c56

Change-Id: I40f0b55ff903dda411a04b088f4ca5b6ab5cf05e
Signed-off-by: Kees Cook <keescook@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/37278
Reviewed-by: Mandeep Singh Baines <msb@chromium.org>

UPSTREAM: r8169: 8168c and later require bit 0x20 to be set in Config2 for PME signaling.

The new 84xx stopped flying below the radars.

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
Cc: Hayes Wang <hayeswang@realtek.com>
(cherry picked from commit d387b427c973974dd619a33549c070ac5d0e089f)

Signed-off-by: Shawn Nematbakhsh <shawnn@google.com>
BUG=chrome-os-partner:14994
TEST=suspend/resume, then check ethernet function on 8168c

Change-Id: I978ff6a80e8575848094ffd85a8aab5cd80ec97a
Reviewed-on: https://gerrit.chromium.org/gerrit/37328
Tested-by: Shawn Nematbakhsh <shawnn@google.com>
Reviewed-by: Todd Broch <tbroch@chromium.org>
Commit-Ready: Shawn Nematbakhsh <shawnn@google.com>

UPSTREAM: r8169: Config1 is read-only on 8168c and later.

Suggested by Hayes.

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
Cc: Hayes Wang <hayeswang@realtek.com>
(cherry picked from commit 851e60221926a53344b4227879858bef841b0477)

Signed-off-by: Shawn Nematbakhsh <shawnn@google.com>
BUG=chrome-os-partner:14994
TEST=suspend/resume, then check ethernet function on 8168c

Change-Id: If48adfd0f35b7727fd0d76764c53291a197963f2
Reviewed-on: https://gerrit.chromium.org/gerrit/37327
Tested-by: Shawn Nematbakhsh <shawnn@google.com>
Reviewed-by: Todd Broch <tbroch@chromium.org>
Commit-Ready: Shawn Nematbakhsh <shawnn@google.com>

Revert "samsung: pm-check: Cleanup chunk exclude code"

This reverts commit a16583b2a0799fffc4f3df22a15a282f851e3735.

BUG=chrome-os-partner:15906
TEST=Run powerd_suspend, should suspend and resume successfully

Change-Id: I5ce64a89a045e1576c45249af2c95ff0cfea06fa
Signed-off-by: Simon Que <sque@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/37312
Reviewed-by: Doug Anderson <dianders@chromium.org>

HID: multitouch: Add 6673 PID for Ideacom

BUG=None
TEST=Boot kernel on machine with ideacom 6673 controller attached.
Verify usbhid module is loaded for the USB device.

Change-Id: I856f2386b889fb88dcd12300080d28626d0ab528
Signed-off-by: Dave Parker <dparker@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/37281
Reviewed-by: Bernie Thompson <bhthompson@chromium.org>

CHROMIUM: ARM: MALI: correct hw counter support locking

Fix locking in the hw counter support.  We cannot use a spinlock
to synchronize the collection work against start/stop because
kbase_instr_hwcnt_dump can block; switch to a mutex.  While here also
synchronize start work which can race in multiple ways with the other
uses of mali_hwcs.  Extend coverage of the mutex in the collection thread
for readability though it's not really needed as the event dispatch work
references only static data which is unaffected by polling being stopped.

While here unique-ify error returns.

Signed-off-by: sleffler@chromium.org
BUG=chrome-os-partner:14556
TEST=verify there are no longer console msgs of the form "BUG: spinlock wrong CPU"; also run concurrent start/stop loops from the shell

Change-Id: Ib74fc5c199212441e0e0ea312f7ca57da7c35278
Reviewed-on: https://gerrit.chromium.org/gerrit/36969
Reviewed-by: Sonny Rao <sonnyrao@chromium.org>
Reviewed-by: Michael Spang <spang@chromium.org>
Tested-by: Sam Leffler <sleffler@chromium.org>
Commit-Ready: Sam Leffler <sleffler@chromium.org>

mwifiex: add multi-queue support

This patch adds support for multiple TX queues inside mwifiex
driver. Four different queues according to WMM access categories
are defined for each virtual interface. When a packet is
received from netdev for transmission, tx pending count for
particular queue is incremented and if tx pending count has
reached upper water-mark, this queue is stopped instead of
stopping all queues. Similarly when a packet is successfully
transmitted from device, tx pending count is decremented per
queue and if pending count falls below lower water-mark, queue
operations are again resumed. This ensures that not all
tranmission is blocked if traffic with particular TOS value
suddenly increases.

Also wake all queues when association/IBSS_join happens to
enable traffic on all queues.

Signed-off-by: Avinash Patil <patila@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
BUG=chrome-os-partner:15536
TEST=pass pre-WiFi Cert. test case 5.2.30

Change-Id: I7b11c6f3f90943f09deaa7ef37b2975982bc02d3
Reviewed-on: https://gerrit.chromium.org/gerrit/36286
Reviewed-by: Paul Stewart <pstew@chromium.org>
Commit-Ready: Bing Zhao <bzhao@marvell.com>
Tested-by: Bing Zhao <bzhao@marvell.com>

CHROMIUM: drm/i915: Add backlight support for Link

This adds device-specific backlight support for Link, and also
enables adaptive backlight by default there.

BUG=chrome-os-partner:13276,chrome-os-partner:15248
TEST=by hand

Change-Id: I9ef546bba9f121657a653aa9cfc6a80bbde55cb0
Reviewed-on: https://gerrit.chromium.org/gerrit/36976
Reviewed-by: Daniel Erat <derat@chromium.org>
Commit-Ready: Stéphane Marchesin <marcheu@chromium.org>
Tested-by: Stéphane Marchesin <marcheu@chromium.org>

CHROMIUM: drm/i915: Remove adaptive backlight from gen6/lvds.

We'll only enable it through the adaptive_backlight device on
Link/gen7/eDP, so prune everything else.

BUG=chrome-os-partner:13276,chrome-os-partner:15248
TEST=by hand

Change-Id: Ib5633a370e736d9afbee441d968f718be814d081
Reviewed-on: https://gerrit.chromium.org/gerrit/36975
Reviewed-by: Daniel Erat <derat@chromium.org>
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Commit-Ready: Stéphane Marchesin <marcheu@chromium.org>
Tested-by: Stéphane Marchesin <marcheu@chromium.org>

CHROMIUM: drm/i915: Move the backlight accessor functions in dev_priv

This abstracts those functions, so that we can plug-in
platform-specific alternatives.

BUG=chrome-os-partner:3276,chrome-os-partner:15248
TEST=by hand: compiles, works

Change-Id: I29dd15722ffed8f5813b34bf0ce4431ce0524cc2
Reviewed-on: https://gerrit.chromium.org/gerrit/36974
Reviewed-by: Daniel Erat <derat@chromium.org>
Commit-Ready: Stéphane Marchesin <marcheu@chromium.org>
Tested-by: Stéphane Marchesin <marcheu@chromium.org>

CHROMIUM: gobi: Keep struct qcusbnet alive while cdev is alive

This fixes an issue where struct qcusbnet is freed while the embedded
character device is still alive.

This uses the ability of cdev to pin an arbitrary kobject by setting
cdev.kobj.parent. We have to switch from kref to kobject to use that.
This is arguably misuse of kobject, but char_dev requires it.t t

BUG=chrome-os-partner:15849
TEST=suspend_stress_test

Change-Id: I309e80c6ec91e491d30ea0bfcb062ae7e55f7198
Signed-off-by: Michael Spang <spang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/37135
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>

UPSTREAM: char_dev: pin parent kobject

In certain cases (for example when a cdev structure is embedded into
another object whose lifetime is controlled by a separate kobject) it is
beneficial to tie lifetime of another object to the lifetime of
character device so that related object is not freed until after
char_dev object is freed.

To achieve this let's pin kobject's parent when doing cdev_add() and
unpin when last reference to cdev structure is being released.

BUG=chrome-os-partner:15849
TEST=suspend_stress_test

Change-Id: I35d1762c94afa8b3eda6329a0bf0cfc7dde98b7b
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Acked-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Michael Spang <spang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/37134
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>

samsung: pm-check: Cleanup chunk exclude code

In a recent change to pm-check we added some more chunks of
memory to the list to exclude.  This cleans up how that works
a little bit:
* We now use a linker script to exclude printk buffers.  This has an
  advange of being generic (can add extra bits to exclusion without
  modifying pm-check) and also segregating printk buffers to their own
  chunk of memory.  If we do CRC on an 8K chunk we now won't exclude
  any extra memory when we exclude the printk buffers.
* Don't exclude *sleep_save_sp.  That was a physical address and is
  also already handled by current_thread_info().
* Don't exclude sleep_save_sp.  Instead: we zero it out before out
  code runs.  This is a transitory value and zeroing it is fine.  Now
  we won't exclude chunks of memory stored near sleep_save_sp.

This also has a few other minor fixes:
* Changes the in_region() function to fix some off-by-one errors and
  also to use physical addresses.
* Skips regions in s3c_pm_makecheck() too.
* Avoids goto in s3c_pm_runcheck() loop.

BUG=chrome-os-partner:15655
TEST=suspend_stress_test

Change-Id: Ic4febb1eb9aaa78d70984c149e26731af738727a
Signed-off-by: Doug Anderson <dianders@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/37031
Reviewed-by: Jon Kliegman <kliegs@chromium.org>
Reviewed-by: Michael Spang <spang@chromium.org>

drm/ptn3460: Add edid emulation selection

This patch adds edid selection to the ptn3460 driver so we can select
the appropriate display resolution from the device tree.

BUG=chrome-os-partner:11158
TEST=Tested on snow, no regressions detected

Change-Id: I19709a5b95d61061d415f3812dc7568d28e8b4ad
Signed-off-by: Sean Paul <seanpaul@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/36700
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>

CHROMIUM: exynos: snow: Add PTN edid emulation

Add an entry in the ptn3460 node to specify which edid emulation value
to use. The value corresponds to 1366x768, which is the resolution of
snow's lcd panel.

BUG=chrome-os-partner:11158
TEST=Tested on snow, verified EDID was 1366x768

Change-Id: Ie0cd390af41cef9a5eb51b1202c91ebc07d35458
Signed-off-by: Sean Paul <seanpaul@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/36699
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>

CHROMIUM: CONFIG: exynos: Use PTN3460 DRM driver

Change CONFIG_PTN3460 to CONFIG_DRM_PTN3460 so we use the PTN3460 driver
that is in gpu/drm/i2c

BUG=chrome-os-partner:11158
TEST=compiles, runs

Change-Id: I3591448c85ced88afb404b6cd4c5829eab87c3b9
Signed-off-by: Sean Paul <seanpaul@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/36792
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>

drm/i2c: Move ptn driver from auxdisplay to drm

In order to properly synchronize setting up the ptn bridge chip with the
displayport drm driver, we need to move ptn into drm.

The original hope was to keep ptn completely seperate from drm, and have
it just act like a regular monitor. Unfortunately, the bridge is
designed such that it cannot be configured until after it asserts
hotplug. This creates a race between the DP driver detecting hotplug and
reading edid, and the ptn driver configuring its edid emulation setting.

By moving it into the drm subsystem, we can hold off reading edid until
the bridge is properly set up.

BUG=chrome-os-partner:11158
TEST=Tested on snow, no regressions detected.

Change-Id: I99f696b4eeabf36f3055c3772482013397da5002
Signed-off-by: Sean Paul <seanpaul@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/36791
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>

CHROMIUM: drm/i915: Check the current edp backlight state before changing it.

This speeds up boot and suspend/resume times.

BUG=chrome-os-partner:13364
TEST=by hand

Change-Id: I8d7b7a22c50bcf32828d492e123ce301cefa297d
Reviewed-on: https://gerrit.chromium.org/gerrit/37047
Reviewed-by: Simon Que <sque@chromium.org>
Commit-Ready: Stéphane Marchesin <marcheu@chromium.org>
Tested-by: Stéphane Marchesin <marcheu@chromium.org>

CHROMIUM: mali-t6xx: Convert mali_dvfs_spinlock to IRQ-safe spinlock

This fixes the following potential deadlock reported by lockdep:

[    2.204853] =================================
[    2.209181] [ INFO: inconsistent lock state ]
[    2.213523] 3.4.0 #101 Not tainted
[    2.216906] ---------------------------------
[    2.221247] inconsistent {HARDIRQ-ON-W} -> {IN-HARDIRQ-W} usage.
[    2.227237] swapper/0/0 [HC1[1]:SC0[0]:HE0:SE1] takes:
[    2.232357]  (&((&mali_dvfs_spinlock))->rlock){?.+...}, at: [<402cef00>] kbase_platform_dvfs_event+0x28/0x54
[    2.242165] {HARDIRQ-ON-W} state was registered at:
[    2.247026]   [<4007a77c>] mark_lock+0x290/0x630
[    2.251626]   [<4007b17c>] __lock_acquire+0x660/0x180c
[    2.256747]   [<4007c8ac>] lock_acquire+0x104/0x128
[    2.261608]   [<40516464>] _raw_spin_lock+0x3c/0x4c
[    2.266469]   [<402cdf00>] kbase_platform_dvfs_init+0x64/0xb0
[    2.272198]   [<402ce17c>] kbase_platform_init+0x164/0x1cc
[    2.277666]   [<402ce200>] kbase_platform_exynos5_init+0x1c/0x124
[    2.283742]   [<402c7118>] kbasep_platform_device_init+0x48/0x54
[    2.289731]   [<402b9128>] kbase_device_init+0x44/0x400
[    2.294939]   [<402ca6ec>] kbase_platform_device_probe+0x1b4/0x6c8
[    2.301102]   [<402d56fc>] platform_drv_probe+0x24/0x28
[    2.306310]   [<402d42b4>] driver_probe_device+0xd8/0x214
[    2.311691]   [<402d4468>] __driver_attach+0x78/0x9c
[    2.316639]   [<402d28e4>] bus_for_each_dev+0x60/0x9c
[    2.321673]   [<402d3d88>] driver_attach+0x28/0x30
[    2.326448]   [<402d38f0>] bus_add_driver+0xc0/0x248
[    2.331395]   [<402d49d8>] driver_register+0xac/0x138
[    2.336429]   [<402d5a18>] platform_driver_register+0x54/0x68
[    2.342158]   [<40767a94>] kbase_driver_init+0xb4/0xd0
[    2.347279]   [<4000869c>] do_one_initcall+0xa0/0x174
[    2.352314]   [<4074db90>] kernel_init+0xf8/0x1b4
[    2.357001]   [<4000f240>] kernel_thread_exit+0x0/0x8
[    2.362036] irq event stamp: 41284
[    2.365421] hardirqs last  enabled at (41281): [<4000f304>] default_idle+0x34/0x40
[    2.372973] hardirqs last disabled at (41282): [<4000db74>] __irq_svc+0x34/0x60
[    2.380264] softirqs last  enabled at (41284): [<40032af8>] _local_bh_enable+0x1c/0x20
[    2.388163] softirqs last disabled at (41283): [<40033340>] irq_enter+0x50/0x7c
[    2.395454]
[    2.395455] other info that might help us debug this:
[    2.401964]  Possible unsafe locking scenario:
[    2.401967]
[    2.407866]        CPU0
[    2.410296]        ----
[    2.412727]   lock(&((&mali_dvfs_spinlock))->rlock);
[    2.417674]   <Interrupt>
[    2.420278]     lock(&((&mali_dvfs_spinlock))->rlock);
[    2.425399]
[    2.425401]  *** DEADLOCK ***
[    2.425403]
[    2.431302] no locks held by swapper/0/0.
[    2.435295]
[    2.435296] stack backtrace:
[    2.439642] [<40015580>] (unwind_backtrace+0x0/0xec) from [<4050e2ec>] (dump_stack+0x20/0x24)
[    2.448146] [<4050e2ec>] (dump_stack+0x20/0x24) from [<4050f5ac>] (print_usage_bug+0x250/0x2b8)
[    2.456826] [<4050f5ac>] (print_usage_bug+0x250/0x2b8) from [<4007a844>] (mark_lock+0x358/0x630)
[    2.465593] [<4007a844>] (mark_lock+0x358/0x630) from [<4007b0f8>] (__lock_acquire+0x5dc/0x180c)
[    2.474359] [<4007b0f8>] (__lock_acquire+0x5dc/0x180c) from [<4007c8ac>] (lock_acquire+0x104/0x128)
[    2.483387] [<4007c8ac>] (lock_acquire+0x104/0x128) from [<40516464>] (_raw_spin_lock+0x3c/0x4c)
[    2.492154] [<40516464>] (_raw_spin_lock+0x3c/0x4c) from [<402cef00>] (kbase_platform_dvfs_event+0x28/0x54)
[    2.501876] [<402cef00>] (kbase_platform_dvfs_event+0x28/0x54) from [<402c6498>] (dvfs_callback+0x28/0x74)
[    2.511511] [<402c6498>] (dvfs_callback+0x28/0x74) from [<400514ec>] (__run_hrtimer+0x144/0x220)
[    2.520276] [<400514ec>] (__run_hrtimer+0x144/0x220) from [<400522f4>] (hrtimer_interrupt+0x120/0x248)
[    2.529566] [<400522f4>] (hrtimer_interrupt+0x120/0x248) from [<400244d8>] (exynos4_mct_tick_isr+0x58/0x60)
[    2.539289] [<400244d8>] (exynos4_mct_tick_isr+0x58/0x60) from [<40091968>] (handle_irq_event_percpu+0x8c/0x248)
[    2.549442] [<40091968>] (handle_irq_event_percpu+0x8c/0x248) from [<40091b70>] (handle_irq_event+0x4c/0x6c)
[    2.559251] [<40091b70>] (handle_irq_event+0x4c/0x6c) from [<40094b50>] (handle_fasteoi_irq+0xe4/0x150)
[    2.568624] [<40094b50>] (handle_fasteoi_irq+0xe4/0x150) from [<40091110>] (generic_handle_irq+0x3c/0x50)
[    2.578173] [<40091110>] (generic_handle_irq+0x3c/0x50) from [<4000efc4>] (handle_IRQ+0x88/0xc8)
[    2.586939] [<4000efc4>] (handle_IRQ+0x88/0xc8) from [<40008564>] (gic_handle_irq+0x44/0x68)
[    2.595358] [<40008564>] (gic_handle_irq+0x44/0x68) from [<4000db84>] (__irq_svc+0x44/0x60)
[    2.603688] Exception stack(0x40797f18 to 0x40797f60)
[    2.608723] 7f00:                                                       00000001 00000004
[    2.616883] 7f20: 00000000 00000000 40796000 4080a1c8 4051a564 407be97c 4000406a 410fc0f4
[    2.625042] 7f40: 00000000 40797f6c 40797f30 40797f60 4007d1b0 4000f308 20000013 ffffffff
[    2.633204] [<4000db84>] (__irq_svc+0x44/0x60) from [<4000f308>] (default_idle+0x38/0x40)
[    2.641363] [<4000f308>] (default_idle+0x38/0x40) from [<4000f570>] (cpu_idle+0xb4/0x10c)
[    2.649523] [<4000f570>] (cpu_idle+0xb4/0x10c) from [<404fccac>] (rest_init+0xb4/0xdc)
[    2.657421] [<404fccac>] (rest_init+0xb4/0xdc) from [<4074d9e0>] (start_kernel+0x46c/0x524)
[    2.665753] [<4074d9e0>] (start_kernel+0x46c/0x524) from [<40008044>] (0x40008044)
[

BUG=chrome-os-partner:15688
TEST=boot with CONFIG_PROVE_LOCKING, check dmesg for warning

Change-Id: Ie2c9fa5b57de6d8f9769d5969142ad7b5ede6440
Signed-off-by: Michael Spang <spang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/36705
Reviewed-by: Mandeep Singh Baines <msb@chromium.org>
Reviewed-by: Sam Leffler <sleffler@chromium.org>

CHROMIUM: USB: ohci-exynos: initialize registers pointer earlier

In the former code, we have a race condition between the first interrupt
and the regs field initilization in the usb_hcd structure.
If the OHCI irq fires before hcd->regs is set, we are getting a null
pointer dereference in ohci_irq.

When calling usb_add_hcd(), it first executes the reset() callback,
then enables the ohci interrupt, and finally executes the start()
callback. So moving the ohci_init() call which actually initializes the
reg field from start() to reset() should remove the race.

Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
BUG=chrome-os-partner:15731
TEST=run on Spring with USB2.0 on internal PHY on HSIC hub initialized
in the bootloader, and see no kernel panic at startup.

Change-Id: If98070a3cd16b7f735c8cfb62c20cebf410953d2
Reviewed-on: https://gerrit.chromium.org/gerrit/36823
Reviewed-by: Doug Anderson <dianders@chromium.org>
Commit-Ready: Vincent Palatin <vpalatin@chromium.org>
Tested-by: Vincent Palatin <vpalatin@chromium.org>

CHROMIUM: samsung: Enable suspend memory check only on old bios

The bios bug only exists in 2695.90.0 so change default to disabled
and only turn on if the RO bios version is Google_Snow.2695.90.0

Also fix a bug where we don't check if crcs are allocated before trying
to access them on resume.

BUG=chrome-os-partner:15811
TEST=deployed on 2695.90.0 bios, 2695.90.2 and unversioned (dev) bios
  Observed /sys/module/pm_check/paramaters/pm_check_enabled set correctly
  Ran power_Resume test on both devices, saw 1 second difference between them.
  Set CONFIG_SAMSUNG_PM_CHECK=n, build, deploy and suspend/resume
Signed-off-by: Jonathan Kliegman <kliegs@chromium.org>
Change-Id: I7eb89ffc7cbd5c6318d02f21d2e598dc383a68b3
Reviewed-on: https://gerrit.chromium.org/gerrit/37032
Reviewed-by: Doug Anderson <dianders@chromium.org>
Commit-Ready: Jon Kliegman <kliegs@chromium.org>
Tested-by: Jon Kliegman <kliegs@chromium.org>
Reviewed-by: Michael Spang <spang@chromium.org>

ath9k: do not start rx poll work when flushing pending rx frames

__ath_cancel_work has stopped the rx poll work timer, and it needs to
stay stopped until after the reset

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
BUG=chrome-os-partner:15169
TEST=Boot, associate

Change-Id: If65461674dd41507891c0a66f188276a614c6b64
Signed-off-by: Paul Stewart <pstew@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/36829
Reviewed-by: Wade Guthrie <wdg@google.com>
Reviewed-by: Christopher Wiley <wiley@chromium.org>

CHROMIUM: exynos: dts: configure hotplug pin on Spring

Spring does not use the dedicated DP hotplug detection pin but a
standard GPIO.

Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
BUG=chrome-os-partner:14491
TEST=boot on Spring, see the display coming up automatically.

Change-Id: I716b8d0a070503d3d94bfb995d7d7ed2427de41d
Reviewed-on: https://gerrit.chromium.org/gerrit/36827
Commit-Ready: Vincent Palatin <vpalatin@chromium.org>
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Tested-by: Vincent Palatin <vpalatin@chromium.org>

CHROMIUM: drm: exynos: optional DP hotplug on a GPIO

Allow to setup the Display Port hotplug detection on a vanilla GPIO
instead of the dedicated DP_HPD pin.

Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
BUG=chrome-os-partner:14491
TEST=boot on Spring, see the display coming up automatically.

Change-Id: I8eaba89bff0fcdc56d05abf769d7097df05807e1
Reviewed-on: https://gerrit.chromium.org/gerrit/36825
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>
Commit-Ready: Vincent Palatin <vpalatin@chromium.org>
Tested-by: Vincent Palatin <vpalatin@chromium.org>

CHROMIUM: exynos: add DP hotplug gpio configuration

Allow to define an optional gpio in DP platform data used for
DP hotplug detection if the machine is not using the dedicated DP_HPD
pin.

Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
BUG=chrome-os-partner:14491
TEST=build and run on Spring and Snow.

Change-Id: Ib1a552b369bceffd739e7d09aeea70cbc80785df
Reviewed-on: https://gerrit.chromium.org/gerrit/36826
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>
Commit-Ready: Vincent Palatin <vpalatin@chromium.org>
Tested-by: Vincent Palatin <vpalatin@chromium.org>

CHROMIUM: Input: atmel_mxt_ts - make mxt_initialize async

mxt_probe() calles mxt_initialize() to initialize the device, which includes
a soft reset and then msleep for the reset to finish. This has big impact on
the system boot time. This patch makes the mxt_initizlize() call async to
reduce the system boot time.

BUG=chrome-os-partner:15743
TEST=Boot the device and check the kernel timestamp in dmesg to see that
     the device initialization is parallelized.

Signed-off-by: Yufeng Shen <miletus@chromium.org>
Change-Id: If106af37a52a0fa874cdc8255c91fdde36776e1f
Reviewed-on: https://gerrit.chromium.org/gerrit/36964
Reviewed-by: Benson Leung <bleung@chromium.org>
Tested-by: Simon Que <sque@chromium.org>
Commit-Ready: Yufeng Shen <miletus@chromium.org>

CHROMIUM: avoid deadlock in OOM killer

This removes code that prevents a memory starvation in low-memory situations.

select_bad_process may fail to find a victim for the OOM-kill by returning
ERR_PTR(-1).  In theory this should happen only when there is a guarantee
that memory will be freed shortly.  But in some cases this is not true.
If any process tries to allocate memory between setting the PF_EXITING
bit of p->flags and setting p->exit_state to non-zero, it prevents
the OOM-killer from making any progress, and nobody is able to
allocate memory.

I have found a process that does exactly that:

[ 4687.418818]  [<8104512d>] __cond_resched+0x1b/0x2b
[ 4687.418828]  [<813b67a7>] _cond_resched+0x18/0x21
[ 4687.418840]  [<81093940>] shrink_slab+0x224/0x22f
[ 4687.418856]  [<81095a96>] try_to_free_pages+0x1b7/0x2e6
[ 4687.418868]  [<8108df2a>] __alloc_pages_nodemask+0x40a/0x61f
[ 4687.418882]  [<810a9dbe>] read_swap_cache_async+0x4a/0xcf
[ 4687.418894]  [<810a9ea4>] swapin_readahead+0x61/0x8d
[ 4687.418906]  [<8109fff4>] handle_pte_fault+0x310/0x5fb
[ 4687.418918]  [<810a0420>] handle_mm_fault+0xae/0xbd
[ 4687.418932]  [<8101d0f9>] do_page_fault+0x265/0x284
[ 4687.419002]  [<813b7887>] error_code+0x67/0x6c
[ 4687.419060]  [<8102351d>] mm_release+0x1d/0xc3
[ 4687.419070]  [<81026ce9>] exit_mm+0x1d/0xe9
[ 4687.419090]  [<81028082>] do_exit+0x19b/0x640

mm_release gets its page fault in the vicinity of this
code which is related to futexes:

if (unlikely(tsk->robust_list)) {
exit_robust_list(tsk);
tsk->robust_list = NULL;
}

Since robust_list is a userspace structure, the page
fault looks legitimate, and this is likely a design bug
(also see comment about deadlocks earlier in select_bad_process)
and difficult to fix completely.

In any case we're happy to trade spurious OOM kills for no hangs.

BUG=chromium-os:32321
TEST=tested with a load that reliably causes a hang before and none after

Change-Id: I7037e68cc3eef3a36ca355b9535af0f559b3a148
Signed-off-by: Luigi Semenzato <semenzato@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/36953
Reviewed-by: Mandeep Singh Baines <msb@chromium.org>
Reviewed-by: Sonny Rao <sonnyrao@chromium.org>

CHROMIUM: config: enable mali hw counter support

Enable support for collecting mali hw counter data via trace events.

Signed-off-by: sleffler@chromium.org
BUG=chrome-os-partner:14556
TEST=boot on snow and collect data

Change-Id: I6b905f2469004238d2e0c7e1fc6c8362d0416f71
Reviewed-on: https://gerrit.chromium.org/gerrit/36565
Tested-by: Sam Leffler <sleffler@chromium.org>
Reviewed-by: Sonny Rao <sonnyrao@chromium.org>
Commit-Ready: Sam Leffler <sleffler@chromium.org>

CHROMIUM: ASoC: samsung: Fix warnings in daisy_max98095

This fixes the following build warnings:

sound/soc/samsung/daisy_max98095.c: In function 'daisy_init':
sound/soc/samsung/daisy_max98095.c:407:9: warning: assignment makes pointer from integer without a cast
sound/soc/samsung/daisy_max98095.c: In function 'daisy_resume_post':
sound/soc/samsung/daisy_max98095.c:437:1: warning: no return statement in function returning non-void

BUG=none
TEST=build

Change-Id: I287ddfe7f48afd2bb519af3b3ec1ae10ee7f60e8
Signed-off-by: Michael Spang <spang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/36912
Reviewed-by: Dylan Reid <dgreid@chromium.org>

CHROMIUM: block: Fix warning in disk_block_events

This fixes the following warning due to type mismatch for
spin_lock_irqsave() flags.

block/genhd.c: In function 'disk_unblock_events':
block/genhd.c:1504:79: warning: comparison of distinct pointer types lacks a cast

BUG=none
TEST=build

Change-Id: Ia692f29445053eef05e94a8b3e8df9a8b59d193f
Signed-off-by: Michael Spang <spang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/36911
Reviewed-by: Paul Taysom <taysom@chromium.org>

ARM: MALI: connect kbase trace events to system tracing

Adapt kbase's trace event mechanism to system trace events.  This facility
is enabled with a new CONFIG_MALI_SYSTEM_TRACE config knob (hidden under
CONFIG_MALI_EXPERT).

Signed-off-by: sleffler@chromium.org
BUG=chrome-os-partner:14556
TEST=verify kbase trace events are generated when enabled

Change-Id: Ic5df3133619c52ae52e9bcd7264494b892e47c86
Reviewed-on: https://gerrit.chromium.org/gerrit/35994
Commit-Ready: Sam Leffler <sleffler@chromium.org>
Reviewed-by: Sam Leffler <sleffler@chromium.org>
Tested-by: Sam Leffler <sleffler@chromium.org>

acpi: thermal: initialize tz_enabled to 1

In the acpi_thermal_add path, acpi_thermal_get_info gets called before
acpi_thermal_register_thermal_zone.  Since tz_enabled was getting set to
1 only in acpi_thermal_register_thermal_zone, acpi_thermal_get_info
ended up disabling thermal polling.

Moved setting of tz_enabled to 1 into acpi_thermal_add itself.

BUG=chrome-os-partner:15697
TEST=on kiev, run CPU soaker threads ("while true; do true; done") and ensure
that the fan turns on as the CPU heats up.

Change-Id: Ib2ad69621ce32f252a8b913387e1560d0750b822
Signed-off-by: Sameer Nanda <snanda@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/36927
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
Reviewed-by: Bernie Thompson <bhthompson@chromium.org>
Tested-by: Bernie Thompson <bhthompson@chromium.org>

CHROMIUM: ARM: MALI: add hardware performance counter support

Add support for dispatching hardware performance counter data through
system trace events.  To enable a counter the associated trace event must
be enabled before turning on hardware counter polling through sysfs.
Data are collected at each vblank event.  Note the set of counters
collected is fixed at the time polling is enabled; to change the set of
counters first disable polling.

This functionality is enabled by CONFIG_MALI_HWC_TRACE.

Signed-off-by: sleffler@chromium.org
BUG=chrome-os-partner:14556
TEST=enable events (e.g. mali_hwc_FRAG_TRANS_ELIM); cat trace_pipe&; echo 1 > /sys/devices/platform/mali.0/hwc_enable; monitor trace data

Change-Id: I801fbfbd87abe69ba5832a1720532ee331fb0314
Reviewed-on: https://gerrit.chromium.org/gerrit/36564
Reviewed-by: Sonny Rao <sonnyrao@chromium.org>
Reviewed-by: Anush Elangovan <anush@chromium.org>
Commit-Ready: Sam Leffler <sleffler@chromium.org>
Tested-by: Sam Leffler <sleffler@chromium.org>

CHROMIUM: drm: add vblank notifier

Add a notifier for vblank events.

Signed-off-by: sleffler@chromium.org
BUG=chrome-os-partner:14556
TEST=collect hardware counter data (dependent cl)

Change-Id: I314186947f9c93b46d3729ffdd5b8f0b0317b88f
Reviewed-on: https://gerrit.chromium.org/gerrit/36781
Reviewed-by: Anush Elangovan <anush@chromium.org>
Commit-Ready: Sam Leffler <sleffler@chromium.org>
Tested-by: Sam Leffler <sleffler@chromium.org>

CHROMIUM: config: exynos: turn on CONFIG_SAMSUNG_PM_CHECK

Enable checking memory on suspend/resume cycles.

BUG=chrome-os-partner:15600
TEST=Ensure /sys/modules/pm_check exists
  Full pm_check functionality tested in CL 36475
Signed-off-by: Jonathan Kliegman <kliegs@chromium.org>
Change-Id: If13496c68ff97a0d6335d9dbfa9ad58b03af377c
Reviewed-on: https://gerrit.chromium.org/gerrit/36575
Tested-by: Jon Kliegman <kliegs@chromium.org>
Reviewed-by: Doug Anderson <dianders@chromium.org>
Commit-Ready: Jon Kliegman <kliegs@chromium.org>

HACK: samsung: pm-check: Add runtime options to enable and panic

Add runtime options to enable, panic and switch between xor and crc
checks of memory.

Update the exclusions to ignore log_buf and sleep_save_sp

BUG=chrome-os-partner:15600
TEST=Tested that with pm_check_should_panic system panics after an
  error and continues running when not set
  Set and unset pm_check_use_xor - confirmed differences in time and
    difference in values for errors
  Set and unset pm_check_enabled, validated log messages for running or not
  Chanaged pm_check_chunksize, saw size of crc error blocks change
  Confirmed correct error messages with pm_check_printskips
  Confirmed suspend_stress_test still catches errors correctly

Change-Id: I4275782391de32378be2f717bd48852ffde634ce
Signed-off-by: Jonathan Kliegman <kliegs@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/36475
Reviewed-by: Michael Spang <spang@chromium.org>
Reviewed-by: Doug Anderson <dianders@chromium.org>

CHROMIUM: add stack trace on gpu reset

Add a stack trace for every gpu reset.  This is temporary
for helping with problems in the field.

Signed-off-by: sleffler@chromium.org
BUG=chrome-os-partner:1556
TEST=force gpu reset and verify a stack trace is dumped to the console

Change-Id: I4543367a2bf7321a292bb585bd46e7dfe92cb6e0
Reviewed-on: https://gerrit.chromium.org/gerrit/36498
Reviewed-by: Anush Elangovan <anush@chromium.org>
Commit-Ready: Sam Leffler <sleffler@chromium.org>
Tested-by: Sam Leffler <sleffler@chromium.org>

CHROMIUM: iio: isl29018: Support suspend and resume.

The driver leaves the device in power-down state anyway,
so there is nothing to do on suspend.
On resume, we just have to make sure the range and ADC
values are updated in the device since it may have been
powered down in suspend.

BUG=chrome-os-partner:9494
TEST=Verify that lux values are generally unchanged after suspend/resume.
This is a bit difficult to reproduce on snow because we have to simulate
a FET5 power-down on suspend by manually hitting the TPS.  We also have
to modify a range or resolution value that gets pushed to the device, and
we do not normally do that.  This can reproduce the problem:
cd /sys/bus/iio/devices/*
cat in_illuminance0_input
echo 4000 > range
cat in_illuminance0_input # The value changes just a little bit.
i2cset -f -y 4 0x48 0x13 0x02 # Turn off FET5
i2cset -f -y 4 0x48 0x13 0x1f # Turn FET5 back on
powerd_suspend
cat in_illuminance0_input # The value is 4x expected without this patch.

Change-Id: I73e9f9357db3ae3c240a6e1cc5c8acbcf97b6971
Signed-off-by: Bryan Freed <bfreed@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/31662
Reviewed-by: Benson Leung <bleung@chromium.org>

CHROMIUM: Input: cyapa - Add debug messages for bl_exit failure cases.

Print the contents of the three status bytes before concluding
that the current state is CYAPA_STATE_BL_IDLE.

Print these in bl_exit error path after setting
cyapa->debug on the failure to exit bootloader in issue 14206.

Also, fix the debug message for "bl_activate" error path
to be correct. Previously it said "bl_exit"

Signed-off-by: Benson Leung <bleung@chromium.org>
BUG=chrome-os-partner:14206
TEST=Boot the system, and then force a firmware update using
/opt/google/touchpad/firmware/chromeos-touch-firmwareupdate.sh -d cyapa -f
Then, shut the system down to interrupt the firmware update.
Boot the system the next time.
Check dmesg after the recovery firmware update is complete.
Verify that the following appears :
[    0.863010] cyapa 1-0067: bl_exit failed. Now in state BL_IDLE.
[    0.863019] cyapa 1-0067: status[REG_OP_STATUS] = 0x00
[    0.863031] cyapa 1-0067: status[REG_BL_STATUS] = 0x10
[    0.863038] cyapa 1-0067: status[REG_BL_ERROR] = 0x00
[    0.863045] cyapa 1-0067: failed to bl_exit. -11
[    0.863052] cyapa 1-0067: device detected, but not operational, -11

Change-Id: I0680fd06164306f521f052c3027c0389dc501cb6
Reviewed-on: https://gerrit.chromium.org/gerrit/35791
Reviewed-by: Daniel Kurtz <djkurtz@chromium.org>
Commit-Ready: Benson Leung <bleung@chromium.org>
Tested-by: Benson Leung <bleung@chromium.org>

drm/exynos: vidi: Fix compilation error re: pitch

Was getting "'struct exynos_drm_overlay' has no member named 'pitch'"
compilation errors when compiling with VIDI enabled.

BUG=None
TEST=Compiles

Change-Id: I375ba229b538b25601745d01cf1c70dcd751d275
Signed-off-by: Sean Paul <seanpaul@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/36476

CHROMIUM: i915: do not set redundant DP dpms mode

Add a check to intel_dp_dpms() to skip setting the DP's DPMS mode if
the current mode is the same as the new one.

BUG=chrome-os-partner:13364,chrome-os-partner:12423
TEST=Run power_Resume autotest with screen on and screen off on Link.
The resume time should be about the same in each case.

Change-Id: I4269c84bdbd7101b1f71b9fce9935901df917381
Signed-off-by: Simon Que <sque@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/36356
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>

UPSTREAM: cfg80211/mac80211: avoid state mishmash on deauth

Avoid situation when we are on associate state in mac80211 and
on disassociate state in cfg80211. This can results on crash
during modules unload (like showed on this thread:
http://marc.info/?t=134373976300001&r=1&w=2) and possibly other
problems.

Reported-by: Pedro Francisco <pedrogfrancisco@gmail.com>
Cc: stable@vger.kernel.org
Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Paul Stewart <pstew@chromium.org>
BUG=chrome-os-partner:14187
TEST=Boot, run, suspend and resume

Change-Id: Ia0fb26d9c5f4df84d460a7b236d9112998ec08ee
Reviewed-on: https://gerrit.chromium.org/gerrit/36438
Reviewed-by: mukesh agrawal <quiche@chromium.org>
Commit-Ready: Paul Stewart <pstew@chromium.org>
Tested-by: Paul Stewart <pstew@chromium.org>

CHROMIUM: exynos: dts: configure SDIO for Wifi

The Wifi SDIO card is on the MMC1 interface,
and the MMC3 interface is no longer connected to anything.

Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
BUG=chrome-os-partner:14489
TEST=run on Spring and attach to the guest wifi network from the UI

Change-Id: I42b2a96c67b61a6f220e12c1bd1242d7919c5e9a
Reviewed-on: https://gerrit.chromium.org/gerrit/36301
Commit-Ready: Vincent Palatin <vpalatin@chromium.org>
Tested-by: Vincent Palatin <vpalatin@chromium.org>
Reviewed-by: Sam Leffler <sleffler@chromium.org>

CHROMIUM: exynos5: enable/reset Wifi module on Spring

The Wifi SDIO module is controlled by the same GPIOs as Daisy and Snow.

Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
BUG=chrome-os-partner:14490
TEST=run on Spring and attach to the guest wifi network from the UI

Change-Id: I6b0e32c09b25f9b279eb57205a098196753882c5
Reviewed-on: https://gerrit.chromium.org/gerrit/36690
Reviewed-by: Sam Leffler <sleffler@chromium.org>
Commit-Ready: Vincent Palatin <vpalatin@chromium.org>
Tested-by: Vincent Palatin <vpalatin@chromium.org>

CHROMIUM: Input: cyapa - Disable irq before changing power modes

There could be a potential race condition here as the irq may
be active as we are trying to read and write to device registers
to change power modes. In order to resolve this, disable irq
before cyapa_set_power_mode in cyapa_suspend.

BUG=chromium-os:35663
TEST=Builds clean.
Check that suspend with power/wakeup enabled and disabled
both work correctly to disable and enable the touchpad wake.
1. echo disabled > /sys/bus/i2c/devices/1-0067/power/wakeup
2. powerd_suspend
3. check that trackpad cannot wake the system.
4. Wake using keyboard. Check mouse is functional.
5. echo enabled > /sys/bus/i2c/devices/1-0067/power/wakeup
6. powerd_suspend
7. check that trackpad can wake the system.
8. Check that the trackpad is operational after resume

Change-Id: Ifd8715c51ddea138c5c4c28b4a4337af8d6021cc
Signed-off-by: Benson Leung <bleung@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/36409

drm/exynos: mixer: support for 800x600 resolution

Extend the mixer configuration to include all resolutions that can be
generated by the mixer. This patch adds 800x600 resolution with the
help of workaround 3.
Workaround summary:
1. Set the resolution mode as HD_1080.
2. Set HD_1080 and all kinds of destination X with biasing 0x20.
3. Start display as 3D one path mode, wait 1 frame and change
   to 3D two path mode.

BUG=chrome-os-partner:12643
TEST=Tested on snow with mentioned resolution

Change-Id: Ib653291acaf5f2f0c5d4545b876223614be642c5
Signed-off-by: Akshay Saraswat <Akshay.s@samsung.com>
Reviewed-on: https://gerrit.chromium.org/gerrit/34746
Tested-by: Akshay Saraswat <akshay.s@samsung.com>
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Commit-Ready: Akshay Saraswat <akshay.s@samsung.com>

drm/exynos: mixer: support for 1440x900 resolution

Extend the mixer configuration to include all resolutions that can be
generated by the mixer. This patch adds 1440x900 resolution with the
help of workaround 4.
Workaround summary:
1. Set the resolution mode as HD_1080.
2. Set HD_1080 and all kinds of destination X with biasing 0xE0.

BUG=chrome-os-partner:12643
TEST=Tested on snow with mentioned resolution

Change-Id: Iaa0cb9ddfe2995a760b4c72a724807c1b6a2031c
Signed-off-by: Akshay Saraswat <Akshay.s@samsung.com>
Reviewed-on: https://gerrit.chromium.org/gerrit/33860
Tested-by: Akshay Saraswat <akshay.s@samsung.com>
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Commit-Ready: Akshay Saraswat <akshay.s@samsung.com>

UPSTREAM: mwifiex: correction in MSDU padding logic

Padding arithmetic will always work for MSDUs provided first MSDU
ends on 4-byte boundary. Fixing it by making sure that all MSDU ends
on 4-byte boundary.

Signed-off-by: Yogesh Ashok Powar <yogeshp@marvell.com>
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Kiran Divekar <dkiran@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: Paul Stewart <pstew@chromium.org>
(cherry picked from commit bda1b1b77918ea340bbfb2938a0fd7d5203e430c)

BUG=chrome-os-partner:14732
TEST=Over-the-air TCP bitrate increases 14% / 11% downlink/uplink

Change-Id: I7c038453504bb399473f9f3cf0c350fbb2a4094f
Reviewed-on: https://gerrit.chromium.org/gerrit/35914
Tested-by: Paul Stewart <pstew@chromium.org>
Reviewed-by: Bing Zhao <bzhao@marvell.com>
Reviewed-by: mukesh agrawal <quiche@chromium.org>
Commit-Ready: Paul Stewart <pstew@chromium.org>

CHROMIUM: Input: cyapa - Fix warning in cyapa_bl_enter

Fix a warning introduced by https://gerrit.chromium.org/gerrit/#/c/35792.
Used a %d instead of a %s.

Signed-off-by: Benson Leung <bleung@chromium.org>
BUG=chrome-os-partner:14206,chromium-os:35682
TEST=Builds clean.
No warnings when building cyapa.

Change-Id: Ib89e10d60639725098fb6d566c39b0e58f5e8998
Reviewed-on: https://gerrit.chromium.org/gerrit/36452
Reviewed-by: Daniel Kurtz <djkurtz@chromium.org>
Commit-Ready: Benson Leung <bleung@chromium.org>
Tested-by: Benson Leung <bleung@chromium.org>

UPSTREAM: e1000e: add device IDs for i218

i218 is the next-generation LOM that will be available on systems with the
Lynx Point LP Platform Controller Hub (PCH) chipset from Intel.  This patch
provides the initial support of those devices.

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 16e310ae6ed352c4963b1f2413fcd88fa693eeda.)
Signed-off-by: Josh Triplett <josh@joshtriplett.org>
BUG=None
TEST=Build and run on a system with an i218 Ethernet controller; verify
     connectivity.

Change-Id: I3676d75ea99e893f69be0b7acc2a14e383d2f9c2
Reviewed-on: https://gerrit.chromium.org/gerrit/36441
Commit-Ready: Josh Triplett <josh@joshtriplett.org>
Tested-by: Josh Triplett <josh@joshtriplett.org>
Reviewed-by: Olof Johansson <olofj@chromium.org>

CHROMIUM: Input : atkbd - Map standard F14, F15 scancode to keycode 184
(KEY_F14) and 185

Signed-off-by: agnescheng <agnescheng@chromium.org>
BUG=chrome-os-partner:14339

TEST=Stout EC maps PrtSC to F14, Fn to F15.
Test on stout. Run evtest and get below result.

(Press PrtSc)
Event: time 1350876553.697803, type 4 (EV_MSC), code 4 (MSC_SCAN), value
65
Event: time 1350876553.697812, type 1 (EV_KEY), code 184 (KEY_F14),
value 1

(Press Fn)
Event: time 1350876555.477358, type 4 (EV_MSC), code 4 (MSC_SCAN), value
66
Event: time 1350876555.477367, type 1 (EV_KEY), code 185 (KEY_F15),
value 1

Change-Id: Ie5161d5952d98b6db24571db5152b02f856d2d4e
Reviewed-on: https://gerrit.chromium.org/gerrit/36420
Tested-by: Agnes Cheng <agnescheng@chromium.org>
Reviewed-by: Benson Leung <bleung@chromium.org>
Commit-Ready: Agnes Cheng <agnescheng@chromium.org>

CHROMIUM: Input: cyapa - remove unused includes

Signed-off-by: Benson Leung <bleung@chromium.org>
BUG=chromium-os:21047
TEST=builds clean

Change-Id: Ie0e19481dcab432f97e6ba31a45880230dc3b33c
Reviewed-on: https://gerrit.chromium.org/gerrit/36407
Reviewed-by: Daniel Kurtz <djkurtz@chromium.org>
Commit-Ready: Benson Leung <bleung@chromium.org>
Tested-by: Benson Leung <bleung@chromium.org>

CHROMIUM: Input: cyapa - get_state at the beginning of cyapa_bl_enter

Rather than trusting the cached value of cyapa->state at the beginning
of cyapa_bl_enter, get state before deciding to return 0 from bl_enter.

If bl_exit failed in early probe, but the trackpad entered operational
mode later, then we may have a discrepancy in state between trackpad
and driver. This will make sure we sync state before we try to
write bytes to the trackpad in bl_activate.

On i2c cypress trackpads, this could result in the bl_activate array
being written into the operational firmware's registers, causing the
trackpad to stop functioning. Instead, this will successfully
update the trackpad firmware.

Signed-off-by: Benson Leung <bleung@chromium.org>
BUG=chrome-os-partner:14206
TEST=Run a forced firmware update.
/opt/google/touchpad/firmware/chromeos-touch-firmwareupdate.sh -d cyapa -f
Verify that the trackpad is functional afterward and that there are no
error messages in dmesg regarding failed firmware update -11.

Change-Id: Ifaf483f30e3f4788d9ef42d023db54cec5c3a44f
Reviewed-on: https://gerrit.chromium.org/gerrit/35792
Commit-Ready: Benson Leung <bleung@chromium.org>
Reviewed-by: Benson Leung <bleung@chromium.org>
Tested-by: Benson Leung <bleung@chromium.org>

UPSTREAM: staging/gdm72xx: sdio_boot: replace firmware upgrade API

Replace firmware upgrade API in download_image().

Signed-off-by: Macpaul Lin <macpaul.from.taiwan@gmail.com>
Cc: Macpaul Lin <macpaul@gmail.com>
Cc: Paul Stewart <pstew@chromium.org>
Cc: Ben Chan <benchan@chromium.org>
Cc: Sage Ahn <syahn@gctsemi.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
(cherry picked from commit 9e412a0a581e07cf1551bbd9b4ae69654e474a3c)

Change-Id: Ic39b571f4d99709826fcba6384413de76f7aa774
Signed-off-by: Ben Chan <benchan@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/36310
Reviewed-by: Paul Stewart <pstew@chromium.org>

UPSTREAM: staging/gdm72xx: gdm_wimax: fix compile error when enable debug

Fix compile error when enable DEBUG_SDU and DEBUG_HCI.
Replace deprecated NIPQUAD marco to C code.

Signed-off-by: Macpaul Lin <macpaul.from.taiwan@gmail.com>
Cc: Macpaul Lin <macpaul@gmail.com>
Cc: Paul Stewart <pstew@chromium.org>
Cc: Ben Chan <benchan@chromium.org>
Cc: Sage Ahn <syahn@gctsemi.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
(cherry picked from commit 2874762b31d8d0eebcfdf189ec9906be9c1460f6)

Change-Id: I4966ab22bddf2225b39a975abbd50e545b8c1a04
Signed-off-by: Ben Chan <benchan@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/36309
Reviewed-by: Paul Stewart <pstew@chromium.org>

ARM: Update kbase to wk42

Catch up to tip of Barium

BUG=chrome-os-partner:15562
TEST=tested on snow, video playback and webgl aquarium

Change-Id: I5356847b662cafcdbe45ab587f4e79afa666e798
Reviewed-on: https://gerrit.chromium.org/gerrit/36357
Reviewed-by: Anush Elangovan <anush@chromium.org>
Commit-Ready: Anush Elangovan <anush@chromium.org>
Tested-by: Anush Elangovan <anush@chromium.org>

UPSTREAM: ath9k_hw: fix ar9462 selfgen chainmask

When the 9462 is operating in 2G mode and MCI is enabled then
reduce the selfgen chain mask to chain 1. Otherwise poor performance
was reported at short range at Rx side when COEX is enabled.

Signed-off-by: Rajkumar Manoharan <rmanohar@qca.qualcomm.com>
Signed-off-by: Paul Stewart <pstew@chromium.org>
BUG=chrome-os-partner:15083
TEST=Use bluetooth and WiFi

Change-Id: I48b2bba6637673112c95ca9c67c7f02d0abcd824
Reviewed-on: https://gerrit.chromium.org/gerrit/36145
Commit-Ready: Paul Stewart <pstew@chromium.org>
Reviewed-by: Paul Stewart <pstew@chromium.org>
Tested-by: Paul Stewart <pstew@chromium.org>

ARM: EXYNOS5: Increase granularity of snow's backlight PWM.

Explicitly set max_brightness to 2800 from its default of 255.
This gives increased granularity in PWM duty cycles to better
utilize the panel's LED driver.

BUG=chromium-os:34821
TEST=cat /sys/clall/backlight/pwm-backlight.0/max_brightness
  Also, set the backlight a bit low (like the 4th step from the bottom),
  then cover and uncover the backlight to confirm smooth backlight
  transitions that each last about 2 seconds.

Signed-off-by: Bryan Freed <bfreed@chromium.org>
Change-Id: If0565d3cf48b02bec178ac2f0c998cb257c062bf
Reviewed-on: https://gerrit.chromium.org/gerrit/36278
Reviewed-by: Grant Grundler <grundler@chromium.org>
Reviewed-by: Doug Anderson <dianders@chromium.org>
Tested-by: Bryan Freed <bfreed@chromium.org>
Commit-Ready: Bryan Freed <bfreed@chromium.org>

CHROMIUM: exynos: dts: configure eDP/LVDS bridge

Put correct settings for Parade PS8622 eDP/LVDS :
- GPIO settings for control signals.
- re-configure properly MMC blocks as MMC2 is now used for GPIO.
- the MAX77686 LDO17 regulator is used to provide 1.2v to the bridge.

Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
BUG=chrome-os-partner:14491
TEST=on Spring, see screen coming up at startup.

Change-Id: Ic305d0b6f6aaf6739ad19818dbfcd8cda90df44a
Reviewed-on: https://gerrit.chromium.org/gerrit/36234
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Commit-Ready: Vincent Palatin <vpalatin@chromium.org>
Tested-by: Vincent Palatin <vpalatin@chromium.org>

CHROMIUM: exynos5: add support for LCD on Spring

Allow to configure DP output for either NXP or Parade eDP-LVDS bridge.
Set sensible LCD panel settings for Spring.

Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
BUG=chrome-os-partner:14491
TEST=run on Spring and Snow and see display coming up on ChromeOS
startup on both machines.

Change-Id: I3cd46327a9d8fd6010c8d308111c6c0267fedae4
Reviewed-on: https://gerrit.chromium.org/gerrit/36233
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Commit-Ready: Vincent Palatin <vpalatin@chromium.org>
Tested-by: Vincent Palatin <vpalatin@chromium.org>

CHROMIUM: config: enable Parade bridge

Compile Parade PS8622 eDP/LVDS.

Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
BUG=chrome-os-partner:14491
TEST=on Spring, boot and see the ps8622 initializing with the right
device tree configuration.

Change-Id: I7eb8cb13ee2a742cb5dd025266e4e1db52ee7656
Reviewed-on: https://gerrit.chromium.org/gerrit/36232
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Commit-Ready: Vincent Palatin <vpalatin@chromium.org>
Tested-by: Vincent Palatin <vpalatin@chromium.org>

CHROMIUM: auxdisplay/ps8620: Add driver for Parade PS8622 bridge

Add a new driver to handle the Parade PS8622 eDP/LVDS bridge chip.
The driver sets properly the SLP_N and RST_N pins and configure the
bridge through I2C.

Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
BUG=chrome-os-partner:14491
TEST=Tested on Spring, see the screen coming up on boot.

Change-Id: I3fd15f9780abe45b1cc93f9234db7d79c65ac5a3
Reviewed-on: https://gerrit.chromium.org/gerrit/36231
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Commit-Ready: Vincent Palatin <vpalatin@chromium.org>
Tested-by: Vincent Palatin <vpalatin@chromium.org>

drm/exynos: Don't enable vblank irq early

This patch fixes a bug where the vsync interrupt was being enabled right
from the win_reset routine. If enable_vblank wasn't called before that,
we would finish pageflip on crtc 0 which would cause tearing on fimd.

BUG=chrome-os-partner:15054
TEST=Tested on snow, tested with boot/hotplug/suspend/resume

Change-Id: I067b2d18dcc48ce5d6cdcfc147d64b7e31f47475
Signed-off-by: Sean Paul <seanpaul@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/35979
Reviewed-by: Mandeep Singh Baines <msb@chromium.org>
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>

drm/exynos: hdmi: support for interlaced mode

This patch adds interlaced mode support for exynos 5 HDMI IP
as per the spec requirement and differing quirk register
values between progressive and interlaced in exynos drm
hdmi driver.

BUG=chrome-os-partner:12642
TEST=Tested with two differnet monitors and
        Agilent Technologies N5998A HDMI
        protocol analyzer and generator
        for 1080i modes at 60Hz and 50Hz
        frequencies.

Change-Id: I61743b91d4553089c72c376cfc94b4bb96f1edf6
Signed-off-by: Shirish S <s.shirish@samsung.com>
Signed-off-by: Akshay Saraswat <Akshay.s@samsung.com>
Reviewed-on: https://gerrit.chromium.org/gerrit/33427
Tested-by: Akshay Saraswat <akshay.s@samsung.com>
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Commit-Ready: Akshay Saraswat <akshay.s@samsung.com>

CHROMIUM: ALSA: hda/ca0132: Stop streaming when turning on/off Crystal Voice.

Similar to the previous fix for switching between microphones, if an
active recording stream is running, it has to be stopped before turning
on/off Crystal Voice, otherwise the parameters won't be applied correctly.

The patch is provided by Creative.

BUG=chrome-os-partner:14920
TEST=use "arecord -f cd -B 10000 -d 20 1.wav"
to record audio, and plug/unplug headphone while recording.

Signed-off-by: Chih-Chung Chang <chihchung@chromium.org>
Change-Id: Ic9296b07942cfaeb66c58cc1024dd20bf924a231
Reviewed-on: https://gerrit.chromium.org/gerrit/36314
Reviewed-by: Dylan Reid <dgreid@chromium.org>

UPSTREAM: mwifiex: minor cleanup and a fix in scan semaphore usage

mwifiex_request_scan() takes care of synchronous internal scan
performed by driver during association.
Currently the semaphore acquired for the scan is unnecessarily
released at the end of different paths. Also, failure paths
returning error code other than "-1" are not considered.

We will release it at the end of routine to fix above issues.

Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Paul Stewart <pstew@chromium.org>
BUG=chrome-os-partner:15390
TEST="iw mlan0 scan trigger; ifconfig mlan0 down"

Change-Id: I5b305ad3eca54659e764008bc2373c6dc7d08be8
Reviewed-on: https://gerrit.chromium.org/gerrit/36160
Reviewed-by: Bing Zhao <bzhao@marvell.com>
Reviewed-by: mukesh agrawal <quiche@chromium.org>
Commit-Ready: Paul Stewart <pstew@chromium.org>
Tested-by: Paul Stewart <pstew@chromium.org>

CHROMIUM: ath9k_hw: fix signal strength / noise measurements on ar9462

The nominal noise floor for most channels is -120, though on some it can
reach -127. Use -120 for converting noise floor to channel noise to avoid
overestimating rx signal strength and channel noise.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Paul Stewart <pstew@chromium.org>
BUG=chrome-os-partner:13074
TEST=Compare RSSI to neighboring device

Change-Id: I88f0e700097a3961ef5e1bb6400f4d84e13d6ffc
Reviewed-on: https://gerrit.chromium.org/gerrit/36149
Commit-Ready: Paul Stewart <pstew@chromium.org>
Reviewed-by: Paul Stewart <pstew@chromium.org>
Tested-by: Paul Stewart <pstew@chromium.org>

CHROMIUM: i915: Allow 0 level when turning on backlight.

Previously, the i915 panel driver would set the backlight to
its maximum level if the level was 0 when turning the
display on.  We don't need this (powerd already does it),
and removing it avoids a potential race where we jump to
100% brightness when the brightness-up key is pressed due to
Chrome turning on the display before powerd sets a non-zero
level.

BUG=chromium-os:31795,chromium-os:32447,chromium-os:35481
TEST=manual: no bright flash when increasing the backlight
     from 0% on lumpy or resuming from suspend
CQ-DEPEND=I26f11562df8e01927c0220cddd91e665fe087be9
CQ-DEPEND=Ia961baae656044c3713fb967d8ad173d317c4217

Signed-off-by: Daniel Erat <derat@chromium.org>
Change-Id: I4bafa1c2e1254b09906245b59c935a4be1088d71
Reviewed-on: https://gerrit.chromium.org/gerrit/36135
Reviewed-by: Simon Que <sque@chromium.org>
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>

UPSTREAM: ath9k: fix queuing MCI work twice

Right now MCI work is being queued twice by profile and status
updation. Queue MCI work once when it is needed.

Signed-off-by: Rajkumar Manoharan <rmanohar@qca.qualcomm.com>
Signed-off-by: Paul Stewart <pstew@chromium.org>
BUG=chrome-os-partner:15083
TEST=Use bluetooth and WiFi

Change-Id: I6b6bebbef160ab0f68ab56c08d57732dab65507f
Reviewed-on: https://gerrit.chromium.org/gerrit/36148
Commit-Ready: Paul Stewart <pstew@chromium.org>
Reviewed-by: Paul Stewart <pstew@chromium.org>
Tested-by: Paul Stewart <pstew@chromium.org>

UPSTREAM: ath9k: move coex param updation within mci work

Update all coex parameters in sigle place. So that we can avoid
redoing the same operation in mutiple places and it eases debugging.

Signed-off-by: Rajkumar Manoharan <rmanohar@qca.qualcomm.com>
Signed-off-by: Paul Stewart <pstew@chromium.org>
BUG=chrome-os-partner:15083
TEST=Use bluetooth and WiFi

Change-Id: Ia694a641418647b90012b891b873aae9b9aa8f26
Reviewed-on: https://gerrit.chromium.org/gerrit/36147
Commit-Ready: Paul Stewart <pstew@chromium.org>
Reviewed-by: Paul Stewart <pstew@chromium.org>
Tested-by: Paul Stewart <pstew@chromium.org>

UPSTREAM: ath9k: flush bt profile whenever it is requested

Before processing BT profiles or status messages, check whether
it is requested to flush BT profile. Otherwise it might increase
number of BT profiles that affects the WLAN performance. Also
flush the profiles when MCI is recovering from broken rx. After
flushing BT profiles, query BT topology to refetch them.

Signed-off-by: Rajkumar Manoharan <rmanohar@qca.qualcomm.com>
Signed-off-by: Paul Stewart <pstew@chromium.org>
BUG=chrome-os-partner:15083
TEST=Use bluetooth and WiFi

Change-Id: If6b1d76594407dbddc0b6549491846247fe1c3ca
Reviewed-on: https://gerrit.chromium.org/gerrit/36146
Commit-Ready: Paul Stewart <pstew@chromium.org>
Reviewed-by: Paul Stewart <pstew@chromium.org>
Tested-by: Paul Stewart <pstew@chromium.org>

UPSTREAM: ath9k: Fix BTCOEX timer triggering comparision

Its more correct to convert btcoex_period to 'us' while
comparing with btcoex_no_stomp which is in 'us'.
Did not find any functionality issues being fixed,
as the generic hardware timer triggers are usually
refreshed with the newer duty cycle.

Cc: Rajkumar Manoharan <rmanohar@qca.qualcomm.com>
Signed-off-by: Mohammed Shafi Shajakhan <mohammed@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: Paul Stewart <pstew@chromium.org>
BUG=chrome-os-partner:15083
TEST=Use bluetooth and WiFi

Change-Id: I98778ad94bf4affd2b7e0443fda6f25950a89488
Reviewed-on: https://gerrit.chromium.org/gerrit/36144
Reviewed-by: mukesh agrawal <quiche@chromium.org>
Commit-Ready: Paul Stewart <pstew@chromium.org>
Tested-by: Paul Stewart <pstew@chromium.org>

CHROMIUM: mwifiex: Abort scan when interface is down

When the interface is marked down, scans should be aborted,
otherwise we risk memory leaks and code paths that may cause
a crash.

Signed-off-by: Paul Stewart <pstew@chromium.org>
BUG=chrome-os-partner:15390
TEST="iw mlan0 scan trigger; ifconfig mlan0 down"

Change-Id: I7a9010ce3086a19fbcd143db60dee27290f1e78b
Reviewed-on: https://gerrit.chromium.org/gerrit/36151
Reviewed-by: Bing Zhao <bzhao@marvell.com>
Reviewed-by: mukesh agrawal <quiche@chromium.org>
Commit-Ready: Paul Stewart <pstew@chromium.org>
Tested-by: Paul Stewart <pstew@chromium.org>

CHROMIUM: gobi: fix a crash due to use-after-free of the qcusbnet object

usbnet_disconnect() calls qcnet_unbind(), which accesses the qcusbnet
object. This patch makes sure that the qcusbnet object is freed after
usbnet_disconnect() is called.

BUG=chrome-os-partner:14889
TEST=Tested the following:
1. emerge-lumpy chromeos-kernel
2. Turn on full SLUB debugging and run suspend_stress_test on a system
   with a Gobi 3000 modem.

Change-Id: I3172565bc0df0c3a283d239424d59b58ae1c262f
Signed-off-by: Ben Chan <benchan@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/36177
Reviewed-by: Olof Johansson <olofj@chromium.org>

CHROMIUM: cdc_ether: handle Novatel E362 modem with updated firmware

This patch is suggested by Novatel Wireless to allow cdc_ether to
correctly detect a Novatel E362 modem with updated firmware.

BUG=chromium-os:15365
TEST=Verified that the cdc_ether driver correctly detects Novatel E362
modem with firmware 1.41 and 4.08.

Change-Id: If880d26140448fcbdaf89ea1fc546ee861882a19
Signed-off-by: Ben Chan <benchan@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/36043
Reviewed-by: Paul Stewart <pstew@chromium.org>

CHROMIUM: usb: option: handle identity morphing on Novatel E362 modem

This patch is suggested by Novatel Wireless to address identity morphing
on Novatel E362 modem such that the correct USB configuration is
selected and detected by the option driver.

BUG=chromium-os:15365
TEST=Verified that the option driver correctly detects Novatel E362
modem with firmware 1.41 and 4.08.

Change-Id: I3449de2327e2b42417e26ff6024f84b72b140853
Signed-off-by: Ben Chan <benchan@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/36042
Reviewed-by: Paul Stewart <pstew@chromium.org>

drm/exynos: Switch debug message DRM_DEBUG_KMS.

BUG=none
TEST=by hand; boots, works, doesn't display debug message.

Change-Id: I5f798f352a5455059dcf9241b1dd1ebc02924352
Reviewed-on: https://gerrit.chromium.org/gerrit/35596
Tested-by: Stéphane Marchesin <marcheu@chromium.org>
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Commit-Ready: Stéphane Marchesin <marcheu@chromium.org>