OSDN Git Service

sagit-ice-cold/kernel_xiaomi_msm8998.git
6 years agoMerge "drm/msm: clear deep color bits in the HDMI controller"
Linux Build Service Account [Thu, 26 Oct 2017 14:35:28 +0000 (07:35 -0700)]
Merge "drm/msm: clear deep color bits in the HDMI controller"

6 years agoMerge "drm/msm: reset sink hdr capabilities during bridge disable"
Linux Build Service Account [Thu, 26 Oct 2017 14:35:27 +0000 (07:35 -0700)]
Merge "drm/msm: reset sink hdr capabilities during bridge disable"

6 years agoMerge "usb: gadget: Add check gadget function bind or not"
Linux Build Service Account [Thu, 26 Oct 2017 14:35:26 +0000 (07:35 -0700)]
Merge "usb: gadget: Add check gadget function bind or not"

6 years agoMerge "defconfig: msm: Enable USB video class support on msm-4.4 kernel"
Linux Build Service Account [Wed, 25 Oct 2017 22:58:59 +0000 (15:58 -0700)]
Merge "defconfig: msm: Enable USB video class support on msm-4.4 kernel"

6 years agoMerge "Merge android-4.4@73a2b70 (v4.4.92) into msm-4.4"
Linux Build Service Account [Wed, 25 Oct 2017 22:58:58 +0000 (15:58 -0700)]
Merge "Merge android-4.4@73a2b70 (v4.4.92) into msm-4.4"

6 years agoMerge "ARM: dts: msm: modify interrupt flag for wled device on PMI8994"
Linux Build Service Account [Wed, 25 Oct 2017 22:58:57 +0000 (15:58 -0700)]
Merge "ARM: dts: msm: modify interrupt flag for wled device on PMI8994"

6 years agodefconfig: msm: Enable USB video class support on msm-4.4 kernel
vpalvai [Sat, 21 Oct 2017 00:36:28 +0000 (17:36 -0700)]
defconfig: msm: Enable USB video class support on msm-4.4 kernel

Enable USB video class for USB camera support on
msm8996-auto devices

Change-Id: I2a385cf4c935d5bc2a4c1171f185baf30322ed1d
Signed-off-by: vpalvai <vpalvai@codeaurora.org>
6 years agoMerge "ath10k: increase the allowed num of tx pending pkts"
Linux Build Service Account [Wed, 25 Oct 2017 14:26:51 +0000 (07:26 -0700)]
Merge "ath10k: increase the allowed num of tx pending pkts"

6 years agoMerge "leds: qpnp-wled: Improve auto calibration algorithm"
Linux Build Service Account [Wed, 25 Oct 2017 14:26:51 +0000 (07:26 -0700)]
Merge "leds: qpnp-wled: Improve auto calibration algorithm"

6 years agoMerge "regulator: qpnp-labibb: Allow 0 ms for IBB PWRUP_DLY1 and PWRDN_DLY2"
Linux Build Service Account [Wed, 25 Oct 2017 14:26:48 +0000 (07:26 -0700)]
Merge "regulator: qpnp-labibb: Allow 0 ms for IBB PWRUP_DLY1 and PWRDN_DLY2"

6 years agoMerge "regulator: qpnp-labibb: Fix LAB PS_CTL configuration for pmi8998"
Linux Build Service Account [Wed, 25 Oct 2017 14:26:47 +0000 (07:26 -0700)]
Merge "regulator: qpnp-labibb: Fix LAB PS_CTL configuration for pmi8998"

6 years agoMerge "ARM: dts: msm: enable camera on SDM636 CDP"
Linux Build Service Account [Wed, 25 Oct 2017 14:26:46 +0000 (07:26 -0700)]
Merge "ARM: dts: msm: enable camera on SDM636 CDP"

6 years agoARM: dts: msm: modify interrupt flag for wled device on PMI8994
Tirupathi Reddy [Wed, 25 Oct 2017 07:03:35 +0000 (12:33 +0530)]
ARM: dts: msm: modify interrupt flag for wled device on PMI8994

The wled driver is not statically handling interrupt flags.
Pass proper flag such as IRQ_TYPE_EDGE_RISING for wled
"sc-irq" interrupt through device node.

Change-Id: I90974bd7995949a97bb7070e86c469cde646096a
Signed-off-by: Tirupathi Reddy <tirupath@codeaurora.org>
6 years agoMerge "leds: qpnp-wled: Fix parameters passed to of_get_address_byname()"
Linux Build Service Account [Wed, 25 Oct 2017 03:02:57 +0000 (20:02 -0700)]
Merge "leds: qpnp-wled: Fix parameters passed to of_get_address_byname()"

6 years agoMerge "soc: qcom: ipc_router_glink_xprt: Change wake source name"
Linux Build Service Account [Wed, 25 Oct 2017 03:02:56 +0000 (20:02 -0700)]
Merge "soc: qcom: ipc_router_glink_xprt: Change wake source name"

6 years agoMerge "net: restore upstream changes which break routing in tunnel scenarios"
Linux Build Service Account [Wed, 25 Oct 2017 03:02:54 +0000 (20:02 -0700)]
Merge "net: restore upstream changes which break routing in tunnel scenarios"

6 years agoMerge "ARM: dts: msm: add support for QBT1000 on SDM630"
Linux Build Service Account [Wed, 25 Oct 2017 03:02:53 +0000 (20:02 -0700)]
Merge "ARM: dts: msm: add support for QBT1000 on SDM630"

6 years agoMerge "drm: msm: fix list corruption problem"
Linux Build Service Account [Wed, 25 Oct 2017 03:02:51 +0000 (20:02 -0700)]
Merge "drm: msm: fix list corruption problem"

6 years agoMerge "msm: vidc: Fix the type of frame size"
Linux Build Service Account [Wed, 25 Oct 2017 03:02:50 +0000 (20:02 -0700)]
Merge "msm: vidc: Fix the type of frame size"

6 years agoMerge "ARM: dts: msm: Use system time timestamp for PPS on msm8996 CV2X"
Linux Build Service Account [Wed, 25 Oct 2017 03:02:49 +0000 (20:02 -0700)]
Merge "ARM: dts: msm: Use system time timestamp for PPS on msm8996 CV2X"

6 years agoMerge "pps-gpio: Add new property to use system time timestamp for PPS GPIO"
Linux Build Service Account [Wed, 25 Oct 2017 03:02:48 +0000 (20:02 -0700)]
Merge "pps-gpio: Add new property to use system time timestamp for PPS GPIO"

6 years agoMerge "defconfig: msm: Disable USELIB and INET_LRO"
Linux Build Service Account [Wed, 25 Oct 2017 03:02:46 +0000 (20:02 -0700)]
Merge "defconfig: msm: Disable USELIB and INET_LRO"

6 years agoMerge "scsi: ufs: Fix stack overflow read in ufs debugfs driver"
Linux Build Service Account [Wed, 25 Oct 2017 03:02:45 +0000 (20:02 -0700)]
Merge "scsi: ufs: Fix stack overflow read in ufs debugfs driver"

6 years agodrm/msm: reset sink hdr capabilities during bridge disable
Abhinav Kumar [Wed, 25 Oct 2017 01:41:01 +0000 (18:41 -0700)]
drm/msm: reset sink hdr capabilities during bridge disable

Reset the HDR capabilities during the bridge disable.
The capabilities are re-programmed the next time
the userspace issues the call to populate the modes.

Change-Id: I04dbaa99c291de13b81e12cb9742cc97d9b02593
Signed-off-by: Abhinav Kumar <abhinavk@codeaurora.org>
6 years agoregulator: qpnp-labibb: Allow 0 ms for IBB PWRUP_DLY1 and PWRDN_DLY2
Subbaraman Narayanamurthy [Fri, 13 Oct 2017 00:43:24 +0000 (17:43 -0700)]
regulator: qpnp-labibb: Allow 0 ms for IBB PWRUP_DLY1 and PWRDN_DLY2

Currently, IBB PWRUP_DLY1 and PWRDN_DLY2 settings are configured
only for values greater than 0. However, 0 ms (no delay) can be
allowed by disabling EN_PWRUP_DLY1 or EN_PWRDN_DLY2 bits. Add
support to allow this configuration.

CRs-Fixed: 2125062
Change-Id: I8ef3b5522b8d4db6befb3fb1c2e58ea24c0f5ea0
Signed-off-by: Subbaraman Narayanamurthy <subbaram@codeaurora.org>
6 years agoregulator: qpnp-labibb: Fix LAB PS_CTL configuration for pmi8998
Subbaraman Narayanamurthy [Wed, 25 Oct 2017 00:35:58 +0000 (17:35 -0700)]
regulator: qpnp-labibb: Fix LAB PS_CTL configuration for pmi8998

PMI8998 LAB pulse skip configuration is different from PMI8996
and earlier PMICs. Fix it by using the right configuration API
to configure LAB PS_CTL along with supporting PM660A properly.
Since the existing LAB PS configuration threshold for pmi8998 is
incorrect in device tree, update it along with this change.

Change-Id: I2686733867ec3e2909d0c071a5881d3d824c713e
Signed-off-by: Subbaraman Narayanamurthy <subbaram@codeaurora.org>
6 years agoleds: qpnp-wled: Improve auto calibration algorithm
Subbaraman Narayanamurthy [Sat, 30 Sep 2017 00:16:28 +0000 (17:16 -0700)]
leds: qpnp-wled: Improve auto calibration algorithm

If CABC is enabled and auto calibration is attempted, then OVP
fault can be seen even with a valid sink configuration and a
brightness level greater than or equal to 0.4 % dimming. This is
because of the modulation done through CABC pin which if low
enough can cause the brightness level to still go further down.

Fix this by disabling CABC on all sinks before running auto
calibration and re-enable it if the feature is enabled. Also,
modify the auto calibration brightness level to 200 so that the
load increase on WLED_VOUT can bring the voltage below OVP
threshold quick enough on a valid sink configuration.

Finally, disable WLED module before switching the sink
configuration and enable it after configuring sinks.

CRs-Fixed: 2125217
Change-Id: I570dc9eba540cd96328afd7c68f4643a112c4962
Signed-off-by: Subbaraman Narayanamurthy <subbaram@codeaurora.org>
6 years agoleds: qpnp-wled: Fix parameters passed to of_get_address_byname()
Subbaraman Narayanamurthy [Thu, 24 Aug 2017 01:44:15 +0000 (18:44 -0700)]
leds: qpnp-wled: Fix parameters passed to of_get_address_byname()

Pass NULL pointer to of_get_address_byname() for size and flags
parameters instead of an integer value 0.

Change-Id: I53984141bb8f1164c27b6a9a5d90853b4d077ee1
Signed-off-by: Subbaraman Narayanamurthy <subbaram@codeaurora.org>
6 years agoleds: qpnp-wled: Run auto-calibration during init
Anirudh Ghayal [Thu, 31 Aug 2017 06:41:06 +0000 (12:11 +0530)]
leds: qpnp-wled: Run auto-calibration during init

If the OVP fault is present during boot-up run
a WLED string auto-calibration.

Change-Id: I4b0e85cd64a1e8bca9bc7de17521d1f83798ed12
Signed-off-by: Anirudh Ghayal <aghayal@codeaurora.org>
6 years agoath10k: increase the allowed num of tx pending pkts
Rakesh Pillai [Mon, 23 Oct 2017 10:45:32 +0000 (16:15 +0530)]
ath10k: increase the allowed num of tx pending pkts

HL1.0 has more number of descriptors and hence we can
queue more packets for tx.

Allow a higher number of tx pending packets for better
throughput in concurrency cases.

CRs-Fixed: 2104072
Change-Id: I8dd98cee04200fbeccf655dcb1d982d037077101
Signed-off-by: Rakesh Pillai <pillair@codeaurora.org>
6 years agodrm/msm: clear deep color bits in the HDMI controller
Abhinav Kumar [Sat, 21 Oct 2017 03:43:23 +0000 (20:43 -0700)]
drm/msm: clear deep color bits in the HDMI controller

Clear the deep color bits in the HDMI controller in case
the mode does not need OR support deep color.

Otherwise it causes issues while switching across modes where
the sink supports deep color in one of the modes but not in the
other.

Change-Id: I8accf00e89868ccaa781a3f18eb9a662d9c9235c
Signed-off-by: Abhinav Kumar <abhinavk@codeaurora.org>
6 years agodrm: msm: fix list corruption problem
Guchun Chen [Fri, 20 Oct 2017 08:51:20 +0000 (16:51 +0800)]
drm: msm: fix list corruption problem

When multiple worker threads compete to update event_list, with current
vblank_ctrl_worker mechanism, there is one risk which can casue list node
is deleted for twice. This is because, due to the protected scope by
the spin_lock, mutex can exit when one thread is transvering the list,
but another queue thread may continue to add new event to the list.
This brings conflict risk. This patch is to correct this.

Change-Id: Ice31462d196c57ce18d7b998c1a1f0b7feeb08fc
Signed-off-by: Xiaowen Wu <wxiaowen@codeaurora.org>
Signed-off-by: Guchun Chen <guchunc@codeaurora.org>
6 years agousb: gadget: Add check gadget function bind or not
Liangliang Lu [Wed, 18 Oct 2017 08:35:18 +0000 (16:35 +0800)]
usb: gadget: Add check gadget function bind or not

For NCM gadget function, net device structure opts->net
will be freed when NCM unbind, but user space related
device node doesn't know it, will access net structre
without checking it exist or not, finally lead to crash.

Fix this by adding check gadget function bind or not.

Change-Id: I6af9e3868f71e7fa923600721e1f716508de9b55
Signed-off-by: Liangliang Lu <luliang@codeaurora.org>
6 years agoscsi: ufs: Fix stack overflow read in ufs debugfs driver
Sayali Lokhande [Thu, 5 Oct 2017 06:06:17 +0000 (11:36 +0530)]
scsi: ufs: Fix stack overflow read in ufs debugfs driver

When getting string from userspace by simple_write_to_buffer
in ufs_qcom_dbg_testbus_cfg_write() function, the string
copied to configuration is not terminated with '\0'. Thus
stack overflow read may occur while copying configuration to
host->testbus.select_major, which will result in information
leak later while printing error message. This change adds null
character at the end of the input string to avoid information
leak.

Change-Id: Ic9a9204def4bd6976f42f5f80ae5c0a9730afeb1
Signed-off-by: Sayali Lokhande <sayalil@codeaurora.org>
6 years agoMerge "cnss2: Decouple WLAN host driver and PCI/MHI for suspend/resume"
Linux Build Service Account [Tue, 24 Oct 2017 00:16:45 +0000 (17:16 -0700)]
Merge "cnss2: Decouple WLAN host driver and PCI/MHI for suspend/resume"

6 years agoARM: dts: msm: Use system time timestamp for PPS on msm8996 CV2X
Gustavo Solaira [Sun, 22 Oct 2017 00:09:17 +0000 (17:09 -0700)]
ARM: dts: msm: Use system time timestamp for PPS on msm8996 CV2X

Use system time timestamp for the PPS GPIO driver on
msm8996 CV2X boards.

Change-Id: Ib7231d9b182c956d3c169d6979dbb5fe7f185353
Signed-off-by: Gustavo Solaira <gustavos@codeaurora.org>
6 years agopps-gpio: Add new property to use system time timestamp for PPS GPIO
Gustavo Solaira [Sat, 21 Oct 2017 23:36:02 +0000 (16:36 -0700)]
pps-gpio: Add new property to use system time timestamp for PPS GPIO

If the device tree property "use-system-time-ts" is defined
then the timestamp used for PPS events will be the system time,
otherwise it will be the monotonic boot time.

Change-Id: I92fab1d8861e1d652fcb68066b417f44c1f8e29b
Signed-off-by: Gustavo Solaira <gustavos@codeaurora.org>
6 years agosoc: qcom: ipc_router_glink_xprt: Change wake source name
Dhoat Harpal [Fri, 13 Oct 2017 10:43:36 +0000 (16:13 +0530)]
soc: qcom: ipc_router_glink_xprt: Change wake source name

Name added for wake source is stored in stack memory. It does
not exist after the stack is gone, this leads to use after free
case.

Added name from Glink xprt struct which persists until
IPCRTR is not removed.

CRs-Fixed: 2124767
Change-Id: I580cf6d6d12682f1f3f25c2506c2b1a969ee89b8
Signed-off-by: Dhoat Harpal <hdhoat@codeaurora.org>
6 years agoMerge "mmc: core: Prevent accessing user space buffer directly"
Linux Build Service Account [Mon, 23 Oct 2017 12:47:58 +0000 (05:47 -0700)]
Merge "mmc: core: Prevent accessing user space buffer directly"

6 years agoMerge "msm: camera: isp: Initialize the isp clock index before enable"
Linux Build Service Account [Mon, 23 Oct 2017 12:47:57 +0000 (05:47 -0700)]
Merge "msm: camera: isp: Initialize the isp clock index before enable"

6 years agoMerge "msm: isp: Clear irq status if irq is set again with same value"
Linux Build Service Account [Mon, 23 Oct 2017 12:47:56 +0000 (05:47 -0700)]
Merge "msm: isp:  Clear irq status if irq is set again with same value"

6 years agoMerge "msm: isp: increase UB size for isp40"
Linux Build Service Account [Mon, 23 Oct 2017 12:47:55 +0000 (05:47 -0700)]
Merge "msm: isp: increase UB size for isp40"

6 years agoMerge "msm: isp: Fix Ub allocation for image WM"
Linux Build Service Account [Mon, 23 Oct 2017 12:47:54 +0000 (05:47 -0700)]
Merge "msm: isp: Fix Ub allocation for image WM"

6 years agodefconfig: msm: Disable USELIB and INET_LRO
Srinivas Ramana [Mon, 23 Oct 2017 09:34:36 +0000 (15:04 +0530)]
defconfig: msm: Disable USELIB and INET_LRO

Disable the below defconfigs for SDM660 to address the
VTS KernelConfig test case failures.
Disable:
CONFIG_USELIB
CONFIG_INET_LRO

Enable:
CONFIG_CGROUP_DEBUG

Change-Id: I20ea6e4f7ac2350ad6a57a61a5b42464cfb273ef
Signed-off-by: Srinivas Ramana <sramana@codeaurora.org>
6 years agoARM: dts: msm: add support for QBT1000 on SDM630
Abir Ghosh [Mon, 23 Oct 2017 08:10:12 +0000 (13:40 +0530)]
ARM: dts: msm: add support for QBT1000 on SDM630

Add QBT100 device for SDM630 with necessary clocks and GPIO
configurations. QBT1000 is fingerprint driver to communicate
with fingerprint trustzone app through QSEECOM. It also
receives interrupts from fingerprint sensor.

Change-Id: I0945f6377013f86ad531d9490ccbfa327bca034e
Signed-off-by: Abir Ghosh <abirg@codeaurora.org>
6 years agoMerge "ARM: dts: msm: Disable ASPM for MDM EP on msm8996 CV2X boards"
Linux Build Service Account [Mon, 23 Oct 2017 03:29:34 +0000 (20:29 -0700)]
Merge "ARM: dts: msm: Disable ASPM for MDM EP on msm8996 CV2X boards"

6 years agoMerge "msm: ipa: Set ep delay on rmnet/mbim tether pipe"
Linux Build Service Account [Sat, 21 Oct 2017 21:11:48 +0000 (14:11 -0700)]
Merge "msm: ipa: Set ep delay on rmnet/mbim tether pipe"

6 years agoMerge changes into msm-4.4
Gerrit - the friendly Code Review server [Sat, 21 Oct 2017 20:10:14 +0000 (13:10 -0700)]
Merge changes  into msm-4.4

6 years agoMerge "ASoC: msm: qdsp6v2: Add MM28 and MM29 for record usecase"
Linux Build Service Account [Sat, 21 Oct 2017 09:30:41 +0000 (02:30 -0700)]
Merge "ASoC: msm: qdsp6v2: Add MM28 and MM29 for record usecase"

6 years agoMerge "ASoC: msm: qdsp6v2: modify API to configure output PCM block"
Linux Build Service Account [Sat, 21 Oct 2017 09:30:40 +0000 (02:30 -0700)]
Merge "ASoC: msm: qdsp6v2: modify API to configure output PCM block"

6 years agoMerge "ASoC: msm: qdsp6v2: Fix information leak in kernel"
Linux Build Service Account [Sat, 21 Oct 2017 09:30:39 +0000 (02:30 -0700)]
Merge "ASoC: msm: qdsp6v2: Fix information leak in kernel"

6 years agoMerge "ASoC: msm: Support Secondary TDM RX interface"
Linux Build Service Account [Sat, 21 Oct 2017 09:30:38 +0000 (02:30 -0700)]
Merge "ASoC: msm: Support Secondary TDM RX interface"

6 years agoMerge "ARM: dts: msm: Support Secondary TDM RX Interface for msm8996"
Linux Build Service Account [Sat, 21 Oct 2017 09:30:37 +0000 (02:30 -0700)]
Merge "ARM: dts: msm: Support Secondary TDM RX Interface for msm8996"

6 years agoMerge "soc: qcom: add SMMU support for virtualized audio ION driver"
Linux Build Service Account [Sat, 21 Oct 2017 09:30:26 +0000 (02:30 -0700)]
Merge "soc: qcom: add SMMU support for virtualized audio ION driver"

6 years agoMerge "ASoC: msm: add soc audio snd virtualization support"
Linux Build Service Account [Sat, 21 Oct 2017 09:30:25 +0000 (02:30 -0700)]
Merge "ASoC: msm: add soc audio snd virtualization support"

6 years agoMerge "ARM: dts: msm: add audio ion vm dev on msm8996 vplatform"
Linux Build Service Account [Sat, 21 Oct 2017 09:30:25 +0000 (02:30 -0700)]
Merge "ARM: dts: msm: add audio ion vm dev on msm8996 vplatform"

6 years agoMerge "Revert "leds: remove unused driver leds-qpnp-flash.c""
Linux Build Service Account [Fri, 20 Oct 2017 23:49:29 +0000 (16:49 -0700)]
Merge "Revert "leds: remove unused driver leds-qpnp-flash.c""

6 years agoMerge "Revert "power: supply: qcom: remove unused drivers""
Linux Build Service Account [Fri, 20 Oct 2017 23:49:28 +0000 (16:49 -0700)]
Merge "Revert "power: supply: qcom: remove unused drivers""

6 years agoMerge "ARM: dts: msm: Enable MPP06 for graphics support in MSM8996 CV2X"
Linux Build Service Account [Fri, 20 Oct 2017 23:49:27 +0000 (16:49 -0700)]
Merge "ARM: dts: msm: Enable MPP06 for graphics support in MSM8996 CV2X"

6 years agoMerge "power: qpnp_smb2: configure to extend DRP.DFP time in try.SNK"
Linux Build Service Account [Fri, 20 Oct 2017 23:49:18 +0000 (16:49 -0700)]
Merge "power: qpnp_smb2: configure to extend DRP.DFP time in try.SNK"

6 years agoMerge "power: smblib: extend wait time after force sink in try.SNK"
Linux Build Service Account [Fri, 20 Oct 2017 23:49:17 +0000 (16:49 -0700)]
Merge "power: smblib: extend wait time after force sink in try.SNK"

6 years agoMerge "power: smblib: SW implementation of typeC try.SNK"
Linux Build Service Account [Fri, 20 Oct 2017 23:49:17 +0000 (16:49 -0700)]
Merge "power: smblib: SW implementation of typeC try.SNK"

6 years agoARM: dts: msm: Disable ASPM for MDM EP on msm8996 CV2X boards
Gustavo Solaira [Fri, 20 Oct 2017 18:35:35 +0000 (11:35 -0700)]
ARM: dts: msm: Disable ASPM for MDM EP on msm8996 CV2X boards

Disable L1/L1ss modes for the RC2 link with MDM9x55 EP on
msm8996 CV2X boards to avoid bursty MSI interrupts and
unstable communication channel.

Change-Id: I12549ec37ceda09ebc9104b0182e53efdcfc8d2f
Signed-off-by: Gustavo Solaira <gustavos@codeaurora.org>
6 years agocnss2: Decouple WLAN host driver and PCI/MHI for suspend/resume
Yue Ma [Fri, 20 Oct 2017 16:54:57 +0000 (09:54 -0700)]
cnss2: Decouple WLAN host driver and PCI/MHI for suspend/resume

Decouple WLAN host driver suspend/resume and PCI link and MHI
suspend/resume so that PCI link and MHI can be suspended/resumeed
separately without WLAN host driver.

Change-Id: I82cb7b8b54c61657c5d0bbaead0ffc1c55ecee8d
Signed-off-by: Yue Ma <yuem@codeaurora.org>
6 years agoMerge android-4.4@73a2b70 (v4.4.92) into msm-4.4
Blagovest Kolenichev [Thu, 19 Oct 2017 23:55:17 +0000 (16:55 -0700)]
Merge android-4.4@73a2b70 (v4.4.92) into msm-4.4

* refs/heads/tmp-73a2b70
  Linux 4.4.92
  ext4: don't allow encrypted operations without keys
  ext4: Don't clear SGID when inheriting ACLs
  ext4: fix data corruption for mmap writes
  sched/cpuset/pm: Fix cpuset vs. suspend-resume bugs
  nvme: protect against simultaneous shutdown invocations
  drm/i915/bios: ignore HDMI on port A
  brcmfmac: setup passive scan if requested by user-space
  uwb: ensure that endpoint is interrupt
  uwb: properly check kthread_run return value
  iio: adc: mcp320x: Fix oops on module unload
  iio: adc: mcp320x: Fix readout of negative voltages
  iio: ad7793: Fix the serial interface reset
  iio: core: Return error for failed read_reg
  staging: iio: ad7192: Fix - use the dedicated reset function avoiding dma from stack.
  iio: ad_sigma_delta: Implement a dedicated reset function
  iio: adc: twl4030: Disable the vusb3v1 rugulator in the error handling path of 'twl4030_madc_probe()'
  iio: adc: twl4030: Fix an error handling path in 'twl4030_madc_probe()'
  xhci: fix finding correct bus_state structure for USB 3.1 hosts
  USB: fix out-of-bounds in usb_set_configuration
  usb: Increase quirk delay for USB devices
  USB: core: harden cdc_parse_cdc_header
  USB: uas: fix bug in handling of alternate settings
  scsi: sd: Do not override max_sectors_kb sysfs setting
  iwlwifi: add workaround to disable wide channels in 5GHz
  HID: i2c-hid: allocate hid buffers for real worst case
  ftrace: Fix kmemleak in unregister_ftrace_graph
  stm class: Fix a use-after-free
  Drivers: hv: fcopy: restore correct transfer length
  driver core: platform: Don't read past the end of "driver_override" buffer
  ALSA: usx2y: Suppress kernel warning at page allocation failures
  ALSA: compress: Remove unused variable
  lsm: fix smack_inode_removexattr and xattr_getsecurity memleak
  USB: g_mass_storage: Fix deadlock when driver is unbound
  usb: gadget: mass_storage: set msg_registered after msg registered
  USB: devio: Don't corrupt user memory
  USB: dummy-hcd: Fix erroneous synchronization change
  USB: dummy-hcd: fix infinite-loop resubmission bug
  USB: dummy-hcd: fix connection failures (wrong speed)
  usb: pci-quirks.c: Corrected timeout values used in handshake
  ALSA: usb-audio: Check out-of-bounds access by corrupted buffer descriptor
  usb: renesas_usbhs: fix usbhsf_fifo_clear() for RX direction
  usb: renesas_usbhs: fix the BCLR setting condition for non-DCP pipe
  usb-storage: unusual_devs entry to fix write-access regression for Seagate external drives
  usb: gadget: udc: atmel: set vbus irqflags explicitly
  USB: gadgetfs: fix copy_to_user while holding spinlock
  USB: gadgetfs: Fix crash caused by inadequate synchronization
  usb: gadget: inode.c: fix unbalanced spin_lock in ep0_write
  ANDROID: binder: init desired_prio.sched_policy before use it
  BACKPORT: net: xfrm: support setting an output mark.
  UPSTREAM: xfrm: Only add l3mdev oif to dst lookups
  UPSTREAM: net: l3mdev: Add master device lookup by index
  Linux 4.4.91
  ttpci: address stringop overflow warning
  ALSA: au88x0: avoid theoretical uninitialized access
  ARM: remove duplicate 'const' annotations'
  IB/qib: fix false-postive maybe-uninitialized warning
  drivers: firmware: psci: drop duplicate const from psci_of_match
  libata: transport: Remove circular dependency at free time
  xfs: remove kmem_zalloc_greedy
  i2c: meson: fix wrong variable usage in meson_i2c_put_data
  md/raid10: submit bio directly to replacement disk
  rds: ib: add error handle
  iommu/io-pgtable-arm: Check for leaf entry before dereferencing it
  parisc: perf: Fix potential NULL pointer dereference
  netfilter: nfnl_cthelper: fix incorrect helper->expect_class_max
  exynos-gsc: Do not swap cb/cr for semi planar formats
  MIPS: IRQ Stack: Unwind IRQ stack onto task stack
  netfilter: invoke synchronize_rcu after set the _hook_ to NULL
  bridge: netlink: register netdevice before executing changelink
  mmc: sdio: fix alignment issue in struct sdio_func
  usb: plusb: Add support for PL-27A1
  team: fix memory leaks
  net/packet: check length in getsockopt() called with PACKET_HDRLEN
  net: core: Prevent from dereferencing null pointer when releasing SKB
  MIPS: Lantiq: Fix another request_mem_region() return code check
  ASoC: dapm: fix some pointer error handling
  usb: chipidea: vbus event may exist before starting gadget
  audit: log 32-bit socketcalls
  ASoC: dapm: handle probe deferrals
  partitions/efi: Fix integer overflow in GPT size calculation
  USB: serial: mos7840: fix control-message error handling
  USB: serial: mos7720: fix control-message error handling
  drm/amdkfd: fix improper return value on error
  IB/ipoib: Replace list_del of the neigh->list with list_del_init
  IB/ipoib: rtnl_unlock can not come after free_netdev
  IB/ipoib: Fix deadlock over vlan_mutex
  tty: goldfish: Fix a parameter of a call to free_irq
  ARM: 8635/1: nommu: allow enabling REMAP_VECTORS_TO_RAM
  iio: adc: hx711: Add DT binding for avia,hx711
  iio: adc: axp288: Drop bogus AXP288_ADC_TS_PIN_CTRL register modifications
  hwmon: (gl520sm) Fix overflows and crash seen when writing into limit attributes
  sh_eth: use correct name for ECMR_MPDE bit
  extcon: axp288: Use vbus-valid instead of -present to determine cable presence
  igb: re-assign hw address pointer on reset after PCI error
  MIPS: ralink: Fix incorrect assignment on ralink_soc
  MIPS: Ensure bss section ends on a long-aligned address
  ARM: dts: r8a7790: Use R-Car Gen 2 fallback binding for msiof nodes
  RDS: RDMA: Fix the composite message user notification
  GFS2: Fix reference to ERR_PTR in gfs2_glock_iter_next
  drm: bridge: add DT bindings for TI ths8135
  drm_fourcc: Fix DRM_FORMAT_MOD_LINEAR #define
  FROMLIST: tracing: Add support for preempt and irq enable/disable events
  FROMLIST: tracing: Prepare to add preempt and irq trace events
  ANDROID: binder: fix transaction leak.
  ANDROID: binder: Add tracing for binder priority inheritance.
  Linux 4.4.90
  fix xen_swiotlb_dma_mmap prototype
  swiotlb-xen: implement xen_swiotlb_dma_mmap callback
  video: fbdev: aty: do not leak uninitialized padding in clk to userspace
  KVM: VMX: use cmpxchg64
  ARM: pxa: fix the number of DMA requestor lines
  ARM: pxa: add the number of DMA requestor lines
  dmaengine: mmp-pdma: add number of requestors
  cxl: Fix driver use count
  KVM: VMX: remove WARN_ON_ONCE in kvm_vcpu_trigger_posted_interrupt
  KVM: VMX: do not change SN bit in vmx_update_pi_irte()
  timer/sysclt: Restrict timer migration sysctl values to 0 and 1
  gfs2: Fix debugfs glocks dump
  x86/fpu: Don't let userspace set bogus xcomp_bv
  btrfs: prevent to set invalid default subvolid
  btrfs: propagate error to btrfs_cmp_data_prepare caller
  btrfs: fix NULL pointer dereference from free_reloc_roots()
  PCI: Fix race condition with driver_override
  kvm: nVMX: Don't allow L2 to access the hardware CR8
  KVM: VMX: Do not BUG() on out-of-bounds guest IRQ
  arm64: fault: Route pte translation faults via do_translation_fault
  arm64: Make sure SPsel is always set
  seccomp: fix the usage of get/put_seccomp_filter() in seccomp_get_filter()
  bsg-lib: don't free job in bsg_prepare_job
  nl80211: check for the required netlink attributes presence
  vfs: Return -ENXIO for negative SEEK_HOLE / SEEK_DATA offsets
  SMB3: Don't ignore O_SYNC/O_DSYNC and O_DIRECT flags
  SMB: Validate negotiate (to protect against downgrade) even if signing off
  Fix SMB3.1.1 guest authentication to Samba
  powerpc/pseries: Fix parent_dn reference leak in add_dt_node()
  KEYS: prevent KEYCTL_READ on negative key
  KEYS: prevent creating a different user's keyrings
  KEYS: fix writing past end of user-supplied buffer in keyring_read()
  crypto: talitos - fix sha224
  crypto: talitos - Don't provide setkey for non hmac hashing algs.
  scsi: scsi_transport_iscsi: fix the issue that iscsi_if_rx doesn't parse nlmsg properly
  md/raid5: preserve STRIPE_ON_UNPLUG_LIST in break_stripe_batch_list
  md/raid5: fix a race condition in stripe batch
  tracing: Erase irqsoff trace with empty write
  tracing: Fix trace_pipe behavior for instance traces
  KVM: PPC: Book3S: Fix race and leak in kvm_vm_ioctl_create_spapr_tce()
  mac80211: flush hw_roc_start work before cancelling the ROC
  cifs: release auth_key.response for reconnect.
  f2fs: catch up to v4.14-rc1
  UPSTREAM: cpufreq: schedutil: use now as reference when aggregating shared policy requests
  ANDROID: add script to fetch android kernel config fragments
  f2fs: reorganize stat information
  f2fs: clean up flush/discard command namings
  f2fs: check in-memory sit version bitmap
  f2fs: check in-memory nat version bitmap
  f2fs: check in-memory block bitmap
  f2fs: introduce FI_ATOMIC_COMMIT
  f2fs: clean up with list_{first, last}_entry
  f2fs: return fs_trim if there is no candidate
  f2fs: avoid needless checkpoint in f2fs_trim_fs
  f2fs: relax async discard commands more
  f2fs: drop exist_data for inline_data when truncated to 0
  f2fs: don't allow encrypted operations without keys
  f2fs: show the max number of atomic operations
  f2fs: get io size bit from mount option
  f2fs: support IO alignment for DATA and NODE writes
  f2fs: add submit_bio tracepoint
  f2fs: reassign new segment for mode=lfs
  f2fs: fix a missing discard prefree segments
  f2fs: use rb_entry_safe
  f2fs: add a case of no need to read a page in write begin
  f2fs: fix a problem of using memory after free
  f2fs: remove unneeded condition
  f2fs: don't cache nat entry if out of memory
  f2fs: remove unused values in recover_fsync_data
  f2fs: support async discard based on v4.9
  f2fs: resolve op and op_flags confilcts
  f2fs: remove wrong backported codes
  FROMLIST: binder: fix use-after-free in binder_transaction()
  UPSTREAM: ipv6: fib: Unlink replaced routes from their nodes

Signed-off-by: Blagovest Kolenichev <bkolenichev@codeaurora.org>
Conflicts:
fs/f2fs/crypto_key.c
fs/f2fs/f2fs_crypto.h
net/wireless/nl80211.c
sound/usb/card.c

Change-Id: I742aeaec84c7892165976b7bea3e07bdd6881d93
Signed-off-by: Blagovest Kolenichev <bkolenichev@codeaurora.org>
6 years agoASoC: msm: Support Secondary TDM RX interface
Hugh Guan [Mon, 18 Sep 2017 09:57:46 +0000 (17:57 +0800)]
ASoC: msm: Support Secondary TDM RX interface

Support Secondary TDM RX interface dai link and export kcontrols to
update Secondary TDM RX interface Channels and Bit Format configuration.

CRs-fixed: 2112052
Change-Id: Idb1d06cd6c0af385a46c775205bb3f6e45d411d7
Signed-off-by: Hugh Guan <hhguan@codeaurora.org>
6 years agoARM: dts: msm: Support Secondary TDM RX Interface for msm8996
Hugh Guan [Mon, 18 Sep 2017 09:54:22 +0000 (17:54 +0800)]
ARM: dts: msm: Support Secondary TDM RX Interface for msm8996

Support Secondary TDM RX Interface for automotive.

CRs-fixed: 2112052
Change-Id: Iae7bde93e8af851e470fb4f81b37d513ce3dfd02
Signed-off-by: Hugh Guan <hhguan@codeaurora.org>
6 years agonet: restore upstream changes which break routing in tunnel scenarios
Blagovest Kolenichev [Thu, 19 Oct 2017 22:28:13 +0000 (15:28 -0700)]
net: restore upstream changes which break routing in tunnel scenarios

This reverts commit:

  3d78e43 net: Revert upstream changes which break routing in
          tunnel scenarios

Restored changes are:

  f1900fb net: Really fix vti6 with oif in dst lookups
  4148987 net: Fix vti use case with oif in dst lookups for IPv6
  42a7b32 xfrm: Add oif to dst lookups

Now it is save to restore the reverted upstream changes. Also
below upstream changes are coming with android@73a2b70 (4.4.92)
which will hit conflict if the above changes are not restored.

  c6a272a BACKPORT: net: xfrm: support setting an output mark.
  511953d UPSTREAM: xfrm: Only add l3mdev oif to dst lookups

Change-Id: Ia34abd5b9e2625d44922c018756d2214e2bd47bf
Signed-off-by: Blagovest Kolenichev <bkolenichev@codeaurora.org>
6 years agof2fs: cleanup not used anymore max_file_size()
Blagovest Kolenichev [Thu, 19 Oct 2017 13:30:23 +0000 (06:30 -0700)]
f2fs: cleanup not used anymore max_file_size()

Remove declaration of function max_file_size(). It was added
with:

63ba8a3 BACKPORT: f2fs: add a max block check for get_data_block_bmap

On merging of android-4.4@d68ba9f (4.4.89) functional part of
the above change was replaced with the upstream's one, which do
not use max_file_size() and also this function was renamed to
max_file_blocks() in its definition with change c1286ff. In
android-4.4 this func was static, hence no declaration line in
f2fs.h and this is why it was left alone in msm-4.4 without to
be noticed.

c1286ff f2fs: backport from (4c1fad64 - Merge tag 'for-f2fs-4.9'
        of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs)

Change-Id: I197e8fa912f645f49acd1f97180944e303f3d314
Signed-off-by: Blagovest Kolenichev <bkolenichev@codeaurora.org>
6 years agoMerge "ALSA: usb-audio: Add length check after string desc copy"
Linux Build Service Account [Thu, 19 Oct 2017 04:38:46 +0000 (21:38 -0700)]
Merge "ALSA: usb-audio: Add length check after string desc copy"

6 years agopower: qpnp_smb2: configure to extend DRP.DFP time in try.SNK
Harry Yang [Tue, 3 Oct 2017 20:56:55 +0000 (13:56 -0700)]
power: qpnp_smb2: configure to extend DRP.DFP time in try.SNK

When both sides are doing DRP and the other side applies Rd near the end
of PMIC DFP cycle, sometimes, PMIC does not lock on into AttachedWait.SRC
and transitions back into Unattached.SNK.

If a resistance ranging (5.1k,300k) pulldown is applied and then 5.1kohms
pulldown is applied, then it takes 10-20ms to detect 5.1kohms.
Before this, if DRP.DFP time expires, then PMI transitions to UFP mode to
avoid extending DRP cycle time.

Fix it by extending DFP time. Note the total DRP time should still be
within max limit (100ms).

Change-Id: I1dd180229bc771542b4f6ca7850aa457478ae5cd
Signed-off-by: Harry Yang <harryy@codeaurora.org>
6 years agopower: smblib: extend wait time after force sink in try.SNK
Harry Yang [Mon, 16 Oct 2017 23:21:19 +0000 (16:21 -0700)]
power: smblib: extend wait time after force sink in try.SNK

In Try.SNK software implementation, after forcing sink, there
is a delay of tDRPTRY + Tccdebounce time, giviing opportunity
to the other side to be a SRC.

The current delay is 100ms < tDRPTRY (90ms) + Tccdebounce (20ms),
too short, which sometimes results in checking Tccdebounce_done
status too early.

Change the delay to 120ms.

Change-Id: If9fe9aa7c405a5ae6416e9373449c4a76295e4d8
Signed-off-by: Harry Yang <harryy@codeaurora.org>
6 years agopower: smblib: SW implementation of typeC try.SNK
Harry Yang [Sat, 9 Sep 2017 07:09:36 +0000 (00:09 -0700)]
power: smblib: SW implementation of typeC try.SNK

Try.SNK allows a DRP that has a policy based preference to be a Sink
when connecting to another DRP to effect a transition from a destined
Source role to the Sink role.

This is a Software implementation for PMI8998.

Exiting from AttachWait.SRC State, a DRP that strongly prefers the
Sink role may optionally transition to Try.SNK instead of Attached.SRC.

When in the Try.SNK state, the port is querying to determine if the
port partner supports the Source role.

The port shall then transition to Attached.SNK when the SNK.Rp state
is detected for at least tTryCCDebounce and VBUS is detected

Alternatively, the port shall transition to TryWait.SRC if SNK.Rp state
is not detected tTryCCDebounce.

The port shall transition to Attached.SRC when SRC.Rd state is detected
on exactly one of the CC pins.

The port shall transition to Unattached.SNK after tDRPTry if neither of
the CC1 or CC2 pins is in the SRC.Rd state.

CRs-Fixed: 1114313
Change-Id: I777332e87df65082333f4e24a61c5ea834784d09
Signed-off-by: Harry Yang <harryy@codeaurora.org>
6 years agosoc: qcom: add SMMU support for virtualized audio ION driver
Honghao Liu [Fri, 25 Aug 2017 16:38:30 +0000 (12:38 -0400)]
soc: qcom: add SMMU support for virtualized audio ION driver

Update the virtualized audio ION driver to add SMMU support.
The virtualized audio ION driver will provide the ION memory
allocated from the guest VM to the physical VM to perform SMMU
mapping.

The virtualized audio ION driver will receive SMMU mapped
device address from physical VM and provide the device address
to its clients. Upon its client frees the memory, the virtualized
audio ION driver will request the physical VM to perform SMMU
un-mapping.

CRs-fixed: 2099245
Change-Id: Idca4ecda65910308d7f7af288381866ab4ce4a7a
Signed-off-by: Honghao Liu <honghaol@codeaurora.org>
6 years agoARM: dts: msm: add audio ion vm dev on msm8996 vplatform
Derek Chen [Wed, 6 Sep 2017 22:34:09 +0000 (18:34 -0400)]
ARM: dts: msm: add audio ion vm dev on msm8996 vplatform

Add msm audio ion vm device node to be compatible with
virtualized audio ion platform driver on MSM8996 vplatform.

CRs-fixed: 2047379
Signed-off-by: Derek Chen <chenche@codeaurora.org>
Change-Id: Ibd6b5cfa7e8539a56b6d1650ea0d2ae15ce82073

6 years agoASoC: msm: add soc audio snd virtualization support
Derek Chen [Wed, 21 Sep 2016 00:14:09 +0000 (20:14 -0400)]
ASoC: msm: add soc audio snd virtualization support

Add soc audio sound card SND_SOC_MSM8996_VM for
virtualization support.
The virtualized audio snd soc is based on legacy
soc msm8996, removing dependency on ARCH_MSM8996
and various soc codec/slimbus and hwdep driver as
well as DTS/SRS PP features.

CRs-fixed: 2047379
Change-Id: I930bf78c2ebf2c605331b50a598b5a43ceed0e0f
Signed-off-by: Derek Chen <chenche@codeaurora.org>
6 years agosoc: qcom: add audio apr and ion virtualization support
Derek Chen [Tue, 20 Sep 2016 22:25:13 +0000 (18:25 -0400)]
soc: qcom: add audio apr and ion virtualization support

Add support for virtualized APR and ION frontend driver
for audio.
Virtualized APR frontend driver resides in guest VM and
supports all legacy APR APIs. It will rely on MSM_HAB
to communicate with APR backend driver, which resides in
physical VM, where legacy SMD is utilized for communication
between APPS and ADSP.
Virtualized ION driver currently is designed to extract
physical memory through hypervisor abstracted ion layer,
which can be shared between APPS and ADSP directly.

CRs-fixed: 2047379
Change-Id: I7dad5aebfd1db00cd32807dc1219818a7a4e62c5
Signed-off-by: Derek Chen <chenche@codeaurora.org>
6 years agoMerge "block: Initialize bd_bdi on inode initialization"
Linux Build Service Account [Wed, 18 Oct 2017 18:08:22 +0000 (11:08 -0700)]
Merge "block: Initialize bd_bdi on inode initialization"

6 years agoMerge "drm/msm/sde: synchronize mdp clk with frame update"
Linux Build Service Account [Wed, 18 Oct 2017 18:08:20 +0000 (11:08 -0700)]
Merge "drm/msm/sde: synchronize mdp clk with frame update"

6 years agoMerge "drm/msm/sde: align bandwidth/clock updates with frame done"
Linux Build Service Account [Wed, 18 Oct 2017 18:08:20 +0000 (11:08 -0700)]
Merge "drm/msm/sde: align bandwidth/clock updates with frame done"

6 years agoMerge "diag: Add mutex protection while reading dci debug statistics"
Linux Build Service Account [Wed, 18 Oct 2017 18:08:19 +0000 (11:08 -0700)]
Merge "diag: Add mutex protection while reading dci debug statistics"

6 years agoMerge "cnss2: Support multiple recoveries when using CORE only firmware"
Linux Build Service Account [Wed, 18 Oct 2017 09:00:40 +0000 (02:00 -0700)]
Merge "cnss2: Support multiple recoveries when using CORE only firmware"

6 years agoMerge "cnss2: Request runtime PM resume when shutdown happens"
Linux Build Service Account [Wed, 18 Oct 2017 09:00:39 +0000 (02:00 -0700)]
Merge "cnss2: Request runtime PM resume when shutdown happens"

6 years agoMerge "msm: ipa: Fix use after free issue"
Linux Build Service Account [Wed, 18 Oct 2017 09:00:38 +0000 (02:00 -0700)]
Merge "msm: ipa: Fix use after free issue"

6 years agoALSA: usb-audio: Add length check after string desc copy
Ajay Agarwal [Tue, 17 Oct 2017 05:58:17 +0000 (11:28 +0530)]
ALSA: usb-audio: Add length check after string desc copy

It might be possible that negative error code is returned
in 'len', when we try to copy USB string desc into the ID
name buf of snd_kcontrol instance. But even in that case,
we are terminating buf with 0 at the 'len' index, which
leads to memory corruption. And for good case where 'len'
is non-negative, usb_string func is terminating the buf
with 0.

Fix this by removing the termination of buf with '0'
in the caller function.

Change-Id: Ie32d395b0fc91d6c3e1cfdbafb76304e21e40577
Signed-off-by: Ajay Agarwal <ajaya@codeaurora.org>
6 years agoblock: Initialize bd_bdi on inode initialization
Jan Kara [Thu, 2 Mar 2017 15:50:13 +0000 (16:50 +0100)]
block: Initialize bd_bdi on inode initialization

So far we initialized bd_bdi only in bdget(). That is fine for normal
bdev inodes however for the special case of the root inode of
blockdev_superblock that function is never called and thus bd_bdi is
left uninitialized. As a result bdev_evict_inode() may oops doing
bdi_put(root->bd_bdi) on that inode as can be seen when doing:

mount -t bdev none /mnt

Fix the problem by initializing bd_bdi when first allocating the inode
and then reinitializing bd_bdi in bdev_evict_inode().

Thanks to syzkaller team for finding the problem.

Reported-by: Dmitry Vyukov <dvyukov@google.com>
Fixes: b1d2dc5659b4 ("block: Make blk_get_backing_dev_info() safe without open bdev")
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Jens Axboe <axboe@fb.com>
Change-Id: I9c34ca321ab311936946187ed21e25d67caf5ba5
Git-repo: git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git
Git-commit: 7ee143f46cb1f58ee07194076b9a3e6ea3f27ad3
[riteshh@codeaurora.org: resolved merge conflicts]
Signed-off-by: Ritesh Harjani <riteshh@codeaurora.org>
6 years agoblock: Make blk_get_backing_dev_info() safe without open bdev
Jan Kara [Thu, 2 Feb 2017 14:56:52 +0000 (15:56 +0100)]
block: Make blk_get_backing_dev_info() safe without open bdev

Currenly blk_get_backing_dev_info() is not safe to be called when the
block device is not open as bdev->bd_disk is NULL in that case. However
inode_to_bdi() uses this function and may be call called from flusher
worker or other writeback related functions without bdev being open
which leads to crashes such as:

[113031.075540] Unable to handle kernel paging request for data at address 0x00000000
[113031.075614] Faulting instruction address: 0xc0000000003692e0
0:mon> t
[c0000000fb65f900c00000000036cb6c writeback_sb_inodes+0x30c/0x590
[c0000000fb65fa10c00000000036ced4 __writeback_inodes_wb+0xe4/0x150
[c0000000fb65fa70c00000000036d33c wb_writeback+0x30c/0x450
[c0000000fb65fb40c00000000036e198 wb_workfn+0x268/0x580
[c0000000fb65fc50c0000000000f3470 process_one_work+0x1e0/0x590
[c0000000fb65fce0c0000000000f38c8 worker_thread+0xa8/0x660
[c0000000fb65fd80c0000000000fc4b0 kthread+0x110/0x130
[c0000000fb65fe30c0000000000098f0 ret_from_kernel_thread+0x5c/0x6c

Signed-off-by: Jens Axboe <axboe@fb.com>
Change-Id: I26955b919bd05fe34dc60aab1797ea2739ad5fd7
Git-repo: git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git
Git-commit: b1d2dc5659b41741f5a29b2ade76ffb4e5bb13d8
[riteshh@codeaurora.org: resolved merge conflicts]
Signed-off-by: Ritesh Harjani <riteshh@codeaurora.org>
6 years agoblock: Dynamically allocate and refcount backing_dev_info
Jan Kara [Thu, 2 Feb 2017 14:56:51 +0000 (15:56 +0100)]
block: Dynamically allocate and refcount backing_dev_info

Instead of storing backing_dev_info inside struct request_queue,
allocate it dynamically, reference count it, and free it when the last
reference is dropped. Currently only request_queue holds the reference
but in the following patch we add other users referencing
backing_dev_info.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Jens Axboe <axboe@fb.com>
Change-Id: Ibcee7b4c014018f9243cd3edbfd9c4a8877c3862
Git-repo: git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git
Git-commit: d03f6cdc1fc422accb734c7c07a661a0018d8631
[riteshh@codeaurora.org: resolved merge conflicts]
Signed-off-by: Ritesh Harjani <riteshh@codeaurora.org>
6 years agoblock: Use pointer to backing_dev_info from request_queue
Jan Kara [Thu, 2 Feb 2017 14:56:50 +0000 (15:56 +0100)]
block: Use pointer to backing_dev_info from request_queue

We will want to have struct backing_dev_info allocated separately from
struct request_queue. As the first step add pointer to backing_dev_info
to request_queue and convert all users touching it. No functional
changes in this patch.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Jens Axboe <axboe@fb.com>
Change-Id: I77fbb181de7e39c83fbfba8cfb128d6ace161f31
Git-repo: git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git
Git-commit: 97419acd22a0bacc52dbc34d5bbc96d315e48acb
[riteshh@codeaurora.org: resolved merge conflicts]
Signed-off-by: Ritesh Harjani <riteshh@codeaurora.org>
6 years agoblock: Unhash block device inodes on gendisk destruction
Jan Kara [Thu, 2 Feb 2017 14:56:49 +0000 (15:56 +0100)]
block: Unhash block device inodes on gendisk destruction

Currently, block device inodes stay around after corresponding gendisk
hash died until memory reclaim finds them and frees them. Since we will
make block device inode pin the bdi, we want to free the block device
inode as soon as the device goes away so that bdi does not stay around
unnecessarily. Furthermore we need to avoid issues when new device with
the same major,minor pair gets created since reusing the bdi structure
would be rather difficult in this case.

Unhashing block device inode on gendisk destruction nicely deals with
these problems. Once last block device inode reference is dropped (which
may be directly in del_gendisk()), the inode gets evicted. Furthermore if
the major,minor pair gets reallocated, we are guaranteed to get new
block device inode even if old block device inode is not yet evicted and
thus we avoid issues with possible reuse of bdi.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Jens Axboe <axboe@fb.com>
Change-Id: I41087d3bc818c4e58e0b4e20876e136ec1cbb07b
Git-repo: git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git
Git-commit: f44f1ab5a2dcd4e16eab850fd08e40ff2d0c28d4
[riteshh@codeaurora.org: resolved merge conflicts]
Signed-off-by: Ritesh Harjani <riteshh@codeaurora.org>
6 years agodrm/msm/sde: synchronize mdp clk with frame update
Abhinav Kumar [Sat, 14 Oct 2017 04:02:23 +0000 (21:02 -0700)]
drm/msm/sde: synchronize mdp clk with frame update

There is a race condition between the commit and
the validate, where the mdp clock or the bandwidth
can be updated for voting before the actual configuration
is taking place.

Fix this issue by caching the performance values in the crtc
till the current hw configuration is on-going.

Change-Id: Icc71c4f58cbc305529d308335f44b8c05702ebee
Signed-off-by: Abhinav Kumar <abhinavk@codeaurora.org>
6 years agodrm/msm/sde: move current performance setting to crtc object
Alan Kwong [Thu, 8 Jun 2017 14:26:46 +0000 (10:26 -0400)]
drm/msm/sde: move current performance setting to crtc object

Current performance setting is maintained in crtc state, and its
update is synchronized to commit cycle. However, the setting may be
committed to clock and bandwidth driver out of sync with respect to
commit cycle, e.g. update at end of frame while another commit is
validating. As a result, requested settings may be missed and result
in older settings being used. Move current performance setting
to crtc object, from crtc state, so it can be updated at the
same time as the setting is committed to clock and bandwidth
driver.

CRs-Fixed: 2048612
Change-Id: I0c3047e8e806460105eaba5d46145798bd98d721
Signed-off-by: Alan Kwong <akwong@codeaurora.org>
Signed-off-by: Abhinav Kumar <abhinavk@codeaurora.org>
6 years agodrm/msm/sde: align bandwidth/clock updates with frame done
Abhinav Kumar [Sat, 14 Oct 2017 01:43:52 +0000 (18:43 -0700)]
drm/msm/sde: align bandwidth/clock updates with frame done

Align clock and bandwidth updates witf frame done event for
video mode panels.

Also move perf state update to atomic check to make perf data
persistent over state duplication.

Change-Id: I5ef02dd78ce6a62cf49f1f105344764fdb3a24da
Signed-off-by: Abhinav Kumar <abhinavk@codeaurora.org>
6 years agocnss2: Support multiple recoveries when using CORE only firmware
Yue Ma [Fri, 13 Oct 2017 17:56:42 +0000 (10:56 -0700)]
cnss2: Support multiple recoveries when using CORE only firmware

After MHI power on is completed, CORE component only firmware boot
can be considered as success. Add a debug parameter in order to
support multiple recoveries when using the firmware without WLAN
component.

Change-Id: I1b6a67a578ff7ffa15dff733b8d691910003ac6f
CRs-fixed: 2124999
Signed-off-by: Yue Ma <yuem@codeaurora.org>
6 years agoMerge "soc: qcom: ipc_router_glink_xprt: Add rx intent timeout for glink"
Linux Build Service Account [Tue, 17 Oct 2017 21:20:45 +0000 (14:20 -0700)]
Merge "soc: qcom: ipc_router_glink_xprt: Add rx intent timeout for glink"

6 years agoMerge "msm: ipa: add null terminator"
Linux Build Service Account [Tue, 17 Oct 2017 21:20:44 +0000 (14:20 -0700)]
Merge "msm: ipa: add null terminator"

6 years agomsm: ipa: Set ep delay on rmnet/mbim tether pipe
Mohammed Javid [Fri, 15 Sep 2017 19:09:46 +0000 (00:39 +0530)]
msm: ipa: Set ep delay on rmnet/mbim tether pipe

As soon we start a gsi channel, there is a chance
of getting packets from USB to IPA before even
we setup flt/rt fules.

To avoid this race condition set ep delay before
start of gsi channel in case of rmnet/mbim tether.

This delay should be cleared by Q6 once required
flt/rt rules in place.

Change-Id: Iac16ba067d4454ffd3f191009ea6eb5998473977
Signed-off-by: Mohammed Javid <mjavid@codeaurora.org>
6 years agoRevert "leds: remove unused driver leds-qpnp-flash.c"
Anirudh Ghayal [Tue, 17 Oct 2017 03:06:08 +0000 (08:36 +0530)]
Revert "leds: remove unused driver leds-qpnp-flash.c"

This reverts commit 5b289d3a649a06d7eb8850796e39412e0d11749e.

Bring back the flash driver for 8996.

Change-Id: I6232ff004fc790e41f73bdf9d90d5455b5cb8a55
Signed-off-by: Anirudh Ghayal <aghayal@codeaurora.org>
6 years agoRevert "power: supply: qcom: remove unused drivers"
Anirudh Ghayal [Tue, 17 Oct 2017 03:01:12 +0000 (08:31 +0530)]
Revert "power: supply: qcom: remove unused drivers"

This reverts commit 941bfaf2dc352b2ba2cd7e50c55843e4585aaffb.

Bring back the charger and fuel gauge drivers supported on
8996.

Change-Id: Ibba07ece69e29a8c7bf26acdc9dda30ac755fb83
Signed-off-by: Anirudh Ghayal <aghayal@codeaurora.org>
6 years agoMerge "rtac: add size check when reading cal data kvaddr buffer"
Linux Build Service Account [Tue, 17 Oct 2017 12:38:29 +0000 (05:38 -0700)]
Merge "rtac: add size check when reading cal data kvaddr buffer"