After enabling the GPU clocks, the GPU can pagefault
when trying to access memory(example the ringbuffer).
This patch addresses the pagefault issue by enabling
the memory retention flags on the GPU core clock.
Change-Id: Ibabecba77501d6a3b188b19c90c172de7d667c8c
Signed-off-by: Sharat Masetty <smasetty@codeaurora.org>
The status debug property lists all of the voters, their enable and
value status, and the effective result. Currently the status is
presented on multiple lines which is difficult to parse. Change the
format to be more parsable.
Print each voter in the following format:
VOTABLE: CLIENT: en=1 v=1000
Print a single summary line at the end in the following format:
VOTABLE: effective=EFFECTIVE_CLIENT type=VOTABLE_TYPE v=1000
Change-Id: I0671cdd8e0911450bafe4869cb49cc48bd622a81
Signed-off-by: Nicholas Troast <ntroast@codeaurora.org>
To avoid leak info of pointers addresses to dmesg,
use %pK instead of %p to print virtual addresses.
Change-Id: I7d8e900d7cd62e9ad3fb9ea3ba9865d6911bdfcb
Acked-by: Chenna Kesava Raju <chennak@qti.qualcomm.com>
Signed-off-by: Tharun Kumar Merugu <mtharu@codeaurora.org>
cnss_genl driver creates a netlink family and multicast groups
to facilitate communication between WLAN driver and userspace.
Define flag CONFIG_CNSS_GENL and set to 'y'(yes) to enable
compilation of the cnss_genl driver inorder to use the same.
Change-Id: I9d81f933e4fe3c9383f213c777d335e1fe51bdd2
Signed-off-by: Srinivas Dasari <dasaris@codeaurora.org>
CRs-Fixed: 1117130
Fix pan control gain param to accommodate Q28 format.
Convert 32bit gain to be inlined with 16bit mixer coefficients
of ADSP.
Change-Id: I85ebd11c49b4af982f34091937acabb9e6580e8f
Signed-off-by: Varun Balaraj <varunb@codeaurora.org>
Few PCIe clients does not have option to call for enumeration.
Update boot option for PCIe such that enumeration is done as
part of PCIe RC probe.
Change-Id: I1bb977b7e21c046ee7757ed509762897e2bcdd8d
Signed-off-by: Rama Krishna Phani A <rphani@codeaurora.org>
Correct FM on/off sequence with following changes ->
- FM slave port close is called after LPASS port close (remove hw_free)
- Use the watermark level of 8 for FM slave ports
- Do not reset the watermark to 0 upon port disable
Change-Id: Ie785715c2b92dae6b46b346bebd45e044f54ea53
Signed-off-by: Rupesh Tatiya <rtatiya@codeaurora.org>
Add tp rev property to expose TP-REV information via
rev-id peripheral.
Change-Id: I0236f84947dbf256f090ea32895904024c1f3fcf
Signed-off-by: Rama Krishna Phani A <rphani@codeaurora.org>
Update scaling factor for RRADC USBIN current
based on USBIN voltage.
Change-Id: I87b2ac56166f2e0812ec56d88c68298094eb86ac
Signed-off-by: Rama Krishna Phani A <rphani@codeaurora.org>
Currently, there is a possibility of using already
freed memory while client reads from diag driver.
The patch adds proper protection to fix the issue.
CRs-Fixed: 2076623
Change-Id: Ic946865ac79f2684c06176be64fd2f3abc6048f7
Signed-off-by: Mohit Aggarwal <maggarwa@codeaurora.org>
Some PMICs software drivers (PM660) require the TP revision
information to implement specific logic. Add an optional
property qcom,tp-rev-valid to check if support for TP REV is
available in the REVID peripheral for the PMIC.
Change-Id: Ia7a344e60bedcc6fb28d9ed9f18dc29a3d7f4fd7
Signed-off-by: Rama Krishna Phani A <rphani@codeaurora.org>
Changes to reflect slight change in HAB API return values.
Change-Id: I8825f6fca6a50449a929f05c317d0beeb6908d26
Signed-off-by: Amit Blay <ablay@codeaurora.org>
MTU defaults to standard CAN frame, however,
since rh850 driver supports CANFD frames MTU
needs to be set accordingly.
CRs-Fixed: 2071638
Change-Id: I5015af6e61ef4b4e021a715a8dec60d42a8a9e5b
Signed-off-by: Alex Yakavenka <ayakav@codeaurora.org>
Currently we use WIPOWER as the type for dc power supply.
Instead use WIRELESS in the type and use WIPOWER in the
real_type property.
Change-Id: I9081e752e5b3ff725a6ba935d6f6df4e36fec3fa
Signed-off-by: Abhijeet Dharmapurikar <adharmap@codeaurora.org>
Add metadata to configure latency mode for DSP transcode
loopback to set legacy or low latency path. Userspace API
introduced to configure the same.
CRs-Fixed: 2092562
Change-Id: I914c68a9e9d8647530b72c42548e571b7508c423
Signed-off-by: Siddartha Shaik <sshaik@codeaurora.org>
Add stream volume control support for DSP transcode
loopback to support mute configuration requirement.
CRs-Fixed: 2092562
Change-Id: I2003e40c9888245c1b12f0e7fbd364170d5008cf
Signed-off-by: Siddartha Shaik <sshaik@codeaurora.org>
p2p_find uses primary interface mac address during active
scan instead of the later created p2p interface on which
wpa_supplicant is running.
wpa_supplicant creates a p2p_device interface if the driver
exposes a dedicated p2p_device interface capability. The mac
address for the created p2p_device interface will be same as of
the primary interface registered with the netdev.
By disabling the dedicated p2p_device interface capabilty
for wcn3990, wpa_supplicant uses the interface passed to it
from commandline interface and hence will use the correct
mac address for the active scan purpose.
CRs-Fixed: 2092523
Change-Id: I42ff819bb195e5f5dd7fef9908786621996e93b7
Signed-off-by: Rakesh Pillai <pillair@codeaurora.org>
Changes to vote for camera clocks so as to maintain clock vote
when early camera transitions from LK to kernel.
Changes to vote off also included once early camera shuts down.
Change-Id: Ic7fc15d706b4c9009abd70db4773c3355e5fbbdd
Signed-off-by: Rahul Sharma <sharah@codeaurora.org>
Add QBT100 device for SDM660 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: I64482d2c36c48837f7ccf68cfbe73da466e3f836
Signed-off-by: Abir Ghosh <abirg@codeaurora.org>
Changes to enable the laser sensor on the sdm660.
Change-Id: I2b0c8dc6454d7e99d2073acbcf58cc2f0f73148a
Signed-off-by: Wei Ding <weiding@codeaurora.org>
Add changes to qbt1000 fingerprint driver to enable
input device to report KEY_VOLUMEDOWN when a finger
is detected. Add changes for supporting CBGE with
retry logic and multiple IPC messages.
Change-Id: I29a52c516a8f9216abc623ca3bfaebe19f89eaa3
Signed-off-by: Abir Ghosh <abirg@codeaurora.org>
If device is created by of_platform_populate and removed by
of_platform_depopulate from parent device driver, platform_data
will be released twice and lead to kernel panic.
of_platform_depopulate will free platform_data of each device driver.
Usually we use devm_kzalloc to malloc for platform_data so it can be
released during driver unregister. So if device is removed by
of_platform_depopulate and platform_data is malloced by devm_kzalloc,
platform_data will be freed twice. To fix it, move this data into
struct usb_bam_ctx_type and then store usb_bam_ctx_type into device
driver data. In this way, it can be get in remove function.
The member regs of usb_bam_ctx_type is not necessary. It is only
used for ioremap which already has been done by virt_addr of
struct sps_bam_props.
Without sps_deregister_bam_device, bam child device will not be
created again during defer probe. So deregister in remove function.
Change-Id: Ibae80745be5810d400a1c05566a99efcc2190020
Signed-off-by: Ziqi Chen <ziqic@codeaurora.org>
In designs without the pd-phy interface the usb
insertion/removal and CC_orientation needs to be
notified to the usb stack by the charger driver.
Add this notification support via the extcon interface.
CRs-Fixed: 2086422
Change-Id: I22eae3f818859c8fbd08066f690517d4210c7dd3
Signed-off-by: Ankit Sharma <ansharma@codeaurora.org>
When LED is changing between fully on and blinking, there is a change
in PWM/LPG channel operational mode. Use pwm_change_mode() API to notify
the mode change to the PWM driver. This way, the operational mode and
the underlying configuration for PWM channel can be updated by PWM driver.
For example, this will help with the following use cases where the mode
can be switched at runtime.
1) When the LED is fully ON, the LPG channel is in PWM mode.
2) When the LED is blinking, the LPG channel is in LPG mode.
CRs-Fixed: 2070945
Change-Id: Idd7fc31de085b7b1f97c9bf823dfd49a46164018
Signed-off-by: Fenglin Wu <fenglinw@codeaurora.org>
commit c8d3578f8b ("pwm: pwm-qpnp: Ignore changing period setting if
it's not requested") had fixed an issue to avoid glitches seen in PWM
output when the PWM period is not changed and the duty cycle alone is
changed. However, that didn't fix the issue completely when the LPG
channel operation mode might be configured differently even when the
duty cycle is changed and the period is unchanged. This will lead to LPG
channel getting configured to operate in a wrong mode. Fix this by
enabling PWM output whenever there is a change in the mode for LPG
channel requested by the client.
CRs-Fixed: 2070945
Change-Id: Ida6145a7cb9016e9e9116fbcd6b8ba3f6c5b3bc1
Signed-off-by: Fenglin Wu <fenglinw@codeaurora.org>
Multiple root complexes can enumerate concurrently.
Add a global lock to prevent race condition when PCI
framework discovers and updates the PCIe bus
driver's global device table.
Change-Id: I9c88f12aef9fce05d95cfb2fa0f5374c6c4ab1e8
Signed-off-by: Tony Truong <truong@codeaurora.org>
Fix a bug for incorrect jumping out of a switch statement when
showing driver status debugfs node. PCIe link up/down description
is also missing for device boot debugfs node. Add them back.
Change-Id: I63f4b61e837643566cd6b802ce94191b7a96ba7e
CRs-fixed: 2059087
Signed-off-by: Yue Ma <yuem@codeaurora.org>
* refs/heads/tmp-9f764bb
Linux 4.4.80
ASoC: dpcm: Avoid putting stream state to STOP when FE stream is paused
scsi: snic: Return error code on memory allocation failure
scsi: fnic: Avoid sending reset to firmware when another reset is in progress
HID: ignore Petzl USB headlamp
ALSA: usb-audio: test EP_FLAG_RUNNING at urb completion
sh_eth: enable RX descriptor word 0 shift on SH7734
nvmem: imx-ocotp: Fix wrong register size
arm64: mm: fix show_pte KERN_CONT fallout
vfio-pci: Handle error from pci_iomap
video: fbdev: cobalt_lcdfb: Handle return NULL error from devm_ioremap
perf symbols: Robustify reading of build-id from sysfs
perf tools: Install tools/lib/traceevent plugins with install-bin
xfrm: Don't use sk_family for socket policy lookups
tools lib traceevent: Fix prev/next_prio for deadline tasks
Btrfs: adjust outstanding_extents counter properly when dio write is split
usb: gadget: Fix copy/pasted error message
ACPI / scan: Prefer devices without _HID/_CID for _ADR matching
ARM: s3c2410_defconfig: Fix invalid values for NF_CT_PROTO_*
ARM64: zynqmp: Fix i2c node's compatible string
ARM64: zynqmp: Fix W=1 dtc 1.4 warnings
dmaengine: ti-dma-crossbar: Add some 'of_node_put()' in error path.
dmaengine: ioatdma: workaround SKX ioatdma version
dmaengine: ioatdma: Add Skylake PCI Dev ID
openrisc: Add _text symbol to fix ksym build error
irqchip/mxs: Enable SKIP_SET_WAKE and MASK_ON_SUSPEND
ASoC: nau8825: fix invalid configuration in Pre-Scalar of FLL
spi: dw: Make debugfs name unique between instances
ASoC: tlv320aic3x: Mark the RESET register as volatile
irqchip/keystone: Fix "scheduling while atomic" on rt
vfio-pci: use 32-bit comparisons for register address for gcc-4.5
drm/msm: Verify that MSM_SUBMIT_BO_FLAGS are set
drm/msm: Ensure that the hardware write pointer is valid
net/mlx4: Remove BUG_ON from ICM allocation routine
ipv6: Should use consistent conditional judgement for ip6 fragment between __ip6_append_data and ip6_finish_output
ARM: dts: n900: Mark eMMC slot with no-sdio and no-sd flags
r8169: add support for RTL8168 series add-on card.
x86/mce/AMD: Make the init code more robust
tpm: Replace device number bitmap with IDR
tpm: fix a kernel memory leak in tpm-sysfs.c
xen/blkback: don't use xen_blkif_get() in xen-blkback kthread
xen/blkback: don't free be structure too early
sched/cputime: Fix prev steal time accouting during CPU hotplug
net: skb_needs_check() accepts CHECKSUM_NONE for tx
pstore: Use dynamic spinlock initializer
pstore: Correctly initialize spinlock and flags
pstore: Allow prz to control need for locking
vlan: Propagate MAC address to VLANs
/proc/iomem: only expose physical resource addresses to privileged users
Make file credentials available to the seqfile interfaces
v4l: s5c73m3: fix negation operator
dentry name snapshots
ipmi/watchdog: fix watchdog timeout set on reboot
libnvdimm, btt: fix btt_rw_page not returning errors
RDMA/uverbs: Fix the check for port number
PM / Domains: defer dev_pm_domain_set() until genpd->attach_dev succeeds if present
sched/cgroup: Move sched_online_group() back into css_online() to fix crash
kaweth: fix oops upon failed memory allocation
kaweth: fix firmware download
mpt3sas: Don't overreach ioc->reply_post[] during initialization
mailbox: handle empty message in tx_tick
mailbox: skip complete wait event if timer expired
mailbox: always wait in mbox_send_message for blocking Tx mode
wil6210: fix deadlock when using fw_no_recovery option
ath10k: fix null deref on wmi-tlv when trying spectral scan
isdn/i4l: fix buffer overflow
isdn: Fix a sleep-in-atomic bug
net: phy: Do not perform software reset for Generic PHY
nfc: fdp: fix NULL pointer dereference
xfs: don't BUG() on mixed direct and mapped I/O
perf intel-pt: Ensure never to set 'last_ip' when packet 'count' is zero
perf intel-pt: Use FUP always when scanning for an IP
perf intel-pt: Fix last_ip usage
perf intel-pt: Fix ip compression
drm: rcar-du: Simplify and fix probe error handling
drm: rcar-du: Perform initialization/cleanup at probe/remove time
drm/rcar: Nuke preclose hook
Staging: comedi: comedi_fops: Avoid orphaned proc entry
Revert "powerpc/numa: Fix percpu allocations to be NUMA aware"
KVM: PPC: Book3S HV: Save/restore host values of debug registers
KVM: PPC: Book3S HV: Reload HTM registers explicitly
KVM: PPC: Book3S HV: Restore critical SPRs to host values on guest exit
KVM: PPC: Book3S HV: Context-switch EBB registers properly
drm/nouveau/bar/gf100: fix access to upper half of BAR2
drm/vmwgfx: Fix gcc-7.1.1 warning
md/raid5: add thread_group worker async_tx_issue_pending_all
crypto: authencesn - Fix digest_null crash
powerpc/pseries: Fix of_node_put() underflow during reconfig remove
net: reduce skb_warn_bad_offload() noise
pstore: Make spinlock per zone instead of global
af_key: Add lock to key dump
ANDROID: binder: Don't BUG_ON(!spin_is_locked()).
Linux 4.4.79
alarmtimer: don't rate limit one-shot timers
tracing: Fix kmemleak in instance_rmdir
spmi: Include OF based modalias in device uevent
of: device: Export of_device_{get_modalias, uvent_modalias} to modules
drm/mst: Avoid processing partially received up/down message transactions
drm/mst: Avoid dereferencing a NULL mstb in drm_dp_mst_handle_up_req()
drm/mst: Fix error handling during MST sideband message reception
RDMA/core: Initialize port_num in qp_attr
ceph: fix race in concurrent readdir
staging: rtl8188eu: add TL-WN722N v2 support
Revert "perf/core: Drop kernel samples even though :u is specified"
perf annotate: Fix broken arrow at row 0 connecting jmp instruction to its target
target: Fix COMPARE_AND_WRITE caw_sem leak during se_cmd quiesce
udf: Fix deadlock between writeback and udf_setsize()
NFS: only invalidate dentrys that are clearly invalid.
Input: i8042 - fix crash at boot time
MIPS: Fix a typo: s/preset/present/ in r2-to-r6 emulation error message
MIPS: Send SIGILL for linked branches in `__compute_return_epc_for_insn'
MIPS: Rename `sigill_r6' to `sigill_r2r6' in `__compute_return_epc_for_insn'
MIPS: Send SIGILL for BPOSGE32 in `__compute_return_epc_for_insn'
MIPS: math-emu: Prevent wrong ISA mode instruction emulation
MIPS: Fix unaligned PC interpretation in `compute_return_epc'
MIPS: Actually decode JALX in `__compute_return_epc_for_insn'
MIPS: Save static registers before sysmips
MIPS: Fix MIPS I ISA /proc/cpuinfo reporting
x86/ioapic: Pass the correct data to unmask_ioapic_irq()
x86/acpi: Prevent out of bound access caused by broken ACPI tables
MIPS: Negate error syscall return in trace
MIPS: Fix mips_atomic_set() with EVA
MIPS: Fix mips_atomic_set() retry condition
ftrace: Fix uninitialized variable in match_records()
vfio: New external user group/file match
vfio: Fix group release deadlock
f2fs: Don't clear SGID when inheriting ACLs
ipmi:ssif: Add missing unlock in error branch
ipmi: use rcu lock around call to intf->handlers->sender()
drm/radeon: Fix eDP for single-display iMac10,1 (v2)
drm/radeon/ci: disable mclk switching for high refresh rates (v2)
drm/amd/amdgpu: Return error if initiating read out of range on vram
s390/syscalls: Fix out of bounds arguments access
Raid5 should update rdev->sectors after reshape
cx88: Fix regression in initial video standard setting
x86/xen: allow userspace access during hypercalls
md: don't use flush_signals in userspace processes
usb: renesas_usbhs: gadget: disable all eps when the driver stops
usb: renesas_usbhs: fix usbhsc_resume() for !USBHSF_RUNTIME_PWCTRL
USB: cdc-acm: add device-id for quirky printer
usb: storage: return on error to avoid a null pointer dereference
xhci: Fix NULL pointer dereference when cleaning up streams for removed host
xhci: fix 20000ms port resume timeout
ipvs: SNAT packet replies only for NATed connections
PCI/PM: Restore the status of PCI devices across hibernation
af_key: Fix sadb_x_ipsecrequest parsing
powerpc/asm: Mark cr0 as clobbered in mftb()
powerpc: Fix emulation of mfocrf in emulate_step()
powerpc: Fix emulation of mcrf in emulate_step()
powerpc/64: Fix atomic64_inc_not_zero() to return an int
iscsi-target: Add login_keys_workaround attribute for non RFC initiators
scsi: ses: do not add a device to an enclosure if enclosure_add_links() fails.
PM / Domains: Fix unsafe iteration over modified list of domain providers
PM / Domains: Fix unsafe iteration over modified list of device links
ASoC: compress: Derive substream from stream based on direction
wlcore: fix 64K page support
Bluetooth: use constant time memory comparison for secret values
perf intel-pt: Clear FUP flag on error
perf intel-pt: Ensure IP is zero when state is INTEL_PT_STATE_NO_IP
perf intel-pt: Fix missing stack clear
perf intel-pt: Improve sample timestamp
perf intel-pt: Move decoder error setting into one condition
NFC: Add sockaddr length checks before accessing sa_family in bind handlers
nfc: Fix the sockaddr length sanitization in llcp_sock_connect
nfc: Ensure presence of required attributes in the activate_target handler
NFC: nfcmrvl: fix firmware-management initialisation
NFC: nfcmrvl: use nfc-device for firmware download
NFC: nfcmrvl: do not use device-managed resources
NFC: nfcmrvl_uart: add missing tty-device sanity check
NFC: fix broken device allocation
ath9k: fix tx99 bus error
ath9k: fix tx99 use after free
thermal: cpu_cooling: Avoid accessing potentially freed structures
s5p-jpeg: don't return a random width/height
ir-core: fix gcc-7 warning on bool arithmetic
disable new gcc-7.1.1 warnings for now
sched/fair: Add a backup_cpu to find_best_target
sched/fair: Try to estimate possible idle states.
sched/fair: Sync task util before EAS wakeup
Revert "sched/fair: ensure utilization signals are synchronized before use"
sched/fair: kick nohz idle balance for misfit task
sched/fair: Update signals of nohz cpus if we are going idle
events: add tracepoint for find_best_target
sched/fair: streamline find_best_target heuristics
UPSTREAM: af_key: Fix sadb_x_ipsecrequest parsing
ANDROID: lowmemorykiller: Add tgid to kill message
Revert "proc: smaps: Allow smaps access for CAP_SYS_RESOURCE"
Conflicts:
drivers/gpu/drm/msm/adreno/adreno_gpu.c
drivers/gpu/drm/msm/msm_ringbuffer.c
drivers/staging/android/lowmemorykiller.c
kernel/sched/fair.c
Change-Id: Ic3b3a522b79b1deb178e513b56b9c39eea48e079
Signed-off-by: Blagovest Kolenichev <bkolenichev@codeaurora.org>
Quaternary MI2S capture dai update to support HDMI In
feature requirement using DSP HW transcode loopback.
CRs-Fixed: 2091424
Change-Id: I144bd8ad8b0e2f708acf78be00e8a93a25a88f2e
Signed-off-by: Siddartha Shaik <sshaik@codeaurora.org>
Add the laser device with the required GPIO and
regulator configuration for sdm660 CDP and MTP.
Change-Id: I763508184eab633df0d45449addcc7d79a366e04
Signed-off-by: Samyukta Mogily <smogily@codeaurora.org>
Fix possible out of bound array access by
putting bounds check in __map_and_update_binfo.
Use u32 to access array to simplify bounds check.
Change-Id: I127fdc76fc273c0bcca727e9264a1d0f208dac1b
Signed-off-by: Surajit Podder <spodder@codeaurora.org>
The moment the online event is notified to userspace, it may happen that
the userspace might enable the WDSP. This causes race between the
enabling of WDSP and SSR handling of WDSP. Change the sequence to notify
online event after all SSR handling is completed.
Change-Id: I3cb5d40034884cdfc35de957fab4dafd42d0697c
Signed-off-by: Bhalchandra Gajare <gajare@codeaurora.org>
Whenever subsystem restart or DSP boot occurs, the manager driver
parses the dsp image and stores the segments in list. During
certain back to back regression tests, it is observed that SSR
is failing due to race between SSR handling and DSP enablement.
Fix this by acquiring the ssr_mutex during enabling DSP and
sequence to list operations such that there is no race conditions.
Change-Id: I15c55bf96737e4ffd7e0faf571a1109ba6c38163
Signed-off-by: Bhalchandra Gajare <gajare@codeaurora.org>
Turn off the GPU power and free all resources allocated during
GPU init in case hardware init fails in adreno_gpu_load. This is
required to make sure further tries to load the GPU again doesn't
fail because of invalid GPU state.
Change-Id: I1d0d68f62be751d76274975e098364131712ca38
Signed-off-by: Deepak Kumar <dkumar@codeaurora.org>
Use mutex lock for all VB2 operations, and use separatelock other
than used for stop streaming operation to avoid dead lock.
Change-Id: Ibb466f81cb2825a77073712419fd17c9759ecde4
Signed-off-by: Rahul Sharma <sharah@codeaurora.org>
If PCIe controller requests exit from L1ss shortly after it brings
rxelecidle_disable high, then it's possible that the PHY pipe clock
will turn back on for a short period of time and then go back off
asychronously. Increase PCIe PHY PLL wakeup delay to avoid this.
Change-Id: I1cace039131879969112e1690d07a8d367c06c6b
Signed-off-by: Tony Truong <truong@codeaurora.org>
Pktlog was not enabled for rx data packets.
Enable pktlog capture for rx data packets.
CRs-Fixed: 2091228
Change-Id: I8f489065081ba4da7ad7f5b8e271272279124abc
Signed-off-by: Rakesh Pillai <pillair@codeaurora.org>
Increase the boost headroom to 200mV. Also add a device tree
property to make the headroom configurable.
Change-Id: Ibc932f191e64824ba948153a7ae80f109ffcdff9
Signed-off-by: Kiran Gunda <kgunda@codeaurora.org>
