Let's switch to OSI mode for QCOM platforms if setting the OSI mode is
supported. We should do this only if the lpm-levels driver is used.
Other VMs using the same kernel but defaulting to ARM cpuidle drivers,
will default to platform coordinated.
Change-Id: I69245bf33361970f37ea2006d669368e30e2ec9d
Signed-off-by: Lina Iyer <ilina@codeaurora.org>
If eud driver is enabled in xbl, there was a explicit need
to enable eud from hlos side also. This change enables eud
if eud_mode_manager2 register is set, and it keeps the check
after tcsr register property check.
Change-Id: I4f99d2fddb36f196514296302c67b75e73545632
Signed-off-by: Komal Bajaj <kbajaj@codeaurora.org>
Signed-off-by: Mukesh Ojha <mojha@codeaurora.org>
For some targets like Bengal, eud enablement expects a TCSR register
to be configured first. This would be one time enablement during
drivers probe. Due to this register disabled, eud was not getting
enabled on Bengal platform.
If this register is not configured for EUD, do not enable
eud-tcsr-check-enable parameter.
Change-Id: Iaabfaecf343090d964d06807dcee5ea6a2dca3dc
Signed-off-by: Shadab Naseem <snaseem@codeaurora.org>
Signed-off-by: Mukesh Ojha <mojha@codeaurora.org>
Commit 05d18ae1cc8a ("scsi: pm: Balance pm_only counter of request queue
during system resume") fixed a problem in the block layer's runtime-PM
code: blk_set_runtime_active() failed to call blk_clear_pm_only().
However, the commit's implementation was awkward; it forced the SCSI
system-resume handler to choose whether to call blk_post_runtime_resume()
or blk_set_runtime_active(), depending on whether or not the SCSI device
had previously been runtime suspended.
This patch simplifies the situation considerably by adding the missing
function call directly into blk_set_runtime_active() (under the condition
that the queue is not already in the RPM_ACTIVE state). This allows the
SCSI routine to revert back to its original form. Furthermore, making this
change reveals that blk_post_runtime_resume() (in its success pathway) does
exactly the same thing as blk_set_runtime_active(). The duplicate code is
easily removed by making one routine call the other.
No functional changes are intended.
Link: https://lore.kernel.org/r/20200706151436.GA702867@rowland.harvard.edu
CC: Can Guo <cang@codeaurora.org>
CC: Bart Van Assche <bvanassche@acm.org>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Bug: 162892567
(cherry picked from commit 8f38f8e0a30e1c33cdfc48eba2bd63ac2eae373a
https://git.kernel.org/pub/scm/linux/kernel/git/mkp/scsi.git 5.9/scsi-queue)
Change-Id: Ibeca9786fb65ee7d782314f48fb39274b82feb57
Signed-off-by: Can Guo <cang@codeaurora.org>
In RT energy aware task placement, cpu utilization is double accounted
to check whether cpu would be overutilized for given task or not.
Fix the double accounting the cpu util.
Change-Id: Id51dd6c952126b870942746c0408fa387ec3bfb1
Signed-off-by: Lingutla Chandrasekhar <clingutla@codeaurora.org>
Propagate following change from 4.14 to 5.4 for SPI.
'commit 0f5e798e98004 ("spi: spi-geni-qcom: Don't
initialize GSI channels for FIFO/SE_DMA mode")'.
Change-Id: I600cb1a522a09c4e9cd2ee1cad3d3f6d4c66a358
Signed-off-by: Ashish Kori <akori@codeaurora.org>
When seek is cancelled before the seek operation completes, Some
time interrupt handler is not called for seek complete and no
SEEK COMPLETE event sent to user space.
Send the SEEK COMPLETE event when seek is cancelled.
CRs-Fixed: 2749392
Change-Id: I61473c325a637e79eb5c7e980921c3ff83381231
Signed-off-by: himta ram <hram@codeaurora.org>
Adding changes for the new target and reading the FW name
from dtsi.
Change-Id: I224ebe001126c33c55fc039e74f42555a4dce1f8
Signed-off-by: suchawla <suchawla@codeaurora.org>
Enable the interconnect driver so that consumers are
able to obtain their path handles properly.
Change-Id: Ic9b11ac283beb5d1c00ba98564377d214d323a82
Signed-off-by: Odelu Kukatla <okukatla@codeaurora.org>
RDS enable is set during scanning and it causing the
incorrect register update for rtc6226 chip and this
leads to no audio issue.
Set the RDS enable bit to 0 before scanning and
set to 1 after scanning finishes.
CRs-Fixed: 2744378
Change-Id: I2c2fae1e01fb1fadccbc9133f472fca996bc2496
Signed-off-by: himta ram <hram@codeaurora.org>
When BT is turned ON on a platform using external power
amplifier(xPA), default voting ON of S3E buck results in
PMIC crash.
Solution is to disable S3E by default in AOP and enable it
dynamically based on the power amplifier confifuration
read from NVM. Handle IOCTL call for setting power
amplifier configuration from BT power driver.
Change-Id: Ibe7ef17481e33456c4692d0820ff44d9f43995de
Signed-off-by: Mahesh Kumar Sharma <smahesh@codeaurora.org>
Add support for PCIE gen switch request from wlan driver using
QMI message to FW.
Change-Id: I6ea4ee234a38162cb9093d36d580c872b4d62644
Signed-off-by: Manikandan Mohan <manikand@codeaurora.org>
Do not check for domains support in kernel as the capability is
queried in userspace itself.
Change-Id: I6b014648ff19d9bfba6f94db078ff748e7de8ce6
Acked-by: Thyagarajan Venkatanarayanan <venkatan@qti.qualcomm.com>
Signed-off-by: Himateja Reddy <hmreddy@codeaurora.org>
If WLAN driver probe fails, its intenal state may not be
proper. Thus check for device powered on state before calling
WLAN driver operations for suspend / resume.
Change-Id: I81da8969e9b0e0525256db8da2c0db25a996b241
Signed-off-by: Manikandan Mohan <manikand@codeaurora.org>
Currently, USB power supply type is set to POWER_SUPPLY_TYPE_USB
in power supply descriptor. power_supply_show_property() would
show POWER_SUPPLY_PROP_TYPE based on psy->desc->type. Userspace
process like healthd would read POWER_SUPPLY_PROP_TYPE to print
the adapter type in logs (e.g. "chg=u" for SDP, "chg=a" for DCP).
Hence update usb_psy_desc type runtime to show the USB power
supply type accordingly. This is useful for testing.
Change-Id: Icd88098f97beba053105d826614b4182f3716144
Signed-off-by: Subbaraman Narayanamurthy <subbaram@codeaurora.org>
To support userspace application and testing, add some more
properties under USB power supply and qcom-battery class.
To show USB connector temperature, following property is added.
- /sys/class/power_supply/usb/temp
To support restricted charging similar to legacy targets,
following parameters are added.
- /sys/class/qcom-battery/restrict_cur
- /sys/class/qcom-battery/restrict_chg
To show whether the Type-C sink/source connected is compliant,
following parameter is added.
- /sys/class/qcom-battery/usb_typec_compliant
To show the custom USB_TYPE for Quick charge adapters like HVDCP,
HVDCP3 etc., following parameter is added.
- /sys/class/qcom-battery/usb_real_type.
Change-Id: Ia8dfba41c6ed3056dda588a1955a60ef19568112
Signed-off-by: Subbaraman Narayanamurthy <subbaram@codeaurora.org>
Switch the clang version to r383902 (clang-11). Update the
ABI snapshot accordingly.
Change-Id: I12e60c1bb3bd1e64f26bdc0b85b08e5e08b72ae9
Signed-off-by: Raghavendra Rao Ananta <rananta@codeaurora.org>
Current code doesn't clean up buffers and dma handles when failure
is seen in the remote invocation to DSP. This will end up leaking
buffers and also dma handles pointing to wrong memory in the fastrpc
kernel. Clean buffers and dma handles even when remote invocation to
DSP returns failure.
Change-Id: I23998a1825f14a4c97380e284626df18f5045ed8
Signed-off-by: Himateja Reddy <hmreddy@codeaurora.org>
The RTB logging was accidently losing the __user tag from I/O operation
addresses which caused the static analysis tools to be upset. Force a
void * address for the logger but leave the original address untouched for
the actual operation.
Change-Id: Ic0dedbad38a49d6bfd9e102f91bbf9ae21c4c14c
Signed-off-by: Jordan Crouse <jcrouse@codeaurora.org>
Add panic notifier, which captures any cache errors
which were not handled, due to interrupts not getting
handled.
Change-Id: I9315a32dcbf6fdfa509c4ab934b75d076fedfa39
Signed-off-by: Neeraj Upadhyay <neeraju@codeaurora.org>
Signed-off-by: Mukesh Ojha <mojha@codeaurora.org>
There is a possibility of null pointer dereference for sk_buff in
the function eth_start_xmit. Fix this by removing sk_buff from if
check at the start of the function eth_start_xmit so that if sk_buff
is null dev_kfree_skb_any will return without proceeding further.
Change-Id: I105e42c59aaa83c8d74cc5a9e894ae4fc87c3c9f
Signed-off-by: Rohith Kollalsi <rkollals@codeaurora.org>
msm_geni_console_setup() can be potentially called from the driver's
probe via uart_add_one_port(). Since it's placed under __init, the
function would be discarded as soon as the module_init portion of the
driver completes, and when called during probe, it causes an instruction
abort. As a result, move the msm_geni_console_setup() out of __init.
Change-Id: I27ebdcd0832bb49d66e85b250b652cc38f4aa333
Signed-off-by: Akash Asthana <akashast@codeaurora.org>
In case of SSR/PDR, when FW comes up, platform driver receives
FW_READY indication. As part of FW_READY, platform driver calls
shutdown and reinit ops of host driver.
Clear SHUTDOWN_DONE flag at right place so that if reinit is not
called SSR/PDR because of FW again went down between calling driver
shutdown and reinit then next time directly call reinit and avoid
calling shutdown.
Change-Id: Icdd259d52db1b84203a9efb2e5948207f3213ce6
Signed-off-by: Naman Padhiar <npadhiar@codeaurora.org>
Change member sid in driver data structure to cap_id.
Fix condition check in hh_vcpu_done_populate_affinity_info for better
readability.
Change-Id: Idc7f8a4fd6b6a7ac8e004eb96c2020a9f1f4444b
Signed-off-by: Sai Harshini Nimmala <snimmala@codeaurora.org>
commit 9acfa80c3d
("BACKPORT: driver core: fw_devlink: Add support for batching fwnode parsing")
changed the way drivers gets probed and by the time hyp core ctl driver tries
to get CPU freq policy, cpufreq driver is not probed which leads to hyp core
ctl driver probe failure. Fix the dependency on CPU Freq policy by moving
freq qos initialization from probe to first write of the min freq node.
Change-Id: I75c628e768ee37a78d1ee54df3e55b6d2389f9a2
Signed-off-by: Satya Durga Srinivasu Prabhala <satyap@codeaurora.org>
Add a new sys node to support runtime dynamic report rate.
Change-Id: Ifa7f4290b648f79cfeec80ae421af4c7c7a36aaf
Signed-off-by: Fei Mao <feim1@codeaurora.org>
-----BEGIN PGP SIGNATURE-----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=A5pn
-----END PGP SIGNATURE-----
Merge 5.4.56 into android11-5.4
Changes in 5.4.56
crypto: ccp - Release all allocated memory if sha type is invalid
media: rc: prevent memory leak in cx23888_ir_probe
sunrpc: check that domain table is empty at module unload.
ath10k: enable transmit data ack RSSI for QCA9884
PCI/ASPM: Disable ASPM on ASMedia ASM1083/1085 PCIe-to-PCI bridge
mm/filemap.c: don't bother dropping mmap_sem for zero size readahead
ALSA: usb-audio: Add implicit feedback quirk for SSL2
ALSA: hda/realtek: enable headset mic of ASUS ROG Zephyrus G15(GA502) series with ALC289
ALSA: hda/realtek: typo_fix: enable headset mic of ASUS ROG Zephyrus G14(GA401) series with ALC289
ALSA: hda/realtek: Fix add a "ultra_low_power" function for intel reference board (alc256)
ALSA: hda/realtek - Fixed HP right speaker no sound
ALSA: hda/hdmi: Fix keep_power assignment for non-component devices
IB/rdmavt: Fix RQ counting issues causing use of an invalid RWQE
vhost/scsi: fix up req type endian-ness
9p/trans_fd: Fix concurrency del of req_list in p9_fd_cancelled/p9_read_work
wireless: Use offsetof instead of custom macro.
ARM: 8986/1: hw_breakpoint: Don't invoke overflow handler on uaccess watchpoints
ARM: dts: imx6sx-sabreauto: Fix the phy-mode on fec2
ARM: dts: imx6sx-sdb: Fix the phy-mode on fec2
ARM: dts: imx6qdl-icore: Fix OTG_ID pin and sdcard detect
virtio_balloon: fix up endian-ness for free cmd id
Revert "drm/amdgpu: Fix NULL dereference in dpm sysfs handlers"
drm/amd/display: Clear dm_state for fast updates
drm/amdgpu: Prevent kernel-infoleak in amdgpu_info_ioctl()
drm/dbi: Fix SPI Type 1 (9-bit) transfer
drm: hold gem reference until object is no longer accessed
rds: Prevent kernel-infoleak in rds_notify_queue_get()
libtraceevent: Fix build with binutils 2.35
net/x25: Fix x25_neigh refcnt leak when x25 disconnect
net/x25: Fix null-ptr-deref in x25_disconnect
xfrm: policy: match with both mark and mask on user interfaces
ARM: dts sunxi: Relax a bit the CMA pool allocation range
xfrm: Fix crash when the hold queue is used.
ARM: dts: armada-38x: fix NETA lockup when repeatedly switching speeds
nvme-tcp: fix possible hang waiting for icresp response
selftests/net: rxtimestamp: fix clang issues for target arch PowerPC
selftests/net: psock_fanout: fix clang issues for target arch PowerPC
selftests/net: so_txtime: fix clang issues for target arch PowerPC
sh/tlb: Fix PGTABLE_LEVELS > 2
sh: Fix validation of system call number
net: hns3: fix a TX timeout issue
net: hns3: fix aRFS FD rules leftover after add a user FD rule
net/mlx5: E-switch, Destroy TSAR when fail to enable the mode
net/mlx5e: Fix error path of device attach
net/mlx5: Verify Hardware supports requested ptp function on a given pin
net/mlx5e: Modify uplink state on interface up/down
net/mlx5e: Fix kernel crash when setting vf VLANID on a VF dev
net: lan78xx: add missing endpoint sanity check
net: lan78xx: fix transfer-buffer memory leak
rhashtable: Fix unprotected RCU dereference in __rht_ptr
mlx4: disable device on shutdown
mlxsw: core: Increase scope of RCU read-side critical section
mlxsw: core: Free EMAD transactions using kfree_rcu()
ibmvnic: Fix IRQ mapping disposal in error path
bpf: Fix map leak in HASH_OF_MAPS map
mac80211: mesh: Free ie data when leaving mesh
mac80211: mesh: Free pending skb when destroying a mpath
arm64/alternatives: move length validation inside the subsection
arm64: csum: Fix handling of bad packets
Bluetooth: fix kernel oops in store_pending_adv_report
net: nixge: fix potential memory leak in nixge_probe()
net: gemini: Fix missing clk_disable_unprepare() in error path of gemini_ethernet_port_probe()
net/mlx5e: fix bpf_prog reference count leaks in mlx5e_alloc_rq
perf tools: Fix record failure when mixed with ARM SPE event
vxlan: fix memleak of fdb
usb: hso: Fix debug compile warning on sparc32
selftests: fib_nexthop_multiprefix: fix cleanup() netns deletion
qed: Disable "MFW indication via attention" SPAM every 5 minutes
selftests: net: ip_defrag: modprobe missing nf_defrag_ipv6 support
nfc: s3fwrn5: add missing release on skb in s3fwrn5_recv_frame
scsi: core: Run queue in case of I/O resource contention failure
parisc: add support for cmpxchg on u8 pointers
net: ethernet: ravb: exit if re-initialization fails in tx timeout
Revert "i2c: cadence: Fix the hold bit setting"
x86/unwind/orc: Fix ORC for newly forked tasks
x86/stacktrace: Fix reliable check for empty user task stacks
cxgb4: add missing release on skb in uld_send()
xen-netfront: fix potential deadlock in xennet_remove()
RISC-V: Set maximum number of mapped pages correctly
drivers/net/wan: lapb: Corrected the usage of skb_cow
KVM: arm64: Don't inherit exec permission across page-table levels
KVM: LAPIC: Prevent setting the tscdeadline timer if the lapic is hw disabled
x86/i8259: Use printk_deferred() to prevent deadlock
perf tests bp_account: Make global variable static
perf env: Do not return pointers to local variables
perf bench: Share some global variables to fix build with gcc 10
Linux 5.4.56
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: I06b0147860242cfcac3dd8a7acefdb211020cab1
