When DEBUG_FS is not defined, the cluster->stats may be initialized to
an error value. Check for valid stats pointer before updating the start
and end times.
Change-Id: I126766a017ebd9a84bd40f19ab3ff47877615429
Signed-off-by: Lina Iyer <ilina@codeaurora.org>
Existing scan logic will start repetitive seek
from lower freq and stops when band limit reached,
with this logic valid lower band is always ignored.
Seek one more time one band limit reached to check
if lower freq channel is valid.
CRs-Fixed: 2683848
Change-Id: I0b2608b544e0402298d1dfdc1bef96ff8ca64123
Signed-off-by: himta ram <hram@codeaurora.org>
avoid event signaled twice.
add ioctl to check if event was signaled.
Change-Id: Ie75b9db86f2badddc08d65eee30ecd6a15795256
Signed-off-by: Amir Samuelov <amirs@codeaurora.org>
If FM is tuned to invalid freq and scan performed then invalid
freq also included in scan list. Post the tune event with
original frequency after the scan complete event posted to
user space to avoid the invalid tuned freq in scan list.
CRs-Fixed: 2681008
Change-Id: I8ac3b43742a4f92272fc5b0283fabcad2e04de5b
Signed-off-by: himta ram <hram@codeaurora.org>
Currently the make targets for CONFIG_USB_F_RNDIS and
CONFIG_USB_F_GSI both include rndis.o. This would work if only
one is enabled, or both are enabled as =Y as rndis.o will get
linked into the vmlinux image once. Otherwise if both are
enabled with at least one as a module, it results in a module
load failure due to duplicate symbols.
A simple fix for this is to let the usb_f_gsi.o module only
include rndis.o if CONFIG_USB_F_RNDIS is disabled.
Change-Id: Id5b96aaa9a728aa26a65b2c66f2341d6e2fb786b
Signed-off-by: Jack Pham <jackp@codeaurora.org>
This reverts commit 56a743ea1e.
Instead of needing to add a callback to the rndis_params structure,
enable/disable flow control by intercepting the RNDIS_MSG_HALT and
RNDIS_MSG_SET messages directly from gsi_rndis_command_complete().
Change-Id: Id1ecf46d9b4da585be49df4447f073b772b30468
Signed-off-by: Jack Pham <jackp@codeaurora.org>
If the device and link are marked as off, suspend routine wouldn't
turn-off the device and link.
But the regulators would be put to lpm mode.
However, the ufs device is still drawing active mode current.
Thus this leads to a OCP in PMIC.
Hence, don't mark the device and link as off in case of error,
if it doesn't reflect the correct state.
Change-Id: I8d5a738a92a1f339c0500a617f55cb77009356b2
Signed-off-by: Asutosh Das <asutoshd@codeaurora.org>
This change adds ipc logging support for qdss function
driver.
Change-Id: I53be137a161bd7a4b0c4be250938e8fa46b2254e
Signed-off-by: Pratham Pratap <prathampratap@codeaurora.org>
Signed-off-by: Mayank Rana <mrana@codeaurora.org>
We observe the IPA RX stall due to the DEAGGR_ERR_IRQ
of PACKET_SIZE > FRAME_SIZE on rndis.
In IPA, some of de-aggregation error handling checks
are dependent on the syspipe_err_detection config bit.
Adding IGNORE_MIN_PKT_ERR bit for rndis as well based
on recommendations.
Change-Id: Iad1c816bf3371696bf665a8702fcf27490e230d0
Signed-off-by: Bojun Pan <bojunp@codeaurora.org>
Create a config flag to add Qualcomm specific code to upstream
MMC drivers.
Change-Id: I551dd96d0fde14f08f3270a0fee5842e7309d28d
Signed-off-by: Bao D. Nguyen <nguyenb@codeaurora.org>
Add the clock scaling support for SD card in the
sdhci-msm layger.
Change-Id: I17a279bea4ee589121785006e1887a6f39e3ce6a
Signed-off-by: Bao D. Nguyen <nguyenb@codeaurora.org>
Add the clock scaling support for SD card to the core layer.
Change-Id: I9bfc8b581aed7013ecd2f8af65643a4e242d4b72
Signed-off-by: Bao D. Nguyen <nguyenb@codeaurora.org>
Add the clock scaling support for SD card in the block layer.
Change-Id: Iac02d68ce01dd2f2437b1b09dba1070c63aa0317
Signed-off-by: Bao D. Nguyen <nguyenb@codeaurora.org>
Commit I4f3c66a1b567976aadb4a4f91a3dd66230386981 removed support for
CFS_BANDWIDTH, but few stubs are left out. Remove remaining stubs.
Change-Id: I670102fa2efb585106e9c7b60277ff142d64465b
Signed-off-by: Shaleen Agrawal <shalagra@codeaurora.org>
Wakeup capable GPIO IRQs routed via PDC are not being migrated when a
CPU is hotplugged. Add affinity callbacks to msmgpio IRQ chip to update the
affinity of wakeup capable IRQs.
Change-Id: Ic0892d583b567c7d6a2c7ab7fa9d75502c3f3a7a
Signed-off-by: Venkata Narendra Kumar Gutta <vnkgutta@codeaurora.org>
Signed-off-by: Maulik Shah <mkshah@codeaurora.org>
Send the actual keylen instead of the shm
bridge allocation length.
Change-Id: I662078b10bf47cb1747fdec8c147faaa64b3e95d
Signed-off-by: Barani Muthukumaran <bmuthuku@codeaurora.org>
1. Fix key invalidate for double keys so that actual
FBE keys can be removed when user is removed.
2. Don't use ICE slot for GP slot.
3. Handle KDF failures gracefully.
Change-Id: I161b96d36206378735ba564430e1f413065ba255
Signed-off-by: Gaurav Kashyap <gaurkash@codeaurora.org>
Currently driver is suppressing the negative temperature
readings from the vadc. Consumers of the thermal zones need
to read the negative temperature too. Don't suppress the
readings.
Change-Id: I147c270176d2cfce63fee71bceb64e784cfb1efa
Signed-off-by: Veera Vegivada <vvegivad@codeaurora.org>
in get_ports_legacy function, if there is no ports and port child nodes
of device node, just return to avoid unnecessary actions.
Change-Id: Ie392f1fe8265cb3dba38ac687a0a1612e772a943
Signed-off-by: Mao Jinlong <jinlmao@codeaurora.org>
Set max size of dma segment to support requests of larger sizes.
Change-Id: I131a36bfc828d80c26bd5f463133edec6707c7fd
Signed-off-by: Prerna Kalla <prernak@codeaurora.org>
APPS and RPM peripherals are communicated using Glink protocol.
Enable Glink over RPM to support this communication.
Change-Id: Iae0e51fb813054fd08435fd8183602347a93cdae
Signed-off-by: Deepak Kumar Singh <deesin@codeaurora.org>
Upstream commit '26c7295be0c5e("kthread: Do not preempt current
task if it is going to call schedule()")' makes parent thread to not
call wait_task_inactive(child) on kthread create completion, which
intern avoids calling schedule_hrtimeout() and reduce bootup latencies.
The commit 'a9e87164b8984(sched: reduce sleep duration in
wait_task_inactive())' reduces wait time to 1msec to avoid bootup
latencies. Which can be reverted now.
Change-Id: I4b2fd0abe5a264f4d662bdeab4160d1977ef6c22
Signed-off-by: Lingutla Chandrasekhar <clingutla@codeaurora.org>
when we create a kthread with ktrhead_create_on_cpu(),the child thread
entry is ktread.c:ktrhead() which will be preempted by the parent after
call complete(done) while schedule() is not called yet,then the parent
will call wait_task_inactive(child) but the child is still on the runqueue,
so the parent will schedule_hrtimeout() for 1 jiffy,it will waste a lot of
time,especially on startup.
parent child
ktrhead_create_on_cpu()
wait_fo_completion(&done) -----> ktread.c:ktrhead()
|----- complete(done);--wakeup and preempted by parent
kthread_bind() <------------| |-> schedule();--dequeue here
wait_task_inactive(child) |
schedule_hrtimeout(1 jiffy) -|
So we hope the child just wakeup parent but not preempted by parent, and the
child is going to call schedule() soon,then the parent will not call
schedule_hrtimeout(1 jiffy) as the child is already dequeue.
The same issue for ktrhead_park()&&kthread_parkme().
This patch can save 120ms on rk312x startup with CONFIG_HZ=300.
Change-Id: Ia457e8ef58444acf7eb65c4a27ceca1d6bb6ab6f
Signed-off-by: Liang Chen <cl@rock-chips.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Link: https://lkml.kernel.org/r/20200306070133.18335-2-cl@rock-chips.com
Git-commit: 26c7295be0c5e6da3fa45970e9748be983175b1b
Git-repo: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
Signed-off-by: Lingutla Chandrasekhar <clingutla@codeaurora.org>
This reverts commit 3c3dd68c48e83aca55dc72a33260a71608510160.
In QC platforms, backend is not teardown during pause state.
Backend prepare is not needed when resume from pause state.
Change-Id: Ia9f8ab36f5061f9909d9026271bac4a2f6d3e79b
Signed-off-by: Laxminath Kasam <lkasam@codeaurora.org>
Do not release CVP persist buffer until all outstanding frames
are handled by firmware.
Change-Id: I4cbff7898f57db82bd3ab2ee8a9a15873e458f2d
Signed-off-by: George Shen <sqiao@codeaurora.org>
idle_tick is a periodic timer which sends the interrupt to cpu
for every jiffy. If next event is read before stopping the cpu
idle_tick it gives the next tick expiry time as the next event.
Current logic reading the cpu next event time before stopping
the idle tick gives shallower sleep length recorded. This prevents
cluster low power mode to get selected since next wakeup is seen
either very near to current time or as already expired.
Fix this to read next event after idle_tick has been stopped.
Change-Id: I8016c7911dc48450668a8a6bc7a6d01d4423da32
Signed-off-by: Raghavendra Kakarla <rkakarla@codeaurora.org>
Add voltage voting support for GCC CPUSS AHB clock upto 100MHz.
Change-Id: I87c06750355c51c3211dc42cb93a87c933870f6b
Signed-off-by: Jagadeesh Kona <jkona@codeaurora.org>
EPSS hardware generates the GT interrupt when LMH throttle occurs.
Read back the state register to get the throttled frequency and update
the scheduler accordingly. When the throttle is removed the state
register frequency would move back to the SW voted frequency.
Change-Id: I75fbbab63e88b539bc3314e7c0085b9f6249699c
Signed-off-by: Taniya Das <tdas@codeaurora.org>
New version of interface for mapping and unmapping buffer
to remote process on DSP with additional mapping control flags.
All mappings of user allocated memory are tagged with file descriptor.
Change-Id: I2fb20df3f39476f92dbe0a6ec10e0b6065f90749
Acked-by: Viswanatham Paduchuri <vpaduchu@qti.qualcomm.com>
Signed-off-by: Himateja Reddy <hmreddy@codeaurora.org>
kgsl_pool_alloc_pages should return -ENOMEM if we cannot allocate
the required pages.
Change-Id: I481f81ab7ef03828290caebd8dd12cb396f98cd3
Signed-off-by: Raghu Ananya Arabolu <rarabolu@codeaurora.org>
make changes to support enlarged qsee logs support.
Change-Id: I15908f7d2bee2dce0b35db751340a71f340ffeea
Signed-off-by: Zhen Kong <zkong@codeaurora.org>
The I2C bus becomes unusable for transactions after it enters
suspend. Ensure that such transactions are not attempted after
the qcom_pm8008-regulator devices have entered suspend.
Change-Id: Ib13bc2a4a12a79bfe9c8de51ba2dc1ea23fb39c5
Signed-off-by: David Collins <collinsd@codeaurora.org>
Add support to synchronize PCIe L1ss sleep state between APPS
and DRV subsystem. This will remove any ambiguity of PCIe link
state for APPS when trying to enable/disable DRV.
Change-Id: Idb27a631a6eca3aa32a372e2e30235ef0ca10bf4
Signed-off-by: Tony Truong <truong@codeaurora.org>
If HVC_HAVEN_CONSOLE is set, then kernel may be using the RM console for
printk. Thus, when userspace accesses the "self" console, we should not
be re-opening or closing this console along with userspace. In open
case, it is incorrect to open console when it is already open and may
result in an error from RM. In close case, this could cause printk
messages to be lost.
Change-Id: Ie73cc0d401c9feb45f952a3dcaaf7ec8cbfccbc9
Signed-off-by: Elliot Berman <eberman@codeaurora.org>
