Destroy mutex before file free, to avoid use after free of mutex.
Change-Id: I4ff73dc17b15043eacbb299219a379bfd1a8efa6
Acked-by: Himateja Reddy <hmreddy@qti.qualcomm.com>
Signed-off-by: Tharun Kumar Merugu <mtharu@codeaurora.org>
This implementation facilitates a framework to report and recover the
errors in display hardware pipeline.
Change-Id: Ie4059dd8bcdd83f73b5a5165ae5a8363ae4efb94
Signed-off-by: Rahul Sharma <rahsha@codeaurora.org>
When print UTP HCI testbus data, a redzone overwritten error is triggered
because the max select minor value goes across its border. This change
fixes it.
Change-Id: Ib085288f3770aa0ab3683926247fb95f5e50cb08
Signed-off-by: Can Guo <cang@codeaurora.org>
Correct the config control indices for TX DAIs of MI2S interface.
Change-Id: Ie8bf646a0c1fbd30abdb34703ae1f90ed9b484e8
Signed-off-by: Satish Babu Patakokila <sbpata@codeaurora.org>
On automotive targets, we do not want the system to wakeup if the
connected peripheral is issuing remote wakeup when the system is
PM suspended. Do not enable the system wakeup capability of the
PHY IRQs if the target is automotive.
Change-Id: I0bd986fdfab8f978692b605a44301086de3d0a87
Signed-off-by: Ajay Agarwal <ajaya@codeaurora.org>
Andrey Konovalov reported a possible out-of-bounds problem for the
cdc_parse_cdc_header function. He writes:
It looks like cdc_parse_cdc_header() doesn't validate buflen
before accessing buffer[1], buffer[2] and so on. The only check
present is while (buflen > 0).
So fix this issue up by properly validating the buffer length matches
what the descriptor says it is.
Change-Id: I9c65dc98778aa357911e554c8d28deef2bf9a5b3
Reported-by: Andrey Konovalov <andreyknvl@google.com>
Tested-by: Andrey Konovalov <andreyknvl@google.com>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Git-commit: 2e1c42391ff2556387b3cb6308b24f6f65619feb
Git-repo: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
Signed-off-by: Srinivasa Rao Kuppala <srkupp@codeaurora.org>
Change the AVI infoframe read to check avi_info_raw bit in
hdmi lvl raw status register instead of new_avi_info_raw
in hdmi edg raw status register. This is required so as not
to miss AVI infoframes if we disconnect and reconnect the
HDMI cable or close and reopen the application. Also return
error if no AVI infoframe is found and change some error
logs to be more specific.
Change-Id: If30ba820d255149f1b54f422b8e075e634271aaf
Signed-off-by: Suprith Malligere Shankaregowda <supgow@codeaurora.org>
The dependencies were impossible to handle preventing
drivers for CDC devices not which are not network drivers
from using the common parser.
Change-Id: Ied0d1d037bfb8b60bc82236382ef894755755cef
Signed-off-by: Oliver Neukum <ONeukum@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Git-commit: e4c6fb779498243ec001c5547b3504fe6b1993ec
Git-repo: https://android.googlesource.com/kernel/common
Signed-off-by: Srinivasa Rao Kuppala <srkupp@codeaurora.org>
Signed-off-by: Ajay Agarwal <ajaya@codeaurora.org>
If MSM_QDSP6_NOTIFIER is not enabled for a target, audio_notifier
feature will not be available for that target. Whenever a call is
made to audio_notifier when its not supported, returning error from
stubbed function may confuse the error with real failure (even though
the feature is not enabled on the target). Avoid returning error code
from stubbed function, when feature is disabled.
Change-Id: I37eb2e25e45b25948aff55f2a231195a22b16f05
Signed-off-by: Banajit Goswami <bgoswami@codeaurora.org>
This mutex lock will help to synchronise the usage of iommu domain
structure which is getting used by multiple threads to perform
iommu attach and iommu detach in case of any error.
Without this change we are seeing page poisoning - write after free
when running stress tests using iommu-debug test cases.
Change-Id: I1033808a510786577c41201c2a6a330e5465d4d1
Acked-by: Ankur Saxena <c_ankusa@qti.qualcomm.com>
Signed-off-by: Vivek Gautam <vivek.gautam@codeaurora.org>
In 'commit 60de9d752e ("usb: misc: Add snapshot of diag_bridge
driver")', we used wrong keyword in the Kconfig file. Fix that
by using "Qualcomm Technologies".
Change-Id: I6caecf19ed67257db7eda0289257118549d46836
Signed-off-by: Ajay Agarwal <ajaya@codeaurora.org>
q6lsm_pack_params requires param hdr size
to be filled and sent as argument.
Change is to update size with param_hdr_v3 struct size.
Change-Id: Id2380f3ceb887e1fc73b7f5f8e28ac93029082e4
Signed-off-by: Soumya Managoli <smanag@codeaurora.org>
Moved some value checks to right positions to prevent
buffer flow, which may be possible before. Previously
these value checks are in an else statement which may
not be executed.
Change-Id: I02dbecd074183581a6bdae6377097bc004bd3d3c
CRs-fixed: 1102936
Signed-off-by: Sriharsha Allenki <sallenki@codeaurora.org>
Added condition check for suspend only if sensor state is power
down and for resume only if sensor state is power up in
pm functions on sensor driver to control.
Change-Id: Ia31f34dc103428423a86efe8f8012db18b5a75f1
Signed-off-by: Suprith Malligere Shankaregowda <supgow@codeaurora.org>
This commits reverts the commit a6566710adaa ("pinctrl:
qcom: Don't clear status bit on irq_unmask").
Above change causes an unexpected wakeup on the gpios when the
interrupt is enabled. The gpio interrupt status bit is getting set
after the irq is disabled and causing an immediate interrupt after
enablling the irq. So, revert this change to avoid unexpected/spurious
wakeups.
The above change came in as part of the merge commit 3a75d7a947
("Merge 4.4.59 into android-4.4")
Change-Id: Ie58116c8ac5c81bc007d2c4398f5edfcbf9988d6
Signed-off-by: Srinivas Ramana <sramana@codeaurora.org>
PM660L v2.0 has the NCP power-on issue fixed. This allows enabling the
short-circuit IRQ. Enable the same.
Change-Id: I4c0523ef8c46dd9bac190d4ad69190ef777ef14b
Signed-off-by: Kiran Gunda <kgunda@codeaurora.org>
Set default configuration of I2C_8 instance to disabled
state, for MSM8996 vplatform, such that clients can overwrite
and enable respective instance as per need.
Change-Id: I2e873cb27ba028323d6dc1a0aa302f2c32e2b69f
Signed-off-by: Shrey Vijay <shreyv@codeaurora.org>
Added the QSEECOM ION heap node, to allow QSEECOM driver and QSEECOM
API library to allocate ION memory from QSEECOM heap.
Change-Id: I72162f8e45bfb8dd3d54e259947aa9604618a7ed
Signed-off-by: Jomana Artul <jartul@codeaurora.org>
Currently, reallocated mask update buffers are not
updated if the received mask range is more than the
mask update buffer length. Update the reallocated buffer
address before writing the mask to peripherals.
CRs-Fixed: 2266693
Change-Id: I6b506ce68e17b7da61926b0f9543157812a8c555
Signed-off-by: Manoj Prabhu B <bmanoj@codeaurora.org>
When splash handoff is enabled, it should skip the hardware init
function. However, the software context should setup properly to
reflect actual hardware state.
Change-Id: I2a69bf5fadab726459e874597022cf5b54503102
Signed-off-by: Camus Wong <camusw@codeaurora.org>
