The padding VMA is never inserted into the VMA tree; therefore we
don't need to have the mmap lock in exclusive mode to modify it.
Test: v2/android-gki/ack_platform_integration_main_cf_arm64_boot_test
on kernel_virt_debug_aarch64
Bug: 346741763
Bug: 352463814
Change-Id: I4ca3ed22dab45b6bb895cb41c5c6792344188b61
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
(cherry picked from commit a8b3ebe7f90afe0ad0d9a8b0b90f266da06a7007)
Increase pimem master priority to P3 from P2 based on post silicon
recommendation from tlm team.
Compared to blair, pimem traffic is more on pitti and increasing the
priority is helping to fix pimem timeouts.
Change-Id: Ib951e7e0d7f2c29b2c1cb519c1b08396d974961a
Signed-off-by: Raviteja Laggyshetty <quic_rlaggysh@quicinc.com>
With Early Ethernet enabled Link is not up after S2R.
reset the PHY at S2R to fix the Link issue.
Change-Id: I8549d4af7c4dcdc84abcf8deb136028925db5dae
Signed-off-by: Uppalamarthi Sowmya <quic_usowmya@quicinc.com>
* commit 'cb6c9fcea75d596f3ea030f1179093d4cedced60':
ANDROID: GKI: Update symbol list for mtk
Change-Id: I758cb38a9b9d7dad2dd131cbde1ed992eacb9c8b
Signed-off-by: keystone-kernel-automerger <keystone-kernel-automerger@google.com>
During Host mode RT suspend, we remove PM QoS requests at the
beginning of msm_suspend(). However, if msm_prepare_suspend()
fails and we bail out after resuming xHCI. And we won't add
the perf votes again from msm_resume(), since msm_suspend()
never succeeded and the in_lpm flag was not set.
Consequently, during the next dwc3_msm_suspend(), we remove the
perf votes again results in warnings in kernel logs.
To address this issue:
- Move the perf_vote_enable(false) operation after completing
dwc3_msm_prepare_suspend().
- moving pm_qos_add() or pm_qos_remove() to msm_probe and
msm_remove respectively.
Change-Id: Id064b0fe55e03905ba6c60c533519d5e0268b1c1
Signed-off-by: Rajashekar kuruva <quic_kuruva@quicinc.com>
commit: 9bbb1b4dc059a introduced new conditional check in
msm_probe function, results in cyclomatic complexity
Move IRQ registration to new function dwc3_msm_register_interrupt
Fix cyclomatic complexity for function dwc3_msm_probe().
Change-Id: Iad10117156dc2b0b86583f1cdafa31ecdbc55ef6
Signed-off-by: Rajashekar kuruva <quic_kuruva@quicinc.com>
Currently i2c geni driver doesn't have null pointer check condition
in event call back function. If any invalid event is coming from GSI,
i2c geni driver accessing null pointer which is causing crash.
To solve this added null pointer checks in event call back functions.
Change-Id: Ie14a40eee846c0ea29bec512d6320e9548c509b5
Signed-off-by: Anil Veshala Veshala <quic_aveshala@quicinc.com>
Due to the permissions are not provided to the apps to read
the secure SE registers. Instead, consider reading the RO
registers that are available.
Change-Id: I350c23eea1f319a50e7fdd61151041bebdbfce7c
Signed-off-by: Aniket Randive <quic_arandive@quicinc.com>
The hab team recommends that the maximum message size that the
hab client can send is 64k. We increase the size of
OUT_LARGE_BUF_SIZE to 64K so that the memory pool can cover
all messages of the recommended size.
In non-stress scenarios, this helps to avoid jitter in the
performance of sending messages due to memory requests.
This change does not result in a large amount of additional
reserved memory, because OUT_LARGE_BUF_NUM is 10, and each
pchan group will only reserves (64k- 50k) * 10 = 140k more memory.
Change-Id: Ia26f408230e1121d6f927dd084f551cc64656e96
Signed-off-by: Zelong Ren <quic_zeloren@quicinc.com>
The TCSR register alone cannot be relied for confirming the D0
state, adding another smp2p bit to confirm the D0 state.
Change-Id: Ibfc3cf0b1c47450026b820e5c99ddc2c615224e1
Signed-off-by: Gokul krishna Krishnakumar <quic_gokukris@quicinc.com>
Signed-off-by: Kamati Srinivas <quic_kamasrin@quicinc.com>
The debugfs creating call should return error number if failed, need
assign null to debugfs ptr for releasing stage.
Change-Id: I45a979014cb083cf9c322af71b15d977700e3bfc
Signed-off-by: Kasin Li <quic_donglil@quicinc.com>
Add some missing USB opcodes to ensure they are aligned
correctly with the USB opcodes defined on CHGFW side. This is
needed to ensure that a new USB opcode which will be added does
not overlap with any of the older existing opcodes.
Change-Id: Ibf05acec9346786193745e1f9db7c1b40e8651b7
Signed-off-by: Jishnu Prakash <quic_jprakash@quicinc.com>
Currently there is support upto three distinct sdhci host
controllers, whose host index is parsed from device tree.
But while allocting pointers to host str, only two instances
are declared.
This will cause error if three host controllers are present
on a SoC.
So, declare three instances of pointers to host str.
Change-Id: Id1d673205c75ee40fadcc249720692a5f38bfeca
Signed-off-by: Ram Kumar Dwivedi <quic_rdwivedi@quicinc.com>
Add and enable qti epm hardware driver and thermal zone
debug feature driver in config and bazel for anorak.
Change-Id: Ib262762b46c37ba01c8ed177291a411011af6001
Signed-off-by: Nitesh Kumar <quic_nitekuma@quicinc.com>
CONFIG_QCOM_MEMORY_DUMP_V2 is required by cnss2 to
support latest wlan chips for wlan fw dump.
Change-Id: I47e42054614979192e5713e1b6c36720c5a32266
Signed-off-by: Zou Shunxiang <quic_shunxian@quicinc.com>
Make hab open as uninterruptible to maintain same atomic
context to avoid file system sync issues.
This is snapshot from msm-5.4 branch with commit ids
commit a099f7774d07 ("crypto-qti-virt: Proper handling of hab close call"),
commit 8aa1befa10e8 ("crypto-qti-virt: Make fbe hab request as
uninterruptible"),
commit 947ff2fe45e2 ("crypto-qt-virt: Remove kthread mechanism while
sending request").
test: VTS kernel encryption tests passed.
Change-Id: Iad8eed6cd8cc8569aea4b1a42ba4b9923b6ac9ab
Signed-off-by: Santosh Dronamraju <quic_sdronamr@quicinc.com>
ETF may fail to re-enable after reading, and driver->reading will
not be set to false, this will cause failure to enable/disable to ETF.
This change set driver->reading to false even if re-enabling fail.
Change-Id: Ie06b49379c200d0bdd3c3f4acb1e7e612c03ad1a
Signed-off-by: Yuanfang Zhang <quic_yuanfang@quicinc.com>
Signed-off-by: Mao Jinlong <quic_jinlmao@quicinc.com>
Enable GKI based build system for autoghgvm.
Change-Id: Ieecd9110d5e78eaa5d54edbef6ba99218cf43bac
Signed-off-by: Singa Reddy Dasari <quic_singredd@quicinc.com>
Enable support for Ethernet Link UP KPI marker for
as part of system resume.
Currenlty Ethernet Link up KPI marker is printed
once & print flag is set to true to avoid further
Link UP KPI print. This flag need to reset as part of
suspend to print the Ethernet Link UP KPI marker at resume.
Change-Id: I849a76520c4c619255ba650fff546a32a05adc2e
Signed-off-by: Suraj Jaiswal <quic_jsuraj@quicinc.com>
