This change will move SE_DMA_TX_ATTR define from serial driver to common
header file so that other protocol drivers also can use the address
definition from common header file.
Change-Id: I60b7e0051bddd9772a30385238c2d1fdbc991396
Signed-off-by: Chandana Kishori Chiluveru <quic_cchiluve@quicinc.com>
In i3c transfer address phase during ACK we are seeing 1.8us
delay due to mixed bus counter values, this we can reduce max
up to 1us by updating counter values as per HPG. Updated counter
values as per HPG.
Change-Id: I03bf24091d3a21452fb291bfa73c42415bd88407
Signed-off-by: Anil Veshala Veshala <quic_aveshala@quicinc.com>
Add kaslr config check to avoid compile error when
CONFIG_RANDOMIZE_BASE is not defined.
Change-Id: Ic5616c56a11e3c6214e33fe796841bc86642bde5
Signed-off-by: Huang Yiwei <quic_hyiwei@quicinc.com>
Extend the default SHM timeout and also add the default timeout
to subsys required timeout to avoid conflict with SSR.
Change-Id: Ic0309904c984bdba779a2e6b37ba0f431c5dda73
Signed-off-by: Huang Yiwei <quic_hyiwei@quicinc.com>
When a SHM check is already in process, there's no need to check it
again, just refuse the new request to avoid problems. Use check_state
instead of check_count/report_count to make it more clear.
Change-Id: Ia650a669179cfc294e08db99920cd22b7e4ee885
Signed-off-by: Jingyi Wang <quic_jingyw@quicinc.com>
Signed-off-by: Huang Yiwei <quic_hyiwei@quicinc.com>
tmp_hma_info->connected is not sufficient to distinguish if hma
info is not initialied or hma is going to shutdown, usually we
handle reg req after receive QCOM_SSR_AFTER_POWERUP and it works
fine, but we may meet corner case such as QCOM_SSR_AFTER_POWERUP->
QCOM_SSR_BEFORE_SHUTDOWN->shm_reg_req_handler, which lead to
duplicate subsystem or spurious health check during shutdown.
Also, update hma_info to newest even we found duplicate subsystem.
Change-Id: I61e570618b490bd13bc022e6cfc3a6c6fa10a0da
Signed-off-by: Jingyi Wang <quic_jingyw@quicinc.com>
Use cpuhp_setup_state_nocalls_cpuslocked instead of
cpuhp_setup_state_nocalls as cpus_read_lock is
already taken.
Using cpuhp_setup_state_nocalls_cpuslocked will avoid
potential recursive locking.
Change-Id: I4639510aa417d6cec73903e6d08ba7b527006314
Signed-off-by: badris <quic_badris@quicinc.com>
Commit 9204f9fa1e ("bus: mhi: host: Disable preemption
while processing data events") replaced spin_lock_bh/unlock_bh
with spin_lock_spin_lock_irqsave/restore to disable preemption
to allow processing of all data ring elements with
parse_xfer_event() API. As part of processing data ring
elements uses read_lock_bh()/read_unlock_bh().
a. spin_lock_irqsave() -> disable interrupt on CPU, store into flags
b. read_lock_bh() -> calls local_irq_disable()
c. process data event ring
d. read_unlock_bh() -> calls local_irq_enable()
e. spin_lock_irqrestore() -> restores interrupt using flags before
enabling interrupt
When CONFIG_TRAC_IRQFLAGS Kconfig is enabled,
raw_check_bogus_irq_restore() is checked to detect situation
where interrupts are already enable while trying to restore it.
Due to above nested usage of locking, there is WARN_ONCE() seen.
Change-Id: I3579c2bbb92f6ea9149a7c9550dedce36466eb52
Signed-off-by: Vivek Pernamitta <quic_vpernami@quicinc.com>
During boot-up if the clock driver put across a initial high vote then
there is no way to remove these votes which can cause the corresponding
votes to be left always high till the CPU suspends causing power
regression. Thus introduce a device property which can remove the
initial votes from the clock driver and leave it up to the client to
keep a vote.
Change-Id: I068c87d06df908ee0ed0e58f69d46933b301ea6f
Signed-off-by: Kalpak Kawadkar <quic_kkawadka@quicinc.com>
-m will block compile by "kernel modules are built but not copied".
Set to -y.
Change-Id: I8fa380f3e60520511a4097d7e7671a1b7615cb30
Signed-off-by: ruoyyang <quic_ruoyyang@quicinc.com>
Configured and enabled Serdes driver for SGMMI/USXGMII interface.
Change-Id: I110694184bba7d5fb1266a8488fe8b39d4a0454a
Signed-off-by: Sarosh Hasan <quic_sarohasa@quicinc.com>
Sometimes tty_insert_flip_string fails when trying to send
large data due to memory allocation failure in tty layer.
Allocation in tty layer can fail due to higher order page
request, hence retry sending data to tty layer in chunks
of 512 bytes which uses zero order pages.
Change-Id: I99dcde0421e07eda61fac19fdc0f45f986c1434c
Signed-off-by: Visweswara Tanuku <quic_vtanuku@quicinc.com>
To enhance tuning capability, ensure that the idle_enough and
util_thres_pct control nodes can be tuned per cluster, enabling
different levels of demand and util comparison per cluster.
These nodes are used only in the packing_cpu fastpath decision
making, for rt and cfs.
Change-Id: I29131d9dbce0e29b74cef69fe153bd009eb4fd58
Signed-off-by: Stephen Dickey <quic_dickey@quicinc.com>
In Android GKI, CONFIG_FAIR_GROUP_SCHED is enabled [1] to help
prioritize important work. Given that CPU shares of root cgroup
can't be changed, leaving the tasks inside root cgroup will give
them higher share compared to the other tasks inside important
cgroups. This is mitigated by moving all tasks inside root cgroup to
a different cgroup after Android is booted. However, there are many
kernel tasks stuck in the root cgroup after the boot.
It is possible to relax kernel threads and kworkers migrations under
certain scenarios. However the patch [2] posted at upstream is not
accepted. Hence add a restricted vendor hook to notify modules when a
kernel thread is requested for cgroup migration. The modules can relax
the restrictions forced by the kernel and allow the cgroup migration.
[1] f08f049de1
[2] https://lore.kernel.org/lkml/1617714261-18111-1-git-send-email-pkondeti@codeaurora.org
Bug: 184594949
Bug: 301261126
Change-Id: I445a170ba797c8bece3b4b59b7a42cdd85438f1f
Signed-off-by: Pavankumar Kondeti <quic_pkondeti@quicinc.com>
[quic_dickey@quicinc.com: port to android-mainline kernel]
Signed-off-by: Stephen Dickey <quic_dickey@quicinc.com>
(cherry picked from commit 7551a1a2a1)
Add minidump support on pineapple_tuivm platform.
Change-Id: I946d640616b0a9abe7be6bebf95fed6b29b6b7bd
Signed-off-by: Fei Yang <quic_feiyan@quicinc.com>
* commit '5c1ed513eded33d20e713b1d346e5766251ac98b':
ANDROID: GKI: Update symbol list for xiaomi
ANDROID: vendor_hooks: export cgroup_threadgroup_rwsem
ANDROID: GKI: Update symbol list for xiaomi
ANDROID: vendor_hooks:vendor hook for percpu-rwsem
Change-Id: If18accea139ba3bbac0db069fad65553ed39b2f5
Signed-off-by: keystone-kernel-automerger <keystone-kernel-automerger@google.com>
Enable base memory modules for niobe by enabling configs and adding
relevant modules for first stage loading.
Change-Id: Ia71f1263b814976d326b782305895358dad2602e
Signed-off-by: Pratyush Brahma <quic_pbrahma@quicinc.com>
Change base address for DMA and MTL registers for direwolf.
A new function ethqos_configure_mac_v3() has been created for
HSR sequence programming for Gen4. ethqos_rgmii_macro_init()
function has been modularised by creating new functions to
configure rgmii macro for different speeds.
Change-Id: I1bc925dd4a9af6013d6838ebaa1ba25073f9b64b
Signed-off-by: Sarosh Hasan <quic_sarohasa@quicinc.com>
There was a debug feature on minidump to collect content around
pointers present in panicked cpu registers from X0-X30. PC, LR,
SP. and it was having check present before copy the contents about
address need to be kernel mapped and not hyp assigned. We should
be able to do most of the checking present as well but tracking
the buffers which are assigned to other VMID through
qcom_scm_assign_mem() and since it is upstream API and adding magic
value to each PFN and restore it if there is failure is kind of
overhead for such a debug feature, ideally it should have done on
client side but doing at the client is very difficult task to do
as there are many users of this API.
For now, we are cleaning this up, unless we know the user of this
information and it is worth doable.
Change-Id: I638877063a4c8c6dabb7d16bcc4eeb63f5108fba
Signed-off-by: Mukesh Ojha <quic_mojha@quicinc.com>
Assign the power management operations defined in the pm8xxx_rtc_pm_ops
structure to the platform driver pm field.
Change-Id: Ifd1eeaf0ae106037a73ff4ef3f7155dfa9894caa
Signed-off-by: Darshankumar Jagdishchandra Thakkar <quic_djagdish@quicinc.com>
Add spss to the list of rprocs, who usually send message up
and down notification message to AOP and also move
out common code from qcom_q6v5_pas to qcom_common
so that it can be used by SPSS rproc driver.
Change-Id: Idbe0774c57d9959a15873e2c71a604dc98435313
Signed-off-by: Sindhura Ganda <quic_sganda@quicinc.com>
This change adds the ability to partially initialize
the MMC card by using card Sleep/Awake sequence (CMD5).
Card will be sent to Sleep state during runtime/system
suspend and will be woken up during runtime/system resume.
By using this sequence the card doesn't need full initialization
which gives time reduction in system/runtime resume path.
Change-Id: Ic4fcc1563cee2503513e6a0f9d3f00eb7b58ad51
Signed-off-by: Maya Erez<quic_merez@quicinc.com>
Signed-off-by: Veerabhadrarao Badiganti <vbadigan@codeaurora.org>
Signed-off-by: Ram Prakash Gupta <quic_rampraka@quicinc.com>
Signed-off-by: Sarthak Garg <quic_sartgarg@quicinc.com>
Add changes to create cooling device name as per alias defined in
devicetree.
Change-Id: I8d9d119372e2daaeadeab6fb97a0f5c053f0200d
Signed-off-by: Priyansh Jain <quic_priyjain@quicinc.com>
Add specific directory for minidump module in blair.bzl.
Change-Id: Idb36eb6e21e7e4ea9f0b305e205c554abb8bcdec
Signed-off-by: Song Xue <quic_songxue@quicinc.com>
