If dmabuf deferred_free_thread is preempted then system heap alloc
path is taking longer time due to mutex lock contention. To avoid
preemption due to low priority updating with spin lock.
Change-Id: I63cd224543dc42c0710bc9202bebed9ced2bf341
Signed-off-by: Faiyaz Mohammed <quic_faiyazm@quicinc.com>
Update variable type to ensure all the valid values can be
accomodated in variable.
Change-Id: Ieaf0301d6ad524bfb87ff019aec41063ebc4c4c4
Signed-off-by: Ashay Jaiswal <quic_ashayj@quicinc.com>
This change sets SPEC_FENCE_FLAG_ARRAY_BIND flag in speculative
fence to indicate bind ioctl is triggered.
Change-Id: I718dfe8c6278a585e58bb59811de05b72106d6d3
Signed-off-by: Jayaprakash Madisetty <quic_jmadiset@quicinc.com>
Add memory related modules to first stage on neo.
Change-Id: Ie1d0234223741f0b43dc79d8dc6c30bd7568bd03
Signed-off-by: Jaskaran Singh <quic_jasksing@quicinc.com>
When the ADSP is down, for particular slimbus message code and
message type transactions we are waiting for 1sec for the ADSP
to be up. Blocking function for 1sec is causing issues in the
audio layer.
Added a dtsi based flag whether to wait for adsp to be up or return
error immediately.
Change-Id: I09a3e135806d4044cdb95e2cc35092d6f5c2f68b
Signed-off-by: Anvesh Salveru <quic_asalveru@quicinc.com>
Free memory only during error while opening a port.
In success case, memory will be freed during port
closure.
CRs-Fixed: 3113018
Change-Id: Ice2172c7865e5df5898b3ac927f842db1516fd96
Signed-off-by: Satish Kumar Kodishala <quic_skodisha@quicinc.com>
We no longer use "qcom,dwc-usb3-msm-tx-fifo-size" property to know
the TXFIFO size instead depend on the value present in DWC3_GHWPARAMS7
register.
Change-Id: I39a95a51b1e0954b91946b602d6754512c97c522
Signed-off-by: Pavankumar Kondeti <quic_pkondeti@quicinc.com>
Compression before encryption to AP minidump can save a lot of
disk size, so disable the AP minidump encryption by default, and
leave the choice to customers.
Change-Id: Ib7a3c71cf6f62466a219a5d86f065bb0f760fe45
Signed-off-by: Huang Yiwei <quic_hyiwei@quicinc.com>
This change does restructure of GPIO PDC mapping.
Change-Id: Ib02cd25b55505dd8f7f3f2d87b482e69df05902c
Signed-off-by: Tushar Nimkar <quic_tnimkar@quicinc.com>
On some endpoints there is no PCIe driver in bootloader to update
MHI registers due to that there is some garbage values present at
miscoffset register so in the sw if there is no valid value
updating it to zero.
Change-Id: Id230da97484f7042166e2cd68fc18730857374c3
Signed-off-by: Krishna chaitanya chundru <quic_krichai@quicinc.com>
Currently, HAPTICS_EN bit is toggled whenever a FIFO pattern is played
followed by setting Vmax. However, this would issue a SWR_VMAX handshake
and then follow a SPMI_VMAX handshake in HW if haptics effect is played
over SWR bus in the background. This back to back Vmax handshake would
potentially cause the HW not responding and no vibration being played.
To address this, set the Vmax before toggling HAPTICS_EN bit to avoid
back to back Vmax handshakes when SWR and FIFO are played concurrently.
Change-Id: I0b8924ab7f369db446458fe2510a0da01519a978
Signed-off-by: Fenglin Wu <quic_fenglinw@quicinc.com>
As per the hardware recommendation, MOD_STATUS_XT need to be programmed
before programming MOD_STATUS_SEL to get the correct status data in all
cases. Update it.
Change-Id: Ie66b12cdc702f32ebfad78a72396ad3ac4d14adf
Signed-off-by: Fenglin Wu <quic_fenglinw@quicinc.com>
Currently, when waiting for hBoost ready times out, error code -EBUSY is
returned. In addition to that, reset the FIFO play logic so that the
following FIFO requests can still be served.
Meanwhile, add 100us before updating HAPTICS_EN for hBoost to have
enough time to handle its power transition as per the hardware
recommendation.
Change-Id: Iebfc98b589bda747e4c2b6fdf6ff78d2f46ece5d
Signed-off-by: Fenglin Wu <quic_fenglinw@quicinc.com>
Add dispcc-neo to modules.list.neo to enable
it to load during first stage init.
Change-Id: I57a15a36de9afc45241bea2e19aa7468ac846872
Signed-off-by: Kalpak Kawadkar <quic_kkawadka@quicinc.com>
Add debug-regulator.ko to modules.list.neo to enable
it to load during first stage init.
Change-Id: If67a8a592d9ced62b4063a96cb3e0c4920b6de1c
Signed-off-by: Ashish Chavan <quic_ashichav@quicinc.com>
This change halts cancel command operation if IOS lines not in
good state for i2c due to Power or slave side issue. Perform cancel
once IO lines comes to good state. Till that i2c transfer will not
be servied.
This is a workaround fix if the abort is failing after cancel and
never recovers.
Change-Id: Icca5904d0d3df7432b948343ef9396cb25d93cd0
Signed-off-by: Mukesh Kumar Savaliya <msavaliy@codeaurora.org>
Signed-off-by: Ajay Prathi <quic_aprathi@quicinc.com>
Make sure to free the temporary buffer that holds the clients message
and header in the sendmsg path for gunyah vsock.
Change-Id: I398d5ec00f869b5f63e5f6daf8a609dcbbb13537
Signed-off-by: Chris Lew <quic_clew@quicinc.com>
