Avoid reading uninitalized variables.
Initialize it to default value before reading.
Change-Id: Ic34dfec16357cbc4992912e883f21f014cf8a66c
Signed-off-by: Dilip Kota <dkota@codeaurora.org>
Signed-off-by: Shrey Vijay <shreyv@codeaurora.org>
Service locator framework may call callback with LOCATOR_UP even if
service is not located. Avoid registering with service-notifier in
that case.
Change-Id: Ib73570d1703803dbe0d769d9d43f9d7d2b257770
Signed-off-by: Sagar Dharia <sdharia@codeaurora.org>
During power up operation check whether power up
failure is due to SSR or noise over slimbus.
If power up failure is due to SSR prevent retry
operation.
This change also do allocate DMA memory for slimbus
bulk trasactions. Memory should be DMA for slimbus
transactions as using non-DMA memory will end up
in creating bounce buffers and affect the performance.
Change-Id: Iaca520d92c4ee2d73f8e751699779e6cfc9ddb87
Signed-off-by: Dilip Kota <dkota@codeaurora.org>
Slimbus master component driver typically resides in a domain on remote
processor (ADSP). If that domain restarts, slimbus master driver
restarts reinitializing the slimbus HW.
Handle such process domain restart, and use subsystem restart if the
process domain restart is not available.
Change-Id: I253c33d07aad513a6621571e870e1d3f66a1b6a1
Signed-off-by: Sagar Dharia <sdharia@codeaurora.org>
Signed-off-by: Girish Mahadevan <girishm@codeaurora.org>
In port de-allocate operation free the resources
only if port is connected as memory allocates while
connecting the port.
Change-Id: I743a27025e83f0f4bb8a47c20b2bb2d4cff80caa
Signed-off-by: Dilip Kota <dkota@codeaurora.org>
Capability exchange may timeout during boot due to noise on slimbus
lines. Introduce retry mechanism to recover from this state. If the
HW noise happens just after capability exchange is successful, make
sure slimbus satellite HW powered on again by checking HW status
during message transfer.
Change-Id: I6ebf0ec78201551152bf6ac0d4372f36734a4708
Signed-off-by: Sagar Dharia <sdharia@codeaurora.org>
Subsystem will restart within short timeframe.
Synchronise subsytem up/down callback notifications
to avoid functionality failures.
Use mutex locks to achieve synchronization.
Change-Id: I5881c7d468507bb8402a2e9f8178b9c31e57e8a5
Signed-off-by: Dilip Kota <dkota@codeaurora.org>
Make sure to use pipe handle from the TX
endpoint rather than using a stored handle.
The stored handle may become invalid if SSR
happens between storing of that handle
and locking the mutex.
Change-Id: I53e8f860c2d50061b6c4e6ce2f944c1e2ed8dff2
Signed-off-by: Dilip Kota <dkota@codeaurora.org>
Use wait_for_event_interruptible instead of setting current task to
TASK_INTERRUPTIBLE to avoid associated warning.
Change-Id: I576959d88ff8278f890ab71963fcec5f547c264d
Signed-off-by: Sagar Dharia <sdharia@codeaurora.org>
Support retention by checking interrupt status rather than logical
address register. During retention, interrupt status is zero'ed but
logical address may be retained to avoid report-present generation.
Change-Id: I9e7f24c5f4eb722643bf3fac2d5c898ad107dd24
Signed-off-by: Sagar Dharia <sdharia@codeaurora.org>
Add API in slimbus driver for clients to get the
matching ID table which helps in accessing driver
data and name fields.
CRs-fixed: 975738
Change-Id: I09c9f1de74e348b032d215cbb0fb9ba6c7aecf18
Signed-off-by: Phani Kumar Uppalapati <phaniu@codeaurora.org>
Currently SSR framework seems to make callbacks in the order
they were registered, or based on pririties.
Due to this, APR's callback (done in its probe) may be executed
before slimbus driver's callback (only registered after children
are probed). This still works, but may take longer since codec
may try sending some disconnection commands over slimbus if APR
is notified about SSR before codec was notified by slimbus.
This failure may result in some delay in SSR handling which is
more than the KPI numbers expected for this recovery mechanism.
Priority is added to slimbus SSR notifier registration to make
sure its callback is called before APR's.
CRs-Fixed: 996871
Change-Id: Ib7f002d52310dc13b60a18972a6899ebae21ba24
Signed-off-by: Sagar Dharia <sdharia@codeaurora.org>
Slimbus master is present in different subsystem
on different tragets.
Register with specific subsystem as mentioned in
the device tree.
Change-Id: I1bee7fdd7578deedca8e4e43af9055b41b96d652
Signed-off-by: Dilip Kota <dkota@codeaurora.org>
Framer is responsible for clocking slimbus, and when it reboots, state
of all associated data channels is reset to be consistent with HW
state.
Change-Id: Ie74ace2b20ad6c255440ff6ce593f7d5503bf388
Signed-off-by: Sagar Dharia <sdharia@codeaurora.org>
If a bulk message times out due to interrupt starvation, a late
coming interrupt may access structures that are not valid.
Improve timeout error handling by resetting variables, and changing
relevant logging.
Change-Id: Ie0ee2fd9c2ce1628c69e1975fe713a2a26d9e716
Signed-off-by: Sagar Dharia <sdharia@codeaurora.org>
Every bus instance can have 2 NGD devices. Fix the instance number
determination based on NGD device-id.
Change-Id: I515f4f09d58a7abc0c0322ca714b88fb847a0909
Signed-off-by: Sagar Dharia <sdharia@codeaurora.org>
If low-power mode is not entered due to timeout, it is possible that
HW has actually been turned off by ADSP but the response was not
received by driver on apps on time.
Change device state to sleep to make sure HW is woken up for
subsequent runtime-resume operation.
Change-Id: I4e454f7e90e1244985c86286b95202eecfc50a7e
Signed-off-by: Sagar Dharia <sdharia@codeaurora.org>
During SSR, all channels are removed and ports are disconnected since
slimbus HW is not functional. Return success to clients for such
calls so that client can mark channel/port state constistent with HW
Change-Id: I9b18f8e7ffe233090c9612d0811a7d36645d0329
Signed-off-by: Sagar Dharia <sdharia@codeaurora.org>
QMI timeout can happen if other processor doesn't respond within the
time expected by slimbus driver on apps. It has been noticed that QMI
call to power-on slimbus rarely takes more than 500msec. Timeout
is increased to allow that possibility. Additionally, QMI channel is
drained out of left-over messages and runtime-PM state check is
modified to allow graceful handling of timeout if at all any timeout
still happens.
Change-Id: I392a4fa761735135a261fb543815823cee5eba41
Signed-off-by: Sagar Dharia <sdharia@codeaurora.org>
Remove the use of PM_RUNTIME feature flag which is obsoleted on newer
kernel versions. Instead use PM feature flag to condiitionally compile
RPM callbacks.
Change-Id: I775c89f79b698bf3f20fdb655216027e58e7059d
Signed-off-by: Girish Mahadevan <girishm@codeaurora.org>
