Assorted cleanups and fixes:
* A break in a switch statement should line up with the rest of
the statements in the case.
* queue_work() can't return one indicating the work is already
queued if the work is dynamically allocated.
* Remove unnecessary header includes.
* Clean up multi-line comments.
Change-Id: Ie5c544ee34e9f4eaeb9af5ff6d4c39975238cde3
Signed-off-by: Stephen Boyd <sboyd@codeaurora.org>
Add a struct subsys_device for the internal stuff we don't want
to expose. Propagate that struct throughout the code. Add a new
restart API, subsystem_restart_dev() that takes the subsys_device
pointer to restart the processor. This saves having to iterate
through the list of devices to find the subsystem (saving a
spinlock for callers in atomic context). Since we still have some
callers of the old string based API, leave subsystem_restart() in
place.
While we're here, remove code that isn't doing anything. We always
know to use the current subsystem if we only have one so the
single_restart_list is useless. Plus, if someone is passing a
NULL descriptor or not initializing the fields of the descriptor
when they register with SSR, they will soon realize that their
code doesn't work. Remove these checks as they'll impede future
patches that don't want to supply all possible ops.
Change-Id: Ia53ffd530982390d1871bd62bfe31e3a0a512fbe
Signed-off-by: Stephen Boyd <sboyd@codeaurora.org>
This initcall was put in place to shut down the processors during
late init to save power. Now that we use the driver model we can
have the affected driver(s) do this in their probe instead and
save calling shutdown for all processor from the core PIL code.
Change-Id: If575589060f1c68177879b6c953f264892e91776
Signed-off-by: Stephen Boyd <sboyd@codeaurora.org>
If an SSR driver is detached there is no way to free the ramdump
device that is allocated in create_ramdump_device(). Add a
destroy_ramdump_device() function to clean up memory. Admittedly
this doesn't try to solve the race condition in destroying the
device while do_ramdump() is in progress, but it is a good first
start.
Change-Id: Ib46ac8589a9ae7405441328eb9bfa0353362513a
Signed-off-by: Stephen Boyd <sboyd@codeaurora.org>
When session CLOSE command is sent right before session RUN command
is acknowledged, callback function can mistakenly think that
the next received acknowledgement is for CLOSE command instead of
RUN command. This triggers driver to send memory unmap command to
the Q6 while it is still processing the CLOSE command. Eventually,
this leads to an invalid memory access and causes Q6 crash.
Change-Id: Ib5d560fbcb7e8ced79cc1075a9f6bea3b55a86b6
CRs-Fixed: 377281
Signed-off-by: Jay Wang <jaywang@codeaurora.org>
Add a function which logs the parameters in each request being
sent to the RPM immediately before each request is sent. Also
add a module parameter to control this logging.
rpm-smd debug_mask supports the following bits:
BIT(0) = Show pretty formatting in RPM request log messages
BIT(1) = Show raw formatting in RPM request log messages
BIT(2) = Show SMD message ID in RPM request log messages
Example log messages for each possible debug_mask value:
0 = (no message)
1 = RPM req: s=act ldoa 9; uv=1800000
2 = RPM req: s=act, rsc_type=0x616F646C, rsc_id=9;
[key=0x00007675, value=0x40 77 1B 00]
3 = RPM req: s=act, rsc_type=0x616F646C (ldoa), rsc_id=9;
[key=0x00007675 (uv), value=0x40 77 1B 00 (1800000)]
4 = (no message)
5 = RPM req: msg_id=8, s=act ldoa 9; uv=1800000
6 = RPM req: msg_id=8, s=act, rsc_type=0x616F646C, rsc_id=9;
[key=0x00007675, value=0x40 77 1B 00]
7 = RPM req: msg_id=8, s=act, rsc_type=0x616F646C (ldoa),
rsc_id=9;
[key=0x00007675 (uv), value=0x40 77 1B 00 (1800000)]
Change-Id: If7e0f6128a065396e703a6c9f8708303491ba5fe
Signed-off-by: David Collins <collinsd@codeaurora.org>
Camera IRQ router changes for 8974
1) Add device tree match struct for camirq_router driver.
This is required for 8974 since platform device node is
added in device tree.
2) Probe camera IRQ router before other modules by
changing probe order. This is required since irq router
needs to register irq information with cam server before
other modules update its irq information
3) Assign proper cam_irq_hw for ISPIF during register
subdev. Otherwise, cam_irq_hw will be used uninitialised.
Change-Id: Ida29abb91264a178d45d000fe78c9555b55c0fe3
Signed-off-by: Sreesudhan Ramakrish Ramkumar <srramku@codeaurora.org>
Replace the clk_enable() with clk_prepare_enable() and replace
clk_disable() with clk_disable_unprepare() functions.
Change-Id: I63479090eccbeac46f091bf95faeb857139d23a4
Signed-off-by: Joel Nider <jnider@codeaurora.org>
On Fusion 2.2 platform, GSBI9 UARTDM is used for primary IPC
mechanism. GSBI9 can be used on different platform for other
configuration mode (i.e. SPI/I2C). Hence add GSBI9 UARTDM device
only for Fusion 2.2 platform.
Change-Id: I5b4dcd259f7966fcf244b47deb8d603a5728af60
Signed-off-by: Mayank Rana <mrana@codeaurora.org>
TSENS driver needs to be invoked early in the kernel
initialization for clients to read the temperature to
perform kernel thermal mitigation.
Change-Id: I3f6c5a716aa36033676523ce941402fc3032dcfa
Signed-off-by: Siddartha Mohanadoss <smohanad@codeaurora.org>
MSM8974 TSENS block supports 11 temperature
sensors that are used for thermal mitigation.
The TSENS driver is part of the thermal sysfs
framework that allows clients to set thresholds
and receive notification on temperature thresholds
being crossed through thermal sysfs.
The driver provides ability to thermal clients
to set upper/lower thresholds for each sensor
and receive individual notification for each
sensor once the threshold is crossed.
All sensors are enabled by default during
driver initialization.
The driver is initialized early during kernel
initialization to provide capability to the
thermal kernel driver to read temperature
from the TSENS sensors. To support this feature
a kernel API is provide to read the TSENS
temperature.
Change-Id: If17daae81cfb522a9cea786b5db0c920adb5ed2d
Signed-off-by: Siddartha Mohanadoss <smohanad@codeaurora.org>
Define the channel and CRCI used by the default TSIF configuration.
This configuration must match the TrustZone image, which is set up
to use TSIF2 on CRCI1 by default.
Change-Id: I86a2e100b29d888c3d6167e4b34b9c4271ada165
Signed-off-by: Joel Nider <jnider@codeaurora.org>
This change fixes a crash when running VCAP in de-interlace mode.
The problem was noticed in the introduction of kernel 3.4. VCAP
was improperly using the videobuf2 framework APIs. Since VCAP is
already reliant on queue structure provided by videobuf2 framework,
we created our own basic version of the qbuf and dqbuf calls that
have some similarities to the videobuf2 framework. This addition
was necessary since we are calling qbuf from a kernel work thread.
Since moving to kernel 3.4 there has been a change in videobuf2
qbuf that dereferences the current process.s mm_struct, from a
kernel work thread this struct is NULL, which was causing a crash.
When we are de-interlacing a captured video we are cycling the
buffers internally in the driver, which is why we needed to create
our own q/dq buffer functionality.
Change-Id: Id51b775ae7f0a58bbac8a581e07fa02417802a30
Signed-off-by: Terence Hampson <thampson@codeaurora.org>
Export the epm header files for clients from userspace
to send commands to the driver.
Change-Id: I9b413f461da58cbf92aefeca9285ac6798468ea2
Signed-off-by: Siddartha Mohanadoss <smohanad@codeaurora.org>
Split Target based USB-Gadget driver (tcm_usb_gadget.c) into composite
driver (tcm_usb_gadget.c) and Function driver (f_tcm.c). This is
required for other composite drivers like Android to include Target
based USB function (UASP) along with other functions.
Composite drivers can include f_tcm.c like any other function and
register a callback function. This callback function is called
when Target Fabric is enabled via configfs. This callback can be
used to enable enumeration with host.
Change-Id: I36beea07bb3641a224694c157849ecec347d5943
Signed-off-by: Pavankumar Kondeti <pkondeti@codeaurora.org>
BAM driver provides a callback mechanism for global BAM device
error or when a AHB access error occurs. This change adds and
registers the callback along with the necessary error handling
and cleanup.
Change-Id: Ib5028ea4b189010da87f6669f0292a146fc05838
Signed-off-by: Krishna Konda <kkonda@codeaurora.org>
Add a NONE alias to the RETENTION voltage corner value in enum
rpm_regulator_voltage_corner in order to satisfy the clock driver
naming scheme.
Change-Id: I29c4039c445a88bc6ac6039d7f37eee16d287695
Signed-off-by: David Collins <collinsd@codeaurora.org>
As part of LPAII, the power profiling showed that having 100msec timeout
is letting the AXI switch to low frequency as soon as the APROC collapses.
Having this timeout as 1 sec is pushing the AXI to stay high for longer
time hence causing more power drain. Hence with this change this time
out is updated to 100msec.
Change-Id: I4aaf8e1ba5c177e4a77291f13a8ecd0fbc6e6130
Signed-off-by: Krishna Vanka <kvanka@codeaurora.org>
Update the clock lookup device mappings so that the TSIF driver
can control the TSIF clocks.
Change-Id: I596d579fa4db9e72f0406f5e134c7c092402eda4
Signed-off-by: Joel Nider <jnider@codeaurora.org>
Amend the input and output buffer calculation to
reflect what the Venus core expects.
Change-Id: Ib3c600d0e838525e3a82a1b2c05803f86075ba15
Signed-off-by: Deva Ramasubramanian <dramasub@codeaurora.org>
GUSB3PIPECTL[22] bit is for enabling/disabling receiver detection in
either U3 or Rx.Detect states. It should be '0' by default to enable
receiver detection. But it is overridden with '1' on cores revision <
2.30a. Fix this in software.
CRs-Fixed: 377848
Change-Id: I9500f687f386330ed3c686f46365c08b8c8b0d28
Signed-off-by: Pavankumar Kondeti <pkondeti@codeaurora.org>
GUCTL[31:22] indicate the time period (in ns) of the reference clock
that is used to generate Isochronous Timestamp Packets (ITP). This
value should be 8ns (125 MHz) by default. But it is overridden with
0 on cores revision < 2.30a. Fix this in software.
CRs-Fixed: 377846
Change-Id: I827540eb9d2f1ffc65b6f30200b6c0e575bc52d2
Signed-off-by: Pavankumar Kondeti <pkondeti@codeaurora.org>
Two changes for clocks:
1> Vote for dfab_clk which is the peripheral NOC on 8974;
2> Do not vote for mem_clk since it does not exist on 8974.
Change-Id: Ia9d5547073219cea4250d3038aa5aa01a47ff4e0
Signed-off-by: Yan He <yanhe@codeaurora.org>
Clock names changes to the tables removed any clock with
connection name mmfpb_a_clk. Add an entry so that we can use
clk_get_sys() to get the clock in the clock driver.
Change-Id: Ie6cdaee9e68833eaf8db86785143ed856f179d27
Signed-off-by: Stephen Boyd <sboyd@codeaurora.org>
Flushing the core is now done with the V4L2_DEC_QCOM_CMD_FLUSH ioctl
rather than V4L2_DEC_CMD_STOP. The latter is confusing and the
arguments specified by the V4L2 spec do not allow for flushing the input
and and output buffers seperately.
Change-Id: I53e18e06c4e09ee8496bdcc337b4798f22911708
Signed-off-by: Deva Ramasubramanian <dramasub@codeaurora.org>
USB Video Class(UVC) Driver is required to work with
any USB Video devices (e.g. USB Webcamera). Hence enable
UVC driver for the same.
Change-Id: Ibb4fb51b111a69be0ca17817172385281d5e58ca
Signed-off-by: Mayank Rana <mrana@codeaurora.org>
The current code registers multiple devices for i2c clients. This
results in code repetition. instead its better to register single
i2c client and register codec based on device id.
Change-Id: Ifb4adbd47b7a2bf7d32be637fd0641545cd84286
Signed-off-by: Asish Bhattacharya <asishb@codeaurora.org>
Update the placeholder/conservative voltage data to allow vdd_dig to
drop to SVS levels when the L2 is running at its slowest. The values
will be updated again when real data is available.
Change-Id: I483865261354e1f397238cfb165d0389c1fd2102
Signed-off-by: Matt Wagantall <mattw@codeaurora.org>
Update the DTS bindings documentation to include the
vdd-supply regulator property.
Change-Id: I006dbbcf55cfdf07604c64380ce227f30f6cc75a
Signed-off-by: Stepan Moskovchenko <stepanm@codeaurora.org>
Query SMT and SID mapping information at probe-time instead
of attach-time to allow this information to be
error-checked at an earlier time.
Change-Id: Ib2bbdc8374f9c86c3e6013d298fe8b279b53d83b
Signed-off-by: Stepan Moskovchenko <stepanm@codeaurora.org>
Add device tree support for qcedev and qcrypto device drivers.
Add documentations for associated bindings.
Change-Id: Ib76986af5d9e30263a559329acbca401ec3304d2
Signed-off-by: Mona Hossain <mhossain@codeaurora.org>
"mask" is a new variable added in the I2C register config
structure. It is introduced in the middle of the structure.
This creates issues with MT9M114 camera YUV sensor
using the structure.
Change-Id: I29f6e175f1e1a7a5275ac3dd7c2e9deb7cfa3a49
Signed-off-by: Katta Santhisindhu <kattas@codeaurora.org>
poll should return all the available events
in one call. This change implements this
behavior.
Change-Id: I498b621cc5d8fca820376109ca4a5f1bd4c492b2
Signed-off-by: Vinay Kalia <vkalia@codeaurora.org>
Since the IOMMU ID registers are only accessible by the
secure environment, specify the SMT sizes in device tree
so that the IOMMU driver knows how many SMRs to initialize.
Change-Id: I614a51069c0304f71b0c7d061d97aca0289c17ea
Signed-off-by: Stepan Moskovchenko <stepanm@codeaurora.org>
Modify GMEM Restore procedure as per A3XX GPU spec to support
multiple GPU revisions.
Change-Id: I037cc32c5e259a2ff83cd579b586e9c1f585236f
Signed-off-by: Rajeev Kulkarni <krajeev@codeaurora.org>
Add support for the QDSS clocks on the 8974. These clocks
are controlled by the RPM; represent them as standard RPM
clocks in the 8974 clock driver, allowing the QDSS driver
to use clock APIs to control them.
Change-Id: I4737112f1e778872a2da4694956c9849c37a6afe
Signed-off-by: Vikram Mulukutla <markivx@codeaurora.org>
RPM clocks may have key-value pairs with differing keys.
For example, rate-settable RPM clocks have 'KHz' as a key,
and branch clocks have 'Enab' as a key. To accomodate for
this, store the relevant key in the rpm_clk structure
and pass the key in via the clock defining macros.
Change-Id: I0dc051410e161d1c3f8199d9b17e4983a55185f3
Signed-off-by: Vikram Mulukutla <markivx@codeaurora.org>
Add an rpm clock that controls the MMSSNOC AHB clock source
and vote for the max SVS rate of 40 MHz on this clock in
post_init.
Change-Id: Iccb5f4bc336d8c96b3797628978ff7b4d2240a8b
Signed-off-by: Vikram Mulukutla <markivx@codeaurora.org>
- Fix test-iosched crash when running multiple tests
- Free the BIOs memory when a request is not completed
Change-Id: I1baa916c04ae73c809dee7e67ec63f4546dc71aa
Signed-off-by: Maya Erez <merez@codeaurora.org>
Expose the packing control test.
Test the packing control feature under these scenarios:
- Packing expected: entering write packing state,
staying in write packing state.
- Packing not expected: not entering write packing state
when not supposed to.
- Mixed states: test the ability to shift from packing
to no-packing and back, and the opposite.
Change-Id: I5a93df8b5b7c48d355c3cc881699a3f9592914e6
Signed-off-by: Lee Susman <lsusman@codeaurora.org>
Add method for scaling backlight based off of a user provided ratio and
minimum level for the scaling to be applied.
Change-Id: I74c2329c842dd9cbbc9eddfff38cd56fd0e939a4
Signed-off-by: Carl Vanderlip <carlv@codeaurora.org>
Add a new LUN(lun0) which will be exposed as a CD-ROM having the
attributes ro(read-only) & removable as enabled.
This CD-ROM lun is used for saving read-only files which can be used
for system requirements.
Change-Id: Idff5aea8bce233e8b77a4b689323a2f1eb2523e1
Signed-off-by: Rajkumar Raghupathy <raghup@codeaurora.org>
Move almost all of the SOC based GPU identification code out
of the KGSL driver and into the board files. This cleans up
the GPU side and reduces our usage of the cpu_is and machine_is
macros.
Change-Id: Ic0dedbad34eb034316b7baabf05aa7f8e8aa9ced
Signed-off-by: Jordan Crouse <jcrouse@codeaurora.org>
The pdata structure and associated structs have no business being
in msm_kgsl.h. Move them to include/mach/kgsl.h where they can
thrive in a non-userspace facing environment.
Change-Id: Ic0dedbadfe0d3d7d44f50c8a9a7f4cc59057d44b
Signed-off-by: Jordan Crouse <jcrouse@codeaurora.org>
Provide sysfs access for user space to query and control krait
processor-internal settings via the secure monitor interface.
Change-Id: I318efc28881414d549d2bb65ba0039606b036484
Signed-off-by: Trevor Bourget <tbourget@codeaurora.org>
Return value of msm_timer_get_sclk_time api is int64_t. Storing the
return value in integer would lead to inconsistencies in pm_stats.
CRs-fixed: 375038
Change-Id: I2c73b387ef01db3a7f71cd1b21b135bea2a475fa
Signed-off-by: Anji Jonnala <anjir@codeaurora.org>
The call to enable autosuspend was recently removed from the
qcserial driver. Re-add it here in the RmNet driver as we know
this device will support it.
Change-Id: Iee1d1dbece01eaab99b03c64ebea0fd5ed87eafa
Signed-off-by: Jack Pham <jackp@codeaurora.org>
DMA addresses are configured after Layer Mixer is set to writeback mode.
This leads to IOMMU page faults due to incorrect addresses read from DMA.
Hence DMA is programmed before Mixer setting.
Change-Id: I4c5e41a56f05e3fabb46b4cadde03168589eeb17
Signed-off-by: Ravishangar Kalyanam <rkalya@codeaurora.org>
