This driver adds support for the Tegra CEC IP. It is based on the
NVIDIA drivers/misc/tegra-cec driver in their 3.10 kernel.

This has been converted to the CEC framework and cleaned up.

Tested with my Jetson TK1 board. It has also been tested with the
Tegra X1 in an embedded product.

Note of warning for the Tegra X2: this SoC supports two HDMI outputs,
but only one CEC adapter and the CEC bus is shared between the
two outputs. This is a design mistake and the CEC adapter can
control only one HDMI output. Never hook up both HDMI outputs
to the CEC bus in a hardware design: this is illegal as per the
CEC specification.

The CEC bus can be shared between multiple inputs and zero or one
outputs, but not between multiple outputs.
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Acked-by: NThierry Reding <treding@nvidia.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Using the ARRAY_SIZE macro improves the readability of the code. Also,
it is useless to use a variable to store this constant calculated at
compile time.

Found with Coccinelle with the following semantic patch:
@r depends on (org || report)@
type T;
T[] E;
position p;
@@
(
 (sizeof(E)@p /sizeof(*E))
|
 (sizeof(E)@p /sizeof(E[...]))
|
 (sizeof(E)@p /sizeof(T))
)
Signed-off-by: NJérémy Lefaure <jeremy.lefaure@lse.epita.fr>
Signed-off-by: NSakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

The functions __bo_alloc, __bo_search_and_remove_from_free_rbtree,
__bo_search_by_addr, __bo_search_by_addr_in_range, __bo_break_up and
__bo_merge  are local to the source and do not need to be in the global
scope, so make them static.

Cleans up sparse warnings:

warning: symbol '__bo_alloc' was not declared. Should it be static?
warning: symbol '__bo_search_and_remove_from_free_rbtree' was not declared. Should it be static?
warning: symbol '__bo_search_by_addr' was not declared. Should it be static?
warning: symbol '__bo_search_by_addr_in_range' was not declared. Should it be static?
warning: symbol '__bo_break_up' was not declared. Should it be static?
warning: symbol '__bo_merge' was not declared. Should it be static?
Signed-off-by: NMuhammad Falak R Wani <falakreyaz@gmail.com>
Signed-off-by: NSakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

There was no specific item regarding what should be done to sensor, lens
and flash drivers. Add one, to replace the vague item denoting support
only to particular sensor, lens and flash devices.
Signed-off-by: NSakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

By adding the "atomisp-" prefix to module names (and "ATOMISP_" to Kconfig
options), the staging drivers for e.g. sensors are labelled as being
specific to atomisp, which they effectively are.
Signed-off-by: NSakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

The imx274 is a Sony CMOS image sensor that has 1/2.5 image size.
It supports up to 3840x2160 (4K) 60fps, 1080p 120fps. The interface
is 4-lane MIPI CSI-2 running at 1.44Gbps each.

This driver has been tested on Xilinx ZCU102 platform with a Leopard
LI-IMX274MIPI-FMC camera board.

Support for the following features:
-Resolutions: 3840x2160, 1920x1080, 1280x720
-Frame rate: 3840x2160 : 5 – 60fps
            1920x1080 : 5 – 120fps
            1280x720 : 5 – 120fps
-Exposure time: 16 – (frame interval) micro-seconds
-Gain: 1x - 180x
-VFLIP: enable/disabledrivers/media/i2c/imx274.c
-Test pattern: 12 test patterns
Signed-off-by: NLeon Luo <leonl@leopardimaging.com>
Tested-by: NSören Brinkmann <soren.brinkmann@xilinx.com>
Signed-off-by: NSakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

The TC358743 is an HDMI to MIPI CSI2-2 bridge.
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

When using VDIC with CSI1, make sure to select the correct CSI in
IPU_CONF.

Fixes: f0d9c892 ("[media] media: imx: Add IC subdev drivers")
Suggested-by: NMarek Vasut <marex@denx.de>
Signed-off-by: NTim Harvey <tharvey@gateworks.com>
Acked-by: NSteve Longerbeam <steve_longerbeam@mentor.com>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

We can remove the check for if "state->cec_adap" is NULL.  The
cec_allocate_adapter() function never returns NULL and also we verified
that "state->cec_adap" is an error pointer.
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Make vb2_ops const as they are only stored in the const field of a
vb2_queue structure. Make the declarations const too.

Done using Coccinelle.
Signed-off-by: NBhumika Goyal <bhumirks@gmail.com>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Make vb2_ops const as they are only stored in the const field of a
vb2_queue structure. Make the declarations const too.

Done using Coccinelle.
Signed-off-by: NBhumika Goyal <bhumirks@gmail.com>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Make this const as it is only used in a copy operation in the file
referencing it. Make the declaration const too.

Done using Coccinelle.
Signed-off-by: NBhumika Goyal <bhumirks@gmail.com>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Make this const as it is only passed to the const argument of the
function videobuf_queue_sg_init in the file referencing it.
Also, make the declaration in the header const.

Structure found using Coccinelle and changes done by hand.
Signed-off-by: NBhumika Goyal <bhumirks@gmail.com>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

kasprintf() can fail here and we must check its return value.
Signed-off-by: NArvind Yadav <arvind.yadav.cs@gmail.com>
Reviewed-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

It should be "LP-11".
Signed-off-by: NJacob Chen <jacob-chen@iotwrt.com>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

- Delete seven error messages for a failed memory allocation
- Adjust seven checks for null pointers
- Use common error handling code in tm6000_usb_probe()
- Adjust jump targets so that the function "kfree" will be always called
  with a non-null pointer.
- Delete an initialisation for the local variable "dev"
  which became unnecessary with this refactoring.
Signed-off-by: NMarkus Elfring <elfring@users.sourceforge.net>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

The kAPI cec-pin.h header also defined data structures that did
not belong here but were private to the CEC core code.

Split that part off into a cec-pin-priv.h header.
Signed-off-by: NHans Verkuil <hansverk@cisco.com>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

This addresses the wrong behavior of decoder stop command by
rewriting it. These new implementation enqueue an empty buffer
on the decoder input buffer queue to signal end-of-stream. The
client should stop queuing buffers on the V4L2 Output queue
and continue queuing/dequeuing buffers on Capture queue. This
process will continue until the client receives a buffer with
V4L2_BUF_FLAG_LAST flag raised, which means that this is last
decoded buffer with data.
Signed-off-by: NStanimir Varbanov <stanimir.varbanov@linaro.org>
Tested-by: NNicolas Dufresne <nicolas.dufresne@collabora.com>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Cc: <stable@vger.kernel.org>      # for v4.13 and up
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

In case of error, the function platform_device_register_data() returns
ERR_PTR() and never returns NULL. The NULL test in the return value check
should be replaced with IS_ERR().
Signed-off-by: NWei Yongjun <weiyongjun1@huawei.com>
Acked-by: NHelen Koike <helen.koike@collabora.com>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

This fixes wrongly filled bytesused field of v4l2_plane structure
by include data_offset in the plane, Also fill data_offset and
bytesused for capture type of buffers only.
Signed-off-by: NStanimir Varbanov <stanimir.varbanov@linaro.org>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Cc: <stable@vger.kernel.org>      # for v4.13 and up
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

To improve the readability of code, split the format array into two,
one for the format description, other for the register configuration.
Meanwhile, add the flag member to indicate the format can be achieved
from the sensor or be produced by the controller, and rename members
related to the register configuration.

Also add more formats support: GREY, ARGB444, ARGB555 and ARGB32.
Signed-off-by: NWenyou Yang <wenyou.yang@microchip.com>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Remove the memeber *config from the isc_subdev_entity struct,
the member is useless afterward.
Signed-off-by: NWenyou Yang <wenyou.yang@microchip.com>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

To meet the relationship, enable the HCLOCK and ispck during the
device probe, "isc_pck frequency is less than or equal to isc_ispck,
and isc_ispck is greater than or equal to HCLOCK."
Meanwhile, call the pm_runtime_enable() in the right place.
Signed-off-by: NWenyou Yang <wenyou.yang@microchip.com>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

A software write operation to the ISC_CLKEN or ISC_CLKDIS register
requires double clock domain synchronization and is not permitted
when the ISC_SR.SIP is asserted. So add the .prepare and .unprepare
ops to make sure the ISC_CLKSR.SIP is unasserted before the write
operation to the ISC_CLKEN or ISC_CLKDIS register.
Signed-off-by: NWenyou Yang <wenyou.yang@microchip.com>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Add the spin lock for the clock enable and disable operations.
Signed-off-by: NWenyou Yang <wenyou.yang@microchip.com>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

This change will fix an issue with dma_free size found with
DMA API debug enabled.
Signed-off-by: NStanimir Varbanov <stanimir.varbanov@linaro.org>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Cc: <stable@vger.kernel.org>      # for v4.13 and up
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Instead of relying on a mix of runtime PM and the s_power() callback, drop
the s_power() callback altogether and use runtime PM solely.

As device access is required during device power-on and power-off
sequences, runtime PM alone cannot tell whether the device is available.
Thus the "active" field is introduced in struct smiapp_sensor to tell
whether it is safe to write to the device.

Consequently there is no need to power on the device whenever a file
handle is open. This functionality is removed as well. The user may still
control the device power management through sysfs. Autosuspend remains
enabled, with 1 s delay.
Signed-off-by: NSakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Use unlocked __v4l2_ctrl_handler_setup() in order to make the
control setup atomic.
Signed-off-by: NSakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

The script “checkpatch.pl” pointed information out like the following.

Comparison to NULL could be written...

Thus fix the affected source code places.
Signed-off-by: NMarkus Elfring <elfring@users.sourceforge.net>
Signed-off-by: NSylwester Nawrocki <s.nawrocki@samsung.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Replace the specification of a data structure by a pointer dereference
as the parameter for the operator "sizeof" to make the corresponding size
determination a bit safer according to the Linux coding style convention.

This issue was detected by using the Coccinelle software.
Signed-off-by: NMarkus Elfring <elfring@users.sourceforge.net>
Signed-off-by: NSylwester Nawrocki <s.nawrocki@samsung.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Omit an extra message for a memory allocation failure in s5p_mfc_probe()
function.  This issue was detected by using the Coccinelle software.
Signed-off-by: NMarkus Elfring <elfring@users.sourceforge.net>
Signed-off-by: NSylwester Nawrocki <s.nawrocki@samsung.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

The hardware rotation limits of gsc depends on SOC (Exynos
5250/5420/5433). Distinguish them and add them to the driver data.

[s.nawrocki@samsung.com: corrected num_entities in 5420 variant data]
Signed-off-by: NHoegeun Kwon <hoegeun.kwon@samsung.com>
Signed-off-by: NSylwester Nawrocki <s.nawrocki@samsung.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

drivers/media/platform/rockchip/rga/rga-hw.c:383:6: warning: no previous prototype for 'rga_cmd_set' [-Wmissing-prototypes]
 void rga_cmd_set(struct rga_ctx *ctx)
      ^~~~~~~~~~~
drivers/media/platform/rockchip/rga/rga.c:359:17: warning: no previous prototype for 'rga_fmt_find' [-Wmissing-prototypes]
 struct rga_fmt *rga_fmt_find(struct v4l2_format *f)
                 ^~~~~~~~~~~~
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Rockchip RGA is a separate 2D raster graphic acceleration unit. It
accelerates 2D graphics operations, such as point/line drawing, image
scaling, rotation, BitBLT, alpha blending and image blur/sharpness

The driver supports various operations from the rendering pipeline.
 - copy
 - fast solid color fill
 - rotation
 - flip
 - alpha blending

The code in rga-hw.c is used to configure regs according to operations
The code in rga-buf.c is used to create private mmu table for RGA.
Signed-off-by: NJacob Chen <jacob-chen@iotwrt.com>
Signed-off-by: NHans Verkuil <hansverk@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

We have seen NULL-pointer dereference crashes in tty->disc_data when the
N_TTY fallback driver failed to open during hangup.  The immediate cause
of this open to fail has been addressed in the preceding patch to
vmalloc(), but this code could be more robust.

As Alan pointed out in commit 8a8dabf2 ("tty: handle the case where
we cannot restore a line discipline"), the N_TTY driver, historically
the safe fallback that could never fail, can indeed fail, but the
surrounding code is not prepared to handle this.  To avoid crashes he
added a new N_NULL driver to take N_TTY's place as the last resort.

Hook that fallback up to the hangup path.  Update tty_ldisc_reinit() to
reflect the reality that n_tty_open can indeed fail.

Link: http://lkml.kernel.org/r/20171004185959.GC2136@cmpxchg.orgSigned-off-by: NJohannes Weiner <hannes@cmpxchg.org>
Cc: Alan Cox <alan@llwyncelyn.cymru>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Michal Hocko <mhocko@suse.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

When I execute numactl -H (which reads /sys/devices/system/node/nodeX/cpumap
and displays cpumask_of_node for each node), I get different result
on X86 and arm64.  For each numa node, the former only displayed online
CPUs, and the latter displayed all possible CPUs.  Unfortunately, both
Linux documentation and numactl manual have not described it clear.

I sent a mail to ask for help, and Michal Hocko replied that he
preferred to print online cpus because it doesn't really make much sense
to bind anything on offline nodes.

Will said:
 "I suspect the vast majority (if not all) code that reads this file was
  developed for x86, so having the same behaviour for arm64 sounds like
  something we should do ASAP before people try to special case with
  things like #ifdef __aarch64__. I'd rather have this in 4.14 if
  possible."

Link: http://lkml.kernel.org/r/1506678805-15392-2-git-send-email-thunder.leizhen@huawei.comSigned-off-by: NZhen Lei <thunder.leizhen@huawei.com>
Acked-by: NMichal Hocko <mhocko@suse.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Tianhong Ding <dingtianhong@huawei.com>
Cc: Hanjun Guo <guohanjun@huawei.com>
Cc: Libin <huawei.libin@huawei.com>
Cc: Kefeng Wang <wangkefeng.wang@huawei.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The function only sends the flush command to the IOMMU(s),
but does not wait for its completion when it returns. Fix
that.

Fixes: 601367d7 ('x86/amd-iommu: Remove iommu_flush_domain function')
Cc: stable@vger.kernel.org # >= 2.6.33
Signed-off-by: NJoerg Roedel <jroedel@suse.de>

We need to call reservation_object_reserve_shared() in both cases, but
this wasn't happening in the _NO_IMPLICIT submit case.

Fixes: f0a42bb5 ("drm/msm: submit support for in-fences")
Reported-by: NJordan Crouse <jcrouse@codeaurora.org>
Signed-off-by: NRob Clark <robdclark@gmail.com>

If we fail to attach iommu, gpu->aspace could be IS_ERR()..
Signed-off-by: NRob Clark <robdclark@gmail.com>

Using CONFIG_OF_DYNAMIC=y uncovered an imbalance in the usecount of the
node being passed to of_fwnode_graph_get_port_parent(). Preserve the
usecount by using of_get_parent() instead of of_get_next_parent() which
don't decrement the usecount of the node passed to it.

Fixes: 3b27d00e ("device property: Move fwnode graph ops to firmware specific locations")
Signed-off-by: NNiklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Acked-by: NSakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: NRob Herring <robh@kernel.org>