Add a parameter to change the sensor device CSI-2 link frequency to
one of the following values: 450Mhz (default), 447Mhz, or 453Mhz.

Signed-off-by: Naushir Patuck <naush@raspberrypi.com>

Add support for three different usable link frequencies (default 450Mhz,
447Mhz, and 453MHz) for the IMX708 camera sensor. The choice of
frequency is handled thorugh the "link-frequency" overlay parameter.

Signed-off-by: Naushir Patuck <naush@raspberrypi.com>

Setting both the Drop and Add bits on the input context prevents the
corruption of split transactions seen with the BCM2711 XHCI controller,
which is a dwc3 variant.

This is a downstream feature that allows usbhid to restrict polling
intervals on mice and keyboards, and was only tested on a VL805 which
didn't complain about the fact the endpoint got added twice.

Signed-off-by: Jonathan Bell <jonathan@raspberrypi.com>

We seem to be getting increased OOM failures with MGLRU
enabled on 32-bit kernels.
Let's disable it by default
(it can still be manually enabled at runtime).

See: https://github.com/raspberrypi/linux/issues/5395


Signed-off-by: Dom Cobley <popcornmix@gmail.com>

Upstream change [1] has deleted the NET_CLS_TCINDEX Kconfig symbol.
Regenerate our defconfigs without it.

Signed-off-by: Phil Elwell <phil@raspberrypi.com>

[1] 3abebc50 ("net/sched: Retire tcindex classifier")

Since [1], the fbdev deferred IO framework is careful to cancel
pending updates on close to prevent dirty pages being accessed after
they may have been reused. However, this is not necessary in the case
that the pagelist is empty, and drivers that don't make use of the
pagelist may have wanted updates cancelled for no good reason.

Avoid penalising fbdev drivers that don't make use of the pagelist by
making the cancelling of deferred IO on close conditional on there
being a non-empty pagelist.

See: https://github.com/raspberrypi/linux/issues/5398



Signed-off-by: Phil Elwell <phil@raspberrypi.com>

[1] 3efc61d9 ("fbdev: Fix invalid page access after closing deferred I/O devices")

This reverts commit b70f2a88

.

Signed-off-by: Phil Elwell <phil@raspberrypi.com>

Now upstream have accepted patches adding a mono version of
IMX290, and IMX327, update the compatible strings in the overlays
to match.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>

Should be upstream commit 3c3f66b8125a

Both sensors are quite similar. Their specs only differ regarding LVDS
and parallel output but are identical regarding MIPI-CSI-2 interface.
But they use a different init setting of hard-coded values, taken from
the datasheet.

Signed-off-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>

Should be upstream commit 5fcaecd6b5aa

imx290_start_streaming logs what failed, but not the error
code from that function. Add it into the log message.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Reviewed-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>

Should be upstream commit 42765b80f8c6

The sensor supports H & V flips, so add the relevant hooks for
V4L2_CID_HFLIP and V4L2_CID_VFLIP to configure them.

Note that the Bayer order is maintained as the readout area
shifts by 1 pixel in the appropriate direction (note the
comment about the top margin being 8 pixels whilst the bottom
margin is 9). The V4L2_SEL_TGT_CROP region is therefore
adjusted appropriately.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>

Should be upstream commit b8b86dfe1aee

The sensor supports either a 37.125 or 74.25MHz external, but
the driver only supported 37.125MHz.

Add the relevant register configuration for either clock
frequency option.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Reviewed-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>

Should be upstream commit 454ce4c6d77b

IMX290_CTRL_07 was written from both imx290_global_init_settings
and imx290_1080p_settings and imx290_720p_settings.

Remove it from imx290_global_init_settings as the setting varies
based on the mode.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>

Should be upstream commit b3aa351ea88a

The default VMAX for 60fps in 720p mode is 750 according to the
datasheet, however the driver always left it at 1125 thereby stopping
60fps being achieved.

Make VMAX (and therefore V4L2_CID_VBLANK) mode dependent so that 720p60
can be achieved.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>

Should be upstream commit 97792a11021b

The driver exposed V4L2_CID_VBLANK as a read only control to allow
for exposure calculations and determination of the frame rate.

Convert to a read/write control so that the frame rate can be
controlled.
V4L2_CID_VBLANK also sets the limits for the exposure control,
therefore exposure ranges have to be updated when vblank changes
(either via s_ctrl, or via changing mode).

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Acked-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>

Should be upstream commit 8cab2bd86307

The driver exposed V4L2_CID_HBLANK as a read only control to allow
for exposure calculations and determination of the frame rate.

Convert to a read/write control so that the frame rate can be
controlled.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>

Should be upstream commit 34819ba0b450

Commit "98e0500e

 media: i2c: imx290: Add configurable link frequency
and pixel rate" added support for the increased link frequencies
on 2 data lanes, but didn't update the CSI timing registers in
accordance with the datasheet.

Use the specified settings.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>

Should be upstream commit 7d4f22ba99cc

Commit "97589ad6 media: i2c: imx290: Add support for 2 data lanes"
added support for running in two lane mode (instead of 4), but
without changing the link frequency that resulted in a max of 30fps.

Commit "98e0500e

 media: i2c: imx290: Add configurable link frequency
and pixel rate" then doubled the link frequency when in 2 lane mode,
but didn't undo the correction for running at only 30fps, just extending
horizontal blanking instead.

Remove the 30fps limit on 2 lane by correcting the register config
in accordance with the datasheet for 60fps operation over 2 lanes.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>

Should be upstream commit 4381df0c756f

The datasheet lists the link frequency changes between
1080p and 720p modes. This is correct that the link frequency
changes as measured on an oscilloscope.

Link frequency is not necessarily the same as pixel rate.

The datasheet gives standard configurations for 1080p and 720p
modes at a number of frame rates.
Looking at the 1080p mode it gives:
HMAX = 0x898 = 2200
VMAX = 0x465 = 1125
2200 * 1125 * 60fps = 148.5MPix/s

Looking at the 720p mode it gives:
HMAX = 0xce4 = 3300
VMAX = 0x2ee = 750
3300 * 750 * 60fps = 148.5Mpix/s

This driver currently scales the pixel rate proportionally to the
link frequency, however the above shows that this is not the
correct thing to do, and currently all frame rate and exposure
calculations give incorrect results.

Correctly report the pixel rate as being 148.5MPix/s under any
mode.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>

Should be upstream commit 5f9dee39ab55

Any V4L2 subdevice that implements controls and declares
V4L2_SUBDEV_FL_HAS_DEVNODE should also declare V4L2_SUBDEV_FL_HAS_EVENTS
and implement subscribe_event and unsubscribe_event hooks.

This driver didn't and would therefore fail v4l2-compliance
testing.

Add the relevant hooks.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>

Should be upstream commit 7843b2eb081b

The colorspace fields were left untouched in imx290_set_fmt
which lead to a v4l2-compliance failure.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>

Should be upstream commit 76d1bd4f0f45

Fix up a couple of coding style issues regarding missing blank
lines after declarations, double blank lines, and incorrect
indentation.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>

Should be upstream commit 9bf52c7136d1

The IMX290 module is available as either mono or colour (Bayer).

Update the driver so that it can advertise the correct mono
formats instead of the colour ones.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Tested-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>

Should be upstream commit 971458482809

No need to call fwnode_property_read_u32(dev_fwnode()), when
we have already existing helper. So use it.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>

Should be upstream commit 5d129c4539c3

Since we have a proper endianness converters for LE 24-bit data use them.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>

Should be upstream commit 783ecfe5556a

The imx290 driver can be used for both imx290 and imx327 as they have a
similar register set and configuration. imx327 lacks 8 lanes LVDS and
120 FPS support.

Signed-off-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>

Should be upstream commit f0c720ff2d21

The IMX290 module is available as either monochrome or colour and
the variant is not detectable at runtime.

Add a new compatible string for the monochrome version, based on the
full device name IMX290LLR. For consistency, add a new compatible string
for the colour version based on the IMX290LQR full device name, and
deprecate the current ambiguous compatible string.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Acked-by: Rob Herring <robh@kernel.org>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Tested-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>

Upstream commit a96dfea1



Convert the Sony IMX290 DT binding from text to YAML. Add Manivannan as
a maintainer given that he is listed in MAINTAINERS for the file, as
volunteering myself.

The name of the input clock, "xclk", is wrong as the hardware manual
names it INCK. As the device has a single clock, the name could be
omitted, but that would require a corresponding change to the driver and
is thus a candidate for further patches.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>

Upstream commit 05ef7ec4



Check the error status returned by imx290_set_data_lanes() in its
caller and propagate it.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>

Upstream commit 76c00128



There's no need to check for an incorrect number of data lanes in
imx290_set_data_lanes() as the value is validated at probe() time. Drop
the check.

The PHY_LANE_NUM and CSI_LANE_MODE registers are programmed with a value
equal to the number of lanes minus one. Compute it instead of handling
it in the switch/case.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>

Upstream commit 7d399658



There's no need to configure the data lanes in the runtime PM resume
handler. Do so in imx290_start_streaming() instead.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>

Upstream commit 02852c01



Initializing the subdev before runtime PM means that no subdev
initialization can interact with the runtime PM framework. This can be
problematic when modifying controls, as the .s_ctrl() handler commonly
calls pm_runtime_get_if_in_use(). These code paths are not trivial,
making the driver fragile and possibly causing subtle bugs.

To make the subdev initialization more robust, initialize runtime PM
first.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Acked-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>

Upstream commit a8c3e0c1



Use runtime PM autosuspend to avoid powering off the sensor during fast
stop-reconfigure-restart cycles. This also fixes runtime PM handling in
the probe function that didn't suspend the device, effectively leaving
it resumed forever.

While at it, improve documentation of power management in probe() and
remove().

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Acked-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>

Upstream commit 10591fe6



The imx290_pixfmt structure contains information about formats,
currently limited to the bpp value. Extend it with the register settings
for each format, and rename it to imx290_format_info to make its purpose
clearer. Add a function named imx290_format_info() to look up format
info for a media bus code, and use it through the code. This allows
dropping the imx290 bpp field as the value is now looked up dynamically.

The error handling in imx290_setup_format() can also be dropped, as the
format is guaranteed by imx290_set_fmt() to be valid.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Acked-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>

Upstream commit a2514b9a



Use the V4L2 subdev active state API to store the active format. This
simplifies the driver not only by dropping the imx290 current_format
field, but it also allows dropping the imx290 lock, replaced with the
state lock.

The lock check in imx290_ctrl_update() can be dropped as
imx290_set_fmt() can't be called anywmore with which set to ACTIVE
before controls are initialized.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>

Upstream commit e5d363ca



Move the external clock initialization code from probe() to a separate
function to improve readability. No functional change intended.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>

Upstream commit 63127235



Improve error handling in the probe() function with dev_err_probe().

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>

Upstream commit 6b69c522



Make the probe() function more readable by factoring out the DT parsing
code to a separate function. No functional change intended.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>

Upstream commit ee4ce893



The black level programmed in the BLKLEVEL register depends on the
output format. The black level value computation is currently performed
in imx290_set_ctrl(), in addition to having different black level values
in the output-specific register value tables. Move it to a separate
function to simplify the imx290_set_ctrl() code.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Reviewed-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>