media: rcar-isp: Move driver to own directory

Before extending the driver with functions from the R-Car ISP core that
will span multiple files move the existing driver to a separate
directory. While at it rename the single source file to allow future
files to be grouped by functions.

Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Reviewed-by: Jacopo Mondi <jacopo.mondi@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Link: https://lore.kernel.org/r/20250423163113.2961049-6-niklas.soderlund+renesas@ragnatech.se
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

dt-bindings: media: renesas,isp: Add ISP core function block

Some R-Car ISP instances have in addition to the channel selector (CS)
an ISP core (CORE) to perform operations on an image stream. The core
function is mapped to a different memory region and has a separate
interrupt than CS, extend the bindings to allow describing this.

On the same SoC different instances of the ISP IP may have, or not have,
the CORE functionality. The CS function on all instances on the SoC are
the same and the documentation describes the full ISP (CS + CORE) as a
single IP block. Where instances not having the CORE function simply
lack the functionality to modify the image data. There are dependencies
on the CS functionality while operating the CORE functionality.

In order for the ISP core to function in memory-to-memory mode it needs
to be feed input data from a Streaming Bridge interface. This interface
is provided thru the VSP-X device. Add an optional new property
"renesas,vspx" to provide a phandle to describe this relationship.

While adding mandatory reg-names and interrupt-names breaks existing
bindings the driver itself remains backward compatible and provides CS
functionality if a single unnamed reg and interrupt property is present.
Furthermore all existing users of the bindings are updated in following
work to add these new mandatory properties.

Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Reviewed-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Link: https://lore.kernel.org/r/20250423163113.2961049-2-niklas.soderlund+renesas@ragnatech.se
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: rcar-vin: Fix RAW10

Fix the following to get RAW10 formats working:

In rvin_formats, the bpp is set to 4 for RAW10. As VIN unpacks RAW10 to
16-bit containers, the bpp should be 2.

Don't set VNDMR_YC_THR to the VNDMR register. The YC_THR is "YC Data
Through Mode", used for YUV formats and should not be set for RAW10.

Fixes: 1b7e7240eaf3 ("media: rcar-vin: Add support for RAW10")
Cc: stable@vger.kernel.org
Signed-off-by: Tomi Valkeinen <tomi.valkeinen+renesas@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Tested-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Link: https://lore.kernel.org/r/20250424-rcar-fix-raw-v2-4-f6afca378124@ideasonboard.com
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: rcar-vin: Fix RAW8

On Gen4 we need to set VNMC's EXINF to a different value (1) than in
Gen3 (0). Add a define for this, and set the bit for Gen4.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen+renesas@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Tested-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Link: https://lore.kernel.org/r/20250424-rcar-fix-raw-v2-3-f6afca378124@ideasonboard.com
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: rcar-vin: Remove unnecessary checks

Remove unnecessary checks wrt. formats and interfaces in rvin_setup().
The validity of the formats has already been checked earlier.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen+renesas@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Tested-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Link: https://lore.kernel.org/r/20250424-rcar-fix-raw-v2-2-f6afca378124@ideasonboard.com
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: rcar-vin: Add RCAR_GEN4 model value

Currently Gen4 VINs are marked as RCAN_GEN3 models. Add a new enum
value, RCAR_GEN4, and use it for Gen4 VINs. No functional changes in
this patch.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen+renesas@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Tested-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Link: https://lore.kernel.org/r/20250424-rcar-fix-raw-v2-1-f6afca378124@ideasonboard.com
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: iris: add qcs8300 platform data

Add platform data for QCS8300, which has different capabilities compared
to SM8550. Introduce a QCS8300 header that defines these capabilities.

Reviewed-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Signed-off-by: Vikash Garodia <quic_vgarodia@quicinc.com>
Reviewed-by: Dikshita Agarwal <quic_dikshita@quicinc.com>
Signed-off-by: Bryan O'Donoghue <bod@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: iris: fix the order of compat strings

Fix the order of compatible strings to make it in alpha numeric order.

Reviewed-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Signed-off-by: Vikash Garodia <quic_vgarodia@quicinc.com>
Reviewed-by: Dikshita Agarwal <quic_dikshita@quicinc.com>
Signed-off-by: Bryan O'Donoghue <bod@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

dt-bindings: media: qcom,sm8550-iris: document QCS8300 IRIS accelerator

Document the IRIS video decoder/encoder accelerator found in the QCS8300
platform. It belongs to same iris v3 family as that of SM8550 but is a
downscaled version of SM8550. It has 2 frame processing hardware blocks
while SM8550 has 4. Thereby QCS8300 have fewer capabilities than those
of SM8550.

Reviewed-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Signed-off-by: Vikash Garodia <quic_vgarodia@quicinc.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Bryan O'Donoghue <bod@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: platform: qcom/iris: add sm8650 support

Add support for the SM8650 platform by re-using the SM8550
definitions and using the vpu33 ops.

Move the Sm8650 reset tables that differs in a per-SoC platform
header, that will contain mode SoC specific data when
more codecs are introduced.

The SM8650/vpu33 requires more reset lines, but the H.264
decoder capabilities are identical.

Tested-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org> # x1e Dell
Reviewed-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Reviewed-by: Vikash Garodia <quic_vgarodia@quicinc.com>
Signed-off-by: Bryan O'Donoghue <bod@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: platform: qcom/iris: rename platform_sm8550 to platform_gen2

In order to prepare for supporting the SM8650 SoC, move the
iris_platform_sm8550.c file into iris_platform_gen2.c that will
contain all the common HFI GEN2x structures.

Reviewed-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Tested-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org> # x1e Dell
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Reviewed-by: Vikash Garodia <quic_vgarodia@quicinc.com>
Signed-off-by: Bryan O'Donoghue <bod@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: platform: qcom/iris: add support for vpu33

The IRIS acceleration found in the SM8650 platforms uses the vpu33
hardware version, and requires a slighly different reset and power off
sequences in order to properly get out of runtime suspend.

Tested-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org> # x1e Dell
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Reviewed-by: Vikash Garodia <quic_vgarodia@quicinc.com>
Signed-off-by: Bryan O'Donoghue <bod@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: platform: qcom/iris: rename iris_vpu3 to iris_vpu3x

The vpu33 HW is very close to vpu3, and shares most of the
operations, so rename file to vpu3x since we'll handle all vpu3
variants in it.

Reviewed-by: Dikshita Agarwal <quic_dikshita@quicinc.com>
Tested-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org> # x1e Dell
Reviewed-by: Vikash Garodia <quic_vgarodia@quicinc.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Bryan O'Donoghue <bod@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: platform: qcom/iris: introduce optional controller_rst_tbl

Introduce an optional controller_rst_tbl use to store reset lines
used to reset part of the controller.

This is necessary for the vpu3 support, when the xo reset line
must be asserted separately from the other reset line
on power off operation.

Factor the iris_init_resets() logic to allow requesting
multiple reset tables.

Reviewed-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Tested-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org> # x1e Dell
Reviewed-by: Dikshita Agarwal <quic_dikshita@quicinc.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Reviewed-by: Vikash Garodia <quic_vgarodia@quicinc.com>
Signed-off-by: Bryan O'Donoghue <bod@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: platform: qcom/iris: add power_off_controller to vpu_ops

In order to support the SM8650 iris33 hardware, we need to provide a
specific constoller power off sequences via the vpu_ops callbacks.

Add the callback, and use the current helper for currently supported
platforms.

Reviewed-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Reviewed-by: Dikshita Agarwal <quic_dikshita@quicinc.com>
Tested-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org> # x1e Dell
Reviewed-by: Vikash Garodia <quic_vgarodia@quicinc.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Bryan O'Donoghue <bod@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

dt-bindings: media: qcom,sm8550-iris: document SM8650 IRIS accelerator

Document the IRIS video decoder and encoder accelerator found in the
SM8650 platform, it requires 2 more reset lines in addition to the
properties required for the SM8550 platform.

Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Reviewed-by: Vikash Garodia <quic_vgarodia@quicinc.com>
Reviewed-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Tested-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org> # x1e Dell
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Bryan O'Donoghue <bod@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: rkisp1: Remove unnecessary defines

The effect modes are not shifts but numbers which are already defined a
few lines above. Remove the misleading defines.

Signed-off-by: Stefan Klug <stefan.klug@ideasonboard.com>
Link: https://lore.kernel.org/r/20250227114558.3097101-4-stefan.klug@ideasonboard.com
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

media: nxp: imx8-isi: better handle the m2m usage_count

Currently, if streamon/streamoff calls are imbalanced we can either end up
with a negative ISI m2m usage_count (if streamoff() is called more times
than streamon()) in which case we'll not be able to restart the ISI pipe
next time, or the usage_count never gets to 0 and the pipe is never
switched off.

To avoid that, add a 'streaming' flag to mxc_isi_m2m_ctx_queue_data and use it
in the streamon/streamoff to avoid incrementing/decrementing the usage_count
uselessly, if called multiple times from the same context.

Fixes: cf21f328fcafac ("media: nxp: Add i.MX8 ISI driver")
Cc: stable@vger.kernel.org
Suggested-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurentiu Palcu <laurentiu.palcu@oss.nxp.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Link: https://lore.kernel.org/r/20241023085643.978729-1-laurentiu.palcu@oss.nxp.com
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

media: rc: add keymap for Hauppauge Credit Card RC

Add keymap for the black Hauppauge Credit Card Remote Control.

Signed-off-by: Petja Patjas <pp01415943@gmail.com>
Signed-off-by: Sean Young <sean@mess.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

media: uapi: cec-funcs.h: use CEC_LOG_ADDR_BROADCAST

The cec-funcs.h header sets the destination to 0xf for those
messages that can only be broadcast. Instead of writing:

msg->msg[0] |= 0xf; /* broadcast */

just write:

msg->msg[0] |= CEC_LOG_ADDR_BROADCAST;

which is more descriptive and allows us to drop the comment.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

media: tc358743: ignore video while HPD is low

If the HPD is low (happens if there is no EDID or the
EDID is being updated), then return -ENOLINK in
tc358743_get_detected_timings() instead of detecting video.

This avoids userspace thinking that it can start streaming when
the HPD is low.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Tested-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/linux-media/20240628-stoic-bettong-of-fortitude-e25611@houat/
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

media: platform: use (t,l)/wxh format for rectangle

Standardize reporting of rectangles to (t,l)/wxh.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

media: pci: zoran: use (t,l)/wxh format for rectangle

Standardize reporting of rectangles to (t,l)/wxh.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

media: i2c: imx283: use (t,l)/wxh format for rectangle

Standardize reporting of rectangles to (t,l)/wxh.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

media: vivid: use (t,l)/wxh format for rectangle

Standardize reporting of rectangles to (t,l)/wxh.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

media: usb: em28xx: use (t,l)/wxh format for rectangle

Standardize reporting of rectangles to (t,l)/wxh.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

media: staging: atomisp/starfive: use (t,l)/wxh format for rectangle

Standardize reporting of rectangles to (t,l)/wxh.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

media: omap3isp: drop wait_prepare/finish callbacks

Since commit 88785982a19d ("media: vb2: use lock if wait_prepare/finish
are NULL") it is no longer needed to set the wait_prepare/finish
vb2_ops callbacks as long as the lock field in vb2_queue is set.

Set the queue lock to &video->queue_lock, which makes it possible to drop
the wait_prepare/finish callbacks.

This simplifies the code and this is a step towards the goal of deleting
these callbacks.

Note that the lock field of struct video_device is never set in this
driver, so the core v4l2_ioctl.c function v4l2_ioctl_get_lock() will never
take a lock.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Acked-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

media: gspca: Add error handling for stv06xx_read_sensor()

In hdcs_init(), the return value of stv06xx_read_sensor() needs to be
checked. A proper implementation can be found in vv6410_dump(). Add a
check in loop condition and propergate error code to fix this issue.

Fixes: 4c98834addfe ("V4L/DVB (10048): gspca - stv06xx: New subdriver.")
Cc: stable@vger.kernel.org # v2.6+
Signed-off-by: Wentao Liang <vulab@iscas.ac.cn>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: platform: exynos4-is: Add hardware sync wait to fimc_is_hw_change_mode()

In fimc_is_hw_change_mode(), the function changes camera modes without
waiting for hardware completion, risking corrupted data or system hangs
if subsequent operations proceed before the hardware is ready.

Add fimc_is_hw_wait_intmsr0_intmsd0() after mode configuration, ensuring
hardware state synchronization and stable interrupt handling.

Signed-off-by: Wentao Liang <vulab@iscas.ac.cn>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: adv7511-v4l2: use constants for BT.2020 colorimetry

Replace numeric values with constants from hdmi.h.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: davinci: vpif: Fix memory leak in probe error path

If an error occurs during the initialization of `pdev_display`,
the allocated platform device `pdev_capture` is not released properly,
leading to a memory leak.

Adjust error path handling to fix the leak.

Found by Linux Verification Center (linuxtesting.org) with SVACE.

Fixes: 43acb728bbc4 ("media: davinci: vpif: fix use-after-free on driver unbind")
Cc: stable@vger.kernel.org
Signed-off-by: Dmitry Nikiforov <Dm1tryNk@yandex.ru>
Reviewed-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: dvb: Fix typos bloc -> block

There are some typos in comments / dprintk messages: bloc -> block.

Fix them via codespell.

Signed-off-by: Andrew Kreimer <algonell@gmail.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: vivid: Change the siize of the composing

syzkaller found a bug:

BUG: KASAN: vmalloc-out-of-bounds in tpg_fill_plane_pattern drivers/media/common/v4l2-tpg/v4l2-tpg-core.c:2608 [inline]
BUG: KASAN: vmalloc-out-of-bounds in tpg_fill_plane_buffer+0x1a9c/0x5af0 drivers/media/common/v4l2-tpg/v4l2-tpg-core.c:2705
Write of size 1440 at addr ffffc9000d0ffda0 by task vivid-000-vid-c/5304

CPU: 0 UID: 0 PID: 5304 Comm: vivid-000-vid-c Not tainted 6.14.0-rc2-syzkaller-00039-g09fbf3d50205 #0
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2~bpo12+1 04/01/2014

Call Trace:
 <TASK>
 __dump_stack lib/dump_stack.c:94 [inline]
 dump_stack_lvl+0x241/0x360 lib/dump_stack.c:120
 print_address_description mm/kasan/report.c:378 [inline]
 print_report+0x169/0x550 mm/kasan/report.c:489
 kasan_report+0x143/0x180 mm/kasan/report.c:602
 kasan_check_range+0x282/0x290 mm/kasan/generic.c:189
 __asan_memcpy+0x40/0x70 mm/kasan/shadow.c:106
 tpg_fill_plane_pattern drivers/media/common/v4l2-tpg/v4l2-tpg-core.c:2608 [inline]
 tpg_fill_plane_buffer+0x1a9c/0x5af0 drivers/media/common/v4l2-tpg/v4l2-tpg-core.c:2705
 vivid_fillbuff drivers/media/test-drivers/vivid/vivid-kthread-cap.c:470 [inline]
 vivid_thread_vid_cap_tick+0xf8e/0x60d0 drivers/media/test-drivers/vivid/vivid-kthread-cap.c:629
 vivid_thread_vid_cap+0x8aa/0xf30 drivers/media/test-drivers/vivid/vivid-kthread-cap.c:767
 kthread+0x7a9/0x920 kernel/kthread.c:464
 ret_from_fork+0x4b/0x80 arch/x86/kernel/process.c:148
 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244
 </TASK>

The composition size cannot be larger than the size of fmt_cap_rect.
So execute v4l2_rect_map_inside() even if has_compose_cap == 0.

Fixes: 94a7ad928346 ("media: vivid: fix compose size exceed boundary")
Cc: stable@vger.kernel.org
Reported-by: syzbot+365005005522b70a36f2@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?id=8ed8e8cc30cbe0d86c9a25bd1d6a5775129b8ea3
Signed-off-by: Denis Arefev <arefev@swemel.ru>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: dt-bindings: Document Tegra186 and Tegra194 cec

These are already used in device trees, so describe them here. As the
driver only declares up through Tegra210, these must use a fallback
compatible of tegra210-cec.

Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Aaron Kling <webgeek1234@gmail.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: platform: exynos4-is: Use of_get_available_child_by_name()

Simplify fimc_md_is_isp_available() by using
of_get_available_child_by_name().

Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: platform: cros-ec: Add Moxie to the match table

The Google Moxie device uses the same approach as the Google Brask
which enables the HDMI CEC via the cros-ec-cec driver.

Signed-off-by: Ken Lin <kenlin5@quanta.corp-partner.google.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: stm32-dcmi: use devm_kmemdup_array()

Convert to use devm_kmemdup_array() and while at it, make the size robust
against type changes.

Signed-off-by: Raag Jadav <raag.jadav@intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: atmel-isi: use devm_kmemdup_array()

Convert to use devm_kmemdup_array() and while at it, make the size robust
against type changes.

Signed-off-by: Raag Jadav <raag.jadav@intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: mgb4: Enumerate only the available timings

Enumerate only the available (as given by the input source) timings, not
all theoretically possible.

Signed-off-by: Martin Tůma <martin.tuma@digiteqautomotive.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: nuvoton: npcm-video: Prevent returning unsupported resolutions

To restrict the returned resolution due to Nuvoton SoC hardware
limitations.

Signed-off-by: Michael Chang <zhang971090220@gmail.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: nuvoton: npcm-video: Fix stuck due to no video signal error

Fix the issue when start_frame and detect_resolution
functions are executed at the same time, which may cause driver
stops capturing due to status of no video signal error.

Signed-off-by: Michael Chang <zhang971090220@gmail.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: platform: cros-ec: Add Dirks to the match table

The Google Dirks device uses the same approach as the Google Brask
which enables the HDMI CEC via the cros-ec-cec driver.

Signed-off-by: Kells Ping <kellsping@gmail.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: mgb4: Fix resolution change events triggering

Always generate the resolution change event when the HW reports it and only
discard the streaming termination in case the new resolution is the same as
the old one. The old logic prevented events on
"no signal" -> "valid resolution" transitions as VIDIOC_QUERY_DV_TIMINGS
never updates the timings when there is no signal present.

Signed-off-by: Martin Tůma <martin.tuma@digiteqautomotive.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: cxusb: no longer judge rbuf when the write fails

syzbot reported a uninit-value in cxusb_i2c_xfer. [1]

Only when the write operation of usb_bulk_msg() in dvb_usb_generic_rw()
succeeds and rlen is greater than 0, the read operation of usb_bulk_msg()
will be executed to read rlen bytes of data from the dvb device into the
rbuf.

In this case, although rlen is 1, the write operation failed which resulted
in the dvb read operation not being executed, and ultimately variable i was
not initialized.

[1]
BUG: KMSAN: uninit-value in cxusb_gpio_tuner drivers/media/usb/dvb-usb/cxusb.c:124 [inline]
BUG: KMSAN: uninit-value in cxusb_i2c_xfer+0x153a/0x1a60 drivers/media/usb/dvb-usb/cxusb.c:196
 cxusb_gpio_tuner drivers/media/usb/dvb-usb/cxusb.c:124 [inline]
 cxusb_i2c_xfer+0x153a/0x1a60 drivers/media/usb/dvb-usb/cxusb.c:196
 __i2c_transfer+0xe25/0x3150 drivers/i2c/i2c-core-base.c:-1
 i2c_transfer+0x317/0x4a0 drivers/i2c/i2c-core-base.c:2315
 i2c_transfer_buffer_flags+0x125/0x1e0 drivers/i2c/i2c-core-base.c:2343
 i2c_master_send include/linux/i2c.h:109 [inline]
 i2cdev_write+0x210/0x280 drivers/i2c/i2c-dev.c:183
 do_loop_readv_writev fs/read_write.c:848 [inline]
 vfs_writev+0x963/0x14e0 fs/read_write.c:1057
 do_writev+0x247/0x5c0 fs/read_write.c:1101
 __do_sys_writev fs/read_write.c:1169 [inline]
 __se_sys_writev fs/read_write.c:1166 [inline]
 __x64_sys_writev+0x98/0xe0 fs/read_write.c:1166
 x64_sys_call+0x2229/0x3c80 arch/x86/include/generated/asm/syscalls_64.h:21
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0xcd/0x1e0 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

Reported-by: syzbot+526bd95c0ec629993bf3@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=526bd95c0ec629993bf3
Tested-by: syzbot+526bd95c0ec629993bf3@syzkaller.appspotmail.com
Fixes: 22c6d93a7310 ("[PATCH] dvb: usb: support Medion hybrid USB2.0 DVB-T/analogue box")
Cc: stable@vger.kernel.org
Signed-off-by: Edward Adam Davis <eadavis@qq.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: tw5864: Replace deprecated PCI functions

pcim_iomap_table() and pcim_iomap_regions() have been deprecated.
pcim_iomap_regions(), furthermore, has so far wrongly been passed the
device's name instead of the driver's name, which makes that function's
debug prints useless.

Replace the deprecated function with pcim_iomap_region().

Define the driver name globally and use it where appropriate.

Signed-off-by: Philipp Stanner <phasta@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: solo6x10: Replace deprecated PCI functions

pcim_iomap_table() and pcim_iomap_regions() have been deprecated.
Replace them with pcim_iomap_region().

Signed-off-by: Philipp Stanner <phasta@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: pt3: Replace deprecated PCI functions

pcim_iomap_table() and pcim_iomap_regions() have been deprecated.
Replace them with pcim_iomap_region().

Signed-off-by: Philipp Stanner <phasta@kernel.org>
Tested-by: Akihiro Tsukada <tskd08@gmail.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: cec: extron-da-hd-4k-plus: Fix Wformat-truncation

Fix gcc8 warning:

drivers/media/cec/usb/extron-da-hd-4k-plus/extron-da-hd-4k-plus.c:1014:44: warning: 'DCEC' directive output may be truncated writing 4 bytes into a region of size between 0 and 53 [-Wformat-truncation=]

Resizing the 'buf' and 'cmd' arrays fixed the warning.

Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: atomisp: Fix Wformat-truncation warning

Gcc8 is convinced that we do not have enough space in dot_id_input_bin.
Extend the variable 17 bytes which is just used for debugging.

drivers/staging/media/atomisp/pci/runtime/debug/src/ia_css_debug.c:1336:9: warning: '(pipe' directive output may be truncated writing 5 bytes into a region of size between 1 and 74 [-Wformat-truncation=]

Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Reviewed-by: Andy Shevchenko <andy@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: videobuf2: check constants during build time

There is nothing a driver author can do fix in the driver to make the
global constants match. Since the assertion can be verified at build
time, don't return EINVAL at runtime for it.

Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Acked-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: vivid: Add more webcam resolutions

Add 3 more common resolution for webcams. This is required to increase
the test coverage of unit tests based on vivid.

Co-developed-by: Hidenori Kobayashi <hidenorik@chromium.org>
Signed-off-by: Hidenori Kobayashi <hidenorik@chromium.org>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Tested-by: Hidenori Kobayashi <hidenorik@chromium.org> # v4l2-compliance,
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: vivid: Fix requirement about webcam_intervals

Since commit f0b4a2c037c0 ("media: vivid: Extend FPS rates offered by
simulated webcam") we do not require twice as many intervals as sizes. In
fact, now we have 13 intervals and 6 sizes.

Fix the comment.

Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

docs: media: mgb4: Improve mgb4 driver documentation

Add some basic info about the HW/driver + contact info.

Signed-off-by: Martin Tůma <martin.tuma@digiteqautomotive.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: vidtv: Terminating the subsequent process of initialization failure

syzbot reported a slab-use-after-free Read in vidtv_mux_init. [1]

After PSI initialization fails, the si member is accessed again, resulting
in this uaf.

After si initialization fails, the subsequent process needs to be exited.

[1]
BUG: KASAN: slab-use-after-free in vidtv_mux_pid_ctx_init drivers/media/test-drivers/vidtv/vidtv_mux.c:78 [inline]
BUG: KASAN: slab-use-after-free in vidtv_mux_init+0xac2/0xbe0 drivers/media/test-drivers/vidtv/vidtv_mux.c:524
Read of size 8 at addr ffff88802fa42acc by task syz.2.37/6059

CPU: 0 UID: 0 PID: 6059 Comm: syz.2.37 Not tainted 6.14.0-rc5-syzkaller #0
Hardware name: Google Compute Engine, BIOS Google 02/12/2025
Call Trace:
<TASK>
__dump_stack lib/dump_stack.c:94 [inline]
dump_stack_lvl+0x116/0x1f0 lib/dump_stack.c:120
print_address_description mm/kasan/report.c:408 [inline]
print_report+0xc3/0x670 mm/kasan/report.c:521
kasan_report+0xd9/0x110 mm/kasan/report.c:634
vidtv_mux_pid_ctx_init drivers/media/test-drivers/vidtv/vidtv_mux.c:78
vidtv_mux_init+0xac2/0xbe0 drivers/media/test-drivers/vidtv/vidtv_mux.c:524
vidtv_start_streaming drivers/media/test-drivers/vidtv/vidtv_bridge.c:194
vidtv_start_feed drivers/media/test-drivers/vidtv/vidtv_bridge.c:239
dmx_section_feed_start_filtering drivers/media/dvb-core/dvb_demux.c:973
dvb_dmxdev_feed_start drivers/media/dvb-core/dmxdev.c:508 [inline]
dvb_dmxdev_feed_restart.isra.0 drivers/media/dvb-core/dmxdev.c:537
dvb_dmxdev_filter_stop+0x2b4/0x3a0 drivers/media/dvb-core/dmxdev.c:564
dvb_dmxdev_filter_free drivers/media/dvb-core/dmxdev.c:840 [inline]
dvb_demux_release+0x92/0x550 drivers/media/dvb-core/dmxdev.c:1246
__fput+0x3ff/0xb70 fs/file_table.c:464
task_work_run+0x14e/0x250 kernel/task_work.c:227
exit_task_work include/linux/task_work.h:40 [inline]
do_exit+0xad8/0x2d70 kernel/exit.c:938
do_group_exit+0xd3/0x2a0 kernel/exit.c:1087
__do_sys_exit_group kernel/exit.c:1098 [inline]
__se_sys_exit_group kernel/exit.c:1096 [inline]
__x64_sys_exit_group+0x3e/0x50 kernel/exit.c:1096
x64_sys_call+0x151f/0x1720 arch/x86/include/generated/asm/syscalls_64.h:232
do_syscall_x64 arch/x86/entry/common.c:52 [inline]
do_syscall_64+0xcd/0x250 arch/x86/entry/common.c:83
entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7f871d58d169
Code: Unable to access opcode bytes at 0x7f871d58d13f.
RSP: 002b:00007fff4b19a788 EFLAGS: 00000246 ORIG_RAX: 00000000000000e7
RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f871d58d169
RDX: 0000000000000064 RSI: 0000000000000000 RDI: 0000000000000000
RBP: 00007fff4b19a7ec R08: 0000000b4b19a87f R09: 00000000000927c0
R10: 0000000000000001 R11: 0000000000000246 R12: 0000000000000003
R13: 00000000000927c0 R14: 000000000001d553 R15: 00007fff4b19a840
 </TASK>

Allocated by task 6059:
 kasan_save_stack+0x33/0x60 mm/kasan/common.c:47
 kasan_save_track+0x14/0x30 mm/kasan/common.c:68
 poison_kmalloc_redzone mm/kasan/common.c:377 [inline]
 __kasan_kmalloc+0xaa/0xb0 mm/kasan/common.c:394
 kmalloc_noprof include/linux/slab.h:901 [inline]
 kzalloc_noprof include/linux/slab.h:1037 [inline]
 vidtv_psi_pat_table_init drivers/media/test-drivers/vidtv/vidtv_psi.c:970
 vidtv_channel_si_init drivers/media/test-drivers/vidtv/vidtv_channel.c:423
 vidtv_mux_init drivers/media/test-drivers/vidtv/vidtv_mux.c:519
 vidtv_start_streaming drivers/media/test-drivers/vidtv/vidtv_bridge.c:194
 vidtv_start_feed drivers/media/test-drivers/vidtv/vidtv_bridge.c:239
 dmx_section_feed_start_filtering drivers/media/dvb-core/dvb_demux.c:973
 dvb_dmxdev_feed_start drivers/media/dvb-core/dmxdev.c:508 [inline]
 dvb_dmxdev_feed_restart.isra.0 drivers/media/dvb-core/dmxdev.c:537
 dvb_dmxdev_filter_stop+0x2b4/0x3a0 drivers/media/dvb-core/dmxdev.c:564
 dvb_dmxdev_filter_free drivers/media/dvb-core/dmxdev.c:840 [inline]
 dvb_demux_release+0x92/0x550 drivers/media/dvb-core/dmxdev.c:1246
 __fput+0x3ff/0xb70 fs/file_table.c:464
 task_work_run+0x14e/0x250 kernel/task_work.c:227
 exit_task_work include/linux/task_work.h:40 [inline]
 do_exit+0xad8/0x2d70 kernel/exit.c:938
 do_group_exit+0xd3/0x2a0 kernel/exit.c:1087
 __do_sys_exit_group kernel/exit.c:1098 [inline]
 __se_sys_exit_group kernel/exit.c:1096 [inline]
 __x64_sys_exit_group+0x3e/0x50 kernel/exit.c:1096
 x64_sys_call arch/x86/include/generated/asm/syscalls_64.h:232
 do_syscall_x64 arch/x86/entry/common.c:52 [inline]
 do_syscall_64+0xcd/0x250 arch/x86/entry/common.c:83
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

Freed by task 6059:
 kasan_save_stack+0x33/0x60 mm/kasan/common.c:47
 kasan_save_track+0x14/0x30 mm/kasan/common.c:68
 kasan_save_free_info+0x3b/0x60 mm/kasan/generic.c:576
 poison_slab_object mm/kasan/common.c:247 [inline]
 __kasan_slab_free+0x51/0x70 mm/kasan/common.c:264
 kasan_slab_free include/linux/kasan.h:233 [inline]
 slab_free_hook mm/slub.c:2353 [inline]
 slab_free mm/slub.c:4609 [inline]
 kfree+0x2c4/0x4d0 mm/slub.c:4757
 vidtv_channel_si_init drivers/media/test-drivers/vidtv/vidtv_channel.c:499
 vidtv_mux_init drivers/media/test-drivers/vidtv/vidtv_mux.c:519
 vidtv_start_streaming drivers/media/test-drivers/vidtv/vidtv_bridge.c:194
 vidtv_start_feed drivers/media/test-drivers/vidtv/vidtv_bridge.c:239
 dmx_section_feed_start_filtering drivers/media/dvb-core/dvb_demux.c:973
 dvb_dmxdev_feed_start drivers/media/dvb-core/dmxdev.c:508 [inline]
 dvb_dmxdev_feed_restart.isra.0 drivers/media/dvb-core/dmxdev.c:537
 dvb_dmxdev_filter_stop+0x2b4/0x3a0 drivers/media/dvb-core/dmxdev.c:564
 dvb_dmxdev_filter_free drivers/media/dvb-core/dmxdev.c:840 [inline]
 dvb_demux_release+0x92/0x550 drivers/media/dvb-core/dmxdev.c:1246
 __fput+0x3ff/0xb70 fs/file_table.c:464
 task_work_run+0x14e/0x250 kernel/task_work.c:227
 exit_task_work include/linux/task_work.h:40 [inline]
 do_exit+0xad8/0x2d70 kernel/exit.c:938
 do_group_exit+0xd3/0x2a0 kernel/exit.c:1087
 __do_sys_exit_group kernel/exit.c:1098 [inline]
 __se_sys_exit_group kernel/exit.c:1096 [inline]
 __x64_sys_exit_group+0x3e/0x50 kernel/exit.c:1096
 x64_sys_call arch/x86/include/generated/asm/syscalls_64.h:232
 do_syscall_x64 arch/x86/entry/common.c:52 [inline]
 do_syscall_64+0xcd/0x250 arch/x86/entry/common.c:83
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

Fixes: 3be8037960bc ("media: vidtv: add error checks")
Cc: stable@vger.kernel.org
Reported-by: syzbot+0d33ab192bd50b6c91e6@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=0d33ab192bd50b6c91e6
Signed-off-by: Edward Adam Davis <eadavis@qq.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: pvrusb2: Remove unused pvr2_std_create_enum

pvr2_std_create_enum() has been unused since 2012's
commit c0bb609fdc0b ("[media] pvrusb2: Get rid of obsolete code for video
standard enumeration")

Remove it.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: s5p-mfc: Support for handling RET_ENC_BUFFER_FULL interrupt

When output encoded buffer size provided by userspace
is insufficient with current encoding parameters, it
leads to RET_ENC_BUFFER_FULL interrupt which was not
handled in IRQ handler.

On handling of RET_ENC_BUFFER_FULL interrupt leads to
NAL_ABORT command from host to risc which in turn leads
to RET_NAL_ABORT interrupt. On receiving RET_NAL_ABORT
driver clears workbit and VB2 queues for cleaner closing
of MFC instance.

When user encounters "Call on DQBUF after unrecoverable
error", userspace should close fd and restart with larger
output encoder buffer size.

Signed-off-by: Aakarsh Jain <aakarsh.jain@samsung.com>
Acked-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: vim2m: Add parametized support for multiplanar API

Add support for the mulitiplaner API. The device can now act as
either a multi-planar or a single-planar device depending on a module
parameter, similar to the way vivid behaves.

Multiplanar support was added by implementing the appropate
try/get/set mplane functions, and by modifying the queue_setup() and
buf_prepare() functions to handle multiple planes. Implementation
was inspired by vivid.

Signed-off-by: Matthew Majewski <mattwmajewski@gmail.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: vim2m: Simplify try_fmt

Clean up vidioc_try_fmt with the following changes:

1. remove unsused vim2m_fmt parameter
2. use clamp() macro to restrain width/height bounds
3. use ALIGN() macro to align width/height
4. use v4l2_fill_pixfmt to set bytesperline/sizeimage

Signed-off-by: Matthew Majewski <mattwmajewski@gmail.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: v4l2-common: Add RGBR format info

Add missing RGBR entry in the v4l2_format_info[] table.

RGBR has identical format information to RGBP, as it is a
big endian variant of RGB-5-6-5 pixel encoding according to the
description in videodev2.h.

Signed-off-by: Matthew Majewski <mattwmajewski@gmail.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: remove STA2x11 media pci driver

With commit dcbb01fbb7ae ("x86/pci: Remove old STA2x11 support"), the
STA2X11 Video Input Port driver is not needed and cannot be built anymore.

Remove the driver and its reference in media documentation.

Signed-off-by: Lukas Bulwahn <lukas.bulwahn@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: i2c: imx334: switch to {enable,disable}_streams

Switch from s_stream to enable_streams and disable_streams callbacks.

Signed-off-by: Tarang Raval <tarang.raval@siliconsignals.io>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: i2c: imx334: Use subdev state lock for synchronization

Replace the custom mutex in the imx334 driver with the V4L2 subdev state
lock for control synchronization. Initialize the subdev with
v4l2_subdev_init_finalize in imx334_probe, adding proper cleanup in error
paths and imx334_remove. This aligns the driver with V4L2 standards.

Signed-off-by: Tarang Raval <tarang.raval@siliconsignals.io>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: i2c: imx334: Enable runtime PM before sub-device registration

Runtime PM is fully initialized before calling
v4l2_async_register_subdev_sensor(). Moving the runtime PM initialization
earlier prevents potential access to an uninitialized or powered-down
device.

Signed-off-by: Tarang Raval <tarang.raval@siliconsignals.io>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: i2c: imx334: Fix runtime PM handling in remove function

pm_runtime_suspended() only checks the current runtime PM status and does
not modify it, making it ineffective in this context. This could result in
improper power management if the device remains active when removed.

This patch fixes the issue by introducing a check with
pm_runtime_status_suspended() to determine if the device is already
suspended. If it is not, it calls imx334_power_off() to power down the
device and then uses pm_runtime_set_suspended() to correctly update the
runtime PM status to suspended.

Signed-off-by: Tarang Raval <tarang.raval@siliconsignals.io>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: i2c: imx334: Fix power management and control handling

Some controls may need the sensor to be powered on to update their
values. Currently, only the exposure control does this. To ensure
proper handling, the power-up sequence is moved outside the switch-case.

Additionally, VBLANK control is now processed earlier so its changes
can correctly affect other controls.

Signed-off-by: Tarang Raval <tarang.raval@siliconsignals.io>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: i2c: imx334: Configure lane mode dynamically

Configure the lane mode dynamically from the streaming
function instead of using a hardcoded value.

Signed-off-by: Tarang Raval <tarang.raval@siliconsignals.io>
[Sakari Ailus: Fix checkpatch.pl issue, lower Dynamically.]
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: i2c: imx334: Remove redundant register entries

IMX334_REG_{ADBIT, MDBIT, ADBIT1}: Already written in
imx334_set_framefmt function.

IMX334_REG_THSPREPARE: Unnecessary repeated writes removed.

CCI_REG8(0x3e04): Unnecessary repeated writes removed.

Signed-off-by: Tarang Raval <tarang.raval@siliconsignals.io>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: i2c: imx334: Convert to CCI register access helpers

Use the new common CCI register access helpers to replace the private
register access helpers in the imx334 driver. This simplifies the driver
by reducing the amount of code.

Acked-by: Shravan Chippa <Shravan.Chippa@microchip.com>
Signed-off-by: Tarang Raval <tarang.raval@siliconsignals.io>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: i2c: imx334: Simplify with dev_err_probe()

Error handling in probe() can be a bit simpler with dev_err_probe().

also, Added missing newline characters (\n) in error messages.

Signed-off-by: Tarang Raval <tarang.raval@siliconsignals.io>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: intel/ipu6: Fix dma mask for non-secure mode

We use dma_get_mask() of auxdev device for calculate iova pfn limit.
This is always 32 bit mask as we do not initialize the mask (and we can
not do so, since dev->dev_mask is NULL anyways for auxdev).

Since we need 31 bit mask for non-secure mode use mmu_info->aperture_end
which is properly initialized to correct mask for both modes.

Fixes: daabc5c64703 ("media: ipu6: not override the dma_ops of device in driver")
Cc: stable@vger.kernel.org
Signed-off-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: intel/ipu6: Minor dma_mask clenaup

Remove unused dma_mask field and ipu-dma.h includes.

Signed-off-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: ccs: Don't complain about missing "clock-frequency" property

The clock frequency is often available via the clock itself and not read
by the driver from the "clock-frequency" property. Don't complain if the
property doesn't exist.

Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: ccs: Don't complain about lack of quirks

Generally any deviance from the standard is handled via CCS static data
nowadays and so not having quirks is expected. Don't warn about it.

Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: ccs: Remove I²C write retry hack

The I²C retry hack has been there in order to address transient I²C
register write access issues on a few very old sensors and possibly it has
addressed also first I²C access problems (device not responding until a
certain amount of time has passed) but that is now separately handled. The
retry hack has a good potential for introducing hard to debug problems in
updating sensor settings while streaming. Remove it and instead pass those
rare errors to the user space -- which is also what virtually all other
drivers do.

Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: ccs: Use read_poll_timeout() in reset polling

Use read_poll_timeout() in polling the device after a reset, either hard
or soft. While at it, improve the related error message.

Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: ccs: Try a little longer to access the sensor before giving up

Some sensors take longer to respond after reset than the spec-required
time. Try up to 1 s for the sensor to become accessible.

Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: i2c: imx219: media: i2c: imx219: Enable runtime PM autosuspend

Use pm_runtime_put_autosuspend() instead of pm_runtime_put()
to allow autosuspend. Set a 1000ms autosuspend delay in
imx219_probe() to improve power efficiency.

Signed-off-by: Tarang Raval <tarang.raval@siliconsignals.io>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: i2c: imx219: switch to {enable,disable}_streams

Switch from s_stream to enable_streams and disable_streams callbacks.

Signed-off-by: Tarang Raval <tarang.raval@siliconsignals.io>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: dt-bindings: Convert Analog Devices ad5820 to DT schema

Convert the Analog Devices ad5820 to DT schema format.

Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: David Heidelberg <david@ixit.cz>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: i2c: ov02e10: add OV02E10 image sensor driver

Add in the ov02e10 driver from the Intel IPU6 repository.

Signed-off-by: Jingjing Xiong <jingjing.xiong@intel.com>
Co-developed-by: Hao Yao <hao.yao@intel.com>
Signed-off-by: Hao Yao <hao.yao@intel.com>
Co-developed-by: Jim Lai <jim.lai@intel.com>
Signed-off-by: Jim Lai <jim.lai@intel.com>
Co-developed-by: You-Sheng Yang <vicamo.yang@canonical.com>
Signed-off-by: You-Sheng Yang <vicamo.yang@canonical.com>
Co-developed-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Co-developed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: dt-bindings: Add OmniVision OV02E10

Add bindings for OVO2E10 a two lane MIPI CSI, two megapixel 1080p RGB
sensor.

Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: intel/ipu6: Replace deprecated PCI functions

pcim_iomap_table() and pcim_iomap_regions() have been deprecated.
Furthermore, the "name" parameter in pcim_iomap_regions() and its
successor, pcim_iomap_region(), should always reflect the driver name,
whereas currently it is the device's name.

Replace the deprecated functions with pcim_iomap_region() and pass the
actual driver name.

Signed-off-by: Philipp Stanner <phasta@kernel.org>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: ipu3-cio2: Replace deprecated PCI functions

pcim_iomap_table() and pcim_iomap_regions() have been deprecated.

Replace them with pcim_iomap_region(), and pass the actual driver name
to that function.

Signed-off-by: Philipp Stanner <phasta@kernel.org>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: rcar-csi2: Remove hack to detect NTSC content

In an effort to emulate support for SEQ_{TB,BT} in the R-Car VIN driver
on data captured from a CVBS input a hack was added to detect NTSC vs
PAL. This is ugly and as support for emulated SEQ_{TB,BT} have been
removed from the VIN driver remove the ugly hack.

Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Reviewed-by: Tomi Valkeinen <tomi.valkeinen+renesas@ideasonboard.com>
Tested-by: Tomi Valkeinen <tomi.valkeinen+renesas@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: rcar-vin: Simplify the shutdown process

When shutting down capture extra care was needed to try and complete a
buffer that was involved in the emulated support for SEQ_{TB,BT}. This
was needed as a buffer might be queued once to the driver, but two times
to the hardware using different offsets.

As support for SEQ_{TB,BT} is now removed this shutdown process can be
greatly simplified. And in addition the state keeping of the VIN device
can be reduced to a single boolean value instead of keeping track of
this SEQ_{TB,BT} stopping dance.

Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Reviewed-by: Tomi Valkeinen <tomi.valkeinen+renesas@ideasonboard.com>
Tested-by: Tomi Valkeinen <tomi.valkeinen+renesas@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: rcar-vin: Remove superfluous starting state

The STARTING state is superfluous and can be replaced with a check of
the sequence counter. The design idea is that the first buffer returned
from the driver have to come from the first hardware buffer slot.
Failing this the first 3 buffers queued to the device can be returned
out-of-order.

But it's much clearer to check the sequence counter to only return the
first buffer if it comes from hardware slot 0 then it is to carry around
an extra state just for this. Remove the unneeded state and replace it
with a simpler check.

Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Reviewed-by: Tomi Valkeinen <tomi.valkeinen+renesas@ideasonboard.com>
Tested-by: Tomi Valkeinen <tomi.valkeinen+renesas@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: rcar-vin: Remove superfluous suspended state

The VIN state of suspended is superfluous. The logic was that when the
device were suspended and in a RUNNING state the state was set to
SUSPENDED. And when resuming it checked if the state is SUSPENDED and if
so started the device and changed it to RUNNING.

This can be avoided by simply checking if the device is in a RUNNING
state at both suspend and resume callbacks. Remove the unneeded
complexity.

Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Reviewed-by: Tomi Valkeinen <tomi.valkeinen+renesas@ideasonboard.com>
Tested-by: Tomi Valkeinen <tomi.valkeinen+renesas@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: rcar-vin: Remove emulated SEQ_{TB,BT}

When the driver was converted from soc_camera software support for
V4L2_FIELD_SEQ_TB and V4L2_FIELD_SEQ_BT were added. This was done by
capturing twice to the same VB2 buffer, but at different offsets.

This turned out to be a bad idea and it never really worked properly in
all situations. As the hardware can't support this mode natively remove
trying to emulate it in software. It's still possible to capture TOP or
BOTTOM fields separately or both ALTERNATING. If user-space wants the
same  fields in the same buffer the same hack to capture twice to the
same buffer can be done.

Removing this error prone emulated support pave ways in future work to
simplify the internal buffer handling and making it less fragile, while
enabling adding support for other features the hardware actually
supports.

Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Reviewed-by: Tomi Valkeinen <tomi.valkeinen+renesas@ideasonboard.com>
Tested-by: Tomi Valkeinen <tomi.valkeinen+renesas@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: ti: cal: Add streams support

Add multiplexed streams support. CAL has 8 DMA-engines and can capture 8
separate streams at the same time. The driver filters the streams based
on CSI-2 virtual channel number and datatype. CAL may have (depending on
the SoC) two CSI-2 RX blocks, which share the 8 DMA-engines, so the
number of capturable streams does not change even if there are two CSI-2
RX blocks.

Add 8 video device nodes, each representing a single DMA-engine, and set
the number of source pads on CSI-2 RX blocks to 8. Each video node can be
connected to any of the source pads on either of the CSI-2 RX instances
using media links. CSI-2 RX block's subdevice internal routing is used
to route the incoming CSI-2 streams to one of the 8 source pads.

Only video data streams are supported at the moment.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Reviewed-by: Jacopo Mondi <jacopo.mondi@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: ti: cal: Fix wrong goto on error path

If pm_runtime_resume_and_get() fails, we should unprepare the context,
but currently we skip that as we goto to a later line.

Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: ti: cal: Use printk's fourcc formatting

Use printk's fourcc formatting instead of a custom one.

Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: ipu-bridge: add ACPI HID for lt6911uxe bridge

Lontium lt6911uxe is a HDMI to CSI-2 bridge, without a fixed
link frequency.

Signed-off-by: Dongcheng Yan <dongcheng.yan@intel.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: ov2740: Move pm-runtime cleanup on probe-errors to proper place

When v4l2_subdev_init_finalize() fails no changes have been made to
the runtime-pm device state yet, so the probe_error_media_entity_cleanup
rollback path should not touch the runtime-pm device state.

Instead this should be done from the probe_error_v4l2_subdev_cleanup
rollback path. Note the pm_runtime_xxx() calls are put above
the v4l2_subdev_cleanup() call to have the reverse call order of probe().

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Bingbu Cao <bingbu.cao@intel.com>
Fixes: 289c25923ecd ("media: ov2740: Use sub-device active state")
Cc: stable@vger.kernel.org
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: intel/ipu6: Remove unused dev field from ipu6_isys_queue

With current code base dev is only written but never read. Remove it
as redundant.

Signed-off-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: intel/ipu6: Abstract buf ready function

Extract values needed by ipu6_queue_buf_ready() function from fw abi
structure. This will allow to reuse same buf ready code when fw abi
change.

Signed-off-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: intel/ipu6: Use timestamp value directly

Remove pointer for fw abi structure when setting frame sequence and
time, use timestamp value from the structure directly.

Signed-off-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: intel/ipu6: Make two functions static

Make function used only in one file static and remove from header file.

Signed-off-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: intel/ipu6: Add missing new line character in error message

End error message with new line.

Signed-off-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>

media: intel/ipu6: Remove unused ipu6_isys_subdev_link_validate()

Remove unused declaration.

Signed-off-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>