get compile-test failed report from lkp@intel.com after the driver
submitted.
"ld.lld: error: undefined symbol: v4l2_fh_init"
so, make CONFIG_STM32_DMA2D depends on CONFIG_VIDEO_DEV and
CONFIG_VIDEO_V4L2 to fix this compile error.
Fixes: bff6e3e2f4c9 ("media: stm32-dma2d: STM32 DMA2D driver")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Dillon Min <dillon.minfei@gmail.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
If res-chg, VE_INTERRUPT_MODE_DETECT_WD irq will be raised. But
v4l2_input_status won't be updated to no-signal immediately until
aspeed_video_get_resolution() in aspeed_video_resolution_work().
During the period of time, aspeed_video_start_frame() could be called
because it doesn't know signal becomes unstable now. If it goes with
aspeed_video_init_regs() of aspeed_video_irq_res_change()
simultaneously, it will mess up hw state.
To fix this problem, v4l2_input_status is updated to no-signal
immediately for VE_INTERRUPT_MODE_DETECT_WD irq.
Fixes: d2b4387f3b ("media: platform: Add Aspeed Video Engine driver")
Signed-off-by: Jammy Huang <jammy_huang@aspeedtech.com>
Acked-by: Paul Menzel <pmenzel@molgen.mpg.de>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
The Documentation/driver-api/media/drivers/rkisp1.rst file
wasn't referenced in the index.rst file, so it was never included.
Add it.
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Width and height need to 64 bytes aligned when setting the format.
Need to make sure all is 64 bytes align when use width and height to
calculate buffer size.
Signed-off-by: Yunfei Dong <yunfei.dong@mediatek.com>
Acked-by: Nicolas Dufresne <nicolas.dufresne@collabora.com>
Tested-by: Steve Cho <stevecho@chromium.org>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
If V4L2_CAP_READWRITE is not set, then readbuffers must be set to 0,
otherwise v4l2-compliance will complain.
A note on the Fixes tag below: this patch does not really fix that commit,
but it can be applied from that commit onwards. For older code there is no
guarantee that device_caps is set, so even though this patch would apply,
it will not work reliably.
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Fixes: 049e684f2d (media: v4l2-dev: fix WARN_ON(!vdev->device_caps))
Cc: <stable@vger.kernel.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Improve the documentation for CEC_RECEIVE/TRANSMIT w.r.t. the
sequence and tx/rx_status fields.
Also remove a duplicate tx_status description.
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Fix example code which has missing or double backslash typo.
Signed-off-by: Kwang Son <dev.kwang.son@gmail.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
In the em28xx_init_rev, if em28xx_audio_setup fails, this function fails
to deallocate the media_dev allocated in the em28xx_media_device_init.
Fix this by adding em28xx_unregister_media_device to free media_dev.
BTW, this patch is tested in my local syzkaller instance, and it can
prevent the memory leak from occurring again.
CC: Pavel Skripkin <paskripkin@gmail.com>
Fixes: 37ecc7b127 ("[media] em28xx: add media controller support")
Signed-off-by: Dongliang Mu <mudongliangabcd@gmail.com>
Reported-by: syzkaller <syzkaller@googlegroups.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Convert ADV748X analog video decoder documentation to json-schema.
While converting the bindings extend it to enforce that all port@n nodes
shall be encapsulated inside a ports node. This change does not have an
effect on drivers parsing the ports@n nodes.
Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
'mfc_dev->mem_bitmap' is a bitmap. So use 'bitmap_zalloc()' to simplify
code and improve the semantic.
Also change the corresponding 'kfree()' into 'bitmap_free()' to keep
consistency.
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
The cleanup in tw5864_finidev() neglected to disable the PCI device
after enabling it in tw5864_initdev().
Call pci_disable_device() after releasing the associated resources.
Signed-off-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Some resources are allocated with 'pci_request_regions()', so use
'pci_release_regions()' to free them, instead of a verbose
'release_mem_region()'.
There is no point in calling 'devm_kfree()'. The corresponding resource is
managed, so it will be fried automatically.
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Repalce kthread_create/wake_up_process() with kthread_run()
to simplify the code.
Signed-off-by: Cai Huoqing <caihuoqing@baidu.com>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
aspeed_video_get_resolution() will try to do res-detect again if the
timing got in last try is invalid. But it will always time out because
VE_SEQ_CTRL_TRIG_MODE_DET is only cleared after 1st mode-detect.
To fix the problem, just clear VE_SEQ_CTRL_TRIG_MODE_DET before setting
it in aspeed_video_enable_mode_detect().
Fixes: d2b4387f3b ("media: platform: Add Aspeed Video Engine driver")
Signed-off-by: Jammy Huang <jammy_huang@aspeedtech.com>
Acked-by: Paul Menzel <pmenzel@molgen.mpg.de>
Reviewed-by: Joel Stanley <joel@jms.id.au>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
The open() fops support two types of devices: "acc" and normal
ones. the acc works on a different way, using a different pipe
struct. Not sure yet if it would make sense to setup a run_mode
there. Also, As default_run_mode exists only on normal modes,
we can simplify the logic to check if the device is in normal
mode.
That solves this warning:
../drivers/staging/media/atomisp/pci/atomisp_fops.c:904 atomisp_open() warn: variable dereferenced before check 'asd' (see line 807)
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
The asd->isp was referenced before checking if asd is not
NULL.
This fixes this warning:
../drivers/staging/media/atomisp/pci/atomisp_cmd.c:5548 atomisp_set_fmt_to_snr() warn: variable dereferenced before check 'asd' (see line 5540)
While here, avoid getting the pipe pointer twice.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
The "power" pointer is not initialized on the else path and that would
lead to an Oops.
Link: https://lore.kernel.org/linux-media/20211012082150.GA31086@kili
Fixes: c30f4cb2d4 ("media: atomisp: Refactor PMIC detection to a separate function")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
The vts value should be set before being checked, as otherwise a
warning will arise:
drivers/staging/media/atomisp/i2c/atomisp-ov2680.c: In function 'ov2680_set_fmt':
drivers/staging/media/atomisp/i2c/atomisp-ov2680.c:873:33: warning: 'vts' may be used uninitialized
[-Wmaybe-uninitialized]
873 | if (dev->exposure > vts - OV2680_INTEGRATION_TIME_MARGIN)
Reported-by: Hans Verkuil <hverkuil@xs4all.nl>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Fixes: 62b984359b6f ("media: atomisp-ov2680: Fix ov2680_set_fmt() messing up high exposure settings")
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
As the settings are only applied when the device is powered on,
it should return 0 when the device is not powered.
Not doing that causes a warning:
drivers/staging/media/atomisp/i2c/atomisp-ov2680.c: In function 'ov2680_ioctl':
drivers/staging/media/atomisp/i2c/atomisp-ov2680.c:390:16: warning: 'ret' may be used uninitialized in this
function [-Wmaybe-uninitialized]
390 | return ov2680_set_exposure(sd, coarse_itg, analog_gain, digital_gain);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/staging/media/atomisp/i2c/atomisp-ov2680.c:359:13: note: 'ret' was declared here
359 | int ret;
| ^~~
Reported-by: Hans Verkuil <hverkuil@xs4all.nl>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Fixes: 6b5b60687ada ("media: atomisp-ov2680: Save/restore exposure and gain over sensor power-down")
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
For exposure settings > (lines_per_frame - vts_margin) the VTS register
needs to be programmed to (exposure + vts_margin) rather then being
set to lines_per_frame.
The res->regs register array was clobbering this higher setting causing
high exposure settings to not work. Fix this by letting ov2680_set_fmt()
calculate the vts value, instead of hardcoding it.
This is the last in a series of fixes which fixes exposure and gain
settings not working, with this everything works, so drop the comment
that it does not work.
Link: https://lore.kernel.org/linux-media/20211107171549.267583-12-hdegoede@redhat.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Now that we restore the default or last user set exposure setting on
power_up() there is no need for the registers written by ov2680_set_fmt()
to write to the exposure register.
Not doing so fixes the exposure always being reset to the value from
the res->regs array after a set_fmt().
Link: https://lore.kernel.org/linux-media/20211107171549.267583-11-hdegoede@redhat.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
The shift << 16 of the value in the code path for 16 bit values is
bogus, put_unaligned_be16() takes the lower 16 bits which will not
always be 0.
This was causing __ov2680_set_exposure() to always set the
OV2680_AGC and OV2680_TIMING_VTS registers to 0.
Link: https://lore.kernel.org/linux-media/20211107171549.267583-10-hdegoede@redhat.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Switch to ov2680_read_reg() to read all 24 bits in one go;
and the exposure value sits in bits 4-19 of the 24 bit exposure
register, so we need to shift the read value by 4 to report the
correct value.
Link: https://lore.kernel.org/linux-media/20211107171549.267583-9-hdegoede@redhat.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Some ov2680 registers like exposure are 24 bit,
ov2680_read_reg() already mostly supports this, we just
need to change the return type from u16 to u32.
Link: https://lore.kernel.org/linux-media/20211107171549.267583-8-hdegoede@redhat.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Save/restore exposure and gain over sensor power-down and don't write them
to the sensor when ov2680_set_exposure() is called while the sensor is off.
Link: https://lore.kernel.org/linux-media/20211107171549.267583-7-hdegoede@redhat.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Move ov2680_init_registers() call to power_up(), so that we also
init the registers on code-paths which do not call ov2680_s_power()
like running camorama.
Link: https://lore.kernel.org/linux-media/20211107171549.267583-6-hdegoede@redhat.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
The ov2680_res and N_RES global variables are just hardcoded as aliases
for ov2680_res_preview and N_RES_PREVIEW, remove them.
Link: https://lore.kernel.org/linux-media/20211107171549.267583-5-hdegoede@redhat.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
ov2680_s_power() is the only caller of ov2680_init(), push the input_lock
taking from ov2680_init() up into ov2680_s_power(), this way the new
power_on bool is also protected by it.
Link: https://lore.kernel.org/linux-media/20211107171549.267583-4-hdegoede@redhat.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Add a power_on bool to track if the power is on, and make
power_up() a no-op if the power is already on.
This also removes a power_down() call from ov2680_s_config() since
that is a no-op now, this is ok because s_config() is only called
once on probe and the sensor is off at boot.
Besides avoiding to the work in power_up() multiple times this patch
is also a preparation for switching to the clk and regulator frameworks
which keep an enable count, so there we must call enable() and
disable() only once per power-cycle.
Link: https://lore.kernel.org/linux-media/20211107171549.267583-3-hdegoede@redhat.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Remove a couple of variables which where either completely unused,
or only ever got a value assigned to them but were never read.
Link: https://lore.kernel.org/linux-media/20211107171549.267583-2-hdegoede@redhat.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
The atomisp currenyl registers 5 pairs of devices each one
for one different run_mode, plus one for "ACC". The only
one that behaves like a normal V4L2 device is the preview
one. The others are doing weird things, and perhaps are
using some proprietary extensions to the API.
Change the device order to start with the preview one,
e. g:
/dev/video0: ATOMISP ISP PREVIEW output
/dev/video1: ATOMISP ISP CAPTURE output
/dev/video2: ATOMISP ISP VIEWFINDER output
/dev/video3: ATOMISP ISP VIDEO output
/dev/video4: ATOMISP ACC
This way, a normal V4L2 application will get the right
device, as the first one will be the one they should use.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
The atomisp driver originally used the s_parm command to
initialize the run_mode type to the driver. So, before start
setting up the streaming, s_parm should be called.
So, even having 5 "normal" video devices, one meant to be used
for each type, the run_mode was actually selected when
s_parm is called.
Without setting the run mode, applications that don't call
VIDIOC_SET_PARM with a custom atomisp parameters won't work, as
the pipeline won't be set:
atomisp-isp2 0000:00:03.0: can't create streams
atomisp-isp2 0000:00:03.0: __get_frame_info 1600x1200 (padded to 0) returned -22
However, commit 8a7c5594c0 ("media: v4l2-ioctl: clear fields in s_parm")
broke support for it, with a good reason, as drivers shoudn't be
extending the API for their own purposes.
So, as an step to allow generic apps to use this driver, put
the device's run_mode in preview after open.
After this patch, using v4l2grab starts to work on preview
mode (/dev/video2):
$ v4l2grab -f YUYV -x 1600 -y 1200 -d /dev/video2 -n 1 -u
$ feh out000.pnm
So, let's just setup the default run_mode that each video devnode
should assume, setting it at open() time.
Reported-by: Tsuchiya Yuto <kitakar@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
All ISP2401 devices use the new input system. So, get rid
of the remaining definitions, replacing them by runtime
checks for BYT/CHT when applicable.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Now that the pipeline config functions can return errors, change
ia_css_dma_configure_from_info() and callers in order for them
to return errors at pipelines instead of using assert().
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Those functions can internally break, but, as they don't return
errors, internally there are some assert() calls, which is bad,
as it hangs the driver.
So, add return codes there, in preparation for removing such
assert() calls.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
The auto-generated code inside ia_css_isp_configs() is more
complex than it should be. Also, it doesn't return any errors.
However, the functions called by it can mis-configure the pipelines,
but, as there's no way to return errors, it internally calls the
assert() macro.
So, add a return parameter to each routine there, in order to
prepare the code to be more robust.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
The contents of ia_css_isp_params.c is almost identical for
2400 and 2401. The only difference is that, on 2400, there
are some duplicated assignments. So, drop it, unifying this
file.
While here, simplify the Makefile's logic by dropping an
unused define.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Instead of reinventing the wheel, use v4l2_find_nearest_size()
in order to get the closest resolution.
This should address a bug where the wrong resolution was
selected.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Instead of reinventing the wheel, use v4l2_find_nearest_size()
in order to get the closest resolution.
This should address a bug where the wrong resolution was
selected.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Instead of reinventing the wheel, use v4l2_find_nearest_size()
in order to get the closest resolution.
This should address a bug where the wrong resolution was
selected.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Instead of reinventing the wheel, use v4l2_find_nearest_size()
in order to get the closest resolution.
This should address a bug where the wrong resolution was
selected.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Dillon Min