media: i2c: Add support for the OV8865 image sensor
The OV8865 is a 8 Mpx CMOS image sensor producing 3264x2448 at 30 fps.
Other modes (including some with sub-sampling) are available too.
It outputs 10-bit bayer CFA data through a MIPI CSI-2 interface with
up to 4 lanes supported.
Some register initialisation sequences are still needed for this driver,
as they cover registers for which no documentation is available.
This work is based on the first version of the driver submitted by
Kévin L'hôpital, which was adapted to mainline from the Allwinner BSP.
This version is a rewrite of the first version that matches the structure
of the OV5648 driver, with explicit PLL configuration, all the necessary
mode-specific fields, associatied registers and reduced static sequences.
It was tested with the Banana Pi Camera Board v3 and the Banana Pi M3.
media: i2c: Add support for the OV5648 image sensor
The OV5648 is a 5 Mpx CMOS image sensor, connected via MIPI CSI-2
in a one or two lane configuration.
Most of the features of the hardware are supported, including:
- Auto and manual exposition/gain
- Auto and manual white balance
- Horizontal and vertical flip
- Test patterns
But the following are still missing:
- Debanding, based on power source frequency;
- Exposition setting correlated to time units.
Signed-off-by: Paul Kocialkowski <paul.kocialkowski@bootlin.com> Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com> Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Jacopo Mondi [Mon, 21 Dec 2020 17:52:20 +0000 (18:52 +0100)]
media: i2c: ov5670: Fix PIXEL_RATE minimum value
The driver currently reports a single supported value for
V4L2_CID_PIXEL_RATE and initializes the control's minimum value to 0,
which is very risky, as userspace might accidentally use it as divider
when calculating the time duration of a line.
Fix this by using as minimum the only supported value when registering
the control.
Andy Shevchenko [Mon, 14 Dec 2020 15:28:32 +0000 (16:28 +0100)]
media: ipu3-cio2: Build only for x86
According to the original code in the driver it was never assumed to work
with big page sizes: unsigned short type followed by PAGE_SHIFT and
PAGE_MASK which may be different on non-x86 architectures.
Recently LKP found an issue on non-x86 architectures due to above
mentioned limitations. Since Sakari acknowledges that it's not really
useful to be able to compile this elsewhere, mark it x86 only.
Fixes: a31d19f88932 ("media: ipu3: allow building it with COMPILE_TEST on non-x86 archs") Reported-by: kernel test robot <lkp@intel.com> Suggested-by: Sakari Ailus <sakari.ailus@linux.intel.com> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com> Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Sakari Ailus [Sun, 20 Dec 2020 20:17:17 +0000 (21:17 +0100)]
media: v4l: ioctl: Use kmalloc to allocate a small chunk of memory
kvmalloc() was used to allocate the temporary memory buffer that was used
to contain both the IOCTL argument as well as a possible array argument
that could have been large. Now that the two are separated, the IOCTL
argument is known to be small in size. Use kmalloc to allocate it instead
of kvmalloc. Similarly for releasing it.
Sakari Ailus [Sat, 19 Dec 2020 22:29:58 +0000 (23:29 +0100)]
media: v4l: ioctl: Fix memory leak in video_usercopy
When an IOCTL with argument size larger than 128 that also used array
arguments were handled, two memory allocations were made but alas, only
the latter one of them was released. This happened because there was only
a single local variable to hold such a temporary allocation.
Fix this by adding separate variables to hold the pointers to the
temporary allocations.
Sakari Ailus [Thu, 17 Dec 2020 18:27:08 +0000 (19:27 +0100)]
media: uapi: Add an entity type for Image Signal Processors
Add and document a media entity type for Image Signal Processor devices.
Surprisingly we didn't have one, so add one now. More or less all ISP
drivers should use this type instead of what they currently are using (or
not using anything).
Sakari Ailus [Mon, 14 Dec 2020 21:31:21 +0000 (22:31 +0100)]
media: v4l: fwnode: v4l2_async_notifier_parse_fwnode_endpoints is deprecated
Document that v4l2_async_notifier_parse_fwnode_endpoints() is deprecated.
Its functionality has been replaced by other, better functions. Also add a
reference to an example if someone ends up wandering here.
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com> Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Jacopo Mondi [Thu, 19 Nov 2020 16:32:38 +0000 (17:32 +0100)]
media: ov5647: Implement set_fmt pad operation
Now that the driver supports more than a single mode, implement the
.set_fmt pad operation and adjust the existing .get_fmt one to report
the currently applied format.
Jacopo Mondi [Thu, 19 Nov 2020 16:32:37 +0000 (17:32 +0100)]
media: ov5647: Use SBGGR10_1X10 640x480 as default
The SBGGR10_1X10 formats support more resolutions than SBGGR8_1X8.
Make it the default sensor format and set 2x2 binned 640x480 resolution
as default sensor size as it maximizes the FOV and framerate.
Jacopo Mondi [Thu, 19 Nov 2020 16:32:36 +0000 (17:32 +0100)]
media: ov5647: Add SGGBR10_1X10 modes
Add 4 additional sensor modes in SBGGR10_1X10 format.
Add the following resolutions:
- 2592x1944 full resolution
- 1920x1080 1080p cropped
- 1296x972 2x2 binned
- 640x480 2x2 binned, 2x2 subsampled
The register lists and modes definition have been upported from the
RaspberryPi BSP at revision:
commit 581dfda6d0a62 ("media: i2c: ov5647: Advertise the correct exposure range")
Dave Stevenson [Thu, 19 Nov 2020 16:32:34 +0000 (17:32 +0100)]
media: ov5647: Add support for get_selection()
Support the get_selection() pad operation to report the device
full pixel array size, the currently applied analogue crop rectangle and
the active pixel array dimensions.
Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com> Signed-off-by: Jacopo Mondi <jacopo@jmondi.org> Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com> Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Jacopo Mondi [Thu, 19 Nov 2020 16:32:33 +0000 (17:32 +0100)]
media: ov5647: Break out format handling
Break format handling out from the main driver structure.
This commit prepares for the introduction of more sensor formats and
resolutions by instrumenting the existing operation to work on multiple
modes instead of assuming a single supported one.
Jacopo Mondi [Thu, 19 Nov 2020 16:32:30 +0000 (17:32 +0100)]
media: ov5647: Protect s_stream() with mutex
Use the driver mutex to protect s_stream() operations.
This will become more relevant once the sensor will support more formats
and set_format() could be issue concurrently to s_stream().
Jacopo Mondi [Thu, 19 Nov 2020 16:19:35 +0000 (17:19 +0100)]
media: ov5647: Program mode at s_stream(1) time
Rename __sensor_init() function to ov5647_set_mode() as the function
is a regular one and the double underscores prefix shall be removed, and
then move it to program the mode at s_stream(1) time, not at sensor power
up.
Break out from __sensor_init() the stream_off() operation call at sensor
power up to coax the lanes in LP-11 state.
Jacopo Mondi [Thu, 19 Nov 2020 16:19:34 +0000 (17:19 +0100)]
media: ov5647: Fix return value from read/write
The ov5647_read()/ov5647_write() return in case of success the number
of bytes read or written respectively. This requires callers to check
if the return value is less than zero to detect an error. Unfortunately,
in several places, callers directly return the result of a read/write
call, causing issues when the returned valued is checked to be different
from zero to detect an error.
Fix this by returning zero if i2c_master_send() and i2c_master_read()
return a positive value (the number of bytes written or read).
Jacopo Mondi [Thu, 19 Nov 2020 16:19:32 +0000 (17:19 +0100)]
media: ov5647: Fix style issues
The driver has some obvious style issues which are worth fixing before
expanding the driver capabilities.
Fix:
- Variable declaration order
- Function parameters alignment
- Multi-line comments and spurious line breaks
- Use lowercase for hexadecimal values
- > 80 cols lines
Dave Stevenson [Thu, 19 Nov 2020 16:19:30 +0000 (17:19 +0100)]
media: ov5647: Add set_fmt and get_fmt calls.
There's no way to query the subdevice for the supported
resolutions. Add set_fmt and get_fmt implementations. Since there's
only one format supported set_fmt does nothing and get returns single
format.
Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.org> Signed-off-by: Roman Kovalivskyi <roman.kovalivskyi@globallogic.com> Signed-off-by: Jacopo Mondi <jacopo@jmondi.org> Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com> Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Dave Stevenson [Thu, 19 Nov 2020 16:19:28 +0000 (17:19 +0100)]
media: ov5647: Add support for PWDN GPIO.
Add support for an optional GPIO connected to PWDN on the sensor. This
allows the use of hardware standby mode where internal device clock
and circuit activities are halted.
Please note that power is off when PWDN is high.
Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.org> Signed-off-by: Roman Kovalivskyi <roman.kovalivskyi@globallogic.com> Signed-off-by: Jacopo Mondi <jacopo@jmondi.org> Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com> Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Michael Tretter [Wed, 2 Dec 2020 13:30:40 +0000 (14:30 +0100)]
media: allegro: rename stream_id to dst_handle
The dst_handle field in the PUT_STREAM_BUFFER and ENCODE_FRAME_RESPONSE
is used to retrieve the V4L2 CAPTURE buffer. Calling it stream_id is
confusing. Call it dst_handle inspired by src_handle for the OUTPUT
buffer.
Signed-off-by: Michael Tretter <m.tretter@pengutronix.de> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Dan Carpenter [Mon, 14 Dec 2020 11:54:47 +0000 (12:54 +0100)]
media: allegro: Fix use after free on error
The "channel" is added to the "dev->channels" but then if
v4l2_m2m_ctx_init() fails then we free "channel" but it's still on the
list so it could lead to a use after free. Let's not add it to the
list until after v4l2_m2m_ctx_init() succeeds.
Fixes: cc62c74749a3 ("media: allegro: add missed checks in allegro_open()") Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Reviewed-by: Michael Tretter <m.tretter@pengutronix.de> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Hans Verkuil [Tue, 1 Dec 2020 12:44:46 +0000 (13:44 +0100)]
media: cec: add EPOLLPRI in poll() when dev is unregistered
If the CEC device was unregistered, then add EPOLLPRI to
the poll() mask. Otherwise a select() that only waits for
exceptions will not wake up. A select() that waits for
read and/or write events *will* wake up on an EPOLLERR, but
not (for some reason) if it just waits for exceptions.
Strangly the epoll functionality will wakeup on EPOLLERR if
you just wait for an exception, so in this respect select()
and epoll differ.
In the end it doesn't really matter, what matters is that
polling file handles are woken up on device unregistration.
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Hans Verkuil [Tue, 1 Dec 2020 12:44:45 +0000 (13:44 +0100)]
media: v4l2-dev: add EPOLLPRI in v4l2_poll() when dev is unregistered
If the V4L2 device was unregistered, then add EPOLLPRI to
the poll mask. Otherwise a select() that only waits for
exceptions will not wake up. A select() that waits for
read and/or write events *will* wake up on an EPOLLERR, but
not (for some reason) if it just waits for exceptions.
Strangly the epoll functionality will wakeup on EPOLLERR if
you just wait for an exception, so in this respect select()
and epoll differ.
In the end it doesn't really matter, what matters is that
polling file handles are woken up on device unregistration.
It also improves the code a bit if vdev->fops->poll is NULL:
this didn't check for device unregistration.
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Hans Verkuil [Tue, 1 Dec 2020 12:44:44 +0000 (13:44 +0100)]
media: vivid: call v4l2_event_wake_all() on disconnect
When the disconnect error injection control is set, then besides
faking unregistering the device nodes, also call v4l2_event_wake_all()
to ensure any userspace applications will wake up as per a 'normal'
unregister.
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
media: v4l2-mem2mem: always call poll_wait() on queues
do_poll()/do_select() seem to set the _qproc member of poll_table to
NULL the first time they are called on a given table, making subsequent
calls of poll_wait() on that table no-ops. This is a problem for mem2mem
which calls poll_wait() on the V4L2 queues' waitqueues only when a
queue-related event is requested, which may not necessarily be the case
during the first poll.
For instance, a stateful decoder is typically only interested in
EPOLLPRI events when it starts, and will switch to listening to both
EPOLLPRI and EPOLLIN after receiving the initial resolution change event
and configuring the CAPTURE queue. However by the time that switch
happens and v4l2_m2m_poll_for_data() is called for the first time,
poll_wait() has become a no-op and the V4L2 queues waitqueues thus
cannot be registered.
Fix this by moving the registration to v4l2_m2m_poll() and do it whether
or not one of the queue-related events are requested.
media: videobuf2: always call poll_wait() on queues
do_poll()/do_select() seem to set the _qproc member of poll_table to
NULL the first time they are called on a given table, making subsequent
calls of poll_wait() on that table no-ops. This is a problem for vb2
which calls poll_wait() on the V4L2 queues' waitqueues only when a
queue-related event is requested, which may not necessarily be the case
during the first poll.
Fix this by making the call to poll_wait() happen first thing and
unconditionally in vb2_core_poll().
media: tegra-video: Add support for x8 captures with gang ports
Tegra VI/CSI hardware don't have native 8 lane capture support.
Each CSI port has max 4 lanes only. So for x8 captures, consecutive
ports are ganged up for left half and right half captures on to each
x4 ports with buffer offsets based on source image split width to align
side-by-side.
All ports in gang are configured together during the corresponding
video device node streaming for x8 captures.
x8 capture with gang ports are supported with HDMI-to-CSI bridges
where they split 4K image into left half onto one x4 port and
right half onto second x4 port.
Implement V4L2 device notify callback to handle
V4L2_EVENT_SOURCE_CHANGE event from subdevices.
HDMI-to-CSI bridge drivers trigger V4L2_EVENT_SOURCE_CHANGE when
source DV timing changes are detected or when HDMI hotplug happens.
Runtime source parameter changes during active streaming is not
allowed and Tegra video driver calls vb2_queue_error to signal a
fatal error if a notification of this event happens during an active
streaming.
media: tegra-video: Use zero crop settings if subdev has no get_selection
Currently try format implementation doesn't check if subdevice has
get_selection ops implemented and returns -EINVAL on error from
direct v4l2_subdev_call of get_selection.
Selection API's are not mandatory for all V4L2 subdevices.
This patch fixes it by adding v4l2_subdev_has_ops check prior to
calling get_selection ops and continues with try or set format with
zero crop settings for subdevices that don't have get_selection and
returns -EINVAL only for subdevices that has get_selection ops.
Rework the context API exposed to cal-video.c to simplify it. The
configuration and enable steps are all grouped in a single
cal_ctx_start() function, and the DMA stop and IRQ disable are similarly
groupd in cal_ctx_stop(). The cal_ctx_wr_dma_addr() function is renamed
to cal_ctx_set_dma_addr() for consistency with the cal_ctx_ prefix of
the start and stop functions.
media: ti-vpe: cal: Set cal_dmaqueue.pending to NULL when no pending buffer
When a pending buffer becomes active, the cal_dmaqueue.active field is
updated, but the pending field keeps the same value until a new buffer
becomes pending. This requires handling the special case of
pending == active in different places. Simplify the code by setting the
pending field to NULL when the pending buffer becomes active. Buffers
are now simply moved from queue to pending and from pending to active.
media: ti-vpe: cal: Group all DMA queue fields in struct cal_dmaqueue
The cal_dmaqueue structure only contains the list of queued buffers.
Move the other fields that are logically related to the DMA queue
(current and next buffer points, state, wait queue and lock) from
cal_ctx to cal_dmaqueue.
Take this as an opportunity to document the fields usage and to give
them more appropriate names. The 'active' field stored the list of all
queued buffers, not the active buffers, so rename it to 'queue'. The
'cur_frm' and 'next_frm' are respectively renamed to 'active' and
'pending' to better explain their purpose. The 'dma_state' and
'dma_wait' fields are stripped of their 'dma_' prefix as they're now
part of cal_dmaqueue. Finally, 'slock' is renamed to 'lock'.
media: ti-vpe: cal: Remove DMA queue empty check at start streaming time
The vb2 queue ensures that the start streaming operation will only be
called with a minimal number of buffers queued to the driver. There's
thus no need to manually check if the DMA queue is empty. Remove the
check.
The V4L2 read/write API is inefficient and makes little sense for
devices that handle frame-based formats. Applications shouldn't use it,
drop its support from the driver.
media: ti-vpe: cal: Share buffer release code between start and stop
The cal_start_streaming() and cal_stop_streaming() functions duplicate
the same buffer release logic. split it to a separate function to share
the code.
media: ti-vpe: cal: Use spin_lock_irq() when starting or stopping stream
The cal_start_streaming() and cal_stop_streaming() functions are called
with interrupts enabled. spin_lock_irq() can thus be used instead of
spin_lock_irqsave().
media: ti-vpe: cal: Stop write DMA without disabling PPI
When stopping the stream, the driver needs to ensure that ongoing DMA
completes and that no new DMA is started. It does so using a feature of
the PPI that can be stopped on a frame boundary. The downside of this
mechanism is that the DMA can't be stopped independently of the source,
which prevents usage of multiple contexts for the same source (to handle
CSI-2 virtual channels or data types).
Rework the stream stop mechanism to stop the write DMA without disabling
the PPI first. The new mechanism relies on the combination of a state
machine in the driver and shadowing of the CAL_WR_DMA_CTRL_j.MODE field
in the hardware. When a stop is requested, the DMA start interrupt
handler will request the hardware to stop at the end of the current
frame by disabling the write DMA context in the shadowed register, and
flag that a stop is in progress. The next DMA end interrupt will then
report that the stop is complete.
This makes it possible to stop the PPI after stopping the DMA, and fold
the cal_camerarx_ppi_disable() call into cal_camerarx_stop().
media: ti-vpe: cal: Fold PPI enable in CAMERARX .s_stream()
To further decouple the context and CAMERARX components, move the call
to cal_camerarx_ppi_enable() from cal_start_streaming() to the CAMERARX
.s_stream() operation. The DMA destination address has to be set before
starting the CAMERARX, which is desirable anyway.
cal_camerarx_ppi_disable() will be addressed separately.
Interrupts are enabled and disabled by the cal_camerarx_enable_irqs()
and cal_camerarx_disable_irqs(). Despite their name, they deal with all
interrupts, not just the CAMERARX interrupts, and they hardcode the
assumption that the context index is identical to the CAMERARX index.
Split the context-related interrupt management to two new functions,
cal_ctx_enable_irqs() and cal_ctx_disable_irqs(), called from the
cal_start_streaming() and cal_stop_streaming() functions. The explicit
calls to cal_camerarx_enable_irqs() and cal_camerarx_disable_irqs() are
folded with the CAMERARX .s_stream() operation to simplify the CAMERARX
API.
Enabling the OCPO error interrupt is moved to the PM runtime resume
operation, as it's global to the device, not related to a CAMERARX or
context. The VC IRQ enable and disable are removed as they're not used,
the parent interrupt bit (CAL_HL_IRQ_VC_MASK) never being set.
media: ti-vpe: cal: Rename struct cal_fmt to cal_format_info
The cal_fmt structure stores information about a format. Its name is
ambiguous, as it could be understood as storing a format. Furthermore,
instances of the structure are called 'fmt' through the code, which
leads to confusion with the 'format' variables. Rename the structure to
cal_format_info, and the corresponding 'fmt' variables to 'fmtinfo'.
media: ti-vpe: cal: Don't pass format to cal_ctx_wr_dma_config()
The cal_ctx_wr_dma_config() function has access to the context, there's
no need to give it format-related values retrieved from the context.
Access the values internally, and reword internal comments. The comment
regarding the CAL_WR_DMA_OFST register not being well understood is
dropped, as the datasheet explicitly documents it as "Offset between two
consecutive line starts".
media: ti-vpe: cal: Use CAMERARX subdev s_stream op in video device code
Replace calls to cal_camerarx_start() and cal_camerarx_stop() with usage
of the .s_stream() subdev operation. This requires calling the
.set_fmt() operation as the CAMERARX now relies on the format
information set through the subdev API instead of receiving it in the
cal_camerarx_start() function.
This change prepare for exposing the CAMERARX subdev operations to
userspace by using the same API within the kernel.
media: ti-vpe: cal: Implement subdev ops for CAMERARX
Implement subdev operations for the CAMERARX. They will be used to
replace calls to custom CAMERARX functions in the V4L2 video device
code, and will be exposed to userspace.
media: ti-vpe: cal: Iterate over correct number of CAMERARX instances
When performing operations on all CAMERARX instances, code usually
iterates over all cal->phy[] entries and skips the NULL pointers. The
number of available CAMERARX instances is however available through
cal->data->num_csi2_phy. Use it instead.
media: ti-vpe: cal: Rename MAX_(WIDTH|HEIGHT)_* macros with CAL_ prefix
The MAX_WIDTH_BYTES and MAX_HEIGHT_LINES macros have a generic name that
is prone to namespace clashes. Rename them with a CAL_ prefix. While at
it, expand their documentation and add CAL_MIN_(WIDTH|HEIGHT)_* macros
that will be used to implement CAMERARX subdev operations.
media: ti-vpe: cal: Move format handling to cal.c and expose helpers
The cal_formats array contain the description of all formats supported
by the hardware. It's currently used by the V4L2 video device operations
only, but will be needed by the CAMERARX subdev code too. Move it from
cal-video.c to cal.c and add helper functions to access it.
The struct cal_ctx m_fmt field stores the media bus format for the
context input. Only the format 'field' field is used, store it in the
video format instead, and drop m_fmt.
Create and register V4L2 subdevs for the CAMERARX instances, and link
them in the media graph to the sensors and video nodes. The subdev API
is not exposed to userspace at this point, and no subdev operation is
implemented, but the media controller graph is visible to applications.
Linus Torvalds [Sun, 27 Dec 2020 18:56:33 +0000 (10:56 -0800)]
proc mountinfo: make splice available again
Since commit 36e2c7421f02 ("fs: don't allow splice read/write without
explicit ops") we've required that file operation structures explicitly
enable splice support, rather than falling back to the default handlers.
Most /proc files use the indirect 'struct proc_ops' to describe their
file operations, and were fixed up to support splice earlier in commits 40be821d627c..b24c30c67863, but the mountinfo files interact with the
VFS directly using their own 'struct file_operations' and got missed as
a result.
This adds the necessary support for splice to work for /proc/*/mountinfo
and friends.
Reported-by: Joan Bruguera Micó <joanbrugueram@gmail.com> Reported-by: Jussi Kivilinna <jussi.kivilinna@iki.fi> Link: https://bugzilla.kernel.org/show_bug.cgi?id=209971 Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Christoph Hellwig <hch@lst.de> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>