Boris Brezillon [Tue, 4 Jun 2019 07:06:25 +0000 (03:06 -0400)]
media: v4l2: Get rid of ->vidioc_enum_fmt_vid_{cap, out}_mplane
Support for multiplanar and singleplanar formats is mutually exclusive,
at least in practice. In our attempt to unify support for support for
mplane and !mplane in v4l, let's get rid of the
->vidioc_enum_fmt_{vid,out}_cap_mplane() hooks and call
->vidioc_enum_fmt_{vid,out}_cap() instead.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com> Reviewed-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
[hverkuil-cisco@xs4all.nl: fix typos: pirv -> priv and prov -> priv] Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Maxime Ripard [Mon, 3 Jun 2019 11:32:50 +0000 (07:32 -0400)]
media: dt-bindings: Fix vendor-prefixes YAML
Commit 8df39e16877f ("media: dt-bindings: media: Add vendor prefix for
allegro") introduced a new devicetree binding vendors, however with an
improper syntax making the resulting YAML impossible to parse. Let's fix
this.
Fixes: 8df39e16877f ("media: dt-bindings: media: Add vendor prefix for allegro") Signed-off-by: Maxime Ripard <maxime.ripard@bootlin.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Neil Armstrong [Fri, 31 May 2019 10:33:15 +0000 (06:33 -0400)]
media: platform: ao-cec-g12a: disable regmap fast_io for cec bus regmap
With fast_io enabled, spinlock_irq is used for read/write operations,
thus leading to :
BUG: sleeping function called from invalid context at [snip]/ao-cec-g12a.c:379
in_atomic(): 1, irqs_disabled(): 128, pid: 1451, name: irq/14-ff800280
[snip]
Call trace:
dump_backtrace+0x0/0x180
show_stack+0x14/0x1c
dump_stack+0xa8/0xe0
___might_sleep+0xf4/0x104
__might_sleep+0x4c/0x80
meson_ao_cec_g12a_read+0x7c/0x164
regmap_read+0x16c/0x1b0
meson_ao_cec_g12a_irq_thread+0xcc/0x200
irq_thread_fn+0x2c/0x60
irq_thread+0x14c/0x1fc
kthread+0x11c/0x12c
ret_from_fork+0x10/0x18
Simply remove fast_io to use mutexes instead.
Fixes: b7778c46683c ("media: platform: meson: Add Amlogic Meson G12A AO CEC Controller driver") Signed-off-by: Neil Armstrong <narmstrong@baylibre.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Colin Ian King [Thu, 30 May 2019 18:59:24 +0000 (14:59 -0400)]
media: cx23885: remove redundant assignment to err
The variable err is assigned with a value that is never read and
it is re-assigned a new value later on. The assignment is redundant
and can be removed.
Addresses-Coverity: ("Unused value")
Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Hans Verkuil [Thu, 30 May 2019 06:45:48 +0000 (02:45 -0400)]
media: dvb-usb/cxusb-analog.c: fix coccinelle warning, use ktime.h
This patch fixes a coccinelle warning and includes ktime.h instead of
timekeeping.h. The first includes the latter, but the latter doesn't
exist before 3.17, causing problems for our compat build. It's easier
to just use ktime.h instead.
coccinelle warnings: (new ones prefixed by >>)
>> drivers/media/usb/dvb-usb/cxusb-analog.c:1498:41-42: WARNING: Use ARRAY_SIZE
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Reported-by: kbuild test robot <lkp@intel.com> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Kefeng Wang [Thu, 30 May 2019 07:25:49 +0000 (03:25 -0400)]
media: wl128x: Fix some error handling in fm_v4l2_init_video_device()
X-Originating-IP: [10.175.113.25]
X-CFilter-Loop: Reflected
The fm_v4l2_init_video_device() forget to unregister v4l2/video device
in the error path, it could lead to UAF issue, eg,
BUG: KASAN: use-after-free in atomic64_read include/asm-generic/atomic-instrumented.h:836 [inline]
BUG: KASAN: use-after-free in atomic_long_read include/asm-generic/atomic-long.h:28 [inline]
BUG: KASAN: use-after-free in __mutex_unlock_slowpath+0x92/0x690 kernel/locking/mutex.c:1206
Read of size 8 at addr ffff8881e84a7c70 by task v4l_id/3659
Arvind Yadav [Tue, 15 Aug 2017 11:23:42 +0000 (07:23 -0400)]
media: omap3isp: constify platform_device_id
platform_device_id are not supposed to change at runtime. All functions
working with platform_device_id provided by <linux/platform_device.h>
work with const platform_device_id. So mark the non-const structs as
const.
media: staging: media: davinci_vpfe: use __func__ for function names
Checkpatch reported warnings for use of embedded function names.
Use __func__ instead of embedded function names.
[mchehab+samsung@kernel.org: rebased on the top of upstream] Signed-off-by: Diwakar Sharma <sharmalxmail@gmail.com> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Sakari Ailus [Thu, 13 Jul 2017 15:23:44 +0000 (11:23 -0400)]
media: omap3isp: Don't rely on devm for memory resource management
devm functions are fine for managing resources that are directly related
to the device at hand and that have no other dependencies. However, a
process holding a file handle to a device created by a driver for a device
may result in the file handle left behind after the device is long gone.
This will result in accessing released (and potentially reallocated)
memory.
Instead, manage the memory resources in the driver. Releasing the
resources can be later on bound to e.g. by releasing a reference.
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com> Acked-by: Hans Verkuil <hans.verkuil@cisco.com> Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Arushi Singhal [Wed, 29 Mar 2017 15:13:20 +0000 (11:13 -0400)]
media: staging: media: davinci_vpfe: Replace a bit shift
This patch replaces bit shifting on 1 with the BIT(x) macro.
This was done with coccinelle:
@@
constant c;
@@
-1 << c
+BIT(c)
[mchehab+samsung@kernel.org: rebase on the top of upstream] Signed-off-by: Arushi Singhal <arushisinghal19971997@gmail.com> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Shailendra Verma [Fri, 25 Nov 2016 04:57:34 +0000 (23:57 -0500)]
media: staging: media: davinci_vpfe: - Fix for memory leak if decoder initialization fails.
Fix to avoid possible memory leak if the decoder initialization
got failed.Free the allocated memory for file handle object
before return in case decoder initialization fails.
Andrew F. Davis [Fri, 29 Mar 2019 17:34:27 +0000 (13:34 -0400)]
media: videobuf-dma-contig: Use size of buffer in mmap not vma size
The size of the vma can be larger than the size of the backing buffer.
Use the buffer size over vma size to prevent exposing extra memory
to userspace.
Signed-off-by: Andrew F. Davis <afd@ti.com> Reviewed-by: Sakari Ailus <sakari.ailus@linux.intel.com> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Young Xiao [Tue, 28 May 2019 12:38:13 +0000 (08:38 -0400)]
media: cx231xx-dvb: fix memory leak in dvb_fini()
In dvb_init(), dev->dvb is allocated by kzalloc.
Therefore, it must be freed being set to NULL.
Signed-off-by: Young Xiao <92siuyang@gmail.com> Signed-off-by: Sean Young <sean@mess.org> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Chen-Yu Tsai [Mon, 20 May 2019 15:06:36 +0000 (11:06 -0400)]
media: sun6i: Support A83T variant
The A83T SoC has a camera sensor interface (known as CSI in Allwinner
lingo), which is similar to the one found on the A64 and H3. The only
difference seems to be that support of MIPI CSI through a connected
MIPI CSI-2 bridge.
Add support for this variant.
Signed-off-by: Chen-Yu Tsai <wens@csie.org> Acked-by: Maxime Ripard <maxime.ripard@bootlin.com> Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Chen-Yu Tsai [Mon, 20 May 2019 15:06:35 +0000 (11:06 -0400)]
media: dt-bindings: media: sun6i-csi: Add compatible string for A83T variant
The A83T SoC has a camera sensor interface (known as CSI in Allwinner
lingo), which is similar to the one found on the A64 and H3. The only
difference seems to be that support of MIPI CSI through a connected
MIPI CSI-2 bridge.
Add a compatible string for this variant.
Signed-off-by: Chen-Yu Tsai <wens@csie.org> Acked-by: Maxime Ripard <maxime.ripard@bootlin.com> Reviewed-by: Rob Herring <robh@kernel.org> Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Sakari Ailus [Mon, 18 Feb 2019 16:29:13 +0000 (11:29 -0500)]
media: staging: imgu: Address a compiler warning on alignment
Address a compiler warnings on alignment of struct ipu3_uapi_awb_fr_config_s
by adding __attribute__((aligned(32))) to a struct member of that type as
well.
Ezequiel Garcia [Tue, 28 May 2019 17:02:27 +0000 (13:02 -0400)]
media: rockchip/vpu: Add decoder boilerplate
This commit adds the needed boilerplate code to support the VPU
in decoding operation. Two v4l2 interfaces are exposed, one for
encoding and one for decoding, but a single m2m device is shared
by them, so jobs are properly serialized.
Signed-off-by: Ezequiel Garcia <ezequiel@collabora.com> Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Boris Brezillon [Tue, 28 May 2019 17:02:26 +0000 (13:02 -0400)]
media: rockchip/vpu: Prepare things to support decoders
The code in rockchip_vpu_v4l2 was hardcoded for encoder support.
Modify it more generic to support the decoder case so that we can
re-use the same vb2/v4l2 ops for both devices.
Boris Brezillon [Tue, 28 May 2019 17:02:25 +0000 (13:02 -0400)]
media: rockchip/vpu: Provide a helper to reset both src and dst formats
When initializing a context, the core wants to reset both src and dst
formats. Right now the order doesn't matter, but if we want to have a
valid default width/height on the non-coded/raw format side (src in
case of encoders, dst in case of decoders), we need to reset those
formats in the right order: first the coded-format side, then the other,
such that width and height on the raw format side can be taken from
the coded format.
Let's provide a helper that will reset both formats and make sure this
is done in the right order.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Boris Brezillon [Tue, 28 May 2019 17:02:24 +0000 (13:02 -0400)]
media: rockchip/vpu: Move encoder logic to a common place
The V4L2/VB2 implementation for the encoder and decoder logic are very
similar, so let's rename rockchip_vpu_enc.c file into
rockchip_vpu_v4l2.c and remove the _enc_ part in objects/functions
exposed in rockchip_vpu_v4l2.h. We also rename the enc_queue_init()
function (in rockchip_vpu_drv.c) queue_init() since it will be used
to initialize both type of queues.
The implementation itself will be patched to support the decoding case
when decoder support is added.
Suggested-by: Ezequiel Garcia <ezequiel@collabora.com> Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Boris Brezillon [Tue, 28 May 2019 17:02:23 +0000 (13:02 -0400)]
media: rockchip/vpu: Rename rockchip_vpu_common.h into rockchip_vpu_v4l2.h
We're about to add prototypes for the vb2/v4l2 helpers shared by the
encoder/decoder logic in this file, so let's pick a name that reflects
that (rockchip_vpu_common.h was a bit to generic).
Suggested-by: Ezequiel Garcia <ezequiel@collabora.com> Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Ezequiel Garcia [Tue, 28 May 2019 17:02:22 +0000 (13:02 -0400)]
media: rockchip/vpu: Support the Request API
Introduce support for the Request API. Although the JPEG encoder
does not mandate using the Request API, it's perfectly possible to
use it, if the application wants to.
Signed-off-by: Ezequiel Garcia <ezequiel@collabora.com> Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Ezequiel Garcia [Tue, 28 May 2019 17:02:21 +0000 (13:02 -0400)]
media: rockchip/vpu: Open-code media controller register
In preparation to support decoders, using a single memory-to-memory
device, we need to roll our own media controller entities registration.
To do that, we define a rockchip_vpu_func object that embeds the
video_device object plus all the elements that are needed to attach this
vdev to the media device.
Signed-off-by: Ezequiel Garcia <ezequiel@collabora.com> Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Boris Brezillon [Tue, 28 May 2019 17:02:18 +0000 (13:02 -0400)]
media: v4l2-common: Add an helper to apply frmsize constraints
The rockchip VPU driver is open-coding this logic which seems pretty
generic. Let's provide an helper to apply the min/max and alignment
constraints on width/height.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Fix it by checking the return of proc_create_single() before
calling remove_proc_entry().
Signed-off-by: Kefeng Wang <wangkefeng.wang@huawei.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
[hverkuil-cisco@xs4all.nl: use 0444 instead of S_IRUGO]
[hverkuil-cisco@xs4all.nl: use pr_info instead of KERN_INFO] Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Wen Yang [Mon, 6 May 2019 07:05:17 +0000 (03:05 -0400)]
media: mtk-vcodec: fix leaked of_node references
The call to of_find_device_by_node returns a node pointer with refcount
incremented thus it must be explicitly decremented after the last
usage.
Detected by coccinelle with the following warnings:
drivers/media/platform/mtk-vcodec/mtk_vcodec_dec_pm.c:60:3-9: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 38, but without a corresponding object release within this function.
drivers/media/platform/mtk-vcodec/mtk_vcodec_dec_pm.c:63:2-8: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 38, but without a corresponding object release within this function.
drivers/media/platform/mtk-vcodec/mtk_vcodec_dec_pm.c:72:3-9: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 38, but without a corresponding object release within this function.
Signed-off-by: Wen Yang <wen.yang99@zte.com.cn> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Wen Yang [Mon, 6 May 2019 07:05:16 +0000 (03:05 -0400)]
media: mtk-vpu: fix leaked of_node references
The call to of_parse_phandle returns a node pointer with refcount
incremented thus it must be explicitly decremented after the last
usage.
Detected by coccinelle with the following warnings:
drivers/media/platform/mtk-vpu/mtk_vpu.c:477:1-7: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 464, but without a corresponding object release within this function.
Signed-off-by: Wen Yang <wen.yang99@zte.com.cn> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Sparse complains about two issues when building with i386
and COMPILE_TEST:
drivers/staging/media/allegro-dvt/allegro-core.c:1849:36: warning: constant 0xffffffff00000000UL is so big it is unsigned long long
drivers/staging/media/allegro-dvt/allegro-core.c:865:24: error: incompatible types in comparison expression (different type sizes)
Hans Verkuil [Wed, 29 May 2019 12:46:32 +0000 (08:46 -0400)]
media: allegro-dvt: fix build failure
nal-h264.h is a local header, not a global one. Use "" instead of <>.
Fixed this compile error:
CC drivers/staging/media/allegro-dvt/nal-h264.o
drivers/staging/media/allegro-dvt/nal-h264.c:24:10: fatal error: nal-h264.h: No such file or directory
#include <nal-h264.h>
^~~~~~~~~~~~
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Reviewed-by: Michael Tretter <m.tretter@pengutronix.de> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Hans Verkuil [Thu, 4 Apr 2019 13:15:00 +0000 (09:15 -0400)]
media: videobuf2-vmalloc: get_userptr: buffers are always writable
In vb2_vmalloc_get_userptr() the framevector is created with the
'write' argument set to false when vb2_create_framevec() is called
for OUTPUT buffers. So the pages are marked as read-only.
However, userspace will write to these buffers since it will fill
in the data to output. Since get_userptr is only called if the userptr
of the queued buffer has changed since the last time that same buffer
was queued, this will fail when the buffer contents is updated and the
buffer is queued again.
E.g., userspace fills buffer 1 with the output video and queues it.
The first time get_userptr is called and the pages are grabbed and
pinned in memory and marked read-only. The second time buffer 1 is
filled with different video data and queued again. Since the userptr
hasn't changed the get_userptr() callback isn't called again. Since
the pages were marked as read-only the new contents isn't updated.
Just always call vb2_create_framevec() with FOLL_WRITE to always
allow writing to the buffers.
Using USERPTR streaming with OUTPUT devices is almost never done. And
when it is done it is via v4l2-compliance and a driver like vim2m. But
since v4l2-compliance doesn't actually inspect the capture buffer and
compare it to the original output buffer, this issue was never noticed.
But the vicodec driver actually needs to parse the bitstream in the
OUTPUT buffers and any errors there will be immediately noticed. So
this time v4l2-compliance failed the USERPTR streaming test.
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Michael Tretter [Tue, 28 May 2019 17:11:20 +0000 (13:11 -0400)]
media: allegro: add SPS/PPS nal unit writer
The allegro hardware encoder does not write SPS/PPS nal units into the
encoded video stream. Therefore, we need to write the units in software.
The implementation follows Rec. ITU-T H.264 (04/2017) to allow to
convert between a C struct and the RBSP representation of the SPS and
PPS nal units.
The allegro driver writes the nal units into the v4l2 capture buffer in
front of the actual video data which is written at an offset by the IP
core. The remaining gap is filled with a filler nal unit.
[hverkuil-cisco@xs4all.nl: added missing @dev description for nal_h264_read_sps()]
[mchehab+samsung@kernel.org: removed uneeded "-ccflags-y $(src)" from Makefile]
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+samsung@kernel.org>
Michael Tretter [Tue, 28 May 2019 17:11:19 +0000 (13:11 -0400)]
media: allegro: add Allegro DVT video IP core driver
Add a V4L2 mem-to-mem driver for Allegro DVT video IP cores as found in
the EV family of the Xilinx ZynqMP SoC. The Zynq UltraScale+ Device
Technical Reference Manual uses the term VCU (Video Codec Unit) for the
encoder, decoder and system integration block.
This driver takes care of interacting with the MicroBlaze MCU that
controls the actual IP cores. The IP cores and MCU are integrated in the
FPGA. The xlnx_vcu driver is responsible for configuring the clocks and
providing information about the codec configuration.
The driver currently only supports the H.264 video encoder.
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+samsung@kernel.org>
Michael Tretter [Tue, 28 May 2019 17:11:18 +0000 (13:11 -0400)]
media: dt-bindings: media: Add vendor prefix for allegro
Add vendor prefix for Allegro DVT, a provider of H.264/AVC, H.265/HEVC,
AVS2, VP9 and AV1 compliance test suites and H.264/AVC, H.265/HEVC, and
VP9 encoder, codec and decoder hardware (RTL) IPs.
Signed-off-by: Michael Tretter <m.tretter@pengutronix.de> Reviewed-by: Rob Herring <robh@kernel.org> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Maxime Ripard [Fri, 24 May 2019 09:20:31 +0000 (05:20 -0400)]
media: cedrus: Add H264 decoding support
Introduce some basic H264 decoding support in cedrus. So far, only the
baseline profile videos have been tested, and some more advanced features
used in higher profiles are not even implemented.
Reviewed-by: Jernej Skrabec <jernej.skrabec@siol.net> Reviewed-by: Paul Kocialkowski <paul.kocialkowski@bootlin.com> Signed-off-by: Maxime Ripard <maxime.ripard@bootlin.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Jernej Skrabec [Sat, 18 May 2019 17:44:15 +0000 (13:44 -0400)]
media: cedrus: Allow different mod clock rates
Some VPU variants may run at higher clock speeds. They actually need
extra speed to be capable of decoding more complex codecs like HEVC or
bigger image sizes (4K).
Expand variant structure with mod_rate information.
Signed-off-by: Jernej Skrabec <jernej.skrabec@siol.net> Acked-by: Maxime Ripard <maxime.ripard@bootlin.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
media: cxusb: add raw mode support for Medion MD95700
This adds raw (unprocessed) BT.656 stream capturing support for the analog
part of Medion 95700.
It can be enabled by setting CXUSB_EXTENDEDMODE_CAPTURE_RAW flag in
parm.capture.extendedmode passed to VIDIOC_S_PARM.
Signed-off-by: Maciej S. Szmigiero <mail@maciej.szmigiero.name> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
media: cxusb: add analog mode support for Medion MD95700
This patch adds support for analog part of Medion 95700 in the cxusb
driver.
What works:
* Video capture at various sizes with sequential fields,
* Input switching (TV Tuner, Composite, S-Video),
* TV and radio tuning,
* Video standard switching and auto detection,
* Radio mode switching (stereo / mono),
* Unplugging while capturing,
* DVB / analog coexistence.
What does not work yet:
* Audio,
* VBI,
* Picture controls.
Signed-off-by: Maciej S. Szmigiero <mail@maciej.szmigiero.name> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
[hverkuil-cisco@xs4all.nl: remove left-over commented-out debug message] Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
media: cxusb: implement Medion MD95700 digital / analog coexistence
This patch prepares cxusb driver for supporting the analog part of
Medion 95700 (previously only the digital - DVB - mode was supported).
Specifically, it adds support for:
* switching the device between analog and digital modes of operation,
* enforcing that only one mode is active at the same time due to hardware
limitations.
Actual implementation of the analog mode will be provided by the next
commit.
Signed-off-by: Maciej S. Szmigiero <mail@maciej.szmigiero.name> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
media: cx25840: set_fmt operation should clamp out-of-range picture sizes
According to V4L2 API set_fmt subdev operation should not return an error
on out-of-range picture sizes, the values should be clamped instead to the
supported range.
The cx25840 datasheet says that the chip is capable of scaling down the
picture width and height, respectively, 16 and 8 times.
These values agree with what the old implementation enforced.
Signed-off-by: Maciej S. Szmigiero <mail@maciej.szmigiero.name> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
media: cx25840: add pin to pad mapping and output format configuration
This commit adds pin to pad mapping and output format configuration support
in CX2584x-series chips to cx25840 driver.
This functionality is then used to allow disabling ivtv-specific hacks and
configuration values (called a "generic mode"), so cx25840 driver can be
used for other devices not needing them without risking compatibility
problems.
Signed-off-by: Maciej S. Szmigiero <mail@maciej.szmigiero.name> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
This commit implements g_std operation in cx25840 driver by returning the
last set video standard.
Signed-off-by: Maciej S. Szmigiero <mail@maciej.szmigiero.name> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
cx25840 driver g_std operation queries the currently detected video signal,
however this is what querystd operation should do, so let's rename the
handler.
None of the existing cx25840 driver users ever called the g_std operation,
one of them calls querystd on each of its subdevs but then the result is
only used to implement VIDIOC_QUERYSTD (as it should).
Signed-off-by: Maciej S. Szmigiero <mail@maciej.szmigiero.name> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
cx25840_load_fw() does the same thing as cx25840_reset(), only keeps
"is_initialized" flag so any further invocation of this function besides
the first one is a NOP.
Let's just call cx25840_reset() directly from cx25840_load_fw() instead of
open coding it there.
While we are at it, let's also improve comments about cx25840_load_fw()
so they are current and in the proper style (one of them even referred to a
non-existing cx25840 init operation).
Signed-off-by: Maciej S. Szmigiero <mail@maciej.szmigiero.name> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Steve Longerbeam [Fri, 10 May 2019 21:50:11 +0000 (17:50 -0400)]
media: staging/imx: Improve pipeline searching
Export find_pipeline_pad(), renaming to imx_media_pipeline_pad(), and
extend its functionality to allow searching for video devices in the
enabled pipeline in addition to sub-devices.
As part of this:
- Rename imx_media_find_mipi_csi2_channel() to
imx_media_pipeline_csi2_channel().
- Remove imx_media_find_upstream_pad(), it is redundant now.
- Rename imx_media_find_upstream_subdev() to imx_media_pipeline_subdev()
with an additional boolean argument for searching upstream or downstream.
- Add imx_media_pipeline_video_device() which is analogous to
imx_media_pipeline_subdev() but searches for video devices.
- Remove imxmd pointer arg from all of the functions above, it was
never used in those functions. With that change the i.MX5/6 CSI,
VDIC, and IC sub-devices no longer require the media_device.
Signed-off-by: Steve Longerbeam <slongerbeam@gmail.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Steve Longerbeam [Fri, 10 May 2019 21:50:10 +0000 (17:50 -0400)]
media: staging/imx: Re-organize modules
Re-organize modules, and which objects are linked into those modules, so
that:
- imx6-media (renamed from imx-media) is the media driver module for
imx5/6 only, and has no symbol exports.
- imx6-media-csi (renamed from imx-media-csi) is the subdev driver
module for imx5/6 CSI. It is now linked direcly with imx-media-fim,
since only the imx5/6 CSI makes use of the frame interval monitor.
- imx-media-common now only contains common code between imx5/6 and imx7
media drivers. It contains imx-media-utils, imx-media-of,
imx-media-dev-common, and imx-media-capture. In order to acheive that,
some functions common to imx5/6 and imx7 have been moved out of
imx-media-dev.c and into imx-media-dev-common.c.
Signed-off-by: Steve Longerbeam <slongerbeam@gmail.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Don't propagate the source pad format to the connected capture device.
It's now the responsibility of userspace to call VIDIOC_S_FMT on the
capture device to ensure the capture format and compose rectangle
are compatible with the connected source. To check this, validate
the capture format with the source before streaming starts.
Signed-off-by: Steve Longerbeam <slongerbeam@gmail.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Steve Longerbeam [Fri, 10 May 2019 21:50:08 +0000 (17:50 -0400)]
media: Revert "media: imx: Set capture compose rectangle in capture_device_set_format"
Revert this commit, as imx_media_capture_device_set_format() will be
removed. The arguments to mx_media_mbus_fmt_to_pix_fmt() and
imx_media_capture_device_set_format() in imx7_csi_set_fmt() are also
reverted.
Steve Longerbeam [Fri, 10 May 2019 21:50:06 +0000 (17:50 -0400)]
media: staging/imx: Pass device to alloc/free_dma_buf
Allocate and free a DMA coherent buffer in imx_media_alloc/free_dma_buf()
from the given device. This allows DMA alloc and free using a device
that is backed by real hardware, which for the imx5/6/7 CSI is the CSI
unit, and for the internal IPU sub-devices, is the parent IPU.
Signed-off-by: Steve Longerbeam <slongerbeam@gmail.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Steve Longerbeam [Fri, 10 May 2019 21:50:05 +0000 (17:50 -0400)]
media: staging/imx: Switch to sync registration for IPU subdevs
Because the IPU sub-devices VDIC and IC are not present in the
device-tree, platform devices were created for them instead. This
allowed these sub-devices to be added to the media device's async
notifier and registered asynchronously along with the other
sub-devices that do have a device-tree presence (CSI and devices
external to the IPU and SoC).
But that approach isn't really necessary. The IPU sub-devices don't
actually require a backing device (sd->dev is allowed to be NULL).
And that approach can't get around the fact that the IPU sub-devices
are not part of a device hierarchy, which makes it awkward to retrieve
the parent IPU of these devices.
By registering them synchronously, they can be registered from the CSI
async bound notifier, so the init function for them can be given the CSI
subdev, who's dev->parent is the IPU. That is a somewhat cleaner way
to retrieve the parent IPU.
So convert to synchronous registration for the VDIC and IC task
sub-devices, at the time a CSI sub-device is bound. There is no longer
a backing device for them (sd->dev is NULL), but that's ok. Also
set the VDIC/IC sub-device owner as the IPU, so that a reference can
be taken on the IPU module.
Since the VDIC and IC task drivers are no longer platform drivers,
they are now statically linked to imx-media module.
Signed-off-by: Steve Longerbeam <slongerbeam@gmail.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Vandana BN [Wed, 22 May 2019 08:34:15 +0000 (04:34 -0400)]
media: usb:zr364xx:Fix KASAN:null-ptr-deref Read in zr364xx_vidioc_querycap
SyzKaller hit the null pointer deref while reading from uninitialized
udev->product in zr364xx_vidioc_querycap().
==================================================================
BUG: KASAN: null-ptr-deref in read_word_at_a_time+0xe/0x20
include/linux/compiler.h:274
Read of size 1 at addr 0000000000000000 by task v4l_id/5287
Niklas Söderlund [Sat, 18 May 2019 00:46:54 +0000 (20:46 -0400)]
media: vimc: Remove unneeded return statement in vimc_sen_s_stream()
The other subdevice implementations in vimc (debayer and scaler) which
share their code structure with the sensor do not have an explicit
return statement at the end of the s_stream(0) code path. Align the
sensor subdevice by dropping the return statement.
Helen Koike [Fri, 17 May 2019 17:20:11 +0000 (13:20 -0400)]
media: vimc: fix component match compare
If the system has other devices being registered in the component
framework, the compare function will be called with a device that
doesn't belong to vimc.
This device is not necessarily a platform_device, nor have a
platform_data (which causes a NULL pointer dereference error) and if it
does have a pdata, it is not necessarily type of struct vimc_platform_data.
So casting to any of these types is wrong.
Instead of expecting a given pdev with a given pdata, just expect for
the device it self. vimc-core is the one who creates them, we know in
advance exactly which object to expect in the match.
Fixes: 4a29b7090749 ("[media] vimc: Subdevices as modules") Signed-off-by: Helen Koike <helen.koike@collabora.com> Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com> Tested-by: Boris Brezillon <boris.brezillon@collabora.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>