Todor Tomov [Tue, 8 Aug 2017 13:30:05 +0000 (09:30 -0400)]
media: camss: Add VFE files
These files control the VFE module. The VFE has different input
interfaces. The PIX input interface feeds the input data to an image
processing pipeline. Three RDI input interfaces bypass the image
processing pipeline. The VFE also contains the AXI bus interface which
writes the output data to memory.
RDI interfaces are supported in this version. PIX interface is not
supported.
Sakari Ailus [Tue, 8 Aug 2017 13:29:58 +0000 (09:29 -0400)]
media: v4l: Add packed Bayer raw12 pixel formats
These formats are compressed 12-bit raw bayer formats with four different
pixel orders. They are similar to 10-bit variants. The formats added by
this patch are
Colin Ian King [Tue, 22 Aug 2017 14:21:20 +0000 (10:21 -0400)]
media: em28xx: calculate left volume level correctly
The calculation of the left volume looks suspect, the value of
0x1f - ((val << 8) & 0x1f) is always 0x1f. The debug prior to the
assignment of value[1] prints the left volume setting using the
calculation 0x1f - (val >> 8) & 0x1f which looks correct to me.
Fix the left volume by using the correct expression as used in
the debug.
Detected by CoverityScan, CID#146140 ("Wrong operator used")
Fixes: 850d24a5a861 ("[media] em28xx-alsa: add mixer support for AC97 volume controls") Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: Hans Verkuil <hansverk@cisco.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
videobuf_queue_ops are not supposed to change at runtime. All functions
working with videobuf_queue_ops provided by <media/videobuf-core.h> work
with const videobuf_queue_ops. So mark the non-const structs as const.
videobuf_queue_ops are not supposed to change at runtime. All functions
working with videobuf_queue_ops provided by <media/videobuf-core.h> work
with const videobuf_queue_ops. So mark the non-const structs as const.
videobuf_queue_ops are not supposed to change at runtime. All functions
working with videobuf_queue_ops provided by <media/videobuf-core.h> work
with const videobuf_queue_ops. So mark the non-const structs as const.
media: venus: venc: set correct resolution on compressed stream
This change the alignment restriction for output type of buffers
only, also set corect input resolution and fill bidirectional
vb2 queue flag in order to map output type buffers read/write.
The last is needed by encoder firmware to add padding at the
bottom of output (input buffers).
This change is intended to give to the v4l2 drivers a choice to
change the default behavior of the v4l2-core DMA mapping direction
from DMA_TO/FROM_DEVICE (depending on the buffer type CAPTURE or
OUTPUT) to DMA_BIDIRECTIONAL during queue_init time.
Initially the issue with DMA mapping direction has been found in
Venus encoder driver where the hardware (firmware side) adds few
lines padding on bottom of the image buffer, and the consequence
is triggering of IOMMU protection faults.
This will help supporting venus encoder (and probably other drivers
in the future) which wants to map output type of buffers as
read/write.
Signed-off-by: Stanimir Varbanov <stanimir.varbanov@linaro.org> Acked-by: Marek Szyprowski <m.szyprowski@samsung.com> Signed-off-by: Hans Verkuil <hansverk@cisco.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Hugues Fruchet [Tue, 22 Aug 2017 14:41:11 +0000 (10:41 -0400)]
media: stm32-dcmi: g_/s_selection crop support
Implements g_/s_selection crop support by using DCMI crop
hardware feature.
User can first get the maximum supported resolution of the sensor
by calling g_selection(V4L2_SEL_TGT_CROP_BOUNDS).
Then user call to s_selection(V4L2_SEL_TGT_CROP) will reset sensor
to its maximum resolution and crop request is saved for later usage
in s_fmt().
Next call to s_fmt() will check if sensor can do frame size request
with crop request. If sensor supports only discrete frame sizes,
the frame size which is larger than user request is selected in
order to be able to match the crop request. Then s_fmt() resolution
user request is adjusted to match crop request resolution.
Uniformize "pixfmt" variables to "pix".
Change "current_fmt" & "dcmi_fmt" variables to variables
with "sd_" prefix to explicitly refer to subdev format.
Sakari Ailus [Mon, 14 Aug 2017 10:43:07 +0000 (06:43 -0400)]
media: v4l: fwnode: Use a less clash-prone name for MAX_DATA_LANES macro
Avoid using a generic name such as MAX_DATA_LANES in a header file widely
included in drivers. Instead, call it V4L2_FWNODE_CSI2_MAX_DATA_LANES.
Fixes: 4ee236219f6d ("media: v4l2-fwnode: suppress a warning at OF parsing logic") Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Sakari Ailus [Mon, 14 Aug 2017 10:22:14 +0000 (06:22 -0400)]
media: v4l: fwnode: The clock lane is the first lane in lane_polarities
The clock lane is the first lane in the lane_polarities array. Reflect this
consistently by putting the number of data lanes after the number of clock
lanes.
Fixes: 4ee236219f6d ("media: v4l2-fwnode: suppress a warning at OF parsing logic") Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
fwnode_property_read_u32_array() only returns the number of array elements
if the array argument is NULL. The assumption that it always did so lead to
lane-polarities properties never being read.
Fixes: 4ee236219f6d ("media: v4l2-fwnode: suppress a warning at OF parsing logic") Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Arvind Yadav [Sat, 19 Aug 2017 19:20:47 +0000 (15:20 -0400)]
media: ths8200: constify i2c_device_id
i2c_device_id are not supposed to change at runtime. All functions
working with i2c_device_id provided by <linux/i2c.h> work with
const i2c_device_id. So mark the non-const structs as const.
Arvind Yadav [Sat, 19 Aug 2017 19:20:46 +0000 (15:20 -0400)]
media: tc358743: constify i2c_device_id
i2c_device_id are not supposed to change at runtime. All functions
working with i2c_device_id provided by <linux/i2c.h> work with
const i2c_device_id. So mark the non-const structs as const.
Arvind Yadav [Sat, 19 Aug 2017 19:20:45 +0000 (15:20 -0400)]
media: saa7127: constify i2c_device_id
i2c_device_id are not supposed to change at runtime. All functions
working with i2c_device_id provided by <linux/i2c.h> work with
const i2c_device_id. So mark the non-const structs as const.
Arvind Yadav [Sat, 19 Aug 2017 19:20:44 +0000 (15:20 -0400)]
media: adv7842: constify i2c_device_id
i2c_device_id are not supposed to change at runtime. All functions
working with i2c_device_id provided by <linux/i2c.h> work with
const i2c_device_id. So mark the non-const structs as const.
Arvind Yadav [Sat, 19 Aug 2017 19:20:43 +0000 (15:20 -0400)]
media: adv7511: constify i2c_device_id
i2c_device_id are not supposed to change at runtime. All functions
working with i2c_device_id provided by <linux/i2c.h> work with
const i2c_device_id. So mark the non-const structs as const.
i2c_device_id are not supposed to change at runtime. All functions
working with i2c_device_id provided by <linux/i2c.h> work with
const i2c_device_id. So mark the non-const structs as const.
Hans Verkuil [Fri, 4 Aug 2017 10:41:54 +0000 (06:41 -0400)]
media: stih-cec: use CEC_CAP_DEFAULTS
Use the new CEC_CAP_DEFAULTS define in this driver.
This also adds the CEC_CAP_RC capability which was missing here
(and this is also the reason for this new define, to avoid missing
such capabilities).
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com> Acked-by: Benjamin Gaignard <benjamin.gaignard@linaro.org> Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Hans Verkuil [Sun, 20 Aug 2017 11:15:25 +0000 (07:15 -0400)]
media: vivid: add CEC pin monitoring emulation
Add support to emulate CEC pin monitoring. There are few hardware devices
that support this, so being able to emulate it here helps developing
software for this.
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Hans Verkuil [Sun, 20 Aug 2017 10:53:10 +0000 (06:53 -0400)]
media: cec: replace pin->cur_value by adap->cec_pin_is_high
The current CEC pin value (0 or 1) was part of the cec_pin struct,
but that assumes that CEC pin monitoring can only be used with
a driver that uses the low-level CEC pin framework.
But hardware that has both a high-level API and can monitor the
CEC pin at low-level at the same time does not need to depend on
the cec pin framework.
To support such devices remove the cur_value field from struct cec_pin
and add a cec_pin_is_high field to cec_adapter. This also makes it
possible to drop the '#ifdef CONFIG_CEC_PIN' in cec-api.c.
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Hans Verkuil [Sun, 20 Aug 2017 11:41:41 +0000 (07:41 -0400)]
media: cec: ensure that adap_enable(false) is called from cec_delete_adapter()
When the adapter is removed the cec_delete_adapter() call attempts
to set the physical address to INVALID by calling __cec_s_phys_addr()
and so disabling the adapter.
However, __cec_s_phys_addr checks if the device node was unregistered
and just returns in that case.
This means that the adap_enable callback is never called with 'false'
to disable the CEC adapter. Most drivers don't care, but some need
to do cleanup here.
Change the test so the adapter is correctly disabled, even when the
device node is unregistered.
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com> Reported-by: Tomi Valkeinen <tomi.valkeinen@ti.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
It was noticed that, if the output load is highly capacitive, a different approach
is recomended when energizing the LNBf.
Also, it is possible to detect if a current overload is happening, by checking an
special flag.
Add support for it.
Tested on Skystar S2. Changes respect override_or option so should still work fine
on cx88 based cards which disable dynamic current limit.
Changes since v1:
v2 - fixed incorrect checking of i2c return values
v3 - fix if logic to check if dcl needs re-enabling
- respect override_or values which aim to disable dcl
- only do long sleep on overload if dcl enabled
- add short sleep before re-enabling dcl
- only check overload and potentially return EINVAL if device is on
v4 - revert v3 sleep logic changes to remove tuning delays
Daniel Scheller [Sun, 20 Aug 2017 12:59:11 +0000 (08:59 -0400)]
media: stv0910: declare global list_head stvlist static
Cleans up smatch warning:
symbol 'stvlist' was not declared. Should it be static?
Patch originally submitted by Colin Ian King <colin.king@canonical.com>,
remainder after the merge of all other stv0910 fixes.
Cc: Colin Ian King <colin.king@canonical.com> Signed-off-by: Daniel Scheller <d.scheller@gmx.net> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Sean Young [Mon, 7 Aug 2017 20:20:58 +0000 (16:20 -0400)]
media: rc: rename RC_TYPE_* to RC_PROTO_* and RC_BIT_* to RC_PROTO_BIT_*
RC_TYPE is confusing and it's just the protocol. So rename it.
Suggested-by: Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by: Sean Young <sean@mess.org> Acked-by: Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Hans Verkuil [Mon, 7 Aug 2017 13:31:24 +0000 (09:31 -0400)]
media: cec: fix remote control passthrough
The 'Press and Hold' operation was not correctly implemented, in
particular the requirement that the repeat doesn't start until
the second identical keypress arrives. The REP_DELAY value also
had to be adjusted (see the comment in the code) to achieve the
desired behavior.
The 'enabled_protocols' field was also never set, fix that too. Since
CEC is a fixed protocol the driver has to set this field.
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com> Acked-by: Sean Young <sean@mess.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Sean Young [Wed, 9 Aug 2017 17:19:16 +0000 (13:19 -0400)]
media: rc: per-protocol repeat period
CEC needs a keypress timeout of 550ms, which is too high for the IR
protocols. Also fill in known repeat times, with 50ms error margin.
Also, combine all protocol data into one structure.
Signed-off-by: Sean Young <sean@mess.org> Suggested-by: Hans Verkuil <hans.verkuil@cisco.com> Acked-by: Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Sean Young [Mon, 7 Aug 2017 13:21:29 +0000 (09:21 -0400)]
media: rc: ensure we do not read out of bounds
If rc_validate_filter() is called for CEC or XMP, then we would read
beyond the end of the array.
Suggested-by: Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by: Sean Young <sean@mess.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Sean Young [Mon, 7 Aug 2017 12:38:10 +0000 (08:38 -0400)]
media: rc: simplify ir_raw_event_store_edge()
Since commit 12749b198fa4 ("[media] rc: saa7134: add trailing space for
timely decoding"), the workaround of inserting reset events is no
longer needed.
Note that the initial reset is not needed either; other rc-core drivers
that don't use ir_raw_event_store_edge() never call this at all.
Verified on a HVR-1150 and Raspberry Pi.
Fixes: 3f5c4c73322e ("[media] rc: fix ghost keypresses with certain hw") Signed-off-by: Sean Young <sean@mess.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Sean Young [Mon, 7 Aug 2017 12:30:18 +0000 (08:30 -0400)]
media: rc: saa7134: add trailing space for timely decoding
The gpio-ir-recv driver adds timeouts which the saa7134 lacks; this
causes keypress not to arrive, and to only arrive once more IR is
received. This is what the commit below calls "ghost keypresses",
and that commit does not solve the issue completely.
This makes the IR on the HVR-1150 much more reliable and responsive.
Fixes: 3f5c4c73322e ("[media] rc: fix ghost keypresses with certain hw") Signed-off-by: Sean Young <sean@mess.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
The gpio-ir-recv driver does many wakeups (once per edge); the saa7134
driver has special handling to only wakeup 15ms after the first edge.
Make this part of rc-core so gpio-ir-recv also benefits from
this (so a rc-5 keypress now causes 3 wakeups rather than 24).
Signed-off-by: Sean Young <sean@mess.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
media: rc: ir-nec-decoder: move scancode composing code into a shared function
The NEC scancode composing and protocol type detection in
ir_nec_decode() is generic enough to be a shared function. Let's create
an inline function in rc-core.h, so that other remote control drivers
can reuse this function to save some code.
Signed-off-by: Shawn Guo <shawn.guo@linaro.org> Signed-off-by: Sean Young <sean@mess.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Philipp Zabel [Wed, 19 Jul 2017 15:25:41 +0000 (11:25 -0400)]
media: rc: sunxi-cir: explicitly request exclusive reset control
Commit a53e35db70d1 ("reset: Ensure drivers are explicit when requesting
reset lines") started to transition the reset control request API calls
to explicitly state whether the driver needs exclusive or shared reset
control behavior. Convert all drivers requesting exclusive resets to the
explicit API call so the temporary transition helpers can be removed.
No functional changes.
Cc: Maxime Ripard <maxime.ripard@free-electrons.com> Cc: Chen-Yu Tsai <wens@csie.org> Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de> Signed-off-by: Sean Young <sean@mess.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Philipp Zabel [Wed, 19 Jul 2017 15:25:39 +0000 (11:25 -0400)]
media: st-rc: explicitly request exclusive reset control
Commit a53e35db70d1 ("reset: Ensure drivers are explicit when requesting
reset lines") started to transition the reset control request API calls
to explicitly state whether the driver needs exclusive or shared reset
control behavior. Convert all drivers requesting exclusive resets to the
explicit API call so the temporary transition helpers can be removed.
No functional changes.
Cc: Patrice Chotard <patrice.chotard@st.com> Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de> Signed-off-by: Sean Young <sean@mess.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Transmit support in this driver was never tested and based on the code
it can't work. Just one example:
The buffer provided to nvt_tx_ir holds unsigned int values in
micro seconds: First value is for a pulse, second for a pause, etc.
Bytes in this buffer are copied as-is to the chip FIFO what can't work
as the chip-internal format is totally different. See also conversion
done in nvt_process_rx_ir_data.
Even if we would try to fix this we have the issue that we can't test
it. There seems to be no device on the market using IR transmit with
one of the chips supported by this driver.
To facilitate maintenance of the driver I'd propose to remove the
rudimentary transmit support.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com> Signed-off-by: Sean Young <sean@mess.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
media: lirc_zilog: Clean up lirc zilog error codes
According the coding style guidelines, the ENOSYS error code must be
returned in case of a non existent system call. This code has been
replaced with the ENOTTY error code indicating a missing functionality.
Signed-off-by: Yves Lemée <yves.lemee.kernel@gmail.com> Signed-off-by: Sean Young <sean@mess.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
attribute_groups are not supposed to change at runtime. All functions
working with attribute_groups provided by <linux/sysfs.h> work with const
attribute_group. So mark the non-const structs as const.
File size before:
text data bss dec hex filename
11605 880 20 12505 30d9 drivers/media/rc/rc-main.o
File size After adding 'const':
text data bss dec hex filename
11797 720 20 12537 30f9 drivers/media/rc/rc-main.o
Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com> Signed-off-by: Sean Young <sean@mess.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
attribute_groups are not supposed to change at runtime. All functions
working with attribute_groups provided by <linux/sysfs.h> work with const
attribute_group. So mark the non-const structs as const.
File size before:
text data bss dec hex filename
18551 2256 77 20884 5194 drivers/media/rc/imon.o
File size After adding 'const':
text data bss dec hex filename
18679 2160 77 20916 51b4 drivers/media/rc/imon.o
Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com> Signed-off-by: Sean Young <sean@mess.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
media: sir_ir: remove unnecessary static in sir_interrupt()
Remove unnecessary static on local variable delt.
Such variable is initialized before being used,
on every execution path throughout the function.
The static has no benefit and, removing it reduces
the code size.
This issue was detected using Coccinelle and the following semantic patch:
@bad exists@
position p;
identifier x;
type T;
@@
static T x@p;
...
x = <+...x...+>
@@
identifier x;
expression e;
type T;
position p != bad.p;
@@
-static
T x@p;
... when != x
when strict
?x = e;
In the following log you can see the difference in the code size. Also,
there is a significant difference in the bss segment. This log is the
output of the size command, before and after the code change:
before:
text data bss dec hex filename
5009 3456 576 9041 2351 drivers/media/rc/sir_ir.o
after:
text data bss dec hex filename
4988 3400 512 8900 22c4 drivers/media/rc/sir_ir.o
Signed-off-by: Gustavo A. R. Silva <garsilva@embeddedor.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Sean Young [Sun, 2 Jul 2017 19:37:58 +0000 (15:37 -0400)]
media: rc-core: do not depend on MEDIA_SUPPORT
There is no dependency between the two, so remove the dependency in
Kconfig files.
Signed-off-by: Sean Young <sean@mess.org> Acked-by: Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Sean Wang [Fri, 30 Jun 2017 06:03:07 +0000 (02:03 -0400)]
media: rc: mtk-cir: add MAINTAINERS entry for MediaTek CIR driver
I work for MediaTek on maintaining the MediaTek CIR driver
for the existing SoCs and adding support for the following
SoCs.
Signed-off-by: Sean Wang <sean.wang@mediatek.com> Signed-off-by: Sean Young <sean@mess.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Sean Wang [Fri, 30 Jun 2017 06:03:06 +0000 (02:03 -0400)]
media: rc: mtk-cir: add support for MediaTek MT7622 SoC
This patch adds driver for CIR controller on MT7622 SoC. It has similar
handling logic as the previously MT7623 does, but there are some
differences in the register and field definition. So for ease portability
and maintenance, those differences all are being kept inside the platform
data as other drivers usually do. Currently testing successfully on NEC
and SONY remote controller.
Signed-off-by: Sean Wang <sean.wang@mediatek.com> Signed-off-by: Sean Young <sean@mess.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Sean Wang [Fri, 30 Jun 2017 06:03:05 +0000 (02:03 -0400)]
media: rc: mtk-cir: add platform data to adapt into various hardware
This patch is the preparation patch in order to adapt into various
hardware through adding platform data which holds specific characteristics
and differences among MediaTek supported CIR devices instead of the old
way defining those data in the static way as macro has. And the existing
logic would be slightly changed to operate on those data which the actual
device depends on.
Signed-off-by: Sean Wang <sean.wang@mediatek.com> Signed-off-by: Sean Young <sean@mess.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Sean Wang [Fri, 30 Jun 2017 06:03:04 +0000 (02:03 -0400)]
media: dt-bindings: media: mtk-cir: Add support for MT7622 SoC
Document the devicetree bindings for CIR on MediaTek MT7622
SoC.
Signed-off-by: Sean Wang <sean.wang@mediatek.com> Acked-by: Rob Herring <robh@kernel.org> Signed-off-by: Sean Young <sean@mess.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
David Härdeman [Thu, 22 Jun 2017 19:23:54 +0000 (15:23 -0400)]
media: rc-core: consistent use of rc_repeat()
The NEC decoder and the Sanyo decoders check if dev->keypressed is true
before calling rc_repeat (without holding dev->keylock).
Meanwhile, the XMP and JVC decoders do no such checks.
This patch makes sure all users of rc_repeat() do so consistently by
removing extra checks in NEC/Sanyo and modifying the check a bit in
rc_repeat() so that no input event is generated if the key isn't pressed.
Signed-off-by: David Härdeman <david@hardeman.nu> Signed-off-by: Sean Young <sean@mess.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Kieran Bingham [Fri, 4 Aug 2017 16:32:44 +0000 (12:32 -0400)]
media: v4l: vsp1: Allow entities to participate in the partition algorithm
The configuration of the pipeline and entities directly affects the
inputs required to each entity for the partition algorithm. Thus it
makes sense to involve those entities in the decision making process.
Extend the entity ops API to provide an optional .partition() operation.
This allows entities that affect the partition window to adapt the
window based on their configuration.
Entities implementing this operation must update the window parameter in
place, which will then be passed up the pipeline. This creates a process
whereby each entity describes what is required to satisfy the required
output to its predecessor in the pipeline.
Kieran Bingham [Fri, 4 Aug 2017 16:32:43 +0000 (12:32 -0400)]
media: v4l: vsp1: Provide UDS register updates
Provide register definitions required for UDS phase and partition
algorithm support. The registers and bits defined here are available on
Gen3 hardware only.
Kieran Bingham [Fri, 4 Aug 2017 16:32:42 +0000 (12:32 -0400)]
media: v4l: vsp1: Move partition rectangles to struct and operate directly
As we develop the partition algorithm, we need to store more information
per partition to describe the phase and other parameters.
To keep this data together, further abstract the existing v4l2_rect
into a partition specific structure. As partitions only have horizontal
coordinates, store the left and width values only.
When generating the partition windows, operate directly on the partition
struct rather than copying and duplicating the processed data
The vsp1_pipe object context variables for div_size and
current_partition allowed state to be maintained through processing the
partitions during processing.
Now that the partition tables are calculated during stream on, there is
no requirement to store these variables in the pipe object.
Utilise local variables for the processing as required.
Kieran Bingham [Fri, 4 Aug 2017 16:32:40 +0000 (12:32 -0400)]
media: v4l: vsp1: Calculate partition sizes at stream start
Previously the active window and partition sizes for each partition were
calculated for each partition every frame. This data is constant and
only needs to be calculated once at the start of the stream.
Extend the vsp1_pipe object to dynamically store the number of partitions
required and pre-calculate the partition sizes into this table.
Kieran Bingham [Fri, 4 Aug 2017 16:32:39 +0000 (12:32 -0400)]
media: v4l: vsp1: Move vsp1_video_pipeline_setup_partitions() function
Separate the code change from the function move so that code changes can
be clearly identified. This commit has no functional change.
The partition algorithm functions will be changed, and
vsp1_video_pipeline_setup_partitions() will call vsp1_video_partition().
To prepare for that, move the function without any code change.
Kieran Bingham [Fri, 4 Aug 2017 16:32:38 +0000 (12:32 -0400)]
media: v4l: vsp1: Release buffers in start_streaming error path
Presently any received buffers are only released back to vb2 if
vsp1_video_stop_streaming() is called. If vsp1_video_start_streaming()
encounters an error, we will be warned by the vb2 handlers that buffers
have not been returned.
Move the buffer cleanup code to its own function to prevent duplication
and call from both vsp1_video_stop_streaming() and the error path in
vsp1_video_start_streaming().
Previously, vblank range given to user was too big, falling outside
of permissible range for a given resolution. Sometimes, too low vblank
resulted in errors.
Now, limit vblank to only permissible range for a given resolution.
This change limits lower-bounds of vblank, doesn't affect upper bounds.
Previously, vblank range given to user was too big, falling outside
of permissible range for a given resolution. Sometimes, too low vblank
resulted in errors.
Now, limit vblank to only permissible range for a given resolution.
This change limits lower-bounds of vblank, doesn't affect upper bounds.
Sakari Ailus [Wed, 16 Aug 2017 07:28:16 +0000 (03:28 -0400)]
media: et8ek8: Decrease stack usage
The et8ek8 driver combines I²C register writes to a single array that it
passes to i2c_transfer(). The maximum number of writes is 48 at once,
decrease it to 8 and make more transfers if needed.
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com> Tested-by: Pavel Machek <pavel@ucw.cz> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Julia Lawall [Tue, 8 Aug 2017 10:58:30 +0000 (06:58 -0400)]
media: mt9m111: constify video_subdev structures
The v4l2_subdev_ops structure is only passed as the third argument
of v4l2_i2c_subdev_init, which is const, so the v4l2_subdev_ops
structure can be const as well. The other structures are only
stored in the v4l2_subdev_ops structure, all the fields of which are
const, so these structures can also be const.
Done with the help of Coccinelle.
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr> Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
The v4l2_subdev_ops structure is only passed as the third argument of
v4l2_i2c_subdev_init, which is const, so the v4l2_subdev_ops structure
can be const as well. The other structures are only stored in the
v4l2_subdev_ops structure, all the fields of which are const, so these
structures can also be const.
Done with the help of Coccinelle.
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr> Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com> Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
projects / mirror_ubuntu-kernels.git / log