This message is generated daily by a cron job that builds media_tree for
the kernels and architectures in the list below.
Results of the daily build of media_tree:
date: Sat Jun 30 05:00:16 CEST 2018
media-tree git hash:3c4a737267e89aafa6308c6c456d2ebea3fcd085
media_build gi
As discussed on IRC, memory-to-memory need to be modeled
properly in order to be supported by the media controller
framework, and thus to support the Request API.
First commit introduces a register/unregister API,
that creates/destroys all the entities and pads needed,
and links them.
The second
From: Hans Verkuil
Request API requires a media node. Add one to the vim2m driver so we can
use requests with it.
Signed-off-by: Hans Verkuil
Signed-off-by: Ezequiel Garcia
---
drivers/media/platform/vim2m.c | 41 ++
1 file changed, 37 insertions(+), 4 deletion
A memory-to-memory pipeline device consists in three
entities: two DMA engine and one video processing entities.
The DMA engine entities are linked to a V4L interface.
This commit add a new v4l2_m2m_{un}register_media_controller
API to register this topology.
For instance, a typical mem2mem devic
On Fri, 2018-06-15 at 16:07 -0300, Ezequiel Garcia wrote:
> Fourth spin of the series posted by Hans:
>
> https://www.mail-archive.com/linux-media@vger.kernel.org/msg131363.ht
> ml
>
> There aren't any changes from v3, aside from rebasing
> and re-ordering the patches as requested by Hans.
>
> S
Hey Hans,
On Fri, 2018-06-29 at 09:03 +0200, Hans Verkuil wrote:
> On 06/28/2018 09:25 PM, Ezequiel Garcia wrote:
> > When creating a second node via copy or assignment:
> >
> > node2 = node
> >
> > The node being assigned to, i.e. node2, obtains the fd.
> > This causes a later call to node2
On Fri, 2018-06-29 at 09:03 +0200, Hans Verkuil wrote:
> On 06/28/2018 09:25 PM, Ezequiel Garcia wrote:
> > When creating a second node via copy or assignment:
> >
> > node2 = node
> >
> > The node being assigned to, i.e. node2, obtains the fd.
> > This causes a later call to node2.media_open
On 29 June 2018 at 08:43, Hans Verkuil wrote:
> From: Hans Verkuil
>
> The use of 'DTV' is very confusing since it normally refers to Digital
> TV e.g. DVB etc.
>
> Instead use 'DV' (Digital Video), which nicely corresponds to the
> DV Timings API used to configure such receivers and transmitters
From: Jacopo Mondi
Now that data and clock lanes start in LP11 no need to issue a start/stop
sequence at power on time to 'coax the lanes in LP11 state'.
Signed-off-by: Jacopo Mondi
---
drivers/media/i2c/ov5640.c | 14 --
1 file changed, 14 deletions(-)
diff --git a/drivers/media/
The sensor MIPI interface is initialized with settings that require a start/stop
sequence at power-up time in order to force lanes into LP11 state, as they are
initialized in LP00 when in 'sleep mode' which I assume to be the sensor
manual definition for the D-PHY stop mode.
The stream start/stop
From: Jacopo Mondi
Change the MIPI CSI-2 interface startup sequence to the following:
Initialization:
0x3019 = 0x70 : Lane1, Lane2 and clock in LP11 when in 'sleep mode'
0x300e = 0x58 : 2 lanes mode, power down TX and RX, MIPI CSI-2 off
0x4800 = 0x20 : Gate clock when not transmitting, LP00 when
On Thu, Jun 28, 2018 at 06:20:37PM +0200, Marco Felsch wrote:
> Add two helper functions to check if two rectangles have the same
> position (top/left) and if two rectangles equals (same size and
> same position).
>
> Signed-off-by: Marco Felsch
Acked-by: Sakari Ailus
--
Sakari Ailus
sakari.a
Hi ov5640 people,
I'm happy to finally jump on the bandwagon...
I had a few days to test the sensor driver (on 2 platforms) and unfortunately
this series breaks MIPI capture for me as well..
On Fri, Jun 01, 2018 at 04:05:58PM -0700, Sam Bobrowicz wrote:
> >> On May 21, 2018, at 12:39 AM, Maxi
Hi Mauro,
Here's a new driver for the ak7375 VCM and improvements for imx274.
Please pull.
The following changes since commit 3c4a737267e89aafa6308c6c456d2ebea3fcd085:
media: ov5640: fix frame interval enumeration (2018-06-28 09:24:38 -0400)
are available in the git repository at:
ssh://
On 06/29/18 09:06, Hans Verkuil wrote:
> On 06/28/2018 09:29 PM, Ezequiel Garcia wrote:
>> Compliance test for device /dev/video2:
>>
>> Driver Info:
>> Driver name : vim2m
>> Card type: vim2m
>> Bus info : platform:vim2m
>> Driver version : 4.18.0
>>
The hardware codec is not colorspace aware. We should trust userspace to
set the correct colorimetry information on the OUTPUT queue and mirror
the exact same setting on the CAPTURE queue.
There is no reason to restrict colorspace to JPEG for JPEG images, if
userspace injects the correct colorspac
Padding the bitstream buffer is not enough to reliably avoid prefetch
failures. Picture runs with the next buffer's header already visible to
the CODA7541 succeed much more reliably, so always queue two JPEG frames
into the bitstream buffer.
Signed-off-by: Philipp Zabel
---
drivers/media/platfo
This patch adds the i.MX6 CODA960 firmware versions 2.1.9
(revision 32515) to the list of supported firmware versions.
Signed-off-by: Philipp Zabel
---
drivers/media/platform/coda/coda-bit.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/media/platform/coda/coda-bit.c
b/drivers/med
Explicitly clear DEC_SEQ_JPG_THUMB_EN during sequence initialization.
Not clearing the register does not cause problems, since the only other
codec (MPEG-4 decode) that writes to this register happens to always
write 0 as well.
Signed-off-by: Philipp Zabel
---
drivers/media/platform/coda/coda-bi
If new buffers are queued after streamoff, the flag will be cleared
anyway, so this is mostly for the purpose of correctness.
Signed-off-by: Philipp Zabel
---
drivers/media/platform/coda/coda-common.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/media/platform/coda/coda-common.c
Em Fri, 29 Jun 2018 12:26:50 +0200
Hans Verkuil escreveu:
> On 06/29/18 11:54, Mauro Carvalho Chehab wrote:
> > Em Thu, 28 Jun 2018 15:12:00 +0200
> > Hans Verkuil escreveu:
> >
> >> From: Hans Verkuil
> >>
> >> Document the new struct media_v2_pad 'index' field.
> >>
> >> Signed-off-by: Han
From: Hans Verkuil
Document the new struct media_v2_entity 'flags' field.
Signed-off-by: Hans Verkuil
Acked-by: Sakari Ailus
---
.../media/uapi/mediactl/media-ioc-g-topology.rst | 10 +-
1 file changed, 9 insertions(+), 1 deletion(-)
diff --git a/Documentation/media/uapi/mediac
From: Hans Verkuil
Mention that IDs should not be hardcoded in applications and that the
entity name must be unique within the media topology.
Signed-off-by: Hans Verkuil
---
.../uapi/mediactl/media-ioc-enum-entities.rst | 9 +---
.../uapi/mediactl/media-ioc-g-topology.rst| 22 +++
From: Hans Verkuil
The v2 entity structure never exposed the entity flags, which made it
impossible to detect connector or default entities.
It is really trivial to just expose this information, so implement this.
Signed-off-by: Hans Verkuil
Acked-by: Sakari Ailus
---
drivers/media/media-dev
From: Hans Verkuil
Document the new struct media_v2_pad 'index' field.
Signed-off-by: Hans Verkuil
Acked-by: Sakari Ailus
---
.../media/uapi/mediactl/media-ioc-g-topology.rst | 12 ++--
1 file changed, 10 insertions(+), 2 deletions(-)
diff --git a/Documentation/media/uapi/mediact
From: Hans Verkuil
These two drivers both have function MEDIA_ENT_F_DV_ENCODER.
Signed-off-by: Hans Verkuil
Acked-by: Sakari Ailus
---
drivers/media/i2c/ad9389b.c | 1 +
drivers/media/i2c/adv7511.c | 1 +
2 files changed, 2 insertions(+)
diff --git a/drivers/media/i2c/ad9389b.c b/drivers/med
From: Hans Verkuil
Several drivers in media/i2c do not set the entity function.
Correct this.
Signed-off-by: Hans Verkuil
Acked-by: Sakari Ailus
---
drivers/media/i2c/et8ek8/et8ek8_driver.c | 1 +
drivers/media/i2c/mt9m032.c | 1 +
drivers/media/i2c/mt9p031.c | 1 +
From: Hans Verkuil
Make it clearer that the index starts at 0, and that it won't change
since future new pads will be added at the end.
Signed-off-by: Hans Verkuil
---
Documentation/media/uapi/mediactl/media-ioc-enum-links.rst | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --gi
From: Hans Verkuil
The entity function was ORed with the flags field instead of
assigned to the function field. Correct this.
Signed-off-by: Hans Verkuil
Acked-by: Sakari Ailus
---
drivers/media/i2c/adv7180.c | 2 +-
drivers/media/i2c/tvp514x.c | 2 +-
drivers/media/i2c/tvp7002.c | 2 +-
3 fi
From: Hans Verkuil
This series is v5 of my previous attempt:
https://www.mail-archive.com/linux-media@vger.kernel.org/msg133111.html
The goal is to fix the inconsistencies between the 'old' and 'new'
MC API. I hate the terms 'old' and 'new', there is nothing wrong IMHO with
using an 'old' API
From: Hans Verkuil
Keep the function defines in numerical order: 0x6000 comes after
0x2000, so move it back.
Signed-off-by: Hans Verkuil
---
include/uapi/linux/media.h | 14 +++---
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git a/include/uapi/linux/media.h b/include/uapi/l
From: Hans Verkuil
The v2 pad structure never exposed the pad index, which made it impossible
to call the MEDIA_IOC_SETUP_LINK ioctl, which needs that information.
It is really trivial to just expose this information, so implement this.
Signed-off-by: Hans Verkuil
Acked-by: Sakari Ailus
---
From: Hans Verkuil
The use of 'DTV' is very confusing since it normally refers to Digital
TV e.g. DVB etc.
Instead use 'DV' (Digital Video), which nicely corresponds to the
DV Timings API used to configure such receivers and transmitters.
We keep an alias to avoid breaking userspace application
From: Hans Verkuil
Add a new function for digital video encoders such as HDMI transmitters.
Signed-off-by: Hans Verkuil
Acked-by: Sakari Ailus
---
Documentation/media/uapi/mediactl/media-types.rst | 7 +++
include/uapi/linux/media.h| 3 ++-
2 files changed, 9 inser
Hi!
> > > > On Nokia N900, I have similar problems as Intel IPU3 hardware.
> > > >
> > > > Meeting notes say that pure software implementation is not fast
> > > > enough, but that it may be useful for debugging. It would be also
> > > > useful for me on N900, and probably useful for processing "ra
On 06/29/18 11:54, Mauro Carvalho Chehab wrote:
> Em Thu, 28 Jun 2018 15:12:00 +0200
> Hans Verkuil escreveu:
>
>> From: Hans Verkuil
>>
>> Document the new struct media_v2_pad 'index' field.
>>
>> Signed-off-by: Hans Verkuil
>> Acked-by: Sakari Ailus
>> ---
>> .../media/uapi/mediactl/media-i
When you switch from auto to manual mode for an auto-cluster (e.g.
autogain+gain controls), then the current HW value has to be copied
to the current control value. However, has_changed was never set to
true, so new_to_cur didn't actually copy this value.
Signed-off-by: Hans Verkuil
Reported-by:
Em Thu, 28 Jun 2018 15:11:58 +0200
Hans Verkuil escreveu:
> From: Hans Verkuil
>
> This series is v4 of my previous attempt:
>
> https://www.mail-archive.com/linux-media@vger.kernel.org/msg132942.html
Patches 1 and 3-10 looks OK to me.
Thanks,
Mauro
Em Thu, 28 Jun 2018 15:12:00 +0200
Hans Verkuil escreveu:
> From: Hans Verkuil
>
> Document the new struct media_v2_pad 'index' field.
>
> Signed-off-by: Hans Verkuil
> Acked-by: Sakari Ailus
> ---
> .../media/uapi/mediactl/media-ioc-g-topology.rst | 11 +--
> 1 file changed, 9
In the vivid driver you want gain to continuous change while autogain
is on. However, dev->jiffies_vid_cap doesn't actually change. It probably
did in the past, but changes in the code caused this to be a fixed value
that is only set when you start streaming.
Replace it by jiffies, which is always
Hi Pavel,
On Fri, Jun 29, 2018 at 6:18 PM Pavel Machek wrote:
>
> Hi!
>
> > > On Nokia N900, I have similar problems as Intel IPU3 hardware.
> > >
> > > Meeting notes say that pure software implementation is not fast
> > > enough, but that it may be useful for debugging. It would be also
> > > us
Hi!
> > On Nokia N900, I have similar problems as Intel IPU3 hardware.
> >
> > Meeting notes say that pure software implementation is not fast
> > enough, but that it may be useful for debugging. It would be also
> > useful for me on N900, and probably useful for processing "raw" images
> > from d
Codec devices have two m2m devices in the media topology: one
for decoding, one for encoding. Since the entity names were the
same for both, this was invalid. Also the major/minor numbers
were not set for the I/O entities.
Signed-off-by: Hans Verkuil
---
Ezequiel, just fold this in your own patc
On 06/29/2018 09:01 AM, Hans Verkuil wrote:
> Hi Ezequiel,
>
> I added support for this to a codec and I discovered that we are missing a
> 'name'
> argument to v4l2_m2m_register_media_controller(): a typical codec driver has
> two
> m2m video nodes: one for encoding, one for decoding. That work
On 06/28/2018 09:29 PM, Ezequiel Garcia wrote:
> Hi Hans,
>
> On Thu, 2018-06-28 at 11:19 +0200, Hans Verkuil wrote:
>> On 06/27/18 22:35, Ezequiel Garcia wrote:
>>> As discussed on IRC, memory-to-memory need to be modeled
>>> properly in order to be supported by the media controller
>>> framework
On 06/28/2018 09:25 PM, Ezequiel Garcia wrote:
> v4l_queue_export_bufs uses the v4l_fd type when calling
> EXPBUF ioctl. However, this doesn't work on mem2mem
> where there are one capture queue and one output queue
> associated to the device.
>
> The current code calls v4l_queue_export_bufs with
On 06/28/2018 09:25 PM, Ezequiel Garcia wrote:
> When creating a second node via copy or assignment:
>
> node2 = node
>
> The node being assigned to, i.e. node2, obtains the fd.
> This causes a later call to node2.media_open to close()
> the fd, thus unintendenly closing the original node fd,
Hi Ezequiel,
I added support for this to a codec and I discovered that we are missing a
'name'
argument to v4l2_m2m_register_media_controller(): a typical codec driver has two
m2m video nodes: one for encoding, one for decoding. That works fine, except
that
the names of the source, sink and proc
48 matches
Mail list logo
