[libcamera-devel,00/25] media: ov5647: Support RaspberryPi Camera Module v1
mbox series

Message ID 20200622171910.608894-1-jacopo@jmondi.org
Headers show
Series
  • media: ov5647: Support RaspberryPi Camera Module v1
Related show

Message

Jacopo Mondi June 22, 2020, 5:18 p.m. UTC
Hello,
  this series improves and expand the existing ov5647 sensor driver to
the same feature level as found in RaspberryPi BSP.

It is based on recent media tree master and the sensor bindings conversion
to dt-schema I sent out a few days ago:
"[PATCH 0/2] dt-bidings: media: ov5647 bindings + small fix"

The first patches in the series has been sent by Roman as part of
"[PATCH v2 0/6] ov5647 driver improvement"
I took his patches in, addressed review comments and rebased on top
of the new dt-schema bindings. I kept the extensive receiver list he had
in his series for this reason.

The series continues by polishing and cleaning up the driver and expanding
its functionalities to support multiple modes and image formats.

The series has been tested with libcamera and the driver functionalities
compared with the BSP driver ones, and tested stand-alone by capturing
raw frames with yavta.

Thanks
   j

Dave Stevenson (8):
  media: ov5647: Add support for PWDN GPIO.
  media: ov5647: Add support for non-continuous clock mode
  media: ov5647: Add set_fmt and get_fmt calls.
  media: ov5647: Add support for get_selection()
  media: ov5647: Set V4L2_SUBDEV_FL_HAS_EVENTS flag
  media: ov5647: Support V4L2_CID_PIXEL_RATE
  media: ov5647: Support V4L2_CID_VBLANK control
  media: ov5647: Advertise the correct exposure range

David Plowman (1):
  media: ov5647: Support gain, exposure and AWB controls

Jacopo Mondi (16):
  dt-bindings: media: ov5647: Document pwdn-gpios
  dt-bindings: media: ov5647: Document clock-noncontinuous
  media: ov5647: Fix format initialization
  media: ov5647: Fix style issues
  media: ov5647: Replace license with SPDX identifier
  media: ov5647: Fix return value from read/write
  media: ov5647: Program mode at s_stream(1) time
  media: ov5647: Implement enum_frame_size()
  media: ov5647: Protect s_stream() with mutex
  media: ov5647: Rationalize driver structure name
  media: ov5647: Break out format handling
  media: ov5647: Rename SBGGR8 VGA mode
  media: ov5647: Add SGGBR10_1X10 modes
  media: ov5647: Implement set_fmt pad operation
  media: ov5647: Program mode only if it has changed
  media: ov5647: Support V4L2_CID_HBLANK control

 .../devicetree/bindings/media/i2c/ov5647.yaml |   11 +
 drivers/media/i2c/ov5647.c                    | 1267 +++++++++++++++--
 2 files changed, 1126 insertions(+), 152 deletions(-)

--
2.27.0

Comments

Jacopo Mondi June 22, 2020, 5:26 p.m. UTC | #1
My ISP has rejected the rest of the series: too many emails :(
Has it ever happened to anyone else ? How did you solved this ?

I'll send the rest of the series tomorrow. Sorry about this.

On Mon, Jun 22, 2020 at 07:18:45PM +0200, Jacopo Mondi wrote:
> Hello,
>   this series improves and expand the existing ov5647 sensor driver to
> the same feature level as found in RaspberryPi BSP.
>
> It is based on recent media tree master and the sensor bindings conversion
> to dt-schema I sent out a few days ago:
> "[PATCH 0/2] dt-bidings: media: ov5647 bindings + small fix"
>
> The first patches in the series has been sent by Roman as part of
> "[PATCH v2 0/6] ov5647 driver improvement"
> I took his patches in, addressed review comments and rebased on top
> of the new dt-schema bindings. I kept the extensive receiver list he had
> in his series for this reason.
>
> The series continues by polishing and cleaning up the driver and expanding
> its functionalities to support multiple modes and image formats.
>
> The series has been tested with libcamera and the driver functionalities
> compared with the BSP driver ones, and tested stand-alone by capturing
> raw frames with yavta.
>
> Thanks
>    j
>
> Dave Stevenson (8):
>   media: ov5647: Add support for PWDN GPIO.
>   media: ov5647: Add support for non-continuous clock mode
>   media: ov5647: Add set_fmt and get_fmt calls.
>   media: ov5647: Add support for get_selection()
>   media: ov5647: Set V4L2_SUBDEV_FL_HAS_EVENTS flag
>   media: ov5647: Support V4L2_CID_PIXEL_RATE
>   media: ov5647: Support V4L2_CID_VBLANK control
>   media: ov5647: Advertise the correct exposure range
>
> David Plowman (1):
>   media: ov5647: Support gain, exposure and AWB controls
>
> Jacopo Mondi (16):
>   dt-bindings: media: ov5647: Document pwdn-gpios
>   dt-bindings: media: ov5647: Document clock-noncontinuous
>   media: ov5647: Fix format initialization
>   media: ov5647: Fix style issues
>   media: ov5647: Replace license with SPDX identifier
>   media: ov5647: Fix return value from read/write
>   media: ov5647: Program mode at s_stream(1) time
>   media: ov5647: Implement enum_frame_size()
>   media: ov5647: Protect s_stream() with mutex
>   media: ov5647: Rationalize driver structure name
>   media: ov5647: Break out format handling
>   media: ov5647: Rename SBGGR8 VGA mode
>   media: ov5647: Add SGGBR10_1X10 modes
>   media: ov5647: Implement set_fmt pad operation
>   media: ov5647: Program mode only if it has changed
>   media: ov5647: Support V4L2_CID_HBLANK control
>
>  .../devicetree/bindings/media/i2c/ov5647.yaml |   11 +
>  drivers/media/i2c/ov5647.c                    | 1267 +++++++++++++++--
>  2 files changed, 1126 insertions(+), 152 deletions(-)
>
> --
> 2.27.0
>
Eugeniu Rosca June 23, 2020, 10:30 a.m. UTC | #2
Hi Jacopo,

Many thanks for your precious contribution.

On Mon, Jun 22, 2020 at 07:26:14PM +0200, Jacopo Mondi wrote:
> My ISP has rejected the rest of the series: too many emails :(
> Has it ever happened to anyone else ? How did you solved this ?

I guess leaving 5-10 seconds between sending individual patches should
overcome this? I wonder if git provides a built-in command for that?
Jacopo Mondi June 23, 2020, 10:49 a.m. UTC | #3
Hi Eugeniu,

On Tue, Jun 23, 2020 at 12:30:02PM +0200, Eugeniu Rosca wrote:
> Hi Jacopo,
>
> Many thanks for your precious contribution.
>
> On Mon, Jun 22, 2020 at 07:26:14PM +0200, Jacopo Mondi wrote:
> > My ISP has rejected the rest of the series: too many emails :(
> > Has it ever happened to anyone else ? How did you solved this ?
>
> I guess leaving 5-10 seconds between sending individual patches should
> overcome this? I wonder if git provides a built-in command for that?
>

git send-email does provide the --batch-size --relogin-delay options,
as Ezequiel suggested me in #v4l.

I tried re-sending with a 10 email batch and a 5 seconds delay but I
got the same failure. I was not able to find any description of the
email number limits for the SMTP server I'm using, so I could only go
and try. I think the extensive CC list of this series which I got from
Roman's series plays a role, so I can't try just by sending to
myself... I wonder if I should send the series in chunks, the first 10
patches went out (2 times '-.- ) already...

> --
> Best regards,
> Eugeniu Rosca
Eugeniu Rosca June 23, 2020, 12:17 p.m. UTC | #4
Hi Jacopo,

On Tue, Jun 23, 2020 at 12:49:03PM +0200, Jacopo Mondi wrote:
> On Tue, Jun 23, 2020 at 12:30:02PM +0200, Eugeniu Rosca wrote:
> > On Mon, Jun 22, 2020 at 07:26:14PM +0200, Jacopo Mondi wrote:
> > > My ISP has rejected the rest of the series: too many emails :(
> > > Has it ever happened to anyone else ? How did you solved this ?
> >
> > I guess leaving 5-10 seconds between sending individual patches should
> > overcome this? I wonder if git provides a built-in command for that?
> >
> 
> git send-email does provide the --batch-size --relogin-delay options,
> as Ezequiel suggested me in #v4l.
> 
> I tried re-sending with a 10 email batch and a 5 seconds delay but I
> got the same failure. I was not able to find any description of the
> email number limits for the SMTP server I'm using, so I could only go
> and try. I think the extensive CC list of this series which I got from
> Roman's series plays a role, so I can't try just by sending to
> myself... I wonder if I should send the series in chunks, the first 10
> patches went out (2 times '-.- ) already...

Any time you send a new batch, please call 'git send-email' with
'--in-reply-to=<cover-letter-id>' to allow LKML front-ends identify
the patches as belonging to one topic and make it less of a pain
for people to go through these discussions later on.
Jacopo Mondi June 24, 2020, 7:47 a.m. UTC | #5
Hi Eugeniu,

On Tue, Jun 23, 2020 at 02:17:49PM +0200, Eugeniu Rosca wrote:
> Hi Jacopo,
>
> On Tue, Jun 23, 2020 at 12:49:03PM +0200, Jacopo Mondi wrote:
> > On Tue, Jun 23, 2020 at 12:30:02PM +0200, Eugeniu Rosca wrote:
> > > On Mon, Jun 22, 2020 at 07:26:14PM +0200, Jacopo Mondi wrote:
> > > > My ISP has rejected the rest of the series: too many emails :(
> > > > Has it ever happened to anyone else ? How did you solved this ?
> > >
> > > I guess leaving 5-10 seconds between sending individual patches should
> > > overcome this? I wonder if git provides a built-in command for that?
> > >
> >
> > git send-email does provide the --batch-size --relogin-delay options,
> > as Ezequiel suggested me in #v4l.
> >
> > I tried re-sending with a 10 email batch and a 5 seconds delay but I
> > got the same failure. I was not able to find any description of the
> > email number limits for the SMTP server I'm using, so I could only go
> > and try. I think the extensive CC list of this series which I got from
> > Roman's series plays a role, so I can't try just by sending to
> > myself... I wonder if I should send the series in chunks, the first 10
> > patches went out (2 times '-.- ) already...
>
> Any time you send a new batch, please call 'git send-email' with
> '--in-reply-to=<cover-letter-id>' to allow LKML front-ends identify
> the patches as belonging to one topic and make it less of a pain
> for people to go through these discussions later on.

Thanks for the suggestion, I hope I got it right ;) The series has now
been resent.

Thanks
   j

>
> --
> Best regards,
> Eugeniu Rosca
Eugeniu Rosca June 24, 2020, 8:04 a.m. UTC | #6
Hi Jacopo,

On Wed, Jun 24, 2020 at 09:47:49AM +0200, Jacopo Mondi wrote:
> On Tue, Jun 23, 2020 at 02:17:49PM +0200, Eugeniu Rosca wrote:
> > Hi Jacopo,
> >
> > On Tue, Jun 23, 2020 at 12:49:03PM +0200, Jacopo Mondi wrote:
> > > On Tue, Jun 23, 2020 at 12:30:02PM +0200, Eugeniu Rosca wrote:
> > > > On Mon, Jun 22, 2020 at 07:26:14PM +0200, Jacopo Mondi wrote:
> > > > > My ISP has rejected the rest of the series: too many emails :(
> > > > > Has it ever happened to anyone else ? How did you solved this ?
> > > >
> > > > I guess leaving 5-10 seconds between sending individual patches should
> > > > overcome this? I wonder if git provides a built-in command for that?
> > > >
> > >
> > > git send-email does provide the --batch-size --relogin-delay options,
> > > as Ezequiel suggested me in #v4l.
> > >
> > > I tried re-sending with a 10 email batch and a 5 seconds delay but I
> > > got the same failure. I was not able to find any description of the
> > > email number limits for the SMTP server I'm using, so I could only go
> > > and try. I think the extensive CC list of this series which I got from
> > > Roman's series plays a role, so I can't try just by sending to
> > > myself... I wonder if I should send the series in chunks, the first 10
> > > patches went out (2 times '-.- ) already...
> >
> > Any time you send a new batch, please call 'git send-email' with
> > '--in-reply-to=<cover-letter-id>' to allow LKML front-ends identify
> > the patches as belonging to one topic and make it less of a pain
> > for people to go through these discussions later on.
> 
> Thanks for the suggestion, I hope I got it right ;) The series has now
> been resent.

That's right. I now see all 25 patches appearing under the same umbrella
of https://patchwork.linuxtv.org/cover/64822/ (via "Related"), even if
the last 15 have been sent out at a later point in time.