Use display information in info not in var for panning

Hi everybody,

Here's the third version of the FBIOPAN_DISPLAY fixes patch series.

As a reminder, while playing with the FBIOPAN_DISPLAY ioctl I noticed that many
drivers use information from the ioctl argument such as the display resolution
when they should use the current settings from the fb_info structure.

These 29 patches fix most of those drivers. Missing drivers are ivtv (for
which a patch has Andy Walls has already pushed a patch), sh_mobile_lcdcfb (for
which the fix is a bit more complex, I will work on it later) and viafb (for
which Florian Tobias Schandinat has already sent a patch).

The patches have been compile-tested on architectures for which I have a
working compiler (x86 and ARM) with allyesconfig. I don't own any hardware
supported by these drivers so I haven't been able to perform additional tests.

I've split the fixes in one patch per driver for easier review and handling of
potential problems. The patches can be squashed together if needed. I've tried
to CC individual driver maintainers when possible, and Acked-by lines received
in reply to the first version have been included.

Compared to v2, I've fixed a compile issue in the atmel_lcdfb driver and added
a couple of SoB lines.

Paul, is it time to push this for 3.1 ?

Laurent Pinchart (29):
  68328fb: use display information in info not in var for panning
  acornfb: Dont BUG() on invalid pan parameters
  acornfb: use display information in info not in var for panning
  arkfb: use display information in info not in var for panning
  atmel_lcdfb: use display information in info not in var for panning
  radeonfb: use display information in info not in var for panning
  fbdev: da8xx: use display information in info not in var for panning
  fbdev: unicore32: use display information in info not in var for
    panning
  g364fb: use display information in info not in var for panning
  gxt4500: use display information in info not in var for panning
  hgafb: use display information in info not in var for panning

55 more lines

On Fri, Jul 08, 2011 at 01:14:14PM +0200, Laurent Pinchart wrote:

Hi everybody,

Here's the third version of the FBIOPAN_DISPLAY fixes patch series.

As a reminder, while playing with the FBIOPAN_DISPLAY ioctl I noticed that many
drivers use information from the ioctl argument such as the display resolution
when they should use the current settings from the fb_info structure.

These 29 patches fix most of those drivers. Missing drivers are ivtv (for
which a patch has Andy Walls has already pushed a patch), sh_mobile_lcdcfb (for
which the fix is a bit more complex, I will work on it later) and viafb (for
which Florian Tobias Schandinat has already sent a patch).

The patches have been compile-tested on architectures for which I have a
working compiler (x86 and ARM) with allyesconfig. I don't own any hardware
supported by these drivers so I haven't been able to perform additional tests.

I've split the fixes in one patch per driver for easier review and handling of
potential problems. The patches can be squashed together if needed. I've tried
to CC individual driver maintainers when possible, and Acked-by lines received
in reply to the first version have been included.

Compared to v2, I've fixed a compile issue in the atmel_lcdfb driver and added
a couple of SoB lines.

Paul, is it time to push this for 3.1 ?

I have the v2 series queued up already, so it would be nice to simply
have the v2->v3 differences as incremental changes that can be piled on.
The signed-off-by lines are not that critical in this case since it's a
largely mechanical change all over the place, so it's primarily the build
fix(es?) that would be nice to have.

Hi Paul,

On Wednesday 13 July 2011 10:24:24 Paul Mundt wrote:

On Fri, Jul 08, 2011 at 01:14:14PM +0200, Laurent Pinchart wrote:

quoted

Hi everybody,

Here's the third version of the FBIOPAN_DISPLAY fixes patch series.

As a reminder, while playing with the FBIOPAN_DISPLAY ioctl I noticed
that many drivers use information from the ioctl argument such as the
display resolution when they should use the current settings from the
fb_info structure.

These 29 patches fix most of those drivers. Missing drivers are ivtv (for
which a patch has Andy Walls has already pushed a patch),
sh_mobile_lcdcfb (for which the fix is a bit more complex, I will work
on it later) and viafb (for which Florian Tobias Schandinat has already
sent a patch).

The patches have been compile-tested on architectures for which I have a
working compiler (x86 and ARM) with allyesconfig. I don't own any
hardware supported by these drivers so I haven't been able to perform
additional tests.

I've split the fixes in one patch per driver for easier review and
handling of potential problems. The patches can be squashed together if
needed. I've tried to CC individual driver maintainers when possible,
and Acked-by lines received in reply to the first version have been
included.

Compared to v2, I've fixed a compile issue in the atmel_lcdfb driver and
added a couple of SoB lines.

Paul, is it time to push this for 3.1 ?

I have the v2 series queued up already, so it would be nice to simply
have the v2->v3 differences as incremental changes that can be piled on.

Ah. Next time please notify me when you queue the patches up. That was too 
soon. This will end up potentially breaking bisections, which is never good.

The signed-off-by lines are not that critical in this case since it's a
largely mechanical change all over the place, so it's primarily the build
fix(es?) that would be nice to have.

I would say they are more than "nice to have". Where have you queued v2 ?

-- 
Regards,

Laurent Pinchart

On Wednesday 13 July 2011 10:24:24 Paul Mundt wrote:

On Fri, Jul 08, 2011 at 01:14:14PM +0200, Laurent Pinchart wrote:

quoted

Hi everybody,

Here's the third version of the FBIOPAN_DISPLAY fixes patch series.

As a reminder, while playing with the FBIOPAN_DISPLAY ioctl I noticed
that many drivers use information from the ioctl argument such as the
display resolution when they should use the current settings from the
fb_info structure.

These 29 patches fix most of those drivers. Missing drivers are ivtv (for
which a patch has Andy Walls has already pushed a patch),
sh_mobile_lcdcfb (for which the fix is a bit more complex, I will work
on it later) and viafb (for which Florian Tobias Schandinat has already
sent a patch).

The patches have been compile-tested on architectures for which I have a
working compiler (x86 and ARM) with allyesconfig. I don't own any
hardware supported by these drivers so I haven't been able to perform
additional tests.

I've split the fixes in one patch per driver for easier review and
handling of potential problems. The patches can be squashed together if
needed. I've tried to CC individual driver maintainers when possible,
and Acked-by lines received in reply to the first version have been
included.

Compared to v2, I've fixed a compile issue in the atmel_lcdfb driver and
added a couple of SoB lines.

Paul, is it time to push this for 3.1 ?

I have the v2 series queued up already, so it would be nice to simply
have the v2->v3 differences as incremental changes that can be piled on.
The signed-off-by lines are not that critical in this case since it's a
largely mechanical change all over the place, so it's primarily the build
fix(es?) that would be nice to have.

I've sent you a fix on top of your v2 series a while ago, and none of these 
patches made it to v3.1-rc1.

Is fbdev maintained at all ?

-- 
Regards,

Laurent Pinchart

Hi Laurent, Paul,

On 08/09/2011 12:35 PM, Laurent Pinchart wrote:

On Wednesday 13 July 2011 10:24:24 Paul Mundt wrote:

quoted

On Fri, Jul 08, 2011 at 01:14:14PM +0200, Laurent Pinchart wrote:

quoted

Hi everybody,

Here's the third version of the FBIOPAN_DISPLAY fixes patch series.

As a reminder, while playing with the FBIOPAN_DISPLAY ioctl I noticed
that many drivers use information from the ioctl argument such as the
display resolution when they should use the current settings from the
fb_info structure.

These 29 patches fix most of those drivers. Missing drivers are ivtv (for
which a patch has Andy Walls has already pushed a patch),
sh_mobile_lcdcfb (for which the fix is a bit more complex, I will work
on it later) and viafb (for which Florian Tobias Schandinat has already
sent a patch).

The patches have been compile-tested on architectures for which I have a
working compiler (x86 and ARM) with allyesconfig. I don't own any
hardware supported by these drivers so I haven't been able to perform
additional tests.

I've split the fixes in one patch per driver for easier review and
handling of potential problems. The patches can be squashed together if
needed. I've tried to CC individual driver maintainers when possible,
and Acked-by lines received in reply to the first version have been
included.

Compared to v2, I've fixed a compile issue in the atmel_lcdfb driver and
added a couple of SoB lines.

Paul, is it time to push this for 3.1 ?

I have the v2 series queued up already, so it would be nice to simply
have the v2->v3 differences as incremental changes that can be piled on.
The signed-off-by lines are not that critical in this case since it's a
largely mechanical change all over the place, so it's primarily the build
fix(es?) that would be nice to have.

I've sent you a fix on top of your v2 series a while ago, and none of these

3 more quoted lines

Yes, it is :)

Well I already wrote to Paul because of this and got a response. Basically he 
had to do other stuff/trees and felt that everything could go in after rc1 as well.
But if the situation persists that he has too much other stuff to do and no time 
left for fbdev I'd be willing to take over maintaining the fbdev stuff.
So you may rest assured that your work will not be lost and at worst be merged 
in the next merge window.


Best regards,

Florian Tobias Schandinat

Hi Florian,

On Wednesday 10 August 2011 23:01:59 Florian Tobias Schandinat wrote:

On 08/09/2011 12:35 PM, Laurent Pinchart wrote:

quoted

On Wednesday 13 July 2011 10:24:24 Paul Mundt wrote:

quoted

On Fri, Jul 08, 2011 at 01:14:14PM +0200, Laurent Pinchart wrote:

quoted

Hi everybody,

Here's the third version of the FBIOPAN_DISPLAY fixes patch series.

As a reminder, while playing with the FBIOPAN_DISPLAY ioctl I noticed
that many drivers use information from the ioctl argument such as the
display resolution when they should use the current settings from the
fb_info structure.

These 29 patches fix most of those drivers. Missing drivers are ivtv
(for which a patch has Andy Walls has already pushed a patch),
sh_mobile_lcdcfb (for which the fix is a bit more complex, I will work
on it later) and viafb (for which Florian Tobias Schandinat has already
sent a patch).

The patches have been compile-tested on architectures for which I have
a working compiler (x86 and ARM) with allyesconfig. I don't own any
hardware supported by these drivers so I haven't been able to perform
additional tests.

I've split the fixes in one patch per driver for easier review and
handling of potential problems. The patches can be squashed together if
needed. I've tried to CC individual driver maintainers when possible,
and Acked-by lines received in reply to the first version have been
included.

Compared to v2, I've fixed a compile issue in the atmel_lcdfb driver
and added a couple of SoB lines.

Paul, is it time to push this for 3.1 ?

I have the v2 series queued up already, so it would be nice to simply
have the v2->v3 differences as incremental changes that can be piled on.
The signed-off-by lines are not that critical in this case since it's a
largely mechanical change all over the place, so it's primarily the
build fix(es?) that would be nice to have.

6 more quoted lines

Great :-)

Well I already wrote to Paul because of this and got a response. Basically
he had to do other stuff/trees and felt that everything could go in after
rc1 as well.

I know how being overwhelmed with work feels, so I'm not blaming Paul at all. 
As the last couple of e-mails I've sent him got no reply, and as the patches 
haven't been pushed to v3.1-rc1, I wanted to check if he was still alive :-) 
(or if a technical issue was preventing his replies for reaching me).

But if the situation persists that he has too much other stuff to do and no
time left for fbdev I'd be willing to take over maintaining the fbdev stuff.
So you may rest assured that your work will not be lost and at worst be
merged in the next merge window.

A delay of one merge window is OK in this case. I'm quite new to fbdev 
development, and still trying to find out how this subsystem is maintained 
(including how fast patches should be pushed, and how to submit urgent fixes 
when required).

-- 
Regards,

Laurent Pinchart