Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1604422 > unrolled thread

Re: [PATCHv2 00/10] Nokia N950 basic display support

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: [PATCHv2 00/10] Nokia N950 basic display support Tomi Valkeinen <tomi.valkeinen@ti.com> - 2017-03-20 12:40 +0100
    Re: [PATCHv2 00/10] Nokia N950 basic display support Tomi Valkeinen <tomi.valkeinen@ti.com> - 2017-03-21 11:00 +0100
    Re: [PATCHv2 00/10] Nokia N950 basic display support Tomi Valkeinen <tomi.valkeinen@ti.com> - 2017-03-21 12:50 +0100

#1604422 — Re: [PATCHv2 00/10] Nokia N950 basic display support

[Multipart message — attachments visible in raw view] — view raw

Hi,

On 05/03/17 01:50, Sebastian Reichel wrote:
> Hi,
> 
> Some of you may remember, that I sent a series for the N950 display
> some time ago. N950 has command mode DSI panel, so the main part of
> the patchset takes care of adding manual display update support in
> omapdrm.
> 
> The N950 also requires display rotation (the panel is mounted vertically
> and bottom-up) and offset. The required bits will be sent separately.
> 
> The patchset is based on 2d62e0768d3c, which is the current commit
> torvald's master branch points to. I tested the patches on N950
> with kernel console (fbcon), Tomi's kmstest and Xorg from Debian sid.
> 
> Rough changelog, most of that work was done by Tony (thanks!)
> 
>  * lots of patches dropped for now
>  * rebased to current omapdrm interface
>  * added OMAP4 support
>  * misc. cleanup

I haven't been able to go through this series with time, but looking at
the patches, there were multiple places where I felt quite uneasy with
the changes. It didn't help that many of the patches did not have
descriptions.

So as an RFC series, it's quite nice and shows that the manual update
can be gotten to work. But I'm not ready to apply this until the patches
touching the omapdrm enable, disable and commit sequences are split up
into smaller patches with proper descriptions.

Also, a minor note that rotating the display on N950 will cause tearing.
Probably not a huge issue, though.

 Tomi

[toc] | [next] | [standalone]

#1605479

[Multipart message — attachments visible in raw view] — view raw

On 21/03/17 11:38, Tomi Valkeinen wrote:
> On 20/03/17 13:29, Tomi Valkeinen wrote:
>> Hi,
>>
>> On 05/03/17 01:50, Sebastian Reichel wrote:
>>> Hi,
>>>
>>> Some of you may remember, that I sent a series for the N950 display
>>> some time ago. N950 has command mode DSI panel, so the main part of
>>> the patchset takes care of adding manual display update support in
>>> omapdrm.
>>>
>>> The N950 also requires display rotation (the panel is mounted vertically
>>> and bottom-up) and offset. The required bits will be sent separately.
>>>
>>> The patchset is based on 2d62e0768d3c, which is the current commit
>>> torvald's master branch points to. I tested the patches on N950
>>> with kernel console (fbcon), Tomi's kmstest and Xorg from Debian sid.
>>>
>>> Rough changelog, most of that work was done by Tony (thanks!)
>>>
>>>  * lots of patches dropped for now
>>>  * rebased to current omapdrm interface
>>>  * added OMAP4 support
>>>  * misc. cleanup
>>
>> I haven't been able to go through this series with time, but looking at
>> the patches, there were multiple places where I felt quite uneasy with
>> the changes. It didn't help that many of the patches did not have
>> descriptions.
>>
>> So as an RFC series, it's quite nice and shows that the manual update
>> can be gotten to work. But I'm not ready to apply this until the patches
>> touching the omapdrm enable, disable and commit sequences are split up
>> into smaller patches with proper descriptions.
> 
> I made a quick test on my OMAP4 SDP (it has similar panel to N950), and
> I did get a picture on the LCD, and kmstest --flip works. It's tearing,
> though, and the fps is ~80, so apparently TE signal is not used for
> synchronization here. I think nobody is calling the enable_te().

Oh, and I also see these.

[   42.550170] omapdrm omapdrm.0: lcd: timeout waiting for disable
[   42.690216] omapdrm omapdrm.0: lcd2: timeout waiting for disable

[  104.280212] omapdrm omapdrm.0: lcd: timeout waiting for enable

 Tomi

[toc] | [prev] | [next] | [standalone]

#1605537

[Multipart message — attachments visible in raw view] — view raw

On 20/03/17 13:29, Tomi Valkeinen wrote:
> Hi,
> 
> On 05/03/17 01:50, Sebastian Reichel wrote:
>> Hi,
>>
>> Some of you may remember, that I sent a series for the N950 display
>> some time ago. N950 has command mode DSI panel, so the main part of
>> the patchset takes care of adding manual display update support in
>> omapdrm.
>>
>> The N950 also requires display rotation (the panel is mounted vertically
>> and bottom-up) and offset. The required bits will be sent separately.
>>
>> The patchset is based on 2d62e0768d3c, which is the current commit
>> torvald's master branch points to. I tested the patches on N950
>> with kernel console (fbcon), Tomi's kmstest and Xorg from Debian sid.
>>
>> Rough changelog, most of that work was done by Tony (thanks!)
>>
>>  * lots of patches dropped for now
>>  * rebased to current omapdrm interface
>>  * added OMAP4 support
>>  * misc. cleanup
> 
> I haven't been able to go through this series with time, but looking at
> the patches, there were multiple places where I felt quite uneasy with
> the changes. It didn't help that many of the patches did not have
> descriptions.
> 
> So as an RFC series, it's quite nice and shows that the manual update
> can be gotten to work. But I'm not ready to apply this until the patches
> touching the omapdrm enable, disable and commit sequences are split up
> into smaller patches with proper descriptions.

I made a quick test on my OMAP4 SDP (it has similar panel to N950), and
I did get a picture on the LCD, and kmstest --flip works. It's tearing,
though, and the fps is ~80, so apparently TE signal is not used for
synchronization here. I think nobody is calling the enable_te().

 Tomi

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web