Re: [PATCH] CHROMIUM: i915: Initialize panel timing registers if VBIOS did not.

Re: [PATCH] CHROMIUM: i915: Initialize panel timing registers if VBIOS did not.

[Chris Wilson]

On Thu, 7 Oct 2010 15:48:14 -0700, Bryan Freed <bfreed@chromium.org> wrote:
> The time between start of the pixel clock and backlight enable is a basic
> panel timing constraint.  If no VBIOS Table is found, and the Panel Power
> On/Off registers are found to be 0, assume we are booting without VBIOS
> initialization and set these registers to something reasonable.

IIRC, the panel sequence registers are meant to be stored in the VBIOS. So
if we add the parsing of those to the driver and add the defaults to
init_vbt_default() then we can check whether PP_ON_DELAYS is valid upon
device init (module load and resume) and fixup in case the BIOS does not.
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre

Re: [PATCH] CHROMIUM: i915: Initialize panel timing registers if VBIOS did not.

[Jesse Barnes]

On Thu, 07 Oct 2010 23:55:38 +0100
Chris Wilson <chris@chris-wilson.co.uk> wrote:

> On Thu, 7 Oct 2010 15:48:14 -0700, Bryan Freed <bfreed@chromium.org> wrote:
> > The time between start of the pixel clock and backlight enable is a basic
> > panel timing constraint.  If no VBIOS Table is found, and the Panel Power
> > On/Off registers are found to be 0, assume we are booting without VBIOS
> > initialization and set these registers to something reasonable.
> 
> IIRC, the panel sequence registers are meant to be stored in the VBIOS. So
> if we add the parsing of those to the driver and add the defaults to
> init_vbt_default() then we can check whether PP_ON_DELAYS is valid upon
> device init (module load and resume) and fixup in case the BIOS does not.

Right, those values should be in the VBT.  Getting them wrong can
actually damage your panel, so we need to take care here.  The whole
point of the power sequencing logic and register lock is to avoid such
damage.

That said, I've abused my panels (both LVDS and eDP) pretty hard and
have yet to damage one in any visible way, so stealing some
conservative (i.e. long) delays from an existing VBT will probably work
ok.

-- 
Jesse Barnes, Intel Open Source Technology Center

Re: [PATCH] CHROMIUM: i915: Initialize panel timing registers if VBIOS did not.

[Bryan Freed]

[-- Attachment #1.1: Type: text/plain, Size: 1178 bytes --]

My change tries to detect the lack of initialization by A) finding no VBT,
and B) finding 0 values in these registers.

But what if there is a VBIOS out there that really wants these values to be
0?  I provide for that case by checking for VBT.

Is this a reasonable case?  If not, I have no problem moving the check
to init_vbt_default().

bryan.

On Thu, Oct 7, 2010 at 3:55 PM, Chris Wilson <chris@chris-wilson.co.uk>wrote:

> On Thu, 7 Oct 2010 15:48:14 -0700, Bryan Freed <bfreed@chromium.org>
> wrote:
> > The time between start of the pixel clock and backlight enable is a basic
> > panel timing constraint.  If no VBIOS Table is found, and the Panel Power
> > On/Off registers are found to be 0, assume we are booting without VBIOS
> > initialization and set these registers to something reasonable.
>
> IIRC, the panel sequence registers are meant to be stored in the VBIOS. So
> if we add the parsing of those to the driver and add the defaults to
> init_vbt_default() then we can check whether PP_ON_DELAYS is valid upon
> device init (module load and resume) and fixup in case the BIOS does not.
> -Chris
>
> --
> Chris Wilson, Intel Open Source Technology Centre
>

[-- Attachment #1.2: Type: text/html, Size: 1661 bytes --]

[-- Attachment #2: Type: text/plain, Size: 159 bytes --]

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH] CHROMIUM: i915: Initialize panel timing registers if VBIOS did not.

[Jesse Barnes]

I don't think 0 is a reasonable value for any of those fields, so
checking them against 0 should be fine.

Jesse

On Thu, 7 Oct 2010 16:48:51 -0700
Bryan Freed <bfreed@chromium.org> wrote:

> My change tries to detect the lack of initialization by A) finding no VBT,
> and B) finding 0 values in these registers.
> 
> But what if there is a VBIOS out there that really wants these values to be
> 0?  I provide for that case by checking for VBT.
> 
> Is this a reasonable case?  If not, I have no problem moving the check
> to init_vbt_default().
> 
> bryan.
> 
> On Thu, Oct 7, 2010 at 3:55 PM, Chris Wilson <chris@chris-wilson.co.uk>wrote:
> 
> > On Thu, 7 Oct 2010 15:48:14 -0700, Bryan Freed <bfreed@chromium.org>
> > wrote:
> > > The time between start of the pixel clock and backlight enable is a basic
> > > panel timing constraint.  If no VBIOS Table is found, and the Panel Power
> > > On/Off registers are found to be 0, assume we are booting without VBIOS
> > > initialization and set these registers to something reasonable.
> >
> > IIRC, the panel sequence registers are meant to be stored in the VBIOS. So
> > if we add the parsing of those to the driver and add the defaults to
> > init_vbt_default() then we can check whether PP_ON_DELAYS is valid upon
> > device init (module load and resume) and fixup in case the BIOS does not.
> > -Chris
> >
> > --
> > Chris Wilson, Intel Open Source Technology Centre
> >


-- 
Jesse Barnes, Intel Open Source Technology Center