[Intel-gfx] [Patch 0/3] DRM/I915: Use the child device array to decide whether the given HDMI/DP should be initialzied

[ykzhao]

On Fri, 2009-09-18 at 03:20 +0800, Adam Jackson wrote:
> I wasn't subscribed at the time (else I'd follow up to the right
> message), but I was recently pointed to this patch series:
> 
> http://lists.freedesktop.org/archives/intel-gfx/2009-July/003634.html
> 
> It seems like it should do the right thing according to my manual dump
> of the VBT on some laptops I've got handy, though I've not tried it yet.
> Some comments though:
Thanks for your comments.
> 
> - HDMI_CLASS/INT_DP_CLASS and the associated PORT and DVO defines should
> really live in intel_bios.h next to the other device types.
Agree. They had better be defined in intel_bios.h together. 
> 
> - TV seems to be device type 1009, and LFP to be 1022, but those aren't
> in intel_bios.h.  Would be nice to resync that with the BIOS sources.
It is also ok to add their definition.(TV_TYPE, LFP_TYPE). But now the
child device array is not used to initialize the TV/LVDS. So it is
unnecessary to add their definitions.
> 
> - The capabilities field in child_device_config seems to be the only
> field there that's all of: interesting, sometimes non-zero, non-obvious,
> and undocumented.  I've seen values of 0, 0x08 and 0x20 so far; what do
> these bits mean?
This is related with the output device type. Sorry that I have no such info about
the definition of this field. In fact the info is not used by
the graphics driver.

thanks.

> 
> - ajax