[haiku-commits] Re: r41411 - in haiku/trunk: build/jam headers/private/graphics/radeon_hd src/add-ons/accelerants/radeon_hd src/add-ons/kernel/drivers/graphics/radeon_hd

  • From: Stephan Aßmus <superstippi@xxxxxx>
  • To: haiku-commits@xxxxxxxxxxxxx
  • Date: Tue, 10 May 2011 16:25:22 +0200

On 10.05.2011 15:40, Axel Dörfler wrote:
Stephan Aßmus<superstippi@xxxxxx>  wrote:
On 10.05.2011 09:20, Axel Dörfler wrote:
Alex von Gluck<kallisti5@xxxxxxxxxxx>   wrote:
The Radeon HD cards seem to inject the preferred mode lines of
the active monitor into their VESA EDID (the vesa driver doesn't
pick it up because it compares the found monitor timings to the VESA
acceptable ones)
That's just how EDID works, and it's actually the monitor which reports it
-- that's exactly what it was made for :-)
VESA doesn't support arbitrary mode lines which is why EDID doesn't always
help in choosing the right correct resolution.
Hm, from your reply I don't know if you maybe misunderstand Alexander.
He is saying that the monitor's preferred mode ends up in the VESA mode
list, but it is still not picked by our VESA driver.

I'm not sure how you can know what Alex really intended to say (that should 
probably be left to him ;-)), but that's not what he wrote: EDID != mode list, 
EDID is what comes from the monitor, the mode list is what comes from the card.

I know, and re-reading what he said, it is bogus in a literal way (no offense, Alex!). However, I believe what he intended to say is that the graphics card reads the EDID information from the monitor by itself, and injects the preferred mode found into its own VESA mode list. At least this is exactly what happens on my GTX 285. The problem that Alex seems to have observed, and which would be in line with my own observations, is that, for whatever reason, this mode is then masked from the available modes in the VESA driver. Alex says the reason is that the VESA driver filters the list against "known" (built into the VESA driver) modes, which I would find a bit strange, but I haven't read the code.

However, I have
also observed that Haiku did not boot in 1920x1080 when I hooked it up
to my TV via HDMI. Maybe I fell victim to the same shortcomming in our
VESA driver and it could have actually worked?

That can only be answered if you provide the EDID output of that monitor, as 
well as the mode list the VESA driver retrieves from your card. In any case, 
it's unlikely that this is a driver issue.

I have no clue what brought Alex to his conclusion, but if he is right, it would be a driver issue after all. One which would be worth fixing since it allows picking the correct native resolution with the VESA driver on graphics cards which exhibit this behaviour.

Best regards,
-Stephan

Other related posts: