[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#268352: 2.6.* radeonfb regression



On Mon, Sep 13, 2004 at 11:44:55AM +0200, Sven Luther wrote:
> On Mon, Sep 13, 2004 at 10:44:32AM +0200, Francesco P. Lovergine wrote:
> > On Sun, Sep 12, 2004 at 09:00:19AM +0200, Sven Luther wrote:
> > > 
> > > > I'm having the same problem on a HP Pavilion zt3020EA, which has an ATI
> > > > Radeon 9000 chipset. Native res is 1280x800. Again blank screen with
> > > > cursor moving on it, and all working fine in 2.4 or X. The problem apparently
> > > > is the same so I'm sending a followup for that.
> > > 
> > > Mmm, Can you clarify if it works fine in 1280x800 mode or not ? I have some
> > > doubts about the wisdom of using an LCD with bigger than native resolution.
> > > 
> > 
> > That's the native resolution and looks very good. This is one of those
> > new displays in 16/9 ratio, used for video playing. As well as 1400x1050
> > is the native resolution for the other box which has a SXVGA unit.
> > All resolutions are stolen from those detected and used under 2.4.
> > The old kernel starts directly in the high resolution mode.
> > BTW, I'm using the same resoltion under X which also appears as the best
> > one for font and textures rendering. 
> 
> Ok, so you have two laptops, one with 1400x1050 native resolution and the
> other with 1280x800 native resolution, exact ?
> 

Right, and I know personally at least other 2-3 people with the same
problem, all with different Compaq/HP laptops.

> In both case, running the fbdev in native resolution worked for 2.4 radeonfb,
> but not for 2.6 radeonfb, and in both case X is working fine.
> 

Right too.

> Furthermore, your laptops displays are not-DDC capable, and thus new_radeonfb
> is doing strange things with monitor timings, and is failing to setup your
> native panel mode.
> 

Definitively yes.

> Benh, any insight on this ? 
> 

I hope so :) Using vesafb with ordinary (non native) resolutions is not so great...

-- 
Francesco P. Lovergine



Reply to: