Re: [PATCH] drm/radeon/kms: Fix oops when set_base is call with no FB

[Andrew M. <ak...@li...>]

On Thu, 12 Nov 2009 12:02:58 +0100
Jerome Glisse <jg...@re...> wrote:

> On Tue, 2009-11-10 at 14:30 -0800, Andrew Morton wrote:
> > On Wed,  4 Nov 2009 20:03:19 +0100
> > Jerome Glisse <jg...@re...> wrote:
> > 
> > > Just do nothings crct_set_base i call with no FB.
> > > 
> > 
> > hmpf.  It's obvious that you spent hours carefully describing this
> > patch for us.
> > 
> 
> Sorry, truth is i don't understand why crtc set_base call back
> can be call with a null fb, i did just replicate what intel kms
> and other part of radeon kms was doing in front of such situation.
> It should go down to 2.6.31, useless before as there is no KMS
> for radeon in earlier version. The oops will happen when user
> switch btw X & vt or in some case when changing mode.
> 
> Will clearly state my ignorance in future patch.
> 

I added a bit more waffle to the changelog.

Is a copy of this oops available anywhere?  Seeing the backtrace would
help people work out what the bug is.