Bug#672644: X on Mobility Radeon HD 3400 Series (Lenovo T400) broken after wheezy upgrade

Michel Dänzer daenzer at debian.org
Tue May 15 00:31:26 PDT 2012


On Mon, 2012-05-14 at 21:19 +0200, Josip Rodin wrote: 
> On Mon, May 14, 2012 at 12:33:08PM +0200, Michel Dänzer wrote:
> > On Mon, 2012-05-14 at 10:56 +0200, Josip Rodin wrote: 
> > > On Sun, May 13, 2012 at 11:00:46PM +0200, Julien Cristau wrote:
> > > > > In any case, I rebooted using SysRq+S+U+B (I think, at least - couldn't see
> > > > > anything) and installed firmware-linux-nonfree, rebooted once again for good
> > > > > measure, and tried startx again, and now it just aborts X saying:
> > > > > 
> > > > > [   137.038] (EE) RADEON(0): [dri] RADEONDRIGetVersion failed because of a version mismatch.
> > > > > [dri] This chipset requires a kernel module version of 1.17.0,
> > > > > [dri] but the kernel reports a version of 2.12.0.
> > > > > [dri] Make sure your module is loaded prior to starting X, and
> > > > > [dri] that this driver was built with support for KMS.
> > > > > [dri] Aborting.
> > 
> > I'm afraid that's as good as it gets. When using a display manager, it
> > should automatically try starting the X server again, which should
> > succeed.
> 
> What about the first upgrade issue, the lack of firmware?

It's hard to say anything about that without seeing at least the dmesg
output and Xorg.0.log corresponding to the problem.


> > > > Means your radeon kernel module is not being properly loaded by udev.
> > > > You need to figure out why.
> > > 
> > > There's nothing strange about it, at least none that I can see.
> > > You can see in the attached dmesg output that it worked fine.
> > 
> > dmesg can't tell you why the kernel module wasn't loaded by udev.
> 
> But it *was* loaded by udev.

If it was, there wouldn't have been a problem.

> I don't run 'modprobe radeon' at any point between the two tests.
> After the broken startx, it's loaded, which is why
> it is in the dmesg output I sent you.

It's loaded by the X driver on demand, which is necessary for UMS. The
problem is it's too late for KMS.


-- 
Earthling Michel Dänzer           |                   http://www.amd.com
Libre software enthusiast         |          Debian, X and DRI developer





More information about the xorg-driver-ati mailing list