Bug#649448: radeon (evergreen): random-looking pattern of pixels when firmware not installed
Hey there,
I ran into the same issue -- and I didn't see a warning either. Installing non-free
firmware fixed it for me (thanks to great help in #debian-next).
Output from reportbug:
-- Package-specific info:
** Version:
Linux version 3.2.0-3-amd64 (Debian 3.2.23-1) (debian-kernel@lists.debian.org) (gcc version 4.6.3 (Debian 4.6.3-8) ) #1 SMP Mon Jul 23 02:45:17 UTC 20
** Command line:
BOOT_IMAGE=/vmlinuz-3.2.0-3-amd64 root=/dev/mapper/debian-root ro quiet
** Not tainted
** Kernel log:
[ 14.061938] iTCO_vendor_support: vendor-support=0
[ 14.070484] iTCO_wdt: Intel TCO WatchDog Timer Driver v1.07
[ 14.070541] iTCO_wdt: Found a H57 TCO device (Version=2, TCOBASE=0x0860)
[ 14.070605] iTCO_wdt: initialized. heartbeat=30 sec (nowayout=0)
[ 14.133441] EDAC MC: Ver: 2.1.0
[ 14.164462] EDAC MC0: Giving out device to 'i7core_edac.c' 'i7 core #0': DEV 0000:ff:03.0
[ 14.164490] EDAC PCI0: Giving out device to module 'i7core_edac' controller 'EDAC PCI controller': DEV '0000:ff:03.0' (POLLED)
[ 14.164499] EDAC i7core: Driver loaded, 1 memory controller(s) found.
[ 14.218000] input: HDA Intel Headphone as /devices/pci0000:00/0000:00:1b.0/sound/card0/input5
[ 14.226474] [drm] Initialized drm 1.1.0 20060810
[ 14.454138] [drm] radeon kernel modesetting enabled.
[ 14.454235] radeon 0000:01:00.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16
[ 14.454241] radeon 0000:01:00.0: setting latency timer to 64
On Mon, Apr 23, 2012 at 07:42:41PM +0100, Ben Hutchings wrote:
> On Mon, Apr 23, 2012 at 09:21:50PM +0300, Touko Korpela wrote:
> > Installer has also a bug report about this (#661069).
> >
> > It would be good if Debian kernel radeon driver learned to fall back
> > sensibly if running on hardware where firmware is required.
>
> The upstream radeon developers insist that the driver tries to do
> this, but in practice it doesn't seem to work for most chips. I am
> inclined to make radeon's probe function fail if the firmware is
> missing, which should result in using a fallback console driver.
>
> > Also installer should warn user about what happens if firmware is missing.
>
> I thought it did...?
>
> Ben.
>
> --
> Ben Hutchings
> We get into the habit of living before acquiring the habit of thinking.
> - Albert Camus
>
Reply to: