[Xen-users] radeon R100 MC microcode problem

[Eric Lindsey <eslindsey@xxxxxxxxx>]

Hi guys, I noticed that at bootup my system was complaining about
missing radeon/r100_cp.bin file, so I tracked it down to Debian's
wheezy package  firmware-linux-nonfree which includes the file.  I
installed this package and now I'm getting all sorts of errors when I
try to start Xen 4.1 (also installed via Debian's packages):

Excerpt from dmesg:

[    6.073729] [drm] radeon kernel modesetting enabled.
[    6.073859] xen: registering gsi 19 triggering 0 polarity 1
[    6.073879] xen: --> pirq=19 -> irq=19 (gsi=19)
[    6.073894] radeon 0000:0e:0d.0: PCI INT A -> GSI 19 (level, low) -> IRQ 19
[    6.074188] [drm] initializing kernel modesetting (RV100
0x1002:0x515E 0x1028:0x01B1).
[    6.074226] [drm] register mmio base: 0xFC2D0000
[    6.074228] [drm] register mmio size: 65536
[    6.074497] radeon 0000:0e:0d.0: VRAM: 128M 0x00000000D8000000 -
0x00000000DFFFFFFF (16M used)
[    6.074500] radeon 0000:0e:0d.0: GTT: 512M 0x00000000B8000000 -
0x00000000D7FFFFFF
[    6.074507] [drm] Supports vblank timestamp caching Rev 1 (10.10.2010).
[    6.074509] [drm] Driver supports precise vblank timestamp query.
[    6.074552] [drm] radeon: irq initialized.
[    6.074555] [drm] Detected VRAM RAM=128M, BAR=128M
[    6.074557] [drm] RAM width 16bits DDR
[    6.074624] [TTM] Zone  kernel: Available graphics memory: 3570914 kiB.
[    6.074626] [TTM] Zone   dma32: Available graphics memory: 2097152 kiB.
[    6.074628] [TTM] Initializing pool allocator.
[    6.074651] [drm] radeon: 16M of VRAM memory ready
[    6.074653] [drm] radeon: 512M of GTT memory ready.
[    6.074670] [drm] GART: num cpu pages 131072, num gpu pages 131072
[    6.096437] [drm] PCIE GART of 512M enabled (table at 0x0000000010A00000).
[    6.096482] radeon 0000:0e:0d.0: WB disabled
[    6.096555] [drm] Loading R100 Microcode
[    6.206973] [drm] radeon: ring at 0x00000000B8001000
[    6.406309] [drm:r100_ring_test] *ERROR* radeon: ring test failed
(scratch(0x15E4)=0xCAFEDEAD)
[    6.406362] [drm:r100_cp_init] *ERROR* radeon: cp isn't working (-22).
[    6.406408] radeon 0000:0e:0d.0: failed initializing CP (-22).
[    6.406452] radeon 0000:0e:0d.0: Disabling GPU acceleration
[    6.600640] [drm:r100_cp_fini] *ERROR* Wait for CP idle timeout,
shutting down CP.
[    6.795013] Failed to wait GUI idle while programming pipes. Bad
things might happen.
[    6.795685] [drm] radeon: cp finalized
[    6.796284] [drm] Radeon Display Connectors
[    6.796286] [drm] Connector 0:
[    6.796287] [drm]   VGA
[    6.796289] [drm]   DDC: 0x60 0x60 0x60 0x60 0x60 0x60 0x60 0x60
[    6.796291] [drm]   Encoders:
[    6.796292] [drm]     CRT1: INTERNAL_DAC1
[    6.867859] [drm] fb mappable at 0xD8040000
[    6.867861] [drm] vram apper at 0xD8000000
[    6.867862] [drm] size 1310720
[    6.867864] [drm] fb depth is 8
[    6.867865] [drm]    pitch is 1280
[    6.867973] fbcon: radeondrmfb (fb0) is primary device
[    7.034522] Console: switching to colour frame buffer device 160x64
[    7.047186] fb0: radeondrmfb frame buffer device
[    7.047188] drm: registered panic notifier
[    7.047194] [drm] Initialized radeon 2.12.0 20080528 for
0000:0e:0d.0 on minor 0

Notice the "ring test failed" line.  I Googled that and apparently it
used to be a problem when trying to run a Radeon graphics card under
Xen a long time ago.  But I wasn't able to find any current
information about it or how to fix it.  This isn't a deal breaker for
me, because my server hardly needs hardware-accelerated graphics.  But
it would be nice not to be able to fix the problem so that all the
hardware is functioning as it should.  I suppose if nobody can help me
fix this I can always remove the CP microcode and just go back to the
complaint about the missing firmware file.

_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxx
http://lists.xen.org/xen-users