Submitted by ernsteiswuerfel @ernsteiswuerfel

Assigned to xf86-video-ati maintainers

Link to original bug (#103147)

Description

Created attachment 134739
kernel log

I know quite a few r300 cards have issues on AGP PPC-hardware, but this is a PCIe card so I decided to post a bug report nevertheless.

It's a Radeon X600 running in a PowerMac G5 11,2 (PCIe) on kernel 4.13.5, xorg-server-1.19.4, libdrm-2.4.83, xf86-video-ati-7.10.0. Kernel log shows no error initializing the card (see dmesg.txt), but when starting Xorg the machine freezes not later than a few secs. It's a complete freeze, accessing the machine via ssh is no longer possible after the freeze happened:

[ 280.815899] radeon 0001:06:00.0: vgaarb: changed VGA decodes: olddecodes=io+mem,decodes=none:owns=none
[ 294.948188] radeon 0001:06:00.0: ring 0 stalled for more than 10133msec
[ 294.948202] radeon 0001:06:00.0: GPU lockup (current fence id 0x0000000000000002 last fence id 0x0000000000000003 on ring 0)
[ 295.181717] radeon: wait for empty RBBM fifo failed! Bad things might happen.
[ 295.412441] Failed to wait GUI idle while programming pipes. Bad things might happen.
[ 295.413597] radeon 0001:06:00.0: Saved 27 dwords of commands on ring 0.
[ 295.413618] radeon 0001:06:00.0: (r300_asic_reset:422) RBBM_STATUS=0x8802C12F
[ 295.913610] radeon 0001:06:00.0: (r300_asic_reset:441) RBBM_STATUS=0x8802C12F
[ 296.409606] radeon 0001:06:00.0: (r300_asic_reset:453) RBBM_STATUS=0x8802C12F
[ 296.409647] radeon 0001:06:00.0: GPU reset succeed
[ 296.409651] radeon 0001:06:00.0: GPU reset succeeded, trying to resume
[ 296.410230] radeon 0001:06:00.0: c000000458090000 unpin not necessary
[ 296.410251] radeon 0001:06:00.0: (r300_asic_reset:422) RBBM_STATUS=0x8802C12F
[ 296.910240] radeon 0001:06:00.0: (r300_asic_reset:441) RBBM_STATUS=0x8802C12F

Attachment 134739, "kernel log":
dmesg.txt