X windows freeze with Mobility Radeon X1400 on Xubuntu 16.04 when returning from blank screen

[Erik Brangs]

Hi,

I'm using Xubuntu 16.04 on my Lenovo Thinkpad T60. It has a Radeon graphics card:

01:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] RV515/M54 [Mobility Radeon X1400]

When I try to use the laptop again after the screen has blanked, XFCE freezes from time to time after restoring the previous screen contents. I can still move the mouse cursor but clicking has no effect. I also can't switch to a terminal. Login via SSH still works.

The only thing that I've been able to find out is that the chance for the freeze to occur seems to depend on the kernel version. With the 4.0.x series, the chance was rather low (less than 1 in 20). With the 4.2.x and 4.4.x series, the chance for a freeze to occur seems to be much higher. The latest kernel that I tested was 4.6-rc4 from the Ubuntu mainline crack repository ( http://kernel.ubuntu.com/~kernel-ppa/mainline/?C=N;O=D ) which also seems to have a high chance for a freeze to occur.

dmesg: http://www.hastebin.com/kagifawete.vhdl
Xorg.log.0: http://hastebin.com/dijajetajo.vhdl

I'm not sure if it's related but I also saw a freeze where the screen didn't reappear before freezing (with 4.6-rc4). In this case, I hadn't used the laptop for some time and the whole dmesg output consisted of messages like:

[18883.528089] radeon 0000:01:00.0: ring 0 stalled for more than 10488msec
[18883.528101] radeon 0000:01:00.0: GPU lockup (current fence id 0x00000000000004f3 last fence id 0x00000000000005d5 on ring 0)
[18883.532747] radeon 0000:01:00.0: Saved 245019 dwords of commands on ring 0.
[18883.532761] radeon 0000:01:00.0: GPU reset succeeded, trying to resume
[18883.532774] radeon 0000:01:00.0: ffff880034ddc000 unpin not necessary
[18883.542676] [drm] radeon: 1 quad pipes, 1 z pipes initialized.
[18883.545556] [drm] PCIE GART of 512M enabled (table at 0x0000000000040000).
[18883.545563] radeon 0000:01:00.0: WB enabled
[18883.545570] radeon 0000:01:00.0: fence driver on ring 0 use gpu addr 0x0000000008000000 and cpu addr 0xffff880034dca000
[18883.545640] [drm] radeon: ring at 0x0000000008001000
[18884.201308] [drm:r100_ring_test [radeon]] *ERROR* radeon: ring test failed (scratch(0x15E8)=0xCAFEDEAD)
[18884.201337] [drm:r100_cp_init [radeon]] *ERROR* radeon: cp isn't working (-22).
[18884.201342] radeon 0000:01:00.0: failed initializing CP (-22).

I'm not sure what I can do to investigate this. Ubuntu provides some information on X freezes at https://wiki.ubuntu.com/X/Troubleshooting/Freeze . I tried using xset but it doesn't currently seem to help in reproducing the problem.


Kind regards,

Erik Brangs