Control: severity -1 important Control: tags -1 + moreinfo Hi,
On Sun, Apr 02, 2023 at 09:56:52PM -0400, A. F. Cano wrote: > Package: src:linux > Version: 6.1.20-1 > Severity: critical > File: nouveau > Justification: breaks the whole system > X-Debbugs-Cc: af...@comcast.net > > When the above message occurs, the system becomes totally unresponsive and > the only way to recover is > a hard power-off via the power button held for about 5 seconds. Upon boot, > the sddm login screen appears > but at 1024x768, which is much less than the monitor is capable of: 1920x1200. > > xrandr > Screen 0: minimum 16 x 16, current 1024 x 768, maximum 32767 x 32767 > XWAYLAND0 connected primary 1024x768+0+0 (normal left inverted right x axis y > axis) 0mm x 0mm > 1024x768 59.92*+ > 800x600 59.86 > 640x480 59.38 > 320x240 59.52 > 720x480 59.71 > 640x400 59.95 > 320x200 58.96 > 1024x576 59.90 > 864x486 59.92 > 720x400 59.55 > 640x350 59.77 > > After login sometimes the screen goes blank (but the backlight remains on). > Hard power off required. > Sometimes the gear wheel stops turning and the system freezes. Hard power > off required. > > I have tried to install the nvidia proprietary driver 304 > (NVIDIA-Linux-x86_64-304.117.run) which is what > this old chip needs but it fails to install. No matter what I do the nouveau > driver is in use and > cannot be removed. > > If it were possible for nouveau and/or X/Wayland to access the whole set of > resolutions of the system > without hard freezes, I'd be happy. Any tricks? Any specific things I could > try to figure out the issue? > > Obviously, in this particular boot the hard freeze did not happen. > > These lines seem to be relevant (from the logs below): > > [ 47.892652] nouveau 0000:00:0 > d.0: bus: MMIO write of 00340001 FAULT at 00b000 > [ 64.113759] nouveau 0000:00:0d.0: bus: MMIO write of 00640001 FAULT at > 00b010 > [ 64.114792] nouveau 0000:00:0d.0: bus: MMIO write of 00310001 FAULT at > 00b020 > [ 69.614326] nouveau 0000:00:0d.0: bus: MMIO write of 00000000 FAULT at > 00b020 > [ 69.614542] nouveau 0000:00:0d.0: bus: MMIO write of 00640001 FAULT at > 00b010 > [ 69.615432] nouveau 0000:00:0d.0: bus: MMIO write of 00310001 FAULT at > 00b020 > [ 70.336843] nouveau 0000:00:0d.0: bus: MMIO write of 00000000 FAULT at > 00b020 > [ 70.337057] nouveau 0000:00:0d.0: bus: MMIO write of 00640001 FAULT at > 00b010 > [ 70.337684] nouveau 0000:00:0d.0: bus: MMIO write of 00660001 FAULT at > 00b020 > [ 70.357387] nouveau 0000:00:0d.0: bus: MMIO write of 00000000 FAULT at > 00b010 > [ 89.666120] nouveau 0000:00:0d.0: bus: MMIO write of 00ca0001 FAULT at > 00b010 > [ 97.330127] nouveau 0000:00:0d.0: bus: MMIO write of 00000000 FAULT at > 00b010 > [ 104.590842] traps: light-locker[4745] trap int3 ip:7f59b65be7d7 > sp:7fff472f8690 error:0 in libglib-2.0.so.0.7400.6[7f59b6580000+8d000] Can you clarify, is this a regression from 6.1.15-1 previously in testing, and now happening first with 6.1.20-1? Looking for reports about the same and similar effects, it looks issues with nouveau and the old eForce 6150SE nForce 430 goes way back several years. Can you please clarify if this is a new regression though between 6.1.15 and 6.1.20. Regards, Salvatore
