GPU crash RX6800
Been having some weird issues recently, and I haven't fully figured out what's been causing it,
Its been manifesting as the display going black, audio still running, but no input possible. switching to tty did not work
Had to hard powerdown and start up again to get a working system again
But I noticed this in the journalctl just now
Aug 18 21:31:04 helios kwin_wayland[4101]: kwin_wayland_drm: Pageflip timed out! This is a bug in the amdgpu kernel driver
Aug 18 21:31:04 helios kwin_wayland[4101]: kwin_wayland_drm: Please report this at https://gitlab.freedesktop.org/drm/amd/-/issues
Aug 18 21:31:04 helios kwin_wayland[4101]: kwin_wayland_drm: With the output of 'sudo dmesg' and 'journalctl --user-unit plasma-kwin_wayland --boot 0'
11 Replies
https://gitlab.freedesktop.org/drm/amd/-/issues/4241
That seems possiblty related
Someone in that thread mentioned it possibly being a bad cable, so am trying that out
Given the issue has been incredibly sporadic, might be a while for it to happen again
I have this issue too on a 9070XT. Im not at my pc anymore but can check journalctl next.time it happens. It did happen like twice in a row though.
Its a weird one alright, I've had 4 hour long game sessions with not a single issue at all, and onces where It'll happen 4 times in 2 hours time
Are you OCing your gpu by any chance? Or undervolting/using lact
Completely stock
Its a ROG TUF RX6800
Interesting. I also see a ton of posts aboht pageflip timed out on the gitlab issue tracker. We dont seem to be the only ones
I suspect there's a lot more that aren't technical enough to troubleshoot it down to this
had a similar problem and it was my ram. it was not faulty but after a bios update my board did for some reason no longer supported the clockspeed of the ram, it worked for good after i downclocked the ram to 3200mhz. but i dont know of course if its something like that in your case
I havent updated my bios recently 🤔
Ill go and check for updates anyway
I'm still evaluating this as an option, last time it happened my memory was too hot to hold when taken out, which definitely can't have been good
Well fck, I just replicated the same issue in a fresh Windows 10 install
Gpu is only just within warranty, so am sending it in for an rma
Gl
I didnt have a crash yesterday.. I did update LACT so maybe
So my pc just crashed again. And it looks like its not booting due to ram issues...
So im not entirely sure if this was related but my cpu died. Im suspecting it might have been cause i was running 4 sticks of ram 6000 mhz cl30 which everyone says is unstable, but ive gone to just 2 sticks now
And another pageflip crash. Im lost