You are not logged in.
Pages: 1
Just joined the forums today so feel free to roast me for lacking information i need below.
My setup has (currently) 5 displays (eventually 7 or 8), because of this i need to use a displayport splitter. Up until about 2 weeks ago this was working fine, then i updated and magically it stopped working. Now what are the symptoms you may ask? As it goes, kde settings sees the displays, and you can set the displays on the splitter just fine, however you would want to normally. Even xorg seems to see them and render them in place correctly. the monitors on the other hand don't get any signal 99.9% of the time. though once without being able to recreate it, the signal passed through to one of the 16:9 1080p monitors, but then it quickly went away after i tried to re-plug some of the other monitors.
I have gone through a bit of troubleshooting including and limited to: trying kde neon to see if it was working still or if the splitter died (the splitter is fine as kde neon worked just fine with it) resetting arch thinking it might be a bad setting somewhere id take like 20 hours to track down. (same issue still.) and as a last effort slowly going through each display individually to see if maybe one of them is just having some unusual issue with only arch or some new package. Also as a note i did try on both x11 and wayland both had issues.
hardware and software info:
asrock radeon rx 6600 (nonxt)
amd 5950x
msi b550-a pro
64 GB (2x32) ddr4 3200 (unknown cas latency as im not too sure how to check at runtime)
splitter: startech mstdp123dp
monitors off splitter: 1 hp l1710
monitors on splitter: 1x: dell e176fp, 2x koorui 22N1
all run monitors run via dp to vga (the dell and hp only have vga, and the koorui's have hdmi but ive ran into a minor issue with hdmi i work arround by using hdmi as little as possible)
kernel: 6.1.2-arch1-1
de: kde plasma (2.26.4)
xorg or wayland: typically xorg (the main theme i use breaks under wayland)
if anyone has any other questions on hardware or software or logs please let me know as i dont know what logs to provide as again, new to forums as a whole.
Offline
Post your xorg log, https://wiki.archlinux.org/title/Xorg#General and try the KTS kernel.
Also check your system journal for errors (typically from amdgpu or drm)
Offline
checked the journal (using -b so its not everything ever) pretty much nothing, only about maybe 30 lines, all seems to be from boot as its all tagged at the exact same time. heres the entire log since last reboot for ya https://pastebin.com/Cr9GGrQw
in the xorg log:
greped for AMDGPU: https://pastebin.com/XmjXsd4t
without grep: https://pastebin.com/xa3uhnbs
not really sure if its telling me all that much due to it spitting out all the modelines everywhere. the only thing i can see is the drmmode_do_ctc_dpms thing
As for the kts kernal Im assuming you may have meant LTS kernal as i cant find anything on the first page of google about a kts kernal so ima go try it and report back
Offline
And I assume you're talking about the LTS kernEl ![]()
But yes.
The journal doesn't say anything because it's only the first page in the pager. Try
sudo journalctl -b | curl -F 'file=@-' 0x0.stto upload it.
Also remove the xf86-video-amdgpu driver (you'll be using the modesetting driver on top of the amdgpu kernel module)
Offline
so lts kernal fixed it.
where would i make a report of the kernel breaking in this way? still pretty new to things like this breaking on linux.
Offline
will remove fx86 as im now seeing a lockup whenever the system shuts down screens to black
Offline
just removed the xf86 amd driver and hopefully that fixes this new issue im having. which is lockup on screen power saver mode
heres the log from journal as requested: https://pastebin.com/Lkk6Ukzn
as for the new issue, im going to test it right now and hopefully the removal of xf86 fixed it, tho maybe not
i just realized i could have kept it at the original url and not had to repaste in pastebin
Last edited by Redd56 (2023-01-03 09:39:38)
Offline
removing xf86 did not solve the issue of system unresponsiveness after power saver, also it does still send usb io just doesnt reconnect, so i may have to make a new post for this one if i cant find anything on it. unless its fine to edit the title and continue here
Offline
i just realized i could have kept it at the original url and not had to repaste in pastebin
Yes, and now the pastebin was removed, so there's no journal available anywhere.
Offline
http://0x0.st/oR4V.txt there we go, just woke up so heres the new log
Offline
That's a journal from the working LTS kernel - hence of course there's no indication of the error either.
We'll need a journal from the broken setup, eg "sudo journalctl -b -5" for 5 boots ago.
Offline
just generated one via just rebooting into the stock kernal http://0x0.st/oRtb.txt
Offline
And as predicted:
Jan 03 18:19:15 archlinux kernel: amdgpu 0000:2d:00.0: [drm] Failed to create MST payload for port 000000004311f9d3: -5
Jan 03 18:19:15 archlinux kernel: amdgpu 0000:2d:00.0: [drm] Failed to create MST payload for port 00000000b2a498af: -5
Jan 03 18:19:15 archlinux kernel: amdgpu 0000:2d:00.0: [drm] Failed to create MST payload for port 000000006a4171bf: -5
Jan 03 18:19:15 archlinux kernel: [drm] DM_MST: stopping TM on aconnector: 00000000e94bf766 [id: 94]Later on kscreen wets itself, but that'S probably just because of the MST condition
Jan 03 18:19:21 archlinux systemd-coredump[2340]: Process 2319 (kscreen_osd_ser) of user 1000 dumped core.The only reference google could come up w/ is actually the introducing patch
https://lore.kernel.org/lkml/2022080823 … edhat.com/
https://github.com/torvalds/linux/commi … 40c5faf92f
You'll probably have to file a bug against the kernel.
https://bugzilla.kernel.org/buglist.cgi … esolution=---
Offline
amdgpu regression in 6.1
https://bugs.archlinux.org/task/76934
https://gitlab.freedesktop.org/drm/amd/-/issues/2171
Offline
Pages: 1