You are not logged in.
Hey all!
This problem seems to be back since updating to 5.17.1
Only difference is that there are no logs whatsoever to go by. It seems more consistent as well, before I could go a couple of days with working suspend, now it seem like every morning I have to hard reset the laptop.
Anyone else experiencing this regression?
KInd regards!
Offline
I can confirm on my T495 as well: Suspend/resume was always a bit unstable, but now I can absolutely never resume from suspend on kernel 5.17.1. Both when docked and when on battery. USB comes online (keyboard, mouse lights up), Display and Ethernet doesn't.
https://bugs.archlinux.org/index.php?do … k_id=74285
Nothing here yet: https://bugzilla.kernel.org/buglist.cgi … esolution=---
Last edited by larsch (2022-04-04 16:05:48)
Offline
Upstream bug report: https://bugzilla.kernel.org/show_bug.cgi?id=215797
Patch to revert offending commit: https://lkml.org/lkml/2022/4/3/35
Offline
I had the same problem on Lenovo Legion 5 15ARH05H, I started using the nouveau driver and the problem disappeared.
Offline
Probably not the same issue, but https://download.nvidia.com/XFree86/Lin … ement.html ?
Offline
Hey all!
This problem seems to be back since updating to 5.17.1
Only difference is that there are no logs whatsoever to go by. It seems more consistent as well, before I could go a couple of days with working suspend, now it seem like every morning I have to hard reset the laptop.
Anyone else experiencing this regression?KInd regards!
Same here, unable to resume from suspend since last kernel update.
Running AMD Ryzen 3700U
No entry in journal
Offline
Upstream bug report: https://bugzilla.kernel.org/show_bug.cgi?id=215797
Patch to revert offending commit: https://lkml.org/lkml/2022/4/3/35
Cool that you found the exact commit!! Incredible what one line of code can do
How did you even get to that post?
This time I'm not going to bother patching and building my own kernel, I downgraded to 5.16 and it is suspending just fine.
I guess 5.18 will fix things?
I had the same problem on Lenovo Legion 5 15ARH05H, I started using the nouveau driver and the problem disappeared.
Hmm, are we talking about the same issue? I don't think this is graphics related, but what do I know...
What I do know is that nouveau has terrible screentearing on my T495 setup, so I want to stick to amdgpu.
Last edited by tubize (2022-04-06 11:40:14)
Offline
https://git.kernel.org/pub/scm/linux/ke … =v5.18-rc1
^When I look at the source of 5.18 mainline, this if statement at line 573 (before the flush cpu cache) is still present.
I guess that means that we won't be able to use 5.18 either?
Offline
Hmm, are we talking about the same issue? I don't think this is graphics related, but what do I know...
What I do know is that nouveau has terrible screentearing on my T495 setup, so I want to stick to amdgpu.
I'm sorry, I thought that fixed the problem, but it really didn't. I managed to avoid this problem by turning off the setting Dim Screen (off) and Screen Blank (Never). And I don't know if it will work tomorrow or in an hour. Also a problem with the brightness of the keyboard backlight, but I do not write a bug report because it is hopeless, no one will fix it
Offline
Offline
my keyboard backlight works but changes brightness level arbitrarily after waking up. also, my laptop is legion 5 (with amd), not legion 7i.
Offline
If you can control it (?, https://wiki.archlinux.org/title/Keyboard_backlight ) you could use some sleep hooks to re/store the value.
As it's rather weird for this to alter during an S3, teh standard question is whether there's a parallel OS (window) and then see the 3rd link below…
Offline
larsch wrote:Upstream bug report: https://bugzilla.kernel.org/show_bug.cgi?id=215797
Patch to revert offending commit: https://lkml.org/lkml/2022/4/3/35
Cool that you found the exact commit!! Incredible what one line of code can do
How did you even get to that post?
Wasn't me who found it the commit. Just googled the commit-id that was reported as offending and found the submission to revert it :-)
Offline
I guess that means that we won't be able to use 5.18 either?
This revert commit is in 5.18-rc2: https://github.com/torvalds/linux/commi … b2f80134d3
Offline
tubize wrote:I guess that means that we won't be able to use 5.18 either?
This revert commit is in 5.18-rc2: https://github.com/torvalds/linux/commi … b2f80134d3
Ah fantastic that rc2 fixes this! Then I can get back up-to-date.
Thanks for the heads up!
Offline
Confirmed fix is also in 5.17.3, which works for me!
Offline
What is currently the best workaround for this issue?
I reverted to 5.17.3-arch1-1. It worked for a couple of weeks, but I've had two freezes in the last day.
Has the mainline (5.18-rc4, as I write) worked?
Last edited by nexus (2022-04-27 15:06:35)
Offline