[Solved] p14s Gen5 (AMD) failes to wake up from sleep

[parasquid]

Hello! I've been running bluefin for the better part of this year on my p14s Gen5 (AMD) system.

Everything has been great except that any new images above

● ostree-image-signed:docker://ghcr.io/ublue-os/bluefin-dx:stable
                   Digest: sha256:2cda1aa45387bd5b9ae92e58f549de9a7d0dd48d77c56db60e28063522c76f2f
                  Version: 42.20250928.1 (2025-09-28T06:05:48Z)

doesn't wake up from sleep :(

I've pinned this current image since this seems to be the last one that allows waking up from sleep.

I've also tried the latest bazzite image

  ostree-unverified-registry:ghcr.io/ublue-os/bazzite-gnome:stable
                   Digest: sha256:d37092ef3f2cc099b8c5a972455826cd5d52fc6dd3459c7adc0db5a1c9c5b267
                  Version: 43.20251102 (2025-11-03T05:04:12Z)

hoping that the newer kernels there would have fixed things, but to no avail.

I've also tried amd_s2idle.py which works well; it sleeps and wakes up but I'm guessing it's because it's setting a wake timer. However suspending (either via laptop lid or through the power menu option or systemctl suspend) makes the machine suspend, but it just keeps sleeping even when I try to wake it up via the keyboard, opening the lid, plugging in a usb keyboard and typing, touchpad, or un/plugging power.

I'm not really sure how to troubleshoot this. Right now I'm good with just selecting the pinned image whenever I reboot (I'll probably set it as default later on if I still can't get this working) so any hints are appreciated.

I'm currently downloading the bluefin iso and later will try to install that on a different drive to see if a fresh install works.

Thanks a lot!

[hanthor]

Have you checked to see if you have any BIOS updates?

[parasquid]

Have you checked to see if you have any BIOS updates?

Yes, I don't have any bios updates on this machine. I have the old nvme that came with the laptop running Windows which I periodically boot up to check if there are any updates from Lenovo Vantage.

[swagromancer]

Having a similar issue on my Razer Blade Stealth 13 Late 2019 (Intel) notebook.

Device info

❯ ujust device-info
[...]
https://paste.centos.org/view/366abb7f

Current Image

❯ rpm-ostree status
[...]
Deployments:
● ostree-image-signed:docker://ghcr.io/ublue-os/bluefin-dx-nvidia-open:stable-20250928
                   Digest: sha256:38db1ff60a7faa7cddef158b1cc24784897c76e131ea582ec815edfdcc2960c3
                  Version: 42.20250928.1 (2025-09-28T06:28:07Z)

If I update to any image after the 20250928 one my notebook won't wake from suspend.

I also tried switching to a current non-nvidia image, after seeing some complaints online about nvidia drivers causing breakages...

ostree-image-signed:docker://ghcr.io/ublue-os/bluefin-dx:stable
                   Digest: sha256:f67b8daf2ea4e4edb9f0543272f2b43cd43d6a41f3755572d6bca4f3221b8165
                  Version: 42.20251106.1 (2025-11-06T12:34:14Z)

However, the issue seems to persist on those images as well. I've rolled back to bluefin-dx-nvidia-open:stable-20250928 and pinned it for now, but this is definitely an annoying problem considering this system has been running fine on bluefin-os for more than a year now :/

Have you checked to see if you have any BIOS updates?
I did check that I have the most recent bios for this system.

[parasquid]

I've just updated the image:

❯ rpm-ostree status
State: idle
AutomaticUpdates: stage; rpm-ostreed-automatic.timer: inactive
Deployments:
● ostree-image-signed:docker://ghcr.io/ublue-os/bluefin-dx:stable
                   Digest: sha256:f67b8daf2ea4e4edb9f0543272f2b43cd43d6a41f3755572d6bca4f3221b8165
                  Version: 42.20251106.1 (2025-11-06T12:34:14Z)

and ran amd_s2idle.py and it was able to successfully sleep and wakup the laptop

I was curious how the script triggers sleep and I found that it will either trigger it through dbus or by writing "mem" to /sys/power/state so I tried replicating that

> sudo su
$ echo "mem" > /sys/power/state

and it sleeps and wakes up no issues! However, suspending via the power menu on the upper right bar fails to actually suspend the laptop (everything goes black but the light on the power button is continuously lit, unlike the fadein/fadeout pattern when the laptop sleeps).

So I'm guessing the problem likely isn't the kernel version but with some kind of interaction with the gui system?

Right now I'm just happy I can just just manually trigger suspend via the command line but it would be nice if suspending via the gui (or by lid) works too.

Update: I may have celebrated too early. While the laptop wakes up from suspend if I wake it up from the keyboard, closing the lid while suspended then waking it up from the lid opening will momentarily wakeup the laptop (I can see the desktop flash) then suddenly hang the machine (screen black, keyboard lights on, but machine is unresponsive) and I have to hard shutdown the laptop via holding the power button for a few seconds :(

The way I work around this is to add an ignore for logind:

❯ cat /etc/systemd/logind.conf.d/ignore-lid.conf 
[Login]
HandleLidSwitch=ignore
HandleLidSwitchDocked=ignore
HandleLidSwitchExternalPower=ignore

This seems to disable the logind handling, but the laptop still wakes up properly from sleep when opening the lid. The only issue is that when it does wake from sleep, the laptop isn't "locked" (i.e. you don't have to enter your password to unlock the desktop) so it's a bit of a security concern if let's say my laptop is stolen while sleeping.

[parasquid]

Not sure if this helps, but sometimes I get these kernel messages whenever I try the echo "mem" command to get the laptop to sleep:

[66610.888740] Freezing user space processes
[66630.891052] Freezing user space processes failed after 20.001 seconds (1 tasks refusing to freeze, wq_busy=0):
[66630.891912] task:rm              state:D stack:0     pid:429540 tgid:429540 ppid:429519 task_flags:0x400000 flags:0x00004004
[66630.891927] Call Trace:
[66630.891932]  <TASK>
[66630.891947]  __schedule+0x2f9/0x7b0
[66630.891962]  ? srso_alias_return_thunk+0x5/0xfbef5
[66630.891977]  schedule+0x27/0x80
[66630.891989]  request_wait_answer+0xce/0x260 [fuse]
[66630.892015]  ? __pfx_autoremove_wake_function+0x10/0x10
[66630.892025]  __fuse_simple_request+0xd5/0x2d0 [fuse]
[66630.892042]  fuse_readdir_uncached+0x155/0x550 [fuse]
[66630.892070]  ? srso_alias_return_thunk+0x5/0xfbef5
[66630.892076]  ? post_alloc_hook+0xc8/0x150
[66630.892087]  ? srso_alias_return_thunk+0x5/0xfbef5
[66630.892093]  ? get_page_from_freelist+0x43b/0x6c0
[66630.892100]  ? srso_alias_return_thunk+0x5/0xfbef5
[66630.892120]  iterate_dir+0xa7/0x270
[66630.892130]  ? srso_alias_return_thunk+0x5/0xfbef5
[66630.892137]  __x64_sys_getdents64+0x7b/0x110
[66630.892145]  ? __pfx_filldir64+0x10/0x10
[66630.892154]  do_syscall_64+0x7e/0x250
[66630.892163]  ? __handle_mm_fault+0x3a7/0x700
[66630.892172]  ? __seccomp_filter+0x41/0x4e0
[66630.892185]  ? srso_alias_return_thunk+0x5/0xfbef5
[66630.892191]  ? count_memcg_events+0x15d/0x230
[66630.892200]  ? srso_alias_return_thunk+0x5/0xfbef5
[66630.892205]  ? handle_mm_fault+0x248/0x360
[66630.892213]  ? srso_alias_return_thunk+0x5/0xfbef5
[66630.892219]  ? do_user_addr_fault+0x21a/0x690
[66630.892230]  ? srso_alias_return_thunk+0x5/0xfbef5
[66630.892235]  ? srso_alias_return_thunk+0x5/0xfbef5
[66630.892241]  ? irqentry_exit_to_user_mode+0x2c/0x1c0
[66630.892249]  entry_SYSCALL_64_after_hwframe+0x76/0x7e
[66630.892256] RIP: 0033:0x7f6bc00864e1
[66630.892331] RSP: 002b:00007ffee17032a0 EFLAGS: 00000246 ORIG_RAX: 00000000000000d9
[66630.892339] RAX: ffffffffffffffda RBX: 000000000000000c RCX: 00007f6bc00864e1
[66630.892343] RDX: 0000000000000800 RSI: 00007f6bc0060038 RDI: 0000000000000003
[66630.892347] RBP: 00007ffee1704def R08: 00007f6bc0060020 R09: 0000000000000000
[66630.892351] R10: 00007f6bc0060aac R11: 0000000000000246 R12: 0000000000000000
[66630.892354] R13: 00007f6bc0060020 R14: 00000000ffffffff R15: 00007f6bc0100b48
[66630.892369]  </TASK>
[66630.892703] OOM killer enabled.
[66630.892706] Restarting tasks: Starting
[66630.926214] Restarting tasks: Done
[66630.926362] random: crng reseeded on system resumption
[66631.029375] PM: suspend exit

the laptop recovers after a bit though, then at that point I just try to sleep again.

[parasquid]

tl;dr docker containers (in this case they all had restart:always set so maybe that was the issue?) are causing the laptop to freeze before entering suspend

I've installed a fresh bluefin onto a spare nvme ssd and it was able to sleep properly. However after I rsynced my /var to this new installation and then rebooted (so basically all of my flatpaks, browser profiles, etc were synced) it wasn't able to sleep again. So it looks like it wasn't the kernel or anything that's part of the image.

At first I thought it was my flatpak apps, but then I remembered sometime ago that docker was causing some issues with sleep on a different laptop (gpd win max 2 also running bluefin) and after a quick docker stop $(docker ps -q) I was able to properly sleep.