AMD CPU stuck in low-speed state after system resume

I just ran into this issue as well and unplugging and plugging the power fixed it for me. I’m on Ubuntu 24.04 with Kernel 6.8.0-45. I see this has been happening for a year and people on earlier versions thought kernel updates would fix things, but that might have just been a placebo unfortunately. Good to know that it’s easy to fix when it happens though.

Just happened to me as well on a Framework 16 (Ryzen 7940). I believe this has been the only time. I recently updated to the 3.05 firmware.

I’m not certain about the sequence of events but, like others, it seems to have been related to unplugging power while in suspend. After resuming, cores were stuck at 544 MHz. Switching power profiles didn’t help and neither did another suspend/resume. Only plugging the power in resolved it. Presumably a reboot would have as well.

Make sure you verify what kernel you are on. For me 6.8.7 fixed it, but your distro kernel might not have the fix.

This also happened to me just now. Same deal, unplugging the power cord while in suspend. Bios 3.05, Kernel 6.12.7-200.fc41.x86_6 on Fedora 41. 7940HS CPU. Rebooting caused the laptop to get stuck at the “Framework” logo which required a hard power off by holding the power button. Came back on just fine after that.

This happens to me fairly often; the sequence is always unplug-then-open-lid, and cores end up at either 400MHz or 544MHz.

BIOS 3.05, 7840HS CPU, kernel Debian 6.12.17-1 (2025-03-01).

@Kevin_Fleming, does that mean that this is a separate problem to: [1] [2]

…or does 0.0.3.7 merely not remediate it? It states that it should:

I ask because I’m attempting to diagnose:

…which I may have misattributed, if the fault was the CPU clock speed.

Unfortunately, because I have been unable to locate an authoritative GitHub issue, IDK how this was discovered, nor what the remediation is.


  1. reddit.com/r/framework/comments/19d5zey/comment/kj54pc0 ↩︎

  2. reddit.com/r/framework/comments/13ehs1t/comment/jjpw6e8 ↩︎

Here’s a kicker. I just found mine back at 0.55 Ghz again, I said on Framework 16 stuck at 0.55 Ghz after windows sleep mode - #16 by Nathan_Hay that it was resolved, but I just noticed that it solved some scenarios, but some still put it at 0.55.

@Nathan_Hay, I suggest that you (or someone else) report this to github.com/FrameworkComputer/SoftwareFirmwareIssueTracker/issues/new?template=laptop-issue.md if you’re able to, and link the report back here when you have:

I had this issue too in my AMD 370 HX. The cores were fixed to lower frequency, after the laptop wakes up from sleep. I discovered that this happens when “boots” is disabled when laptop goes to sleep (I am a Linux user, the “boost“ config is usually disabled when the laptop is unplugged or is on battery, to limit the speed of the cores and save battery, not sure how is that controlled in Windows, but I am almost sure the saver mode should be similar). In the case of my laptop, I just prevented the powermode to disable the boost and that was enough, because the max frequencies are setup low for that mode, so, even with boost enabled the cpu consumption was not affected. Not sure if this has to do with the problem in this post, but I solved my issue in that way now I am able to get back to performance mode after sleep.

I think the new bios update 4.03 solves it.

I seem to be just recently beginning to experience this issue - and have just filed a report.

@Orlando_RT , would you could share how to prevent the powermode from disabling boost to perhaps work around this issue? I’d greatly appreciate it.

There is a new beta bios out that supposedly fixes one of the 545MHz issues. I haven’t yet tried it though.