I had this issue for a while as well and initially thought this was related to the BIOS update I took, but based on the comments in the thread I downgraded the kernel from 6.14.0 to 6.13.5 with no other package changes and the problem went away. That confirms the problem is kernel related.
Should I test 6.14.1 as well or does this issue still need to be bisected?
Update: It does look like 6.14.1 works on my end as reported by a few others.
The connect/disconnect noises is a separate issue that has been reported upstream to upower. I would expect an update to that project to come soon that will address the battery cycling on and off charge, which is what’s causing the notification sounds.
I don’t know how you do it, but it doesn’t work for me 
I have the latest Fedora updates + Linux 6.14.1-100.fc41.x86_64
Tested old upower version and the issue is gone, thx. For anyone else, can install the old package here: upower-1.90.7-1.fc41 | Build Info | koji
I’m going to be staying on that version until the problem is fixed, looks like there’s a lot of discussion here: Charge notification cycle on some laptops since 1.90.8 (#306) · Issues · upower / upower · GitLab
Same… gonna try dropping to 6.13-4 now.
Upgraded to 6.14.0-63.fc42.x86_64 (64-bit) (Fedora 42 beta) to give things a try. After my wifi card (AX210) finally stopped connecting/disconnecting I hit the issue where my 7700S claims to be at 100% load and my fans are also at 100%. This happened after unplugging from the power adapter (180 watt framework adapter), sleeping, and plugging back in. I have the performance power profile selected.
I’m not sure if these errors are relevant but posting anyway:
[ 3838.053336] amdgpu 0000:03:00.0: amdgpu: PSP is resuming...
[ 3838.109631] amdgpu 0000:03:00.0: amdgpu: reserve 0x1300000 from 0x81fc000000 for PSP TMR
[ 3838.204313] amdgpu 0000:03:00.0: amdgpu: RAS: optional ras ta ucode is not available
[ 3838.211836] amdgpu 0000:03:00.0: amdgpu: RAP: optional rap ta ucode is not available
[ 3838.211838] amdgpu 0000:03:00.0: amdgpu: SECUREDISPLAY: securedisplay ta ucode is not available
[ 3838.211840] amdgpu 0000:03:00.0: amdgpu: SMU is resuming...
[ 3838.211844] amdgpu 0000:03:00.0: amdgpu: smu driver if version = 0x00000035, smu fw if version = 0x00000040, smu fw program = 0, smu fw version = 0x00525c00 (82.92.0)
[ 3838.211848] amdgpu 0000:03:00.0: amdgpu: SMU driver if version not matched
[ 3838.252620] amdgpu 0000:03:00.0: amdgpu: SMU is resumed successfully!
[ 3838.253622] [drm] kiq ring mec 3 pipe 1 q 0
[ 3838.259467] amdgpu 0000:03:00.0: amdgpu: MES FW versoin must be larger than 0x63 to support limit single process feature.
[ 3838.259469] amdgpu 0000:03:00.0: amdgpu: failed to change_config.
[ 3838.259470] amdgpu 0000:03:00.0: amdgpu: resume of IP block <mes_v11_0> failed -22
[ 3838.259472] amdgpu 0000:03:00.0: amdgpu: amdgpu_device_ip_resume failed (-22).
[ 3838.259473] amdgpu 0000:03:00.0: PM: dpm_run_callback(): pci_pm_resume returns -22
[ 3838.259482] amdgpu 0000:03:00.0: PM: failed to resume async: error -22
Can confirm no sleep issues, connect/disconnect sounds, or GPU issues on 6.13.4-200
So yesterday I tested a little again, 6.13.5-200.fc41 is still the only version that works for me. 6.14.1-100.fc41 does not allow me to access the graphics card.
Today I received a few updates, including upower; unfortunately I failed to copy all the changes. Anyway, the graphics card is sometimes running with the patch kernel-6.14.1-100.fc41 unfortunately the connect/disconnect sound problem remains; however, it only occurred some time after the restart, i was already hopeful.
With the patch kernel-6.14.2-100.fc41 it no longer works. What an annoying back and forth.
Do you know if there is a BZ created for the AMD gpu issues w/fedora or any other distro?
@Mario_Limonciello as well?
If not, I can bug this - but it seems its been an issue for quite a few kernel releases at this point.
Edit to this:
It should land in 6.15-rc2 and then make its way back to stable kernels as well.
Per https://gitlab.freedesktop.org/drm/amd/-/issues/4083#note_2839425
Here’s the mainline commit for any distro you want to notify to pull it in early.
Has anyone here tested the new Fedora 42 release Matt posted?
I am currently running Fedora 42 with Linux 6.14.2-300.fc42.x86_64 and do not have a working graphics card.
Loud sigh.
Thanks for the update.
As long as this backport will not exist, you will probably have to wait for 6.15…
Arch on 6.14+ seems to be working fine. Other Linux distros either need to install an older kernel that works, or wait for an upgrade.
Fedora 42 KDE, just upgraded to 6.14.2, dGPU available after sleep, but running at 100% utilization.
I have updated to Fedora 42 yesterday and experience the same problem. Usually the dGPU is available and working when booting with a power cable attached. It disappears when unplugging AC power. When booting while running on battery, the dGPU is not available from the start.
On the unpatched 6.14.2-300.fc42.x86_64 kernel, besides the “MES FW versoin must be larger than 0x63 to support limit single process feature.” error (already posted above), there are also the following error messages shown on boot:
Apr 18 10:16:18 ****** kernel: [drm] Register(0) [regUVD_PGFSM_STATUS] failed to reach value 0x00800000 != 0x00c00000n
Apr 18 10:16:18 ****** kernel: amdgpu 0000:03:00.0: [drm:jpeg_v4_0_set_powergating_state [amdgpu]] *ERROR* amdgpu: JPEG enable power gating failed
Apr 18 10:16:18 ****** kernel: [drm:amdgpu_device_ip_set_powergating_state [amdgpu]] *ERROR* set_powergating_state of IP block <jpeg_v4_0> failed -110
and:
Apr 18 10:16:19 ****** kernel: [drm] Register(0) [regUVD_POWER_STATUS] failed to reach value 0x00000001 != 0x00000003n
Apr 18 10:16:19 ****** kernel: [drm] Register(0) [regUVD_POWER_STATUS] failed to reach value 0x00000001 != 0x00000003n
Apr 18 10:16:19 ****** kernel: amdgpu 0000:03:00.0: amdgpu: SMU: response:0xFFFFFFFF for index:43 param:0x00000000 message:PowerDownVcn?
Apr 18 10:16:19 ****** kernel: amdgpu 0000:03:00.0: amdgpu: Failed to power gate VCN instance 0!
Apr 18 10:16:19 ****** kernel: [drm:vcn_v4_0_stop [amdgpu]] *ERROR* Dpm disable uvd failed, ret = -121.
Apr 18 10:16:19 ****** kernel: amdgpu 0000:03:00.0: amdgpu: SMU: response:0xFFFFFFFF for index:36 param:0x00000001 message:SetWorkloadMask?
Apr 18 10:16:19 ****** kernel: amdgpu 0000:03:00.0: amdgpu: Failed to set workload mask 0x00000001
Apr 18 10:16:19 ****** kernel: amdgpu 0000:03:00.0: amdgpu: (-121) failed to disable video power profile mode
When booting the patched kernel, none of those occur.
