Since filing the initial report, I ran a real workload comparison using
my full pytest suite.

On this Metabox system running Ubuntu 24.04, the suite takes about 22 minutes.
On a less powerful Dell Ubuntu laptop, the same suite takes about 11–12 minutes.

The project directory on the Metabox is on the internal NVMe and mounted
as ext4 (`df -T .` shows ext4), so this does not appear to be a Windows-
mounted filesystem issue.

`powerprofilesctl` still reports:
`performance -> Degraded: yes (high-operating-temperature)`

At the time of sampling, temperatures were moderate rather than near critical:
- CPU package: about 60C
- GPU: about 49C
- NVMe: about 38.9C
- fans: about 2200–2700 RPM

The filtered kernel log from this boot shows ACPI BIOS errors and NVIDIA
ACPI DSM warnings, but I did not see obvious thermal-throttling messages
in that filtered output.

I have attached the latest outputs of:
- `powerprofilesctl list`
- `sensors`
- `nvidia-smi`
- `journalctl -k -b | grep -Ei 'thermal|thrott|tcc|pstate|acpi|NVRM'`

No firmware updates are available via `fwupdmgr`.

** Attachment added: "metabox_kernel_log_journalctl_k_b_grep_thermal_acpi.txt"
   
https://bugs.launchpad.net/ubuntu/+source/power-profiles-daemon/+bug/2148099/+attachment/5960538/+files/metabox_kernel_log_journalctl_k_b_grep_thermal_acpi.txt

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2148099

Title:
  power-profiles-daemon reports high-operating-temperature degradation
  at moderate temperatures on Ubuntu 24.04

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/power-profiles-daemon/+bug/2148099/+subscriptions


-- 
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to