Bugzilla – Bug 1219195
/sys/class/drm/card1/device/hwmon/hwmon*/power1_cap_min increased from 6.6. to 6.7 way too much
Last modified: 2024-03-24 21:17:45 UTC
Dear maintainers, when upgrading the kernel in Tumbleweed from 6.611-1.1 to 6.7.1-1.1 the lowest TDP I could set for my AMD RX6600 increased from 0 (might be nonsensical) to 94W (which is close to the default of 100W). I'm referring to the TDP floor of /sys/class/drm/card1/device/hwmon/hwmon*/power1_cap_min where * denotes the device number. On 6.611-1.1, I could happily drive my RX 6600 with a 50W TDP (by means of setting echo '50000000' > '/sys/class/drm/card1/device/hwmon/hwmon3/power1_cap' at boot); once upgraded to 6.7.1-1.1 this no longer works. Have the TDP floors been purposefully increased for some reason? As my 50W TDP setting on 6.611-1.1 works like a charm, I'd be delighted if the previous behaviour could perhaps be restored. Thanks a lot in advance and let me know if I can provide further info.
The change was possibly introduced in: commit 1958946858a62b6b5392ed075aa219d199bcae39 Author: Ma Jun <Jun.Ma2@amd.com> Date: Thu Oct 12 09:33:45 2023 +0800 drm/amd/pm: Support for getting power1_cap_min value Support for getting power1_cap_min value on smu13 and smu11. For other Asics, we still use 0 as the default value. Signed-off-by: Ma Jun <Jun.Ma2@amd.com> Reviewed-by: Kenneth Feng <kenneth.feng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> The above commit may also be an indirect fix for this upstream issue: https://gitlab.freedesktop.org/drm/amd/-/issues/2992
So this sounds like an intentional behavior change in the upstream. But if you find it's still buggy, please report it to the upstream devs, gitlab.freedesktop.org issues.
Thanks for the clarification. Yes, it seems like an upstream change (even though I'm not sure if raising the power cap min was part of the intention). It's been reported at https://gitlab.freedesktop.org/drm/amd/-/issues/3137
This is supposed to be fixed in the upstream, so at least 6.8 should work.
If I understand correctly, this is - unfortunately - intended behaviour, so I'd suggest marking it as wontfix instead of resolved, if that's ok. Thanks!