Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Wrong power limit on custom 7900 XTX (just making sure this is considered) #237

Closed
43615 opened this issue Jan 7, 2024 · 5 comments
Closed

Comments

@43615
Copy link

43615 commented Jan 7, 2024

I have a Sapphire blah blah 7900 XTX, which comes with a default board power limit of 420 W (increasable by 15% (=483 W) using AMD's software on Windows).
In LACT 0.5.1 with kernel 6.6.9, the slider is capped at 339 W. I'm assuming it's using some fallback value as that doesn't even make sense for AMD's stock cards.
Don't worry, I know that you're still working on RDNA 3 support and that some of it will only be possible with kernel 6.7. Fan control is honestly more of a priority for me, but will power limits also be considered?

@ilya-zlobintsev
Copy link
Owner

The power limit API in RDNA3 is the same as in previous generations, so there shouldn't be any changes required for it on LACT's side. However it looks like there are people reporting a similar issue with the power limit being lower than expected: https://gitlab.freedesktop.org/drm/amd/-/issues/2356

It's also probably worth checking if the issue still persists in kernel 6.7, where the bulk of RDNA3 overdrive support is added. But in either case, this seems to be a kernel-side issue

@FerrumMaster
Copy link

In kernel 6.7 you can undervolt and set the power limit lower, but not higher, you need to compile your own kernel with a patch.

@ilya-zlobintsev
Copy link
Owner

There have been some updates regarding the power limit setting in kernel 6.7.3:

drm/amd/pm: update the power cap setting
drm/amd/pm: Fetch current power limit from FW

It's worth checking if that helps with the incorrect limit

Originally posted by @ilya-zlobintsev in #255 (comment)

@43615
Copy link
Author

43615 commented Feb 5, 2024

It changed but is still wrong (goes up to 402 now). Assuming it allows a 15% increase, this suggests an incorrect "default" value of 350 W.

@ilya-zlobintsev
Copy link
Owner

As this is/was a kernel-side problem, i'm going to close this issue. However #255 will remain for tracking changes

@ilya-zlobintsev ilya-zlobintsev closed this as not planned Won't fix, can't repro, duplicate, stale Feb 29, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants