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

nVidia broke nvml_fix on 331.17? #5

Closed
CFSworks opened this issue Oct 24, 2013 · 6 comments
Closed

nVidia broke nvml_fix on 331.17? #5

CFSworks opened this issue Oct 24, 2013 · 6 comments

Comments

@CFSworks
Copy link
Owner

I'm not 100% on this, but it looks like nVidia changed something during the 331.13->331.17 update that broke nvml_fix. I'm opening an issue on my own repository to track the bug until I have a chance to address it.

@MeFri
Copy link

MeFri commented Nov 10, 2013

I can confirm this issue with 331.17 here...

@Moguri
Copy link

Moguri commented Nov 20, 2013

331.20 here, and nvml_fix no longer works for me (nvidia-smi cannot find libnvidia-ml.so).

@vacaloca
Copy link

I can also confirm the same issue... the old nvml_bug.c with added version check returns not supported... NVIDIA must not have liked your 'fix' ;)

@CFSworks
Copy link
Owner Author

CFSworks commented Dec 6, 2013

I've updated this for the 331.xx series on 64-bit. 32-bit coming later... I've tested it on 331.20. I'm going to go ahead and close this, let me know if it's still broken for you.

@CFSworks CFSworks closed this as completed Dec 6, 2013
@vacaloca
Copy link

Seems like NVIDIA broke it again on anything higher than 331.20 (e.g. 331.49)

@cwchan55
Copy link

nvml remains broken for 331.75. In nvml.c.
I see

#elif defined(NVML_PATCH_331)

ifdef i386

error "No i386 support for this version yet!"

else

    fix[187] = 2;
    fix[188] = 1;

endif

How are the fix[] locations and values determined in the NVML API?

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

5 participants