-
Notifications
You must be signed in to change notification settings - Fork 87
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
Deprecation warnings with cuSPARSE 11.8 #1146
Comments
Thanks for the report! We should add some version compile guards here (mirroring the ones in common_trs_kernels.cuh for csrsm*). Does AUR have a mechanism to compile for specific hardware? SM 3.5 is only being enabled if no GPU is detected at all. You might want to list a set of GPU architectures explicitly (increasing compile time and binary size) or find some other way to make it configurable |
I'm explicitly setting -DGINKGO_CUDA_ARCHITECTURES="All" because the built package should be usable on any hardware. But the AUR contains just the PKGBUILD, not the binary package itself. |
In that case, I would suggest listing the non-deprecated SMs instead of |
I could also add the |
That sounds like the best course of action. I'll add the missing compiler guards to deal with the deprecated functionality. |
Building ginkgo with CUDA 11.8 results in many warnings, mostly deprecation warnings from cuSPARSE. I'm attaching a complete build log of the ginkgo-hpc-git AUR package.
ginkgo-hpc-git-r5521.8b036c64f4-1-x86_64-build.log
The text was updated successfully, but these errors were encountered: