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

[CUSOLVER] Add generic routines #2074

Merged
merged 3 commits into from
Oct 17, 2023
Merged

Conversation

amontoison
Copy link
Member

@amontoison amontoison commented Sep 7, 2023

Interface the new generic routines of CUSOLVER.
systrs and trtri were not available before.

@amontoison
Copy link
Member Author

amontoison commented Oct 16, 2023

@maleadt
CUSOLVER added generic routines for LAPACK functions and started to deprecate the legacy routines.
They are doing something similar to CUSPARSE.

The new main routines that I added are sytrs and trtri.
We will be able to add specialized methods for inverting dense triangular matrices and solving dense symmetric / hermitian linear systems.

Except gesvdp, sysevdx and gesvdr, I added tests for all new routines.
I will uncomment gesvdp, sysevdx and gesvdr when I will add the tests but I propose to do that in a different PR.

@maleadt maleadt added enhancement New feature or request cuda libraries Stuff about CUDA library wrappers. labels Oct 17, 2023
Copy link
Member

@maleadt maleadt left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Some minor nits, but LGTM. Thanks!

lib/cusolver/libcusolver.jl Show resolved Hide resolved
lib/utils/call.jl Outdated Show resolved Hide resolved
@maleadt maleadt merged commit 42051eb into JuliaGPU:master Oct 17, 2023
1 check passed
@amontoison amontoison deleted the cusolver_generic branch October 18, 2023 03:46
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
cuda libraries Stuff about CUDA library wrappers. enhancement New feature or request
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

2 participants