-
Notifications
You must be signed in to change notification settings - Fork 70
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
64-bit interface for L3 functions #855
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM! if the tests pass.
clients/gtest/blas3/trtri_gtest.yaml
Outdated
@@ -17,7 +17,7 @@ Tests: | |||
uplo: [ 'L', 'U' ] | |||
diag: [ 'N', 'U' ] | |||
matrix_size: *size_range | |||
api: [ FORTRAN, C ] | |||
api: [ FORTRAN, C , FORTRAN_64, C_64] |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
trtri ILP64 rocBLAS is not done right? If that is the case could we comment this just like other functions?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Good catch, I've removed this. Didn't make other changes for trtri since I'm not sure if we're planning on 64-bit support for it or not.
beta, | ||
dC, | ||
ldc)); | ||
// DAPI_EXPECT(HIPBLAS_STATUS_INVALID_VALUE, |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Should the commented out code be removed?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
uncommented it and moved into if(arg.bad_arg_all)
since cuBLAS documentation is unclear on whether or not this is supported.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is a huge change, and very thorough to include the functions not yet implemented! Although it takes time to review, it will save a lot of time by having all the work in one PR.
Sorry for the giant PR. This adds the interfaces needed for 64-bit support for all L3 functions (with the exception of trtri). It adds rocBLAS backend support for dgmm, trmm, and trsm, and will need fairly minimal changes to include the other functions as they are added to rocBLAS.