Skip to content

Use BLAS naming for VectorHandler methods#411

Merged
pelesh merged 23 commits intodevelopfrom
slaven/vector-handler-cleanup
Mar 3, 2026
Merged

Use BLAS naming for VectorHandler methods#411
pelesh merged 23 commits intodevelopfrom
slaven/vector-handler-cleanup

Conversation

@pelesh
Copy link
Collaborator

@pelesh pelesh commented Feb 25, 2026

Description

Updates names of methods in VecorHandler to match BLAS naming convention.

Closes #410

CC @alexander-novo

Proposed changes

  • Names of vector operations changed to match BLAS naming scheme as close as possible.
  • Updated and corrected Doxygen docs.
  • Scalar arguments passed by value.
  • assert statements added to aid debugging.

Checklist

  • All tests pass (make test and make test_install per testing instructions). Code tested on
    • CPU backend
    • CUDA backend
    • HIP backend
  • I have manually run the non-experimental examples and verified that residuals are close to machine precision. (In your build directory run: ./examples/<your_example>.exe -h to get instructions how to run examples). Code tested on:
    • CPU backend
    • CUDA backend
    • HIP backend
  • Code compiles cleanly with flags -Wall -Wpedantic -Wconversion -Wextra.
  • The new code follows Re::Solve style guidelines.
  • There are unit tests for the new code.
  • The new code is documented.
  • The feature branch is rebased with respect to the target branch.
  • I have updated CHANGELOG.md to reflect the changes in this PR. If this is a minor PR that is part of a larger fix already included in the file, state so.

Further comments

Assert statements in gemv fail for some reason. Need to investigate that more; out of scope here.

@pelesh pelesh self-assigned this Feb 25, 2026
@pelesh pelesh added enhancement New feature or request hip cuda labels Feb 25, 2026
@pelesh pelesh added this to the Release 0.99.3 milestone Feb 25, 2026
@pelesh pelesh changed the title Slaven/vector handler cleanup Use BLAS naming for VectorHandler methods Feb 25, 2026
@alexander-novo
Copy link
Collaborator

Shouldn't the infinity norm be amax? iamax returns the index of the maximum value.

@pelesh
Copy link
Collaborator Author

pelesh commented Feb 26, 2026

Shouldn't the infinity norm be amax? iamax returns the index of the maximum value.

Good catch. Addressed in 8310c9f.

Copy link
Collaborator

@kakeueda kakeueda left a comment

Choose a reason for hiding this comment

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

All tests passed with the CUDA backend after fixing VectorHandlerCuda::scal

Copy link
Collaborator

@shakedregev shakedregev left a comment

Choose a reason for hiding this comment

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

Fix the failing asserts and name convention violations.

Copy link
Collaborator

@shakedregev shakedregev left a comment

Choose a reason for hiding this comment

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

These are important changes. There are a few naming things that slipped.

@pelesh pelesh requested a review from shakedregev March 3, 2026 17:19
Copy link
Collaborator

@shakedregev shakedregev left a comment

Choose a reason for hiding this comment

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

Approved after testing on Cuda and RocM

@pelesh pelesh force-pushed the slaven/vector-handler-cleanup branch from 2a26d63 to 8d6fea5 Compare March 3, 2026 20:51
@pelesh pelesh merged commit cb7ff1b into develop Mar 3, 2026
6 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

cuda enhancement New feature or request hip

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Consider reviewing naming scheme for VectorHandler methods

4 participants