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

Fix gemm() performance for panel matrices #43

Merged
merged 12 commits into from
Oct 15, 2024
Merged

Conversation

mkatliar
Copy link
Owner

@mkatliar mkatliar commented Oct 5, 2024

  • Correct deduction of matrix pointer alignment
  • Added operator[] and removed broadcast() in matrix and vector pointers
  • Increased inlining threshold

@mkatliar mkatliar changed the title Fix gemm() performance for panel matrices [WIP] Fix gemm() performance for panel matrices Oct 5, 2024
@mkatliar mkatliar linked an issue Oct 5, 2024 that may be closed by this pull request
@mkatliar mkatliar force-pushed the panel_gemm_performance branch 2 times, most recently from f852308 to 6114d2e Compare October 6, 2024 12:46
@mkatliar mkatliar force-pushed the panel_gemm_performance branch 3 times, most recently from 38d6e26 to 9711bf4 Compare October 6, 2024 13:04
@mkatliar mkatliar force-pushed the panel_gemm_performance branch from 9711bf4 to 96401d5 Compare October 7, 2024 06:02
@mkatliar mkatliar requested a review from roversch October 8, 2024 17:28
@mkatliar mkatliar changed the title [WIP] Fix gemm() performance for panel matrices Fix gemm() performance for panel matrices Oct 8, 2024
Copy link
Collaborator

@roversch roversch left a comment

Choose a reason for hiding this comment

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

Looks good

@mkatliar mkatliar merged commit 94d8a2f into master Oct 15, 2024
2 checks passed
@mkatliar mkatliar deleted the panel_gemm_performance branch October 15, 2024 08:59
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

Successfully merging this pull request may close these issues.

Panel gemm() performance
2 participants