Skip to content

Explicit unaligned vector load/stores in PPC64LE GEMV kernels #5235

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

Merged

Conversation

quickwritereader
Copy link
Contributor

  • Explicit unaligned vector load/stores in PPC64LE GEMV kernels
  • Formatted code

@quickwritereader
Copy link
Contributor Author

do not we have ppc64le GitHub build and tests?

@quickwritereader
Copy link
Contributor Author

#2166

@martin-frbg
Copy link
Collaborator

The "Jenkins" build runs on both POWER and ZARCH, however the underlying PPC hardware is POWER9 now I think.

@martin-frbg
Copy link
Collaborator

@xry111 reminds me that we have been there two months ago with #5122 (and my subsequent #5125 to revert parts of
#4880 which is where the if defined(POWER8) comes from), so perhaps the code can be further streamlined again. I'll need to do build tests in the GCC Compile Farm when I'm able

@martin-frbg martin-frbg added this to the 0.3.30 milestone Apr 21, 2025
@martin-frbg
Copy link
Collaborator

Thank you very much for the quick fix - it does indeed look like reverting my #5125 will be possible with this fix, restoring optimizations from #4880 that increase accuracy.

@martin-frbg martin-frbg merged commit 1df8738 into OpenMathLib:develop Apr 21, 2025
85 of 86 checks passed
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.

2 participants