-
-
Notifications
You must be signed in to change notification settings - Fork 5.5k
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
Faster matrix sqrt for upper triangular matrices. #31100
Faster matrix sqrt for upper triangular matrices. #31100
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.
Thanks. Would be great if you could also prepare a benchmark for this in BaseBenchmarks. It's not required for this PR but it would be great to have to avoid that this regresses.
No problem, I'll make a PR in BaseBenchmarks for this issue. |
For tracking potential regressions to JuliaLang/julia#31100 .
For tracking potential regressions to JuliaLang/julia#31100 .
@mateuszbaran Could you please rebase this PR, and when tests pass, I'll merge. Or has this been fixed by type inference improvements by now? |
I've actually forgotten about this PR. I will check if this is still relevant. |
Replacing Val(::Bool) in an argument list with an explicit if makes the call type stable. The issue was discussed in JuliaLang#31007.
f1f6857
to
9075235
Compare
OK, this still looks relevant, there is a similar difference in performance on 1.4.1 and a 44 day old Julia master I had around. I think my branch is rebased now. |
Sorry to bother you again, but can you rebase again. It's not your fault, but you rebased onto a commit that had a whitespace issue. |
Nevermind, I'll just merge. The whitespace issue is resolved on master, and this PR doesn't touch the file in question. |
OK, thanks 🙂 |
Replacing Val(::Bool) in an argument list with an explicit if makes the
call type stable. The issue was discussed in #31007.