Skip to content

trsm is not actually implemented yet #2814

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.

Sign up for GitHub

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
merged 1 commit into from
Nov 25, 2025
Merged

trsm is not actually implemented yet #2814

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion src/compiler.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5074,7 +5074,7 @@ function GPUCompiler.compile_unhooked(output::Symbol, job::CompilerJob{<:EnzymeT
"trmv",
"syrk",
"trmm",
"trsm",
# "trsm", Not actually implemented yet
Copy link
Member

@wsmoses wsmoses Nov 25, 2025

Choose a reason for hiding this comment

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

sorry quick post commit review, can this be limited to 1.12+.

It will otherwise slow down other code by making it non-blas, if not differentiated

Copy link
Member Author

@vchuravy vchuravy Nov 25, 2025

Choose a reason for hiding this comment

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

Oof, that's frustrating, shouldn't we ideally not do the blas replacement when a function is not being differentiated? Otherwise this problem occurs for many more functions?

Copy link
Member

@wsmoses wsmoses Nov 25, 2025

Choose a reason for hiding this comment

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

it does indeed occur for more functions (which is why we should be somewhat sparing on the blas replacements).

The issue is that at the time of replacement, we haven't run activity analysis so we don't know what is needed and what isn't

"potrf",
)
ForwardModeTypes = ("s", "d", "c", "z")
Expand Down
Loading