Skip to content

Conversation

@chengjunlu
Copy link
Contributor

The set fast math to the LLVM IR.

@vlad-penkin vlad-penkin linked an issue Oct 18, 2024 that may be closed by this pull request
Copy link
Contributor

@whitneywhtsang whitneywhtsang left a comment

Choose a reason for hiding this comment

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

Since there are UTs failure, IMO it makes sense to guard this change under autotune option first, and create an issue to further investigate.

Comment on lines +221 to +224
for (auto &func : *mod) {
for (auto &bb : func) {
for (auto &inst : bb) {
if (auto *op = dyn_cast<FPMathOperator>(&inst)) {
Copy link
Contributor

Choose a reason for hiding this comment

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

use a walker on FPMathOperator?

context.loadAllAvailableDialects();
});

// May do this after llvm ir according to user fmath flag.
Copy link
Contributor

Choose a reason for hiding this comment

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

Since this PR moves it to LLVMIR, we can remove this comment.

@etiotto etiotto marked this pull request as draft October 21, 2024 14:32
@chengjunlu
Copy link
Contributor Author

Can we close this PR? @whitneywhtsang

@whitneywhtsang
Copy link
Contributor

Can we close this PR? @whitneywhtsang

Sure, we can reopen or create a new PR if needed.

@whitneywhtsang whitneywhtsang deleted the chengjun/update-fast-math-pass branch May 9, 2025 18:16
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.

[Performance] Update the fast math pass

3 participants