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.

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

Jump to bottom

[Codegen] Add f8 to f32 pass for arith.negf #19942

Closed
wants to merge 1 commit into from
Closed

[Codegen] Add f8 to f32 pass for arith.negf #19942

wants to merge 1 commit into from
+89 −0

Conversation

AmosLewis
Copy link
Contributor

Try to fix #19921 (comment) under @pashu123 guidence.

Replace the following sequence

%1 = arith.negf %input : vector<1x2x1x1x1x1xf8E4M3FNUZ>

with

%0 = arith.extf %input : f8E4M3FNUZ to f32
%1 = arith.negf %0 : vector<1x2x1x1x1x1xf32>
%2 = arith.truncf %1 : vector<1x2x1x1x1x1xf8E4M3FNUZ>

@AmosLewis AmosLewis requested a review from pashu123 February 10, 2025 07:00
@AmosLewis AmosLewis requested a review from hanhanW as a code owner February 10, 2025 07:00
@AmosLewis AmosLewis marked this pull request as draft February 10, 2025 07:00
@AmosLewis
[Codegen] Add f8 to f32 pass for arith.negf
582dacd 
Signed-off-by: Chi Liu<Amos_Lewis@foxmail.com>
@pashu123
Copy link
Contributor

pashu123 commented Feb 10, 2025 •
edited
Loading

#19943 Made a new patch here.

@AmosLewis AmosLewis closed this Feb 10, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@pashu123 pashu123 Awaiting requested review from pashu123

@hanhanW hanhanW Awaiting requested review from hanhanW hanhanW is a code owner

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

[ROCm][Codegen] llama 8b fp8 with attention segfault
2 participants
@AmosLewis @pashu123