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

suboptimal shufflevector codegen, shufflevector+trunc #88030

Open
zhengyang92 opened this issue Apr 8, 2024 · 2 comments
Open

suboptimal shufflevector codegen, shufflevector+trunc #88030

zhengyang92 opened this issue Apr 8, 2024 · 2 comments
Assignees
@RKSimon
Labels
llvm:optimizations llvm:SelectionDAG SelectionDAGISel as well missed-optimization

Comments

@zhengyang92
Copy link
Contributor

https://alive2.llvm.org/ce/z/X6DPQJ

https://godbolt.org/z/5GMPGxdKE

This pattern and its variants appear quite a lot in 638.imagick in SPEC CPU17.
@zhengyang92
Copy link
Contributor Author

@regehr @RKSimon @dtcxzyw

@RKSimon RKSimon self-assigned this Apr 8, 2024
RKSimon added a commit that referenced this issue Apr 9, 2024
@RKSimon
[X86] Add test coverage for #88030
866a1bc
@RKSimon RKSimon added the llvm:SelectionDAG SelectionDAGISel as well label Jun 19, 2024
@RKSimon
Copy link
Collaborator

RKSimon commented Jun 19, 2024

I had considered whether we need some sort of ISD::TRUNCATE_VECTOR_INREG node but I think we might be able to make use of ISD::MASKED_COMPRESS once #92289 lands

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@RKSimon RKSimon

Labels
llvm:optimizations llvm:SelectionDAG SelectionDAGISel as well missed-optimization
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@zhengyang92 @RKSimon