-
Notifications
You must be signed in to change notification settings - Fork 11.7k
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
[MLIR][memref] The Normalizing Memref on removing extra input arguments onmemref.load
#82675
Labels
Comments
This is happening because the memref memory operations are not handled when there is a diminsionality ( airity) change. |
DarshanRamakant
added a commit
to DarshanRamakant/llvm-project
that referenced
this issue
Sep 22, 2024
This change will fix the normalization issue with memref.load when the associated affine map is reducing the dimension. This PR fixes llvm#82675
DarshanRamakant
added a commit
to DarshanRamakant/llvm-project
that referenced
this issue
Oct 6, 2024
This change will fix the normalization issue with memref.load when the associated affine map is reducing the dimension. This PR fixes llvm#82675
@llvm/issue-subscribers-mlir-affine Author: Amir Mohammad Tavakkoli (tavakkoliamirmohammad)
Hi. If you want to normalize the `memref` layout that reduces the dimensionality, the `-normalize-memrefs` pass will work perfectly if there is `affine.load`, but in the case that the operation is `memref.load`, the code fails to reduce the inputs to `memref.load`.
Complier explorer link: https://godbolt.org/z/zYT7jxs9n
Reproducing the error
The error
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hi. If you want to normalize the
memref
layout that reduces the dimensionality, the-normalize-memrefs
pass will work perfectly if there isaffine.load
, but in the case that the operation ismemref.load
, the code fails to reduce the inputs tomemref.load
.Complier explorer link: https://godbolt.org/z/zYT7jxs9n
Reproducing the error
The error
The text was updated successfully, but these errors were encountered: