matthias-springer wrote:

> Part of it is folding constantOp into memref.store.

This is not part of the pattern. I just needed a side-effecting op in the test 
case.

> Part of it is the functionality of remove-dead-values, but I believe it ought 
> to be implemented within remove-dead-values.

`remove-dead-values` indeed performs this kind of folding. I'm trying to remove 
this as part of #173505 and this PR is in preparation of that change. I am 
going to send an RFC for that soon.



https://github.com/llvm/llvm-project/pull/173560
_______________________________________________
llvm-branch-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

Reply via email to