This is an archive of the discontinued LLVM Phabricator instance.

BEGIN_PUBLIC [mlir] Add support for unranked case for `tensor_store` and `tensor_load` ops. END_PUBLIC
ClosedPublic

Authored by pifon2a on Aug 7 2020, 5:02 AM.

Diff Detail

Event Timeline

pifon2a created this revision.Aug 7 2020, 5:02 AM
Herald added a project: Restricted Project. · View Herald TranscriptAug 7 2020, 5:02 AM
pifon2a requested review of this revision.Aug 7 2020, 5:02 AM
herhut accepted this revision.Aug 7 2020, 5:15 AM
herhut added inline comments.
mlir/include/mlir/Dialect/StandardOps/IR/Ops.td
2946

Add the inverse to getTensorTypeFromMemRefType?

2959

Accidental?

This revision is now accepted and ready to land.Aug 7 2020, 5:15 AM
pifon2a updated this revision to Diff 283881.Aug 7 2020, 5:31 AM
pifon2a marked 2 inline comments as done.

Address the comments.

This revision was landed with ongoing or failed builds.Aug 7 2020, 5:33 AM
Closed by commit rG9c9490832054: BEGIN_PUBLIC (authored by pifon2a). · Explain Why
This revision was automatically updated to reflect the committed changes.
ftynse added a subscriber: ftynse.Aug 7 2020, 5:35 AM

Please drop additional tags (BEGIN_PUBLIC / END_PUBLIC) from the commit message.

herhut added inline comments.Aug 7 2020, 6:13 AM
mlir/include/mlir/Dialect/StandardOps/IR/Ops.td
2960

Can this ever anything else than TensorType? Both ranked and unranked are TensorType, right?

mlir/lib/Dialect/StandardOps/IR/Ops.cpp
2993

Turns out this is not actually needed, right? Sorry, I was confused.