[PATCH] D144108: [mlir][LinAlg][Transform] Add a transform op for conv2d to im2col

[Thomas Raoux via Phabricator via llvm-commits]

ThomasRaoux accepted this revision.
ThomasRaoux added a comment.

In D144108#4134753 <https://reviews.llvm.org/D144108#4134753>, @nicolasvasilache wrote:

> Thanks for addressing @qcolombet , I am fine with iterating from this and seeing where it goes.
> I can see that it will not be in an ideal form for mapping im2col on device, I expect we will be able to evolve it iteratively.
>
> Please wait for @mravishankar for a final confirmation

+1 in order to do tile and fuse of generic op + conv we will need to have matching iteration space. This can be done by changing the op to use gather semantic (tensor.extract). But I think it is reasonable to do as a second step.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D144108/new/

https://reviews.llvm.org/D144108