[PATCH] D78996: [MLIR][LINALG] Convert Linalg to Linalg

[Alexander Belyaev via Phabricator via llvm-commits]

pifon2a added inline comments.


================
Comment at: mlir/lib/Conversion/LinalgToLinalg/LinalgToLinalg.cpp:74
+
+    rewriter.eraseOp(op);
+    return success();
----------------
shouldn't it be `rewriter.replaceOp(op, <what to replace results with>)`? Please, add a test with two chained Linalg ops.


================
Comment at: mlir/lib/Conversion/LinalgToLinalg/LinalgToLinalg.cpp:103
+
+void TensorLinalgToBufferLinalgPass::runOnOperation() {
+  auto &context = getContext();
----------------
why is the definition of `runOnOperation()` method not inside `TensorLinalgToBufferLinalgPass`? I don't think it improved readability here. Is there some guide that does not allow doing that?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D78996/new/

https://reviews.llvm.org/D78996