[mlir][vector] Address linearization comments (post commit) (#138075)

This PR adds some documentation to address comments in
https://github.com/llvm/llvm-project/pull/136581 

This PR adds a test for linearization across scf.for. This new test
might be considered redundant by more experienced MLIRers, but might
help newer users understand how to linearize scf/cf/func operations
easily

The documentation added in this PR also tightens our definition of
linearization, to now exclude unrolling (which creates multiple ops from
1 op). We hadn't really specified what linearization meant before.

1 files changed, 4 insertions, 6 deletions

diff --git a/mlir/test/lib/Dialect/Vector/TestVectorTransforms.cpp b/mlir/test/lib/Dialect/Vector/TestVectorTransforms.cpp
index 54defd949c26..ccba2e280686 100644
--- a/mlir/test/lib/Dialect/Vector/TestVectorTransforms.cpp
+++ b/mlir/test/lib/Dialect/Vector/TestVectorTransforms.cpp
@@ -17,6 +17,7 @@
 #include "mlir/Dialect/MemRef/IR/MemRef.h"
 #include "mlir/Dialect/NVGPU/IR/NVGPUDialect.h"
 #include "mlir/Dialect/SCF/IR/SCF.h"
+#include "mlir/Dialect/SCF/Transforms/Patterns.h"
 #include "mlir/Dialect/Tensor/IR/Tensor.h"
 #include "mlir/Dialect/Vector/IR/VectorOps.h"
 #include "mlir/Dialect/Vector/Transforms/LoweringPatterns.h"
@@ -836,9 +837,6 @@ struct TestVectorEmulateMaskedLoadStore final
   }
 };
 
-// TODO: move this code into the user project.
-namespace vendor {
-
 /// Get the set of operand/result types to check for sufficiently
 /// small inner-most dimension size.
 static SmallVector<std::pair<Type, unsigned>>
@@ -960,8 +958,6 @@ struct TestVectorBitWidthLinearize final
   }
 };
 
-} // namespace vendor
-
 struct TestVectorLinearize final
     : public PassWrapper<TestVectorLinearize, OperationPass<>> {
   MLIR_DEFINE_EXPLICIT_INTERNAL_INLINE_TYPE_ID(TestVectorLinearize)
@@ -987,6 +983,8 @@ struct TestVectorLinearize final
     vector::populateVectorLinearizeBasePatterns(converter, target, patterns);
     vector::populateVectorLinearizeShuffleLikeOpsPatterns(converter, target,
                                                           patterns);
+    mlir::scf::populateSCFStructuralTypeConversionsAndLegality(
+        converter, patterns, target);
 
     if (failed(applyPartialConversion(getOperation(), target,
                                       std::move(patterns))))
@@ -1067,7 +1065,7 @@ void registerTestVectorLowerings() {
 
   PassRegistration<TestVectorLinearize>();
 
-  PassRegistration<vendor::TestVectorBitWidthLinearize>();
+  PassRegistration<TestVectorBitWidthLinearize>();
 
   PassRegistration<TestEliminateVectorMasks>();
 }