[mlir][linalg] Add reduction tiling transformation

[ThomasRaoux]

Summary

Add a transformation to tile reduction ops into a parallel operation
followed by a merge operation. This is equivalent to the existing
reduction spliting transformation but using loops instead of using
higher dimensions linalg.

[mravishankar]

Wow! Awesome! I was just going to accept this, but realized I have one question on further reading. There seems to be an implicit assumption about there being a single reduction dimension through the different interface methods. Maybe its better to be more explicit, and just make the dimension (or list of dimensions) that is being tiled be an explicit argument to the interface methods. That will remove that implicit assumption through the different methods.

mlir/include/mlir/Interfaces/TilingInterface.td

163

Is there are reason the interface method has to return a tensor. I think if it returns the identity element that should be enough? I am not opposed to doing this.. This might actually be more general, but trying to understand what is the minimum requirement that the interface should expect ops to implement.

mlir/lib/Dialect/Linalg/Transforms/TilingInterfaceImpl.cpp

273

If I am reading this correctly, the setting of the identity element can be moved into the tileReductionUsingSCFForOp method (related to the comment above)

mlir/lib/Dialect/SCF/Transforms/TileUsingInterface.cpp

424

I think SCF already has a depence on Arith dialect. I think you can replace this with getValueOrCreateConstantIndexOp.

[ThomasRaoux]

In D136586#3879727, @mravishankar wrote:

Wow! Awesome! I was just going to accept this, but realized I have one question on further reading. There seems to be an implicit assumption about there being a single reduction dimension through the different interface methods. Maybe its better to be more explicit, and just make the dimension (or list of dimensions) that is being tiled be an explicit argument to the interface methods. That will remove that implicit assumption through the different methods.

Yes I can do that. I was trying to make the interface more forward looking as I think we should be able to support multiple reductions at some point but I agree it is a bit odd. I'll change the interfaces for now.

mlir/include/mlir/Interfaces/TilingInterface.td

163

If we return the identity element we still need to create op generating the tensor which would be dependent on the op we are tiling. (for linalg we would generate linalg.fill but not for other ops). Where would you do that?

mlir/lib/Dialect/Linalg/Transforms/TilingInterfaceImpl.cpp

273

Getting the reductionOp requires code depending on linalg so I'm not sure how we could do that without adding more interface functions.

[vmurali]

Flyby comment: looks like this differential is essentially doing what is being done in https://github.com/iree-org/iree/pull/10537/files#diff-e105ec843efed97f5884ad26c85561c72967f0546029bd772335634f23553c31, which means I can remove that PR and use your new interface. You seem to have gotten rid of the control for inserted index. Do you no longer need that for general purpose reduction tiling?

[ThomasRaoux]

In D136586#3880177, @vmurali wrote:

Flyby comment: looks like this differential is essentially doing what is being done in https://github.com/iree-org/iree/pull/10537/files#diff-e105ec843efed97f5884ad26c85561c72967f0546029bd772335634f23553c31, which means I can remove that PR and use your new interface. You seem to have gotten rid of the control for inserted index. Do you no longer need that for general purpose reduction tiling?

Yes it could be a potential way to replace this and generalize it by supporting dynamic shapes. The control index can be added to this if needed. Right now the code just makes a decision based on the reduction loop index but for more flexibility there should be a control indeed.

[vmurali]

In D136586#3879727, @mravishankar wrote:

Wow! Awesome! I was just going to accept this, but realized I have one question on further reading. There seems to be an implicit assumption about there being a single reduction dimension through the different interface methods. Maybe its better to be more explicit, and just make the dimension (or list of dimensions) that is being tiled be an explicit argument to the interface methods. That will remove that implicit assumption through the different methods.

I think this OP should _not_ take care of multiple reduction dimensions. There's already a separate transformation op or transformation (that Thomas wrote earlier) that combines multiple contiguous reduction dimensions - so we can use that to combine the sequence of innermost reductions into one. If the reduction dimension is not innermost, then there's no need to apply the technique as parallel tiling will already create tiles that can be reduced in parallel.

In other words, I think this differential should be accepted as is, with comments (and checks) denoting that only innermost dimension must be reduced using this transformation.

[pifon2a]

mlir/include/mlir/Interfaces/TilingInterface.td

200

why are mergeReductions, tileToPartialReduction, generateInitialTensorForPartialReduction a part of tiling interface? They work only for Linalg, don't they?

[Hardcode84]

mlir/lib/Dialect/Linalg/Utils/Utils.cpp

1064

maybe return Optional<Attribute> or FailureOr<Attribute> to force users to check validity?

1079

Did you mean getLargest(semantic, false)? Also, why getLargest and not getInf?

[ThomasRaoux]

Move the reduction detection dimension in the common code and address other review comments.

[ThomasRaoux]

more review comments

[ThomasRaoux]

In D136586#3879727, @mravishankar wrote:

Wow! Awesome! I was just going to accept this, but realized I have one question on further reading. There seems to be an implicit assumption about there being a single reduction dimension through the different interface methods. Maybe its better to be more explicit, and just make the dimension (or list of dimensions) that is being tiled be an explicit argument to the interface methods. That will remove that implicit assumption through the different methods.

Thanks Mahesh. As discussed I moved the code detecting the reduction dimension to the common code but creating the tensor is still done by the interface function.

mlir/include/mlir/Interfaces/TilingInterface.td

200

The idea is that other ops than linalg could implement this interface in the future. For instance we did have to do reduction splitting for linalgext topK and that could be an op that would implement this interface. Also since we need to re-use helpers in the tiling code this allows us to do that without adding dependency between the tiling code and linalg.

[ThomasRaoux]

mlir/lib/Dialect/Linalg/Utils/Utils.cpp

1064

Changed it.

1079

You are right this is a bug. Since I'm only moving this code here I would rather not add the bug fix as part of this patch. I'll send another patch to fix this independently of this change,

[herhut]

mlir/include/mlir/Interfaces/TilingInterface.td

200

Is this really tiling the operation? Yes, at some level it moves the computation to working on tiles but those tiles are not "normal" tiles in the sense of all other operations in this interface. Instead, these methods allow to "distribute" a reduction over multiple threads. So I see how having these methods implemented allows to apply this transformation to anything that is like a reduction but they are not useful otherwise. Maybe have a StructuredReductionOp interface?

mlir/lib/Dialect/Linalg/Transforms/TilingInterfaceImpl.cpp

315

Why is omitPartialTileCheck true here? Do we statically know that the dimension tiles?

mlir/test/Dialect/Linalg/transform-tile-reduction.mlir

38

Why is this valid? What happens if ARG0 does not have a size that is divisible by 5?

[mravishankar]

mlir/include/mlir/Interfaces/TilingInterface.td

189

Could we make this an ArrayRef<unsigned>. Also I have been lead to believe that recommended C++ usage is to use int types unless you are doing bit manipulation.

200

I dont see a reason why this needs to be moved out of TilingInterface into any other interface... If an op doesnt support this kind of tiling then it just does not implement these methods. Also I think there is a final state where we can combine tileToPartialReduction and getTiledImplementation, but that isnt clear yet. So there is a separate method added for now. It is meant to handle tiling of loops that are reductions, which are fine. Haing more interfaces just adds strange dependencies, or results in having redundant interface methods across interfaces.

mlir/lib/Dialect/Linalg/Transforms/TilingInterfaceImpl.cpp

315

The implementation of the loops already handles the partial tile handling. That is accounted for in the sizes. There is no need for an op implementation to account for it.

mlir/test/Dialect/Linalg/transform-tile-reduction.mlir

38

This is a good point. I'd expect there to be an affine.min generated by generateTiledLoopNest for this case....

[ThomasRaoux]

Fix bug in insert_slice size

[ThomasRaoux]

mlir/lib/Dialect/Linalg/Transforms/TilingInterfaceImpl.cpp

315

Yes, this is handled by generateTileLoopNest so the sizes are already adjusted.

mlir/test/Dialect/Linalg/transform-tile-reduction.mlir

38

good catch, there was a bug in the size used to insert. It is fixes now and the size is based on affine_min as expected.

[ThomasRaoux]

Address more review comments.

mlir/include/mlir/Interfaces/TilingInterface.td

189

Changed it to ArrayRef<int>

[ThomasRaoux]

Fix alignments

[pifon2a]

mlir/include/mlir/Interfaces/TilingInterface.td

200

I am worried that this change modifies TilingInterface substantially without a proper RFC and it is not clear whether these methods should be a part of the interface at all.

[mravishankar]

mlir/include/mlir/Interfaces/TilingInterface.td

200

This is a strict addition to the TilingInterface, and is an opt-in. So does not impact anyone who is using TilingInterface. So I dont really think it needs an RFC. Can you give more substantial reasoning as to why this should not be part of the TilingInterface? It is reusing the implementation of these methods in other operations, as well as reusing parts of the implementation of tiling using the interface. That is a strong signal to me that it indeed belongs here. I dont see a reason to block this patch. If indeed it turns out to be muddling separation of concerns we can revisit that, but to me this is already layered correctly.

[pifon2a]

mlir/include/mlir/Interfaces/TilingInterface.td

200

These methods are needed only for a subset of tileable operations. Adding it all here is similar to adding everything to LinalgStructuredOp interface and then spending a lot of time extracting DPS interface out of it, because LinalgOps are just a subset of DPS ops.

[mravishankar]

mlir/include/mlir/Interfaces/TilingInterface.td

200

I don't think they are equivalent. The DPS was deemed not a subset of LinalgStructuredOp, but rather a superset. That's why it was pulled out. As you say this is needed only for a subset of tileable operations. If an operation cannot support this way of Tiling, it just doesn't implement those interface methods. Then you cannot apply this transformation on those ops, and that is exactly how it should work.

[ThomasRaoux]

Move code out of TilingInterface

[ThomasRaoux]

mlir/include/mlir/Interfaces/TilingInterface.td

200

The code is moved all into linalg. This requires exposing more scf helpers.

[mravishankar]

I am not on board with exposing these utility functions. The layering of this going into TilingInterface is the right one which will avoid exposing these methods that are really implementation details. Lets hold off till we can reach a consensus.

[ThomasRaoux]

Moved new interface method to a separate interface outside of TilingInterface

[ThomasRaoux]

Based on discussions I moved the methods to a separate interface in parallel of TilingInterface. Please take another look.

[nicolasvasilache]

Bunch of suggestions for improving code and legitbility but nothing worth delaying more.

Nice stuff, thank you @ThomasRaoux !

mlir/include/mlir/Dialect/Linalg/TransformOps/LinalgTransformOps.td

652

Maybe document that your transform here overallocates for the sequential case (i.e. you could only have a tensor<5xf32> and iteratively reduce within the loop) but that this also works for the parallel case?

Do you expect this to be configurable in the future or to always use the larger allocation?

Edit: I realize the regular tiling to scf.for should do the "smaller allocation version". So now I am wondering whether we see an advantage doing it this way in the sequential case ? (i.e. is this related to @dcaballe 's vectorization comments ?)

mlir/include/mlir/Dialect/SCF/Transforms/TileUsingInterface.h

143

better docs here plz

153

nit: merged

154

super nit: column

160

super nit: extra newline

mlir/include/mlir/Interfaces/TilingInterface.td

189

Our default type is int64_t everywhere, any concrete reason to use int (I don't think the memory usage for a few ints jusifies int)?

213

I like this new interface, not all ops can support partial-reduction tiling without quite more effort (e.g. gather/scatter/sort).
But a big class of ops does support this out of the box.

216

to tile reductions using partial reductions + merge ?

223

nit: methode is French :) (here and below)

233

nit: reductionDims (plural) since you take an ArrayRef ?

mlir/lib/Dialect/Linalg/Transforms/TilingInterfaceImpl.cpp

412

OpBuilder::InsertionGuard guard(b);

In anything that takes a builder reference, otherwise you're leaking insertion point changes across function boundaries which creates bugs that are very hard to track.

419

int64_t everywhere, all the time, plz (unless doing bit manipulations).
If you hit a size_t or unsigned you don't control, please static_cast<int64_t>

432

I don't see a new map here ?

448

Please use ShapedType::isDynamic.
We should be close to be able to make ShapedType::kDynamicSize private now.

466

insertion guard

473

Can we add a proper AffineMap.h/cpp helper:

AffineMap::injectResult(int64_t resultPos, AffineExpr newResult)

and just use that directly:

newMaps.back()  = newMaps.back() .injectResult(...);

474

Some comment here re injecting the split dimension (e0 .. ek .. en) -> (e0 .. ek, split_dim, ek+1 .. en)

483

Better comments plz:

Step 1. ...
Step 1. a. ...

489

Plz add a top-of-the-function // TODO: SubsetExtractOpInterface

511

insertion guard

521

I have a personal preference for fewer loops / less code under loops:

SmallVector<StringRef> reductionIteratorTypes(intermRank, getParallelIteratorTypeName());
reductionIteratorTypes[dimToMerge] = `();
AffineMap outputMap = AffineMap::multiDimSomethingMap().drop_result(dimToMerge); // or the equivalent APIs

523

unsigned -> int64_t everywhere plz

mlir/lib/Dialect/SCF/Transforms/TileUsingInterface.cpp

422

Interesting, we don't have a good way of enforcing that an interface implies another.
Can you add a top-level comment that this is assumed?

436

Can we just use an llvm::count and llvm::any_of/all_of ?
The readability improvements would greatly trump the performance gains of fusion..

int64_t numReductionDims = llvm::count/count_if(lambda);
if (numReductionDims != 1) error;

if (llvm::any_of(lambda))
  return b.notifyMatchFailure(op, "only reduction dimensions can have non zero tile sizes");

457

return b.notifyMatchFailure( for better debugging

460

Can you add a TODO at the func declaration that we should use ArrayRef<OpFoldResult> instead of ArrayRef<Value>?

465

something's fishy between this insertion point and the next .. either the next one crashes or this one does nothing.

475

You need to wrap this into something else to get a real OpFoldResult.
However, with the current broken createOrFold API you can never avoir emitting the constant op ..

483

Is the else ever legitimate ?
This needs a comment with a counter-example.

[herhut]

Thanks for refactoring. Having this in a separate interface addresses my concerns about the partial interface. Did not look at the deep details but also do not want to block this.

[mravishankar]

Nicolas already seems to have reviewed in great detail! The overall structure looks fine to me. Thanks @ThomasRaoux for going through the different iterations to reach convergence.

mlir/include/mlir/Interfaces/TilingInterface.td

213

Agreed! sort can actually support it I think :)

[ThomasRaoux]

Address review comments.

mlir/include/mlir/Dialect/Linalg/TransformOps/LinalgTransformOps.td

652

I'm adding more doc, in general the only case where we would want to allocate less is if the reduction size is smaller than the tile size. This could become an affin.min but I'm not sure we would ever want that. We can iterate if needed.

mlir/lib/Dialect/Linalg/Transforms/TilingInterfaceImpl.cpp

412

even if I don't move the insertion point?
Added to all of those right now

432

Yes the comment was outdated. Rewrote it.

mlir/lib/Dialect/SCF/Transforms/TileUsingInterface.cpp

422

Makes sense, added a comment.

465

oops yeah removed this code

483

I don't think there is, I removed the if

mlir/include/mlir/Dialect/Linalg/TransformOps/LinalgTransformOps.td

def SplitReductionOp : Op<Transform_Dialect, "structured.split_reduction",
   let extraClassDeclaration = [{
     ::mlir::DiagnosedSilenceableFailure applyToOne(
         ::mlir::linalg::LinalgOp target, 
         ::llvm::SmallVectorImpl<::mlir::Operation *> &results, 
         ::mlir::transform::TransformState &state);
   }];
 }
 
+
+def TileReductionUsingScfOp : Op<Transform_Dialect, "structured.tile_reduction_using_scf",
+       [FunctionalStyleTransformOpTrait, MemoryEffectsOpInterface,
+        TransformEachOpTrait, TransformOpInterface]> {
+  let description = [{
+    Indicates that the given `target` op should be transformed with the 
+    `tileReduction` transformation with the tile size provided as attribute.
+    
+    This transformation tiles the `target` along the reduction dimensions. It
+    creates a tensor initialized with the identity value. Then it creates nested
+    loops with a parallel version of `target` op inside. The parallel op
+    dimensions are less or equal to the tile size passed by user.
+    After the loop a merge operation is created to do a final reduction with the
+    partial reductions.
+
+    #### Return modes
+
+    This 3 returned handles point to:
+      - the fill op used to initialize the neutral element, 
+      - the parallel tiled op and 
+      - the result-combining op.
+
+    #### Example:
+    
+    ```
+      %red = linalg.generic {indexing_maps = [affine_map<(d0, d1) -> (d0, d1)>,
+                                              affine_map<(d0, d1) -> (d0)>],
+      iterator_types = ["parallel", "reduction"]}
+      ins(%arg0 : tensor<?x?xf32>)
+      outs(%out : tensor<?xf32>) {
+        ^bb0(%arg7: f32, %arg9: f32):
+        %1 = arith.addf %arg7, %arg9 : f32
+        linalg.yield %1 : f32
+      } -> tensor<?xf32>
+      return %red : tensor<?xf32>
+    ```
+  
+    is transformed into:
+    
+    ```
+      %0 = tensor.empty(%dim_1) : tensor<?x5xf32>
+      %1 = linalg.fill ins(%cst : f32) outs(%0 : tensor<?x5xf32>) -> tensor<?x5xf32>
+      %2 = scf.for %arg2 = %c0 to %dim_0 step %c5 iter_args(%arg3 = %1) -> (tensor<?x5xf32>) {
+        %extracted_slice = tensor.extract_slice %1[0, 0] [%dim, 5] [1, 1] : tensor<?x5xf32> to tensor<?x5xf32>
+        %extracted_slice_2 = tensor.extract_slice %arg0[0, %arg2] [%dim, 5] [1, 1] : tensor<?x?xf32> to tensor<?x5xf32>
+        %4 = linalg.generic {indexing_maps = [affine_map<(d0, d1) -> (d0, d1)>, 
+                                              affine_map<(d0, d1) -> (d0, d1)>],
+        iterator_types = ["parallel", "parallel"]}
+        ins(%extracted_slice_2 : tensor<?x5xf32>)
+        outs(%extracted_slice : tensor<?x5xf32>) {
+        ^bb0(%in: f32, %out: f32):
+          %5 = arith.addf %in, %out : f32
+          linalg.yield %5 : f32
+        } -> tensor<?x5xf32>
+        %dim_3 = tensor.dim %1, %c0 : tensor<?x5xf32>
+        %inserted_slice = tensor.insert_slice %4 into %arg3[0, 0] [%dim_3, 5] [1, 1] : tensor<?x5xf32> into tensor<?x5xf32>
+        scf.yield %inserted_slice : tensor<?x5xf32>
+      }
+      %3 = linalg.generic {indexing_maps = [affine_map<(d0, d1) -> (d0, d1)>,
+                                            affine_map<(d0, d1) -> (d0)>],
+      iterator_types = ["parallel", "reduction"]}
+      ins(%2 : tensor<?x5xf32>)
+      outs(%arg1 : tensor<?xf32>) {
+      ^bb0(%in: f32, %out: f32):
+        %4 = arith.addf %in, %out : f32
+        linalg.yield %4 : f32
+      } -> tensor<?xf32>
+    ```
+  }];
+
+  let arguments = (ins PDL_Operation:$target,
+                   DefaultValuedAttr<I64ArrayAttr, "{}">:$tile_sizes);
+  let results = (outs PDL_Operation:$fill_op,
+                      PDL_Operation:$split_linalg_op,
+                      PDL_Operation:$combining_linalg_op);
+
+  let assemblyFormat = "$target attr-dict";
+
+  let extraClassDeclaration = [{
+    ::mlir::DiagnosedSilenceableFailure applyToOne(
+        ::mlir::linalg::LinalgOp target, 
+        ::llvm::SmallVectorImpl<::mlir::Operation *> &results, 
+        ::mlir::transform::TransformState &state);
+  }];
+}
+
 def TileOp : Op<Transform_Dialect, "structured.tile",
        [DeclareOpInterfaceMethods<TransformOpInterface>,
         DeclareOpInterfaceMethods<MemoryEffectsOpInterface>]> {
   let description = [{
     Indicates that the given `target` op should be tiled with the given sizes.
     This transform generates a loop nest with a smaller ("tiled") target
     operation in its body. Currently limited to LinalgOps.
 

mlir/include/mlir/Dialect/Linalg/Utils/Utils.h

 /// of a insert_slice) operation with given offsets, and sizes to its
 /// rank-reduced version. This is only done for the cases where the size is 1
 /// and offset is 0. Strictly speaking the offset 0 is not required in general,
 /// but non-zero offsets are not handled by SPIR-V backend at this point (and
 /// potentially cannot be handled).
 Optional<SmallVector<ReassociationIndices>>
 getReassociationMapForFoldingUnitDims(ArrayRef<OpFoldResult> mixedSizes);
 
+/// Return the identity numeric value associated to the give op.
+Attribute getNeutralElement(Operation *op);
+
 //===----------------------------------------------------------------------===//
 // Fusion / Tiling utilities
 //===----------------------------------------------------------------------===//
 
 /// The type of loops to be generated during tiling.
 enum class LinalgTilingLoopType {
   Loops = 0,
   AffineLoops = 1,

mlir/include/mlir/Dialect/SCF/Transforms/TileUsingInterface.h

tileConsumerAndFuseProducerGreedilyUsingSCFForOp(
     RewriterBase &rewriter, TilingInterface consumer,
     const SCFTileAndFuseOptions &options);
 
 /// Method to lower an `op` that implements the `TilingInterface` to
 /// loops/scalars.
 FailureOr<SmallVector<scf::ForOp>>
 lowerToLoopsUsingSCFForOp(RewriterBase &rewriter, TilingInterface op);
 
+/// Transformation information returned after reduction tiling.
+struct SCFReductionTilingResult {
+  /// The tiled operation generated.
+  Operation *parallelTiledOp;
+  /// The tiled operation generated.
+  Operation *mergeOp;
+  /// Initial op
+  Operation *initialOp;
+  /// The `scf.for` operations that iterate over the tiles.
+  SmallVector<scf::ForOp> loops;
+};
+
+/// Method to tile a reduction and generate a parallel op within a serial loop.
+/// Each of the partial reductions are calculated in parallel. Then after the
+/// loop all the partial reduction are mered into a final reduction.
+/// For example, for the following sequence
+///
+/// ```mlir
+/// %0 = linalg.generic %in ["parallel", "reduction"]
+///   : tensor<7x9xf32> -> tensor<7xf32>
+/// ```
+/// into:
+///
+/// ```mlir
+/// %0 = linalg.fill ... : tensor<7x4xf32>
+/// %1 = scf.for ... iter_args(%arg0 = %0)
+///   %2 = tensor.extract_slice %arg0 : tensor<7x4xf32> -> tensor<7x?xf32>
+///   %3 = tensor.extract_slice %in : tensor<7x9xf32> -> tensor<7x?xf32>
+///   %4 = linalg.generic %2, %3 ["parallel", "parallel"]
+///     : tensor<7x?xf32> -> tensor<7x?xf32>
+///   %5 = tensor.insert_slice %3, %0[0, 0] : tensor<7x4xf32>
+/// }
+/// %6 = linalg.generic %1 ["parallel", "reduction"]
+///   : tensor<7x4xf32> -> tensor<7xf32>
+/// ```
+FailureOr<scf::SCFReductionTilingResult>
+tileReductionUsingScf(PatternRewriter &b, TilingInterface op,
+                      ArrayRef<OpFoldResult> tileSize);
+
 } // namespace scf
 } // namespace mlir
 
 #endif // MLIR_DIALECT_SCF_TRANSFORMS_TILEUSINGINTERFACE_H

mlir/include/mlir/Interfaces/TilingInterface.td

let methods = [
         /*args=*/(ins
             "OpBuilder &":$b,
             "Location ":$loc,
             "ValueRange ":$ivs),
         /*methodBody=*/"",
         /*defaultImplementation=*/[{
           return failure();
         }]
+      >,
+      InterfaceMethod<
+        /*desc=*/[{
+          Methode to generate a tensor initalized with the identity value of the
+          operation reduction. The tensor shape is equal to operation result
+          shape with new dimension for each non zero tile size.
+        }],
+        /*retType=*/"FailureOr<Operation*>",
+        /*methodName=*/"generateInitialTensorForPartialReduction",
+        /*args=*/(ins
+            "OpBuilder &":$b,
+            "Location ":$loc,
+            "ArrayRef<OpFoldResult>":$sizes),
+        /*methodBody=*/"",
+        /*defaultImplementation=*/[{
+          return failure();
+        }]
+      >,
+      InterfaceMethod<
+        /*desc=*/[{
+          Methode to generate a tiled version of the operation where the tiled
+          reduction dimension are converted to parallel dimensions with a size
+          less or equal to the tile size. This is meant to be used with
+          `mergeReductions` methode which will combine the partial reductions.
+        }],
+        /*retType=*/"Operation*",
+        /*methodName=*/"tileToPartialReduction",
+        /*args=*/(ins
+            "OpBuilder &":$b,
+            "Location ":$loc,
+            "ValueRange":$init,
+            "ArrayRef<OpFoldResult>":$offsets,
+            "ArrayRef<OpFoldResult>":$sizes),
+        /*methodBody=*/"",
+        /*defaultImplementation=*/[{
+          return nullptr;
+        }]
+      >,
+      InterfaceMethod<
+        /*desc=*/[{
+          Methode to merge partial reductions for an operation that has been
+          tiled along the reduction dimensions. This will only apply the
+          reduction the operation.
+        }],
+        /*retType=*/"Operation*",
+        /*methodName=*/"mergeReductions",
+        /*args=*/(ins
+            "OpBuilder &":$b,
+            "Location ":$loc,
+            "ValueRange":$partialReduce),
+        /*methodBody=*/"",
+        /*defaultImplementation=*/[{
+          return nullptr;
+        }]
       >
   ];  
 }
 #endif // MLIR_TILINGINTERFACE

mlir/lib/Dialect/Linalg/TransformOps/LinalgTransformOps.cpp

transform::SplitReductionOp::applyToOne(linalg::LinalgOp target,
   results.push_back(splitResult->initOrAlloc);
   results.push_back(splitResult->fillOp);
   results.push_back(splitResult->splitLinalgOp);
   results.push_back(splitResult->resultCombiningLinalgOp);
   return DiagnosedSilenceableFailure(success());
 }
 
 //===----------------------------------------------------------------------===//
+// SplitReductionOp
+//===----------------------------------------------------------------------===//
+
+DiagnosedSilenceableFailure transform::TileReductionUsingScfOp::applyToOne(
+    linalg::LinalgOp target, SmallVectorImpl<Operation *> &results,
+    transform::TransformState &state) {
+  SimpleRewriter rewriter(getContext());
+  rewriter.setInsertionPoint(target);
+  SmallVector<int64_t> tileSizes = extractFromI64ArrayAttr(getTileSizes());
+  SmallVector<OpFoldResult> sizes;
+  for (int64_t size : tileSizes) {
+    sizes.push_back(rewriter.getIndexAttr(size));
+  }
+
+  FailureOr<scf::SCFReductionTilingResult> result = scf::tileReductionUsingScf(
+      rewriter, cast<TilingInterface>(target.getOperation()), sizes);
+
+  if (failed(result))
+    return DiagnosedSilenceableFailure(reportUnknownTransformError(target));
+  results.push_back(result->initialOp);
+  results.push_back(result->parallelTiledOp);
+  results.push_back(result->mergeOp);
+  return DiagnosedSilenceableFailure(success());
+}
+
+//===----------------------------------------------------------------------===//
 // TileOp
 //===----------------------------------------------------------------------===//
 
 DiagnosedSilenceableFailure
 transform::TileOp::apply(TransformResults &transformResults,
                          TransformState &state) {
   SmallVector<int64_t> tileSizes = extractFromI64ArrayAttr(getStaticSizes());
 

mlir/lib/Dialect/Linalg/Transforms/SplitReduction.cpp

 #include "mlir/Dialect/Linalg/Utils/Utils.h"
 #include "mlir/Dialect/Tensor/IR/Tensor.h"
 #include "mlir/Dialect/Tensor/Utils/Utils.h"
 #include "mlir/IR/PatternMatch.h"
 
 using namespace mlir;
 using namespace mlir::linalg;
 
-/// Return the identity numeric value associated to the give op.
-static Attribute getNeutralElement(Operation *op) {
-  // Builder only used as helper for attribute creation.
-  OpBuilder b(op->getContext());
-  Type resultType = op->getResult(0).getType();
-  if (auto floatType = resultType.dyn_cast<FloatType>()) {
-    const llvm::fltSemantics &semantic = floatType.getFloatSemantics();
-    if (isa<arith::AddFOp>(op))
-      return b.getFloatAttr(resultType, llvm::APFloat::getZero(semantic));
-    if (isa<arith::MulFOp>(op))
-      return b.getFloatAttr(resultType, llvm::APFloat(semantic, 1));
-    if (isa<arith::MaxFOp>(op))
-      return b.getFloatAttr(resultType,
-                            llvm::APFloat::getLargest(semantic, true));
-    if (isa<arith::MinFOp>(op))
-      return b.getFloatAttr(resultType,
-                            llvm::APFloat::getLargest(semantic, true));
-    return Attribute();
-  }
-  if (isa<arith::AddIOp, arith::OrIOp, arith::XOrIOp>(op))
-    return b.getIntegerAttr(resultType, 0);
-  if (isa<arith::AndIOp>(op))
-    return b.getIntegerAttr(resultType, -1);
-  if (isa<arith::MaxSIOp>(op))
-    return b.getIntegerAttr(resultType, std::numeric_limits<int64_t>::min());
-  if (isa<arith::MinSIOp>(op))
-    return b.getIntegerAttr(resultType, std::numeric_limits<int64_t>::max());
-  if (isa<arith::MulIOp>(op))
-    return b.getIntegerAttr(resultType, 1);
-  return Attribute();
-}
-
 FailureOr<SplitReductionResult> mlir::linalg::splitReduction(
     PatternRewriter &b, LinalgOp op,
     const ControlSplitReductionFn &controlSplitReductionFn, bool useAlloc) {
   OpBuilder::InsertionGuard guard(b);
   b.setInsertionPoint(op);
 
   SplitReductionOptions control = controlSplitReductionFn(op);
   int64_t ratio = control.ratio;

mlir/lib/Dialect/Linalg/Transforms/TilingInterfaceImpl.cpp

 //===- TilingInterfaceImpl.cpp - Implementation of TilingInterface -------===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
 
 #include "mlir/Dialect/Linalg/Transforms/TilingInterfaceImpl.h"
 
+#include "mlir/Analysis/SliceAnalysis.h"
 #include "mlir/Dialect/Affine/IR/AffineOps.h"
 #include "mlir/Dialect/Arith/IR/Arith.h"
 #include "mlir/Dialect/Arith/Utils/Utils.h"
 #include "mlir/Dialect/Linalg/IR/Linalg.h"
 #include "mlir/Dialect/Linalg/Utils/Utils.h"
 #include "mlir/Dialect/MemRef/IR/MemRef.h"
 #include "mlir/Dialect/Tensor/IR/Tensor.h"
 #include "mlir/Dialect/Utils/StaticValueUtils.h"
for (OpOperand &operand : linalgOp->getOpOperands()) {
       Value load =
           builder.create<memref::LoadOp>(linalgOpLoc, operand.get(), indices);
       indexedValues.push_back(load);
     }
 
     /// Inline the op payload and store the result.
     return inlinePayload(builder, linalgOp, ivs, indexedValues);
   }
+
+  FailureOr<Operation *>
+  generateInitialTensorForPartialReduction(Operation *op, OpBuilder &b,
+                                           Location loc,
+                                           ArrayRef<OpFoldResult> sizes) const {
+    auto linalgOp = cast<LinalgOp>(op);
+    if (linalgOp.hasBufferSemantics())
+      return op->emitOpError("expected operation to have tensor semantics");
+    if (linalgOp->getNumResults() != 1)
+      return op->emitOpError(
+          "only support linalg operation with single return value.");
+    SmallVector<unsigned> dims;
+    linalgOp.getReductionDims(dims);
+    if (dims.size() != 1)
+      return op->emitOpError(
+          "only support linalg operation with one reduction dimension.");
+    for (auto &[idx, size] : llvm::enumerate(sizes)) {
+      if (idx != dims[0] && !isConstantIntValue(size, 0))
+        return op->emitOpError(
+            "Only reduction dimensions can have a non zero tile size.");
+    }
+    // Insert the new parallel dimension based on the index of the reduction
+    // loop. This could be controlled by user for more flexibility.
+    unsigned insertSplitDimension = dims[0];
+
+    SmallVector<Operation *, 4> combinerOps;
+    if (!matchReduction(linalgOp.getRegionOutputArgs(), 0, combinerOps) ||
+        combinerOps.size() != 1)
+      return op->emitOpError("Failed to anaysis the reduction operation.");
+
+    Operation *reductionOp = combinerOps[0];
+    Attribute identity = getNeutralElement(reductionOp);
+    if (!identity)
+      return op->emitOpError(
+          "Failed to get an identity value for the reduction operation.");
+
+    // Calculate the new output map and shape, we insert the new dimension in
+    // the intermediate tensor based on the index returned by
+    // `controlSplitReductionFn`.
+    SmallVector<int64_t> newOutputShape;
+    ArrayRef<int64_t> oldShape =
+        linalgOp.getShape(linalgOp.getOutputOperand(0));
+    SmallVector<Value> dynamicDims;
+    for (unsigned idx : llvm::seq<unsigned>(0, oldShape.size() + 1)) {
+      if (idx == insertSplitDimension) {
+        dispatchIndexOpFoldResults(sizes[idx], dynamicDims, newOutputShape,
+                                   ShapedType::kDynamicStrideOrOffset);
+        continue;
+      }
+      unsigned oldIdx = idx < insertSplitDimension ? idx : idx - 1;
+      int64_t dim = oldShape[oldIdx];
+      newOutputShape.push_back(dim);
+      if (dim == ShapedType::kDynamicSize)
+        dynamicDims.push_back(b.createOrFold<tensor::DimOp>(
+            loc, linalgOp.getOutputOperand(0)->get(), oldIdx));
+    }
+    Value emptyTensor = b.create<tensor::EmptyOp>(
+        loc, newOutputShape, linalgOp.getRegionOutputArgs()[0].getType(),
+        dynamicDims);
+    Value constantOp = b.create<arith::ConstantOp>(loc, identity);
+    auto identityTensor =
+        b.create<linalg::FillOp>(loc, constantOp, emptyTensor);
+    return identityTensor.getOperation();
+  }
+
+  Operation *tileToPartialReduction(Operation *op, OpBuilder &b, Location loc,
+                                    ValueRange init,
+                                    ArrayRef<OpFoldResult> offsets,
+                                    ArrayRef<OpFoldResult> sizes) const {
+    auto linalgOp = cast<LinalgOp>(op);
+    SmallVector<unsigned> dims;
+    linalgOp.getReductionDims(dims);
+    assert(dims.size() == 1);
+    unsigned reductionDim = dims[0];
+    unsigned insertSplitDimension = reductionDim;
+
+    AffineMap oldOutputMap =
+        linalgOp.getMatchingIndexingMap(linalgOp.getOutputOperand(0));
+    SmallVector<AffineExpr> outputExpr;
+    for (auto &[idx, expr] : llvm::enumerate(oldOutputMap.getResults())) {
+      if (idx == insertSplitDimension) {
+        outputExpr.push_back(b.getAffineDimExpr(reductionDim));
+      }
+      outputExpr.push_back(expr);
+    }
+    if (insertSplitDimension == oldOutputMap.getNumResults())
+      outputExpr.push_back(b.getAffineDimExpr(reductionDim));
+
+    SmallVector<Value> valuesToTile = linalgOp.getInputOperands();
+    SmallVector<Value, 4> tiledOperands =
+        makeTiledShapes(b, loc, op, valuesToTile, offsets, sizes, {}, true);
+
+    SmallVector<OpFoldResult> strides(offsets.size(), b.getIndexAttr(1));
+    SmallVector<OpFoldResult> outOffsets(offsets.size(), b.getIndexAttr(0));
+    Value out = b.create<tensor::ExtractSliceOp>(loc, init[0], outOffsets,
+                                                 sizes, strides);
+
+    // Create the new op matching the original op where the reduction dimension
+    // is now parallel.
+    SmallVector<StringRef> newIteratorTypes = linalgOp.getIteratorTypesArray();
+    newIteratorTypes[reductionDim] = getParallelIteratorTypeName();
+    SmallVector<AffineMap> newMaps = linalgOp.getIndexingMapsArray();
+    newMaps.back() = AffineMap::get(newMaps.back().getNumDims(), 0, outputExpr,
+                                    linalgOp.getContext());
+    auto genericOp =
+        b.create<GenericOp>(loc, TypeRange({out.getType()}), tiledOperands,
+                            ValueRange({out}), newMaps, newIteratorTypes);
+    BlockAndValueMapping mapping;
+    op->getRegion(0).cloneInto(&genericOp.getRegion(),
+                               genericOp.getRegion().begin(), mapping);
+    return genericOp.getOperation();
+  }
+
+  Operation *mergeReductions(Operation *op, OpBuilder &b, Location loc,
+                             ValueRange partialReduce) const {
+    auto linalgOp = cast<LinalgOp>(op);
+    SmallVector<unsigned> dims;
+    linalgOp.getReductionDims(dims);
+    assert(dims.size() == 1);
+    unsigned dimToMerge = dims[0];
+
+    // Then create a new reduction that only reduce the newly added dimension
+    // from the previous op.
+    unsigned intermRank =
+        partialReduce[0].getType().cast<ShapedType>().getRank();
+    AffineMap inputMap = b.getMultiDimIdentityMap(intermRank);
+    SmallVector<StringRef> reductionIteratorTypes;
+    SmallVector<AffineExpr> exprs;
+    for (unsigned i : llvm::seq<unsigned>(0, intermRank)) {
+      if (dimToMerge == i) {
+        reductionIteratorTypes.push_back(getReductionIteratorTypeName());
+      } else {
+        exprs.push_back(b.getAffineDimExpr(i));
+        reductionIteratorTypes.push_back(getParallelIteratorTypeName());
+      }
+    }
+    AffineMap outputMap =
+        AffineMap::get(intermRank, 0, exprs, op->getContext());
+    SmallVector<AffineMap> reductionMaps = {inputMap, outputMap};
+
+    SmallVector<Operation *, 4> combinerOps;
+    matchReduction(linalgOp.getRegionOutputArgs(), 0, combinerOps);
+    Operation *reductionOp = combinerOps[0];
+
+    auto reduction = b.create<GenericOp>(
+        loc, op->getResultTypes(), ValueRange({partialReduce[0]}),
+        SmallVector<Value>{linalgOp.getOutputOperands()}, reductionMaps,
+        reductionIteratorTypes,
+        [reductionOp](OpBuilder &b, Location loc, ValueRange inputs) {
+          Operation *clonedReductionOp = b.clone(*reductionOp);
+          clonedReductionOp->setOperand(0, inputs[0]);
+          clonedReductionOp->setOperand(1, inputs[1]);
+          b.create<linalg::YieldOp>(loc, clonedReductionOp->getResult(0));
+        });
+    return reduction.getOperation();
+  }
 };
 
 } // namespace
 
 template <typename OpType>
 static void registerOne(MLIRContext *ctx) {
   OpType::template attachInterface<LinalgOpTilingInterface<OpType>>(*ctx);
 }
 
 /// Variadic helper function.
 template <typename... OpTypes>
 static void registerAll(MLIRContext *ctx) {
   (registerOne<OpTypes>(ctx), ...);
 }
 
 #define GET_OP_LIST
 
 void mlir::linalg::registerTilingInterfaceExternalModels(
     DialectRegistry &registry) {
   registry.addExtension(+[](MLIRContext *ctx, linalg::LinalgDialect *dialect) {
     registerOne<linalg::GenericOp>(ctx);
     registerAll<
 #include "mlir/Dialect/Linalg/IR/LinalgStructuredOps.cpp.inc"
         >(ctx);
   });
 }

mlir/lib/Dialect/Linalg/Utils/Utils.cpp

computeAllSliceParameters(OpBuilder &builder, Location loc, LinalgOp linalgOp,
 
   // Construct (potentially temporary) mins and maxes on which to apply maps
   // that define tile subshapes.
   SmallVector<OpFoldResult> lbs =
       computeTileOffsets(builder, loc, ivs, tileSizes);
   SmallVector<OpFoldResult> subShapeSizes =
       computeTileSizes(builder, loc, tileSizes, sizeBounds);
 
-  assert(static_cast<int64_t>(valuesToTile.size()) ==
+  assert(static_cast<int64_t>(valuesToTile.size()) <=
              linalgOp->getNumOperands() &&
-         "expected one value to tile for every operand");
+         "more value to tile than operands.");
   SmallVector<Optional<SliceParameters>> allSliceParams;
   allSliceParams.reserve(valuesToTile.size());
-  for (OpOperand &opOperand : linalgOp->getOpOperands()) {
-    Value shapedOp = valuesToTile[opOperand.getOperandNumber()];
+  for (auto [opOperand, val] :
+       llvm::zip(linalgOp->getOpOperands(), valuesToTile)) {
+    Value shapedOp = val;
     LLVM_DEBUG(llvm::dbgs() << "makeTiledShapes: for operand " << shapedOp);
     AffineMap map = linalgOp.getMatchingIndexingMap(&opOperand);
     // Use `opOperand` as is if it is not tiled and not an output tensor. Having
     // an extract/insert slice pair for all output tensors simplifies follow up
     // transformations such as padding and bufferization since the
     // extract/insert slice pairs make the accessed iteration argument
     // subdomains explicit.
 
getReassociationMapForFoldingUnitDims(ArrayRef<OpFoldResult> mixedSizes) {
   // When the reassociations are not empty, then fold the remaining
   // unit-dimensions into the last dimension.  If the reassociations so far is
   // empty, then leave it emtpy. This will fold everything to a rank-0 tensor.
   if (!curr.empty() && !reassociation.empty())
     reassociation.back().append(curr.begin(), curr.end());
   return reassociation;
 }
 
+/// Return the identity numeric value associated to the give op.
+Attribute getNeutralElement(Operation *op) {
+  // Builder only used as helper for attribute creation.
+  OpBuilder b(op->getContext());
+  Type resultType = op->getResult(0).getType();
+  if (auto floatType = resultType.dyn_cast<FloatType>()) {
+    const llvm::fltSemantics &semantic = floatType.getFloatSemantics();
+    if (isa<arith::AddFOp>(op))
+      return b.getFloatAttr(resultType, llvm::APFloat::getZero(semantic));
+    if (isa<arith::MulFOp>(op))
+      return b.getFloatAttr(resultType, llvm::APFloat(semantic, 1));
+    if (isa<arith::MaxFOp>(op))
+      return b.getFloatAttr(resultType,
+                            llvm::APFloat::getLargest(semantic, true));
+    if (isa<arith::MinFOp>(op))
+      return b.getFloatAttr(resultType,
+                            llvm::APFloat::getLargest(semantic, true));
+    return Attribute();
+  }
+  if (isa<arith::AddIOp, arith::OrIOp, arith::XOrIOp>(op))
+    return b.getIntegerAttr(resultType, 0);
+  if (isa<arith::AndIOp>(op))
+    return b.getIntegerAttr(resultType, -1);
+  if (isa<arith::MaxSIOp>(op))
+    return b.getIntegerAttr(resultType, std::numeric_limits<int64_t>::min());
+  if (isa<arith::MinSIOp>(op))
+    return b.getIntegerAttr(resultType, std::numeric_limits<int64_t>::max());
+  if (isa<arith::MulIOp>(op))
+    return b.getIntegerAttr(resultType, 1);
+  return Attribute();
+}
+
 } // namespace linalg
 } // namespace mlir

mlir/lib/Dialect/SCF/Transforms/TileUsingInterface.cpp

if (!tilingResult.loops.empty()) {
       llvm::dbgs() << "After tiled implementation :\n";
       tilingResult.loops.front().dump();
       llvm::dbgs() << "\n";
     }
   });
   return tilingResult;
 }
 
+FailureOr<scf::SCFReductionTilingResult>
+mlir::scf::tileReductionUsingScf(PatternRewriter &b, TilingInterface op,
+                                 ArrayRef<OpFoldResult> tileSize) {
+  Location loc = op.getLoc();
+  auto tilingInterfaceOp = cast<TilingInterface>(op.getOperation());
+  SmallVector<Range> iterationDomain = tilingInterfaceOp.getIterationDomain(b);
+
+  SmallVector<Value> tileSizeVector = llvm::to_vector<4>(
+      llvm::map_range(tileSize, [&](OpFoldResult value) -> Value {
+        if (auto attr = value.dyn_cast<Attribute>())
+          return b.create<arith::ConstantOp>(loc, attr);
+        return value.get<Value>();
+      }));
+  if (tileSizeVector.size() < iterationDomain.size()) {
+    auto zero = b.create<arith::ConstantIndexOp>(loc, 0);
+    tileSizeVector.append(iterationDomain.size() - tileSizeVector.size(), zero);
+  }
+
+  // 1. create the inital tensor value.
+  FailureOr<Operation *> identityTensor =
+      op.generateInitialTensorForPartialReduction(b, loc, tileSize);
+
+  if (failed(identityTensor))
+    return failure();
+  // 2. Create the nested loops.
+  SmallVector<OpFoldResult> offsets, sizes;
+  SmallVector<scf::ForOp> loops = generateTileLoopNest(
+      b, loc, iterationDomain, tileSizeVector, offsets, sizes);
+
+  // 3. Generate the tiled implementation within the inner most loop.
+  if (!loops.empty())
+    b.setInsertionPoint(loops.back().getBody()->getTerminator());
+
+  b.setInsertionPoint(loops.back().getBody()->getTerminator());
+  Operation *parallelOp = op.tileToPartialReduction(
+      b, loc, identityTensor.value()->getResults(), offsets, sizes);
+
+  SmallVector<OpFoldResult> resultSizesList;
+  for (int64_t i = 0; i < offsets.size(); i++)
+    resultSizesList.push_back(b.createOrFold<tensor::DimOp>(
+        loc, identityTensor.value()->getResult(0), i));
+  SmallVector<OpFoldResult> outOffsets(offsets.size(), b.getIndexAttr(0));
+  FailureOr<SmallVector<Value>> replacementOr = yieldTiledValues(
+      b, identityTensor.value()->getResults(), parallelOp->getResults(),
+      outOffsets, resultSizesList, loops);
+  if (failed(replacementOr))
+    return b.notifyMatchFailure(op, "failed to yield replacement");
+  if (auto tiledInterfaceOp = dyn_cast<TilingInterface>(parallelOp)) {
+    auto innerMostLoop = loops.back();
+    updateDestinationOperandsForTiledOp(
+        b, tiledInterfaceOp.getDestinationOperands(b),
+        innerMostLoop.getRegionIterArgs());
+  }
+
+  // 4. Apply the merge reduction to combine all the partial values.
+  b.setInsertionPointAfter(*loops.begin());
+  Operation *mergeOp = op.mergeReductions(b, loc, replacementOr.value());
+  b.replaceOp(op, mergeOp->getResults());
+
+  SCFReductionTilingResult results;
+  results.initialOp = identityTensor.value();
+  results.loops = std::move(loops);
+  results.parallelTiledOp = parallelOp;
+  results.mergeOp = mergeOp;
+  return results;
+}
 //===----------------------------------------------------------------------===//
 // tileConsumerAndFuseProducerGreedilyUsingSCFForOp implementation.
 //===----------------------------------------------------------------------===//
 
 /// Return the untiled producer whose slice is used in a tiled consumer. The
 /// method traverses the tile loop nest (`loops`) if needed, and returns the
 /// `iter_args` of the outer most that is encountered. Traversing the iter_args
 /// indicates that this is a destination operand of the consumer. If there was
 /// no loop traversal needed, the second value of the returned tuple is empty.

mlir/test/Dialect/Linalg/transform-tile-reduction.mlir

+// RUN: mlir-opt %s -test-transform-dialect-interpreter -split-input-file -canonicalize | FileCheck %s
+
+func.func @reduction_tile(%arg0: tensor<?x?xf32>, %out: tensor<?xf32>) -> tensor<?xf32> {
+  %red = linalg.generic {indexing_maps = [affine_map<(d0, d1) -> (d0, d1)>,
+                                          affine_map<(d0, d1) -> (d0)>],
+   iterator_types = ["parallel", "reduction"]}
+   ins(%arg0 : tensor<?x?xf32>)
+   outs(%out : tensor<?xf32>) {
+    ^bb0(%arg7: f32, %arg9: f32):
+      %1 = arith.mulf %arg7, %arg7 : f32
+      %2 = arith.addf %1, %arg9 : f32
+      linalg.yield %2 : f32
+    } -> tensor<?xf32>
+  return %red : tensor<?xf32>
+}
+
+transform.sequence failures(propagate) {
+^bb0(%arg1: !pdl.operation):
+  %0 = transform.structured.match ops{["linalg.generic"]} in %arg1
+  %1, %2, %3 = transform.structured.tile_reduction_using_scf %0 { tile_sizes = [0, 5] }
+}
+
+//     CHECK: #[[MAP0:.*]] = affine_map<(d0, d1) -> (d0, d1)>
+//     CHECK: #[[MAP1:.*]] = affine_map<(d0, d1) -> (d0)>
+//     CHECK: func @reduction_tile(%[[ARG0:.+]]: tensor<?x?xf32>, %[[ARG1:.+]]: tensor<?xf32>
+// CHECK-DAG:   %[[I:.*]] = arith.constant 0.000000e+00 : f32
+// CHECK-DAG:   %[[C5:.*]] = arith.constant 5 : index
+// CHECK-DAG:   %[[C1:.*]] = arith.constant 1 : index
+// CHECK-DAG:   %[[C0:.*]] = arith.constant 0 : index
+// CHECK-DAG:   %[[D0:.*]] = tensor.dim %[[ARG0]], %[[C0]] : tensor<?x?xf32>
+// CHECK-DAG:   %[[D1:.*]] = tensor.dim %[[ARG0]], %[[C1]] : tensor<?x?xf32>
+// CHECK-DAG:   %[[D2:.*]] = tensor.dim %[[ARG1]], %[[C0]] : tensor<?xf32>
+//     CHECK:   %[[E:.*]] = tensor.empty(%[[D2]]) : tensor<?x5xf32>
+//     CHECK:   %[[F:.*]] = linalg.fill ins(%[[I]] : f32) outs(%[[E]] : tensor<?x5xf32>) -> tensor<?x5xf32>
+//     CHECK:   %[[L:.*]] = scf.for %[[K:.*]] = %[[C0]] to %[[D1]] step %[[C5]] iter_args(%[[ARG3:.*]] = %[[F]]) -> (tensor<?x5xf32>) {
+//     CHECK:     %[[EXT:.*]] = tensor.extract_slice %[[ARG3]][0, 0] [%[[D0]], 5] [1, 1] : tensor<?x5xf32> to tensor<?x5xf32>
+//     CHECK:     %[[EXT2:.*]] = tensor.extract_slice %[[ARG0]][0, %[[K:.*]]] [%[[D0]], 5] [1, 1] : tensor<?x?xf32> to tensor<?x5xf32>
+//     CHECK:     %[[PR:.*]] = linalg.generic {indexing_maps = [#[[MAP0]], #[[MAP0]]], iterator_types = ["parallel", "parallel"]} ins(%[[EXT2]] : tensor<?x5xf32>) outs(%[[EXT]] : tensor<?x5xf32>) {
+//     CHECK:       arith.mulf
+//     CHECK:       arith.addf
+//     CHECK:       linalg.yield
+//     CHECK:     } -> tensor<?x5xf32>
+//     CHECK:     %[[D3:.*]] = tensor.dim %[[F]], %[[C0]] : tensor<?x5xf32>
+//     CHECK:     %[[INS:.*]] = tensor.insert_slice %[[PR]] into %[[ARG3]][0, 0] [%[[D3]], 5] [1, 1] : tensor<?x5xf32> into tensor<?x5xf32>
+//     CHECK:     scf.yield %[[INS]] : tensor<?x5xf32>
+//     CHECK:   }
+//     CHECK:   %[[R:.*]] = linalg.generic {indexing_maps = [#[[MAP0]], #[[MAP1]]], iterator_types = ["parallel", "reduction"]} ins(%[[L]] : tensor<?x5xf32>) outs(%[[ARG1]] : tensor<?xf32>) {
+//     CHECK:     arith.addf
+//     CHECK:     linalg.yield
+//     CHECK:   } -> tensor<?xf32>
+//     CHECK:   return %[[R]] : tensor<?xf32>
+
+// -----
+
+func.func @reduction_tile_transpose(%arg0: tensor<?x?xf32>, %out: tensor<?xf32>) -> tensor<?xf32> {
+  %red = linalg.generic {indexing_maps = [affine_map<(d0, d1) -> (d0, d1)>,
+                                          affine_map<(d0, d1) -> (d1)>],
+   iterator_types = ["reduction", "parallel"]}
+   ins(%arg0 : tensor<?x?xf32>)
+   outs(%out : tensor<?xf32>) {
+    ^bb0(%arg7: f32, %arg9: f32):
+      %42 = arith.addf %arg7, %arg9 : f32
+      linalg.yield %42 : f32
+    } -> tensor<?xf32>
+  return %red : tensor<?xf32>
+}
+
+transform.sequence failures(propagate) {
+^bb0(%arg1: !pdl.operation):
+  %0 = transform.structured.match ops{["linalg.generic"]} in %arg1
+  %1, %2, %3 = transform.structured.tile_reduction_using_scf %0 { tile_sizes = [5, 0] }
+}
+
+//     CHECK: func @reduction_tile_transpose
+//     CHECK:   tensor.empty(%{{.*}}) : tensor<5x?xf32>
+//     CHECK:   linalg.fill {{.*}} : tensor<5x?xf32>) -> tensor<5x?xf32>
+//     CHECK:   scf.for
+//     CHECK:     linalg.generic
+//     CHECK:     %[[D3:.*]] = tensor.dim %[[F]], %[[C1]] : tensor<5x?xf32>
+//     CHECK:     %[[INS:.*]] = tensor.insert_slice %[[PR]] into %[[ARG3]][0, 0] [5, %[[D3]]] [1, 1] : tensor<5x?xf32> into tensor<5x?xf32>
+//     CHECK:     scf.yield {{.*}} : tensor<5x?xf32>
+//     CHECK:   }
+//     CHECK:   linalg.generic 
+//     CHECK:   return