diff --git a/mlir/include/mlir/Dialect/Linalg/IR/LinalgInterfaces.h b/mlir/include/mlir/Dialect/Linalg/IR/LinalgInterfaces.h
--- a/mlir/include/mlir/Dialect/Linalg/IR/LinalgInterfaces.h
+++ b/mlir/include/mlir/Dialect/Linalg/IR/LinalgInterfaces.h
@@ -33,7 +33,8 @@
/// can be dropped, i.e. the remaining operands can compute the loop
/// bounds of the op.
bool canOpOperandsBeDroppedImpl(linalg::LinalgOp linalgOp,
- ArrayRef<OpOperand *> droppedOperands);
+ ArrayRef<OpOperand *> droppedOperands,
+ MLIRContext *context);
} // namespace detail
/// Checks whether `linalgOp` conforms to ContractionOpInterface.
diff --git a/mlir/include/mlir/Dialect/Linalg/IR/LinalgInterfaces.td b/mlir/include/mlir/Dialect/Linalg/IR/LinalgInterfaces.td
--- a/mlir/include/mlir/Dialect/Linalg/IR/LinalgInterfaces.td
+++ b/mlir/include/mlir/Dialect/Linalg/IR/LinalgInterfaces.td
@@ -610,11 +610,11 @@
}],
/*retTy=*/"AffineMap",
/*methodName=*/"getLoopsToShapesMap",
- /*args=*/(ins),
+ /*args=*/(ins "MLIRContext *":$context),
/*methodBody=*/"",
/*defaultImplementation=*/[{
auto maps = $_op.getIndexingMapsArray();
- return concatAffineMaps(maps);
+ return concatAffineMaps(maps, context);
}]
>,
InterfaceMethod<
@@ -635,10 +635,10 @@
}],
/*retTy=*/"AffineMap",
/*methodName=*/"getShapesToLoopsMap",
- /*args=*/(ins),
+ /*args=*/(ins "MLIRContext *":$context),
/*methodBody=*/"",
/*defaultImplementation=*/[{
- return inversePermutation(getLoopsToShapesMap());
+ return inversePermutation(getLoopsToShapesMap(context));
}]
>,
InterfaceMethod<
@@ -648,10 +648,11 @@
}],
/*retTy=*/"bool",
/*methodName=*/"canOpOperandsBeDropped",
- /*args=*/(ins "ArrayRef<OpOperand *>":$droppedOperands),
+ /*args=*/(ins "MLIRContext *":$context,
+ "ArrayRef<OpOperand *>":$droppedOperands),
/*methodBody=*/"",
/*defaultImplementation=*/[{
- return detail::canOpOperandsBeDroppedImpl($_op, droppedOperands);
+ return detail::canOpOperandsBeDroppedImpl($_op, droppedOperands, context);
}]
>,
InterfaceMethod<
@@ -680,11 +681,11 @@
}],
/*retTy=*/"SmallVector<int64_t, 4>",
/*methodName=*/"getStaticLoopRanges",
- /*args=*/(ins),
+ /*args=*/(ins "MLIRContext *":$context),
/*methodBody=*/"",
/*defaultImplementation=*/[{
SmallVector<int64_t> viewSizes = getStaticShape();
- AffineMap invertedMap = getShapesToLoopsMap();
+ AffineMap invertedMap = getShapesToLoopsMap(context);
assert(invertedMap && "expected a valid Linalg op to call the method");
return invertedMap.compose(viewSizes);
}]
@@ -736,7 +737,7 @@
/// Compute the static loop sizes necessary to vectorize the computation.
/// This is done by applying `getShapesToLoopsMap` to
/// `createFlatListOfOperandStaticDims`.
- SmallVector<int64_t, 4> computeStaticLoopSizes();
+ SmallVector<int64_t, 4> computeStaticLoopSizes(MLIRContext *context);
/// Returns the value that expresses the shape of the output in terms of
/// shape of the input operands where possible
diff --git a/mlir/include/mlir/IR/AffineMap.h b/mlir/include/mlir/IR/AffineMap.h
--- a/mlir/include/mlir/IR/AffineMap.h
+++ b/mlir/include/mlir/IR/AffineMap.h
@@ -535,7 +535,7 @@
/// ```mlir
/// (i, j, k) -> (i, k, k, j, i, j)
/// ```
-AffineMap concatAffineMaps(ArrayRef<AffineMap> maps);
+AffineMap concatAffineMaps(ArrayRef<AffineMap> maps, MLIRContext *context);
/// Returns the map that results from projecting out the dimensions specified in
/// `projectedDimensions`. The projected dimensions are set to 0.
diff --git a/mlir/lib/Dialect/Linalg/IR/LinalgInterfaces.cpp b/mlir/lib/Dialect/Linalg/IR/LinalgInterfaces.cpp
--- a/mlir/lib/Dialect/Linalg/IR/LinalgInterfaces.cpp
+++ b/mlir/lib/Dialect/Linalg/IR/LinalgInterfaces.cpp
@@ -29,7 +29,8 @@
// Interface utility functions
//===----------------------------------------------------------------------===//
bool linalg::detail::canOpOperandsBeDroppedImpl(
- linalg::LinalgOp linalgOp, ArrayRef<OpOperand *> droppedOperands) {
+ linalg::LinalgOp linalgOp, ArrayRef<OpOperand *> droppedOperands,
+ MLIRContext *context) {
SmallVector<AffineMap> indexingMaps;
for (auto &opOperand : linalgOp->getOpOperands()) {
if (llvm::is_contained(droppedOperands, &opOperand))
@@ -41,7 +42,8 @@
// if the op has no loops.
return linalgOp.getNumLoops() == 0;
}
- return inversePermutation(concatAffineMaps(indexingMaps)) != AffineMap();
+ return inversePermutation(concatAffineMaps(indexingMaps, context)) !=
+ AffineMap();
}
//===----------------------------------------------------------------------===//
@@ -504,7 +506,7 @@
}
SmallVector<Range, 4> LinalgOp::createLoopRanges(OpBuilder &b, Location loc) {
- AffineMap map = getLoopsToShapesMap();
+ AffineMap map = getLoopsToShapesMap(b.getContext());
unsigned numDims = map.getNumDims(), numRes = map.getNumResults();
auto viewSizes = createFlatListOfOperandDims(b, loc);
SmallVector<Range, 4> res(numDims);
@@ -520,8 +522,8 @@
return res;
}
-SmallVector<int64_t, 4> LinalgOp::computeStaticLoopSizes() {
- AffineMap map = getLoopsToShapesMap();
+SmallVector<int64_t, 4> LinalgOp::computeStaticLoopSizes(MLIRContext *context) {
+ AffineMap map = getLoopsToShapesMap(context);
unsigned numDims = map.getNumDims(), numRes = map.getNumResults();
SmallVector<int64_t, 4> allShapeSizes = createFlatListOfOperandStaticDims();
SmallVector<int64_t, 4> res(numDims, 0);
@@ -579,7 +581,7 @@
// shapesToLoopsMap = (d0, d2, d2, d3, d4, d5) -> (d0, d3, d2)
// resultShapesFromInputShapes = subMapOfResultDim.compose(shapesToLoopMap)
// = (d0, d1, d2, d3, d4, d5) -> (d0 + d1, d1)
- AffineMap loopsToShapesMap = getLoopsToShapesMap();
+ AffineMap loopsToShapesMap = getLoopsToShapesMap(b.getContext());
// Find the position in the above map that represents the shape of the
// result:dim being inferred.
@@ -591,7 +593,7 @@
resultShapesSubMapPos.first,
resultShapesSubMapPos.second - resultShapesSubMapPos.first);
AffineMap resultShapesFromInputShapesMap =
- loopToResultsShapeMap.compose(getShapesToLoopsMap());
+ loopToResultsShapeMap.compose(getShapesToLoopsMap(b.getContext()));
// Check that the result dim map does not contain the positions corresponding
// to the outputs.
@@ -664,11 +666,12 @@
SmallVector<unsigned> redDims;
linalgOp.getReductionDims(redDims);
- if (!linalgOp.getShapesToLoopsMap())
+ if (!linalgOp.getShapesToLoopsMap(op->getContext()))
return op->emitOpError("expected the shape-to-loops map to be non-null");
// Check if given shapes match to inferred shapes.
- SmallVector<int64_t, 4> endLoopRangeValues = linalgOp.getStaticLoopRanges();
+ SmallVector<int64_t, 4> endLoopRangeValues =
+ linalgOp.getStaticLoopRanges(op->getContext());
SmallVector<int64_t, 4> startLoopRangeValues(endLoopRangeValues.size(), 0);
// Verify only static cases since we can't get exact dimension sizes and loop
diff --git a/mlir/lib/Dialect/Linalg/TransformOps/LinalgTransformOps.cpp b/mlir/lib/Dialect/Linalg/TransformOps/LinalgTransformOps.cpp
--- a/mlir/lib/Dialect/Linalg/TransformOps/LinalgTransformOps.cpp
+++ b/mlir/lib/Dialect/Linalg/TransformOps/LinalgTransformOps.cpp
@@ -907,7 +907,7 @@
Location loc = target.getLoc();
SmallVector<OpFoldResult> allShapeSizes =
target.createFlatListOfOperandDims(b, loc);
- AffineMap map = target.getShapesToLoopsMap();
+ AffineMap map = target.getShapesToLoopsMap(getContext());
if (!map)
return tileSizes;
IRRewriter rewriter(b);
diff --git a/mlir/lib/Dialect/Linalg/Transforms/BubbleUpExtractSlice.cpp b/mlir/lib/Dialect/Linalg/Transforms/BubbleUpExtractSlice.cpp
--- a/mlir/lib/Dialect/Linalg/Transforms/BubbleUpExtractSlice.cpp
+++ b/mlir/lib/Dialect/Linalg/Transforms/BubbleUpExtractSlice.cpp
@@ -90,7 +90,8 @@
auto linalgLoc = linalgOp.getLoc();
SmallVector<OpFoldResult> allShapeSizes =
linalgOp.createFlatListOfOperandDims(rewriter, linalgLoc);
- AffineMap shapeSizesToLoopsMap = linalgOp.getShapesToLoopsMap();
+ AffineMap shapeSizesToLoopsMap =
+ linalgOp.getShapesToLoopsMap(rewriter.getContext());
if (!shapeSizesToLoopsMap) {
return rewriter.notifyMatchFailure(
linalgOp, "failed to get loops map from shape sizes");
diff --git a/mlir/lib/Dialect/Linalg/Transforms/ConstantFold.cpp b/mlir/lib/Dialect/Linalg/Transforms/ConstantFold.cpp
--- a/mlir/lib/Dialect/Linalg/Transforms/ConstantFold.cpp
+++ b/mlir/lib/Dialect/Linalg/Transforms/ConstantFold.cpp
@@ -128,7 +128,8 @@
}
auto linalgOp = cast<LinalgOp>(genericOp.getOperation());
- SmallVector<int64_t, 4> loopBounds = linalgOp.computeStaticLoopSizes();
+ SmallVector<int64_t, 4> loopBounds =
+ linalgOp.computeStaticLoopSizes(rewriter.getContext());
int64_t numElements = outputType.getNumElements();
// Use APInt/APFloat instead of Attribute here for constructing the output.
diff --git a/mlir/lib/Dialect/Linalg/Transforms/DecomposeLinalgOps.cpp b/mlir/lib/Dialect/Linalg/Transforms/DecomposeLinalgOps.cpp
--- a/mlir/lib/Dialect/Linalg/Transforms/DecomposeLinalgOps.cpp
+++ b/mlir/lib/Dialect/Linalg/Transforms/DecomposeLinalgOps.cpp
@@ -108,7 +108,7 @@
Location loc = op.getLoc();
auto allShapesSizes =
cast<LinalgOp>(op.getOperation()).createFlatListOfOperandDims(b, loc);
- AffineMap map = op.getShapesToLoopsMap();
+ AffineMap map = op.getShapesToLoopsMap(b.getContext());
IRRewriter rewriter(b);
return makeComposedFoldedMultiResultAffineApply(rewriter, loc, map,
allShapesSizes);
diff --git a/mlir/lib/Dialect/Linalg/Transforms/DropUnitDims.cpp b/mlir/lib/Dialect/Linalg/Transforms/DropUnitDims.cpp
--- a/mlir/lib/Dialect/Linalg/Transforms/DropUnitDims.cpp
+++ b/mlir/lib/Dialect/Linalg/Transforms/DropUnitDims.cpp
@@ -141,7 +141,7 @@
// Check that the new index maps are invertible. If not, something went
// wrong, so abort.
- if (!inversePermutation(concatAffineMaps(newIndexingMaps)))
+ if (!inversePermutation(concatAffineMaps(newIndexingMaps, context)))
return nullptr;
return ArrayAttr::get(context,
llvm::to_vector<4>(llvm::map_range(
@@ -184,7 +184,8 @@
// Check if any of the iteration dimensions are unit-trip count. They will
// end up being unit-trip count if they are used to index into a unit-dim
// tensor/memref.
- AffineMap invertedMap = inversePermutation(concatAffineMaps(indexingMaps));
+ AffineMap invertedMap = inversePermutation(
+ concatAffineMaps(indexingMaps, rewriter.getContext()));
if (!invertedMap)
return failure();
SmallVector<int64_t> dims = genericOp.getStaticShape();
@@ -404,8 +405,8 @@
// If the indexing maps of the result operation are not invertible (i.e. not
// legal), abort.
- if (!doCanonicalization ||
- !inversePermutation(concatAffineMaps(newIndexingMaps)))
+ if (!doCanonicalization || !inversePermutation(concatAffineMaps(
+ newIndexingMaps, rewriter.getContext())))
return failure();
// If any operand type change, insert a reshape to convert from the original
diff --git a/mlir/lib/Dialect/Linalg/Transforms/ElementwiseOpFusion.cpp b/mlir/lib/Dialect/Linalg/Transforms/ElementwiseOpFusion.cpp
--- a/mlir/lib/Dialect/Linalg/Transforms/ElementwiseOpFusion.cpp
+++ b/mlir/lib/Dialect/Linalg/Transforms/ElementwiseOpFusion.cpp
@@ -281,7 +281,8 @@
for (auto producerResult : llvm::enumerate(producer->getResults())) {
auto outputOperand = producer.getDpsInitOperand(producerResult.index());
if (producer.payloadUsesValueFromOperand(outputOperand) ||
- !producer.canOpOperandsBeDropped(outputOperand) ||
+ !producer.canOpOperandsBeDropped(rewriter.getContext(),
+ outputOperand) ||
llvm::any_of(producerResult.value().getUsers(), [&](Operation *user) {
return user != consumer.getOperation();
})) {
@@ -358,7 +359,7 @@
consumer.getIteratorTypes(),
/*doc=*/nullptr,
/*library_call=*/nullptr);
- if (!fusedOp.getShapesToLoopsMap()) {
+ if (!fusedOp.getShapesToLoopsMap(rewriter.getContext())) {
// Fused op has invalid indexing maps. Typically this means something is off
// in the input, but going ahead here would result in verification errors.
// So cleanup and abort.
@@ -544,7 +545,8 @@
return failure();
AffineMap fusedIndexMap = linalgOp.getMatchingIndexingMap(fusableOpOperand);
- SmallVector<int64_t, 4> originalLoopRange = linalgOp.getStaticLoopRanges();
+ SmallVector<int64_t, 4> originalLoopRange =
+ linalgOp.getStaticLoopRanges(rewriter.getContext());
originalLoopExtent.assign(originalLoopRange.begin(), originalLoopRange.end());
reassociation.clear();
@@ -1655,7 +1657,8 @@
genericOp.getMatchingIndexingMap(outputOperand));
// Check if the operation shapes to loops map is computable.
- if (!inversePermutation(concatAffineMaps(fusedIndexMaps))) {
+ if (!inversePermutation(
+ concatAffineMaps(fusedIndexMaps, rewriter.getContext()))) {
return rewriter.notifyMatchFailure(
genericOp, "fused op loop bound computation failed");
}
diff --git a/mlir/lib/Dialect/Linalg/Transforms/EraseUnusedOperandsAndResults.cpp b/mlir/lib/Dialect/Linalg/Transforms/EraseUnusedOperandsAndResults.cpp
--- a/mlir/lib/Dialect/Linalg/Transforms/EraseUnusedOperandsAndResults.cpp
+++ b/mlir/lib/Dialect/Linalg/Transforms/EraseUnusedOperandsAndResults.cpp
@@ -70,13 +70,15 @@
// Gather information about duplicate input operands.
llvm::SmallDenseMap<unsigned, unsigned> origInsToNewInsPos =
- deduplicateInputOperands(genericOp, droppedOpOperands, newInputOperands,
+ deduplicateInputOperands(rewriter.getContext(), genericOp,
+ droppedOpOperands, newInputOperands,
newIndexingMaps);
// Gather information about the dropped outputs.
llvm::SmallDenseMap<unsigned, unsigned> origOutsToNewOutsPos =
- deduplicateOutputOperands(genericOp, droppedOpOperands,
- newOutputOperands, newIndexingMaps);
+ deduplicateOutputOperands(rewriter.getContext(), genericOp,
+ droppedOpOperands, newOutputOperands,
+ newIndexingMaps);
// Check if there is any change to operands.
if (newInputOperands.size() + newOutputOperands.size() ==
@@ -125,7 +127,7 @@
// the canonicalized op.
// - The preserved input operands list (by reference).
llvm::SmallDenseMap<unsigned, unsigned>
- deduplicateInputOperands(GenericOp genericOp,
+ deduplicateInputOperands(MLIRContext *context, GenericOp genericOp,
SmallVector<OpOperand *> &droppedOpOperands,
SmallVector<Value> &newInputOperands,
SmallVector<AffineMap> &newIndexingMaps) const {
@@ -139,7 +141,7 @@
// Add the current operands to the list of potentially droppable
// operands. If it cannot be dropped, this needs to be popped back.
droppedOpOperands.push_back(inputOpOperand);
- if (genericOp.canOpOperandsBeDropped(droppedOpOperands))
+ if (genericOp.canOpOperandsBeDropped(context, droppedOpOperands))
continue;
droppedOpOperands.pop_back();
}
@@ -169,7 +171,7 @@
// the canonicalized op.
// - The preserved output operands list (by reference).
llvm::SmallDenseMap<unsigned, unsigned>
- deduplicateOutputOperands(GenericOp genericOp,
+ deduplicateOutputOperands(MLIRContext *context, GenericOp genericOp,
SmallVector<OpOperand *> &droppedOpOperands,
SmallVector<Value> &newOutputOperands,
SmallVector<AffineMap> &newIndexingMaps) const {
@@ -205,7 +207,7 @@
// operand for checking. If it cannot be dropped, need to pop the
// value back.
droppedOpOperands.push_back(outputOpOperand.value());
- if (genericOp.canOpOperandsBeDropped(droppedOpOperands)) {
+ if (genericOp.canOpOperandsBeDropped(context, droppedOpOperands)) {
continue;
}
droppedOpOperands.pop_back();
diff --git a/mlir/lib/Dialect/Linalg/Transforms/SplitReduction.cpp b/mlir/lib/Dialect/Linalg/Transforms/SplitReduction.cpp
--- a/mlir/lib/Dialect/Linalg/Transforms/SplitReduction.cpp
+++ b/mlir/lib/Dialect/Linalg/Transforms/SplitReduction.cpp
@@ -42,7 +42,7 @@
op.getReductionDims(dims);
assert(dims.size() == 1);
unsigned reductionDim = dims[0];
- SmallVector<int64_t, 4> loopRanges = op.getStaticLoopRanges();
+ SmallVector<int64_t, 4> loopRanges = op.getStaticLoopRanges(b.getContext());
int64_t reductionDimSize = loopRanges[reductionDim];
if (reductionDimSize == ShapedType::kDynamicSize ||
reductionDimSize % ratio != 0 ||
@@ -265,7 +265,7 @@
return b.notifyMatchFailure(op, "needs at least 1 reduction dimension");
unsigned reductionDimPos = dims[0];
- SmallVector<int64_t> loopRanges = op.getStaticLoopRanges();
+ SmallVector<int64_t> loopRanges = op.getStaticLoopRanges(b.getContext());
int64_t reductionDimSize = loopRanges[reductionDimPos];
if (reductionDimSize == ShapedType::kDynamicSize ||
reductionDimSize % splitFactor != 0 ||
diff --git a/mlir/lib/Dialect/Linalg/Transforms/Tiling.cpp b/mlir/lib/Dialect/Linalg/Transforms/Tiling.cpp
--- a/mlir/lib/Dialect/Linalg/Transforms/Tiling.cpp
+++ b/mlir/lib/Dialect/Linalg/Transforms/Tiling.cpp
@@ -136,7 +136,7 @@
// sizes are computed.
SmallVector<OpFoldResult> allShapes =
op.createFlatListOfOperandDims(b, b.getLoc());
- AffineMap shapesToLoops = op.getShapesToLoopsMap();
+ AffineMap shapesToLoops = op.getShapesToLoopsMap(b.getContext());
SmallVector<OpFoldResult> loopRanges =
makeComposedFoldedMultiResultAffineApply(b, op.getLoc(), shapesToLoops,
allShapes);
@@ -575,7 +575,7 @@
// 1. Build the tiled loop ranges.
SmallVector<OpFoldResult> allShapeSizes =
op.createFlatListOfOperandDims(b, op.getLoc());
- AffineMap shapeSizesToLoopsMap = op.getShapesToLoopsMap();
+ AffineMap shapeSizesToLoopsMap = op.getShapesToLoopsMap(b.getContext());
if (!shapeSizesToLoopsMap)
return failure();
diff --git a/mlir/lib/Dialect/Linalg/Transforms/TilingInterfaceImpl.cpp b/mlir/lib/Dialect/Linalg/Transforms/TilingInterfaceImpl.cpp
--- a/mlir/lib/Dialect/Linalg/Transforms/TilingInterfaceImpl.cpp
+++ b/mlir/lib/Dialect/Linalg/Transforms/TilingInterfaceImpl.cpp
@@ -99,7 +99,7 @@
LinalgOp linalgOp = cast<LinalgOp>(op);
SmallVector<OpFoldResult> allShapesSizes =
linalgOp.createFlatListOfOperandDims(b, loc);
- AffineMap map = linalgOp.getShapesToLoopsMap();
+ AffineMap map = linalgOp.getShapesToLoopsMap(b.getContext());
return llvm::to_vector(
llvm::map_range(map.getResults(), [&](AffineExpr loopExpr) {
diff --git a/mlir/lib/Dialect/Linalg/Transforms/Vectorization.cpp b/mlir/lib/Dialect/Linalg/Transforms/Vectorization.cpp
--- a/mlir/lib/Dialect/Linalg/Transforms/Vectorization.cpp
+++ b/mlir/lib/Dialect/Linalg/Transforms/Vectorization.cpp
@@ -281,7 +281,7 @@
return VectorizationResult{VectorizationStatus::Failure, nullptr};
auto loc = indexOp.getLoc();
// Compute the static loop sizes of the index op.
- auto targetShape = linalgOp.computeStaticLoopSizes();
+ auto targetShape = linalgOp.computeStaticLoopSizes(b.getContext());
// Compute a one-dimensional index vector for the index op dimension.
SmallVector<int64_t> constantSeq =
llvm::to_vector<16>(llvm::seq<int64_t>(0, targetShape[indexOp.getDim()]));
@@ -347,7 +347,7 @@
auto indexVec = bvm.lookup(extractOp.getIndices()[0]);
// Compute the static loop sizes of the extract op.
- auto targetShape = linalgOp.computeStaticLoopSizes();
+ auto targetShape = linalgOp.computeStaticLoopSizes(b.getContext());
SmallVector<Value> gatherIndices;
gatherIndices.push_back(b.create<arith::ConstantIndexOp>(loc, 0));
@@ -523,7 +523,8 @@
// TODO: the common vector shape is equal to the static loop sizes only when
// all indexing maps are projected permutations. For convs and stencils the
// logic will need to evolve.
- SmallVector<int64_t> commonVectorShape = linalgOp.computeStaticLoopSizes();
+ SmallVector<int64_t> commonVectorShape =
+ linalgOp.computeStaticLoopSizes(b.getContext());
// 3. Turn all BBArgs into vector.transfer_read / load.
Location loc = linalgOp.getLoc();
diff --git a/mlir/lib/IR/AffineMap.cpp b/mlir/lib/IR/AffineMap.cpp
--- a/mlir/lib/IR/AffineMap.cpp
+++ b/mlir/lib/IR/AffineMap.cpp
@@ -707,7 +707,10 @@
return AffineMap::get(map.getNumResults(), /*symbolCount=*/0, exprs, context);
}
-AffineMap mlir::concatAffineMaps(ArrayRef<AffineMap> maps) {
+AffineMap mlir::concatAffineMaps(ArrayRef<AffineMap> maps,
+ MLIRContext *context) {
+ if (maps.empty())
+ return AffineMap::get(context);
unsigned numResults = 0, numDims = 0, numSymbols = 0;
for (auto m : maps)
numResults += m.getNumResults();
@@ -720,8 +723,7 @@
numSymbols += m.getNumSymbols();
numDims = std::max(m.getNumDims(), numDims);
}
- return AffineMap::get(numDims, numSymbols, results,
- maps.front().getContext());
+ return AffineMap::get(numDims, numSymbols, results, context);
}
AffineMap mlir::getProjectedMap(AffineMap map,