[Mlir-commits] [mlir] 3393cc4 - [mlir] NFC - Appease GCC 5 again..
Nicolas Vasilache
llvmlistbot at llvm.org
Wed May 20 15:03:27 PDT 2020
Author: Nicolas Vasilache
Date: 2020-05-20T17:58:18-04:00
New Revision: 3393cc4cebf9969db94dc424b7a2b6195589c33b
URL: https://github.com/llvm/llvm-project/commit/3393cc4cebf9969db94dc424b7a2b6195589c33b
DIFF: https://github.com/llvm/llvm-project/commit/3393cc4cebf9969db94dc424b7a2b6195589c33b.diff
LOG: [mlir] NFC - Appease GCC 5 again..
Added:
Modified:
mlir/lib/Conversion/VectorToSCF/VectorToSCF.cpp
Removed:
################################################################################
diff --git a/mlir/lib/Conversion/VectorToSCF/VectorToSCF.cpp b/mlir/lib/Conversion/VectorToSCF/VectorToSCF.cpp
index 4743fe67f764..f703a8c621e6 100644
--- a/mlir/lib/Conversion/VectorToSCF/VectorToSCF.cpp
+++ b/mlir/lib/Conversion/VectorToSCF/VectorToSCF.cpp
@@ -437,15 +437,17 @@ clip(TransferOpTy transfer, MemRefBoundsCapture &bounds, ArrayRef<Value> ivs) {
return clippedScalarAccessExprs;
}
+namespace mlir {
+
template <typename TransferOpTy>
-mlir::VectorTransferRewriter<TransferOpTy>::VectorTransferRewriter(
+VectorTransferRewriter<TransferOpTy>::VectorTransferRewriter(
VectorTransferToSCFOptions options, MLIRContext *context)
: RewritePattern(TransferOpTy::getOperationName(), 1, context),
options(options) {}
/// Used for staging the transfer in a local buffer.
template <typename TransferOpTy>
-MemRefType mlir::VectorTransferRewriter<TransferOpTy>::tmpMemRefType(
+MemRefType VectorTransferRewriter<TransferOpTy>::tmpMemRefType(
TransferOpTy transfer) const {
auto vectorType = transfer.getVectorType();
return MemRefType::get(vectorType.getShape(), vectorType.getElementType(), {},
@@ -496,7 +498,7 @@ MemRefType mlir::VectorTransferRewriter<TransferOpTy>::tmpMemRefType(
/// Performs the rewrite.
template <>
-LogicalResult mlir::VectorTransferRewriter<TransferReadOp>::matchAndRewrite(
+LogicalResult VectorTransferRewriter<TransferReadOp>::matchAndRewrite(
Operation *op, PatternRewriter &rewriter) const {
using namespace mlir::edsc::op;
@@ -569,7 +571,7 @@ LogicalResult mlir::VectorTransferRewriter<TransferReadOp>::matchAndRewrite(
/// TODO(ntv): implement alternatives to clipping.
/// TODO(ntv): support non-data-parallel operations.
template <>
-LogicalResult mlir::VectorTransferRewriter<TransferWriteOp>::matchAndRewrite(
+LogicalResult VectorTransferRewriter<TransferWriteOp>::matchAndRewrite(
Operation *op, PatternRewriter &rewriter) const {
using namespace edsc::op;
@@ -622,10 +624,13 @@ LogicalResult mlir::VectorTransferRewriter<TransferWriteOp>::matchAndRewrite(
return success();
}
-void mlir::populateVectorToSCFConversionPatterns(
+void populateVectorToSCFConversionPatterns(
OwningRewritePatternList &patterns, MLIRContext *context,
const VectorTransferToSCFOptions &options) {
patterns.insert<VectorTransferRewriter<vector::TransferReadOp>,
VectorTransferRewriter<vector::TransferWriteOp>>(options,
context);
}
+
+} // namespace mlir
+
More information about the Mlir-commits
mailing list