[mlir][linalg][NFC] Cleanup: Drop linalg.inplaceable attribute

[springerm]

Summary

This attribute has already been mostly phased out over the last months. bufferization.writable is used in most cases instead. All remaining test cases are updated. Some code that is no longer needed is deleted.

mlir/include/mlir/Dialect/Bufferization/IR/BufferizableOpInterface.td

let extraClassDeclaration = [{
     bool bufferizesToAliasOnly(OpOperand &opOperand,
                                const AnalysisState &state) {
       auto bufferizableOp =
           cast<BufferizableOpInterface>(getOperation());
       return !bufferizableOp.bufferizesToMemoryRead(opOperand, state)
           && !bufferizableOp.bufferizesToMemoryWrite(opOperand, state)
           && !bufferizableOp.getAliasingOpResult(opOperand, state).empty();
     }
-
-    // TODO: This attribute is deprecated. Use `bufferization.writable` or add
-    // a new attribute in a different dialect.
-    /// Attribute name used to mark region arguments that can be bufferized
-    /// in-place during one-shot bufferization.
-    constexpr const static ::llvm::StringLiteral
-        kInplaceableAttrName = "linalg.inplaceable";
   }];
 }
 
 #endif  // BUFFERIZABLE_OP_INTERFACE

mlir/lib/Dialect/Bufferization/IR/BufferizableOpInterface.cpp

 
 #define DEBUG_TYPE "bufferizable-op-interface"
 #define DBGS() (llvm::dbgs() << '[' << DEBUG_TYPE << "] ")
 #define LDBG(X) LLVM_DEBUG(DBGS() << (X))
 
 using namespace mlir;
 using namespace bufferization;
 
-/// Attribute name used to mark region arguments that can be bufferized
-/// in-place during linalg comprehensive bufferization.
-constexpr const ::llvm::StringLiteral
-    bufferization::BufferizableOpInterface::kInplaceableAttrName;
-
 /// Return the owner of the given value.
 static Operation *getOwnerOfValue(Value value) {
   if (auto opResult = value.dyn_cast<OpResult>())
     return opResult.getDefiningOp();
   return value.cast<BlockArgument>().getOwner()->getParentOp();
 }
 
 bool bufferization::allocationDoesNotEscape(OpResult opResult) {

mlir/lib/Dialect/Bufferization/Transforms/OneShotModuleBufferize.cpp

while (!numberCallOpsContainedInFuncOp.empty()) {
     orderedFuncOps.push_back(it->getFirst());
     for (auto callee : calledBy[it->getFirst()])
       numberCallOpsContainedInFuncOp[callee]--;
     numberCallOpsContainedInFuncOp.erase(it);
   }
   return success();
 }
 
-/// Set the attribute that triggers inplace bufferization on a FuncOp argument
-/// `bbArg`.
-static void setInPlaceFuncArgument(BlockArgument bbArg, bool inPlace) {
-  auto funcOp = cast<func::FuncOp>(bbArg.getOwner()->getParentOp());
-  funcOp.setArgAttr(bbArg.getArgNumber(),
-                    BufferizableOpInterface::kInplaceableAttrName,
-                    BoolAttr::get(bbArg.getContext(), inPlace));
-}
-
-/// Annotate the IR with the result of the analysis. For testing/debugging only.
-static void annotateOpsWithBufferizationMarkers(func::FuncOp funcOp,
-                                                const AnalysisState &state) {
-  auto bufferizableOp = cast<BufferizableOpInterface>(funcOp.getOperation());
-  for (BlockArgument bbArg : funcOp.getArguments())
-    if (bbArg.getType().isa<TensorType>())
-      setInPlaceFuncArgument(bbArg, bufferizableOp.isWritable(bbArg, state));
-}
-
 /// Fold return values that are memref casts and update function return types.
 ///
 /// During FuncOp bufferization, the exact type of the returned memrefs (if any)
 /// is not known yet. Therefore, the bufferization uses memref types with the
 /// most generic layout map as function return types. After bufferizing the
 /// entire function body, a more concise memref type can potentially be used for
 /// the return type of the function.
 static void foldMemRefCasts(func::FuncOp funcOp) {
for (func::FuncOp funcOp : orderedFuncOps) {
 
     // Run some extra function analyses.
     if (failed(aliasingFuncOpBBArgsAnalysis(funcOp, state)) ||
         failed(funcOpBbArgReadWriteAnalysis(funcOp, state)))
       return failure();
 
     // Mark op as fully analyzed.
     funcState.analyzedFuncOps[funcOp] = FuncOpAnalysisState::Analyzed;
-
-    // Add annotations to function arguments.
-    if (options.testAnalysisOnly)
-      annotateOpsWithBufferizationMarkers(funcOp, state);
   }
 
   return success();
 }
 
 LogicalResult mlir::bufferization::bufferizeModuleOp(
     ModuleOp moduleOp, const OneShotAnalysisState &analysisState) {
   auto const &options = static_cast<const OneShotBufferizationOptions &>(

mlir/lib/Dialect/Linalg/IR/LinalgDialect.cpp

 #include "mlir/Dialect/Linalg/IR/LinalgStructuredOps.cpp.inc"
       >(namedStructuredOpRegionBuilders);
 
   addInterfaces<LinalgInlinerInterface>();
 }
 
 LogicalResult LinalgDialect::verifyOperationAttribute(Operation *op,
                                                       NamedAttribute attr) {
-  using bufferization::BufferizableOpInterface;
-
-  if (attr.getName() == BufferizableOpInterface::kInplaceableAttrName) {
-    if (!attr.getValue().isa<BoolAttr>()) {
-      return op->emitError()
-             << "'" << BufferizableOpInterface::kInplaceableAttrName
-             << "' is expected to be a boolean attribute";
-    }
-    if (!isa<FunctionOpInterface>(op))
-      return op->emitError() << "expected " << attr.getName()
-                             << " to be used on function-like operations";
-    return success();
-  }
   if (attr.getName() == LinalgDialect::kMemoizedIndexingMapsAttrName)
     return success();
   return op->emitError() << "attribute '" << attr.getName()
                          << "' not supported by the linalg dialect";
 }
 
 #include "mlir/Dialect/Linalg/IR/LinalgOpsEnums.cpp.inc"
 
 #define GET_ATTRDEF_CLASSES
 #include "mlir/Dialect/Linalg/IR/LinalgOpsAttrDefs.cpp.inc"
 
 #include "mlir/Dialect/Linalg/IR/LinalgOpsDialect.cpp.inc"

mlir/test/Dialect/Bufferization/Transforms/one-shot-module-bufferize-analysis.mlir

func.func @write_after_select_no_conflict(
 
   return %f, %w : f32, tensor<?xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @write_to_same_tensor_in_loop_out_of_place(
 func.func @write_to_same_tensor_in_loop_out_of_place(
-    %A : tensor<?xf32> {linalg.inplaceable = true},
-    %B : tensor<?xf32> {linalg.inplaceable = true},
+    %A : tensor<?xf32> {bufferization.writable = true},
+    %B : tensor<?xf32> {bufferization.writable = true},
     %lb : index, %ub : index, %step : index, %sz: index)
   -> (tensor<?xf32>)
 {
   // CHECK: scf.for {{.*}} {
   %r0 = scf.for %i = %lb to %ub step %step iter_args(%t = %A) -> (tensor<?xf32>) {
     %i2 = arith.index_cast %i : index to i32
     %i3 = arith.sitofp %i2 : i32 to f32
     // The tensor.insert is out-of-place because the %B is written multiple
func.func @write_to_same_tensor_in_loop_out_of_place(
 
   return %r0 : tensor<?xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @write_to_same_alloc_tensor_in_place(
 func.func @write_to_same_alloc_tensor_in_place(
-    %A : tensor<?xf32> {linalg.inplaceable = true},
+    %A : tensor<?xf32> {bufferization.writable = true},
     %lb : index, %ub : index, %step : index, %sz: index, %sz2: index)
   -> (tensor<?xf32>)
 {
   %B = bufferization.alloc_tensor(%sz2) : tensor<?xf32>
 
   // CHECK: scf.for {{.*}} {
   %r0 = scf.for %i = %lb to %ub step %step iter_args(%t = %A) -> (tensor<?xf32>) {
     %i2 = arith.index_cast %i : index to i32
func.func @write_to_same_alloc_tensor_in_place(
 
   return %r0 : tensor<?xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @write_to_same_alloc_tensor_out_of_place(
 func.func @write_to_same_alloc_tensor_out_of_place(
-    %A : tensor<?xf32> {linalg.inplaceable = true},
+    %A : tensor<?xf32> {bufferization.writable = true},
     %lb : index, %ub : index, %step : index, %sz: index, %sz2: index, %f: f32)
   -> (tensor<?xf32>)
 {
   %B = bufferization.alloc_tensor(%sz2) : tensor<?xf32>
   %C = tensor.insert %f into %B[%lb] : tensor<?xf32>
 
   // CHECK: scf.for {{.*}} {
   %r0 = scf.for %i = %lb to %ub step %step iter_args(%t = %A) -> (tensor<?xf32>) {

mlir/test/Dialect/Linalg/one-shot-bufferize-analysis-2fill-extract-matmul-all-perms.mlir

 // RUN: mlir-opt %s -one-shot-bufferize="test-analysis-only bufferize-function-boundaries" -split-input-file | FileCheck %s
 
 /// All combinations of matmul(fill(extract(alloc_tensor)), fill(extract(%alloc_tensor)), %arg2)
 /// These should all be inplaceable except the first op.
 
 // -----
 
 // CHECK-LABEL: func @fill_extract_matmul_
 func.func @fill_extract_matmul_1234(
-    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = true})
+    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = true})
   -> tensor<256x256xf32>
 {
   %c0 = arith.constant 0 : index
   %cst = arith.constant 0.000000e+00 : f32
   %cst_0 = arith.constant 1.000000e+00 : f32
   %0 = bufferization.alloc_tensor() : tensor<256x256xf32>
 
   // CHECK: {__inplace_operands_attr__ = ["none", "false"]}
   %1 = linalg.fill ins(%cst : f32) outs(%0 : tensor<256x256xf32>) -> tensor<256x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %2 = linalg.fill ins(%cst_0 : f32) outs(%0 : tensor<256x256xf32>) -> tensor<256x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %3 = tensor.extract_slice %1[0, 0] [256, 16] [1, 1] : tensor<256x256xf32> to tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %4 = tensor.extract_slice %2[0, 0] [16, 256] [1, 1] : tensor<256x256xf32> to tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true", "true", "true"]}
   %5 = linalg.matmul ins(%3, %4 : tensor<256x16xf32>, tensor<16x256xf32>) outs(%arg2 : tensor<256x256xf32>) -> tensor<256x256xf32>
   return %5 : tensor<256x256xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @fill_extract_matmul_
 func.func @fill_extract_matmul_1243(
-    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = true})
+    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = true})
   -> tensor<256x256xf32>
 {
   %c0 = arith.constant 0 : index
   %cst = arith.constant 0.000000e+00 : f32
   %cst_0 = arith.constant 1.000000e+00 : f32
   %0 = bufferization.alloc_tensor() : tensor<256x256xf32>
 
   // CHECK: {__inplace_operands_attr__ = ["none", "false"]}
   %1 = linalg.fill ins(%cst : f32) outs(%0 : tensor<256x256xf32>) -> tensor<256x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %2 = linalg.fill ins(%cst_0 : f32) outs(%0 : tensor<256x256xf32>) -> tensor<256x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %4 = tensor.extract_slice %2[0, 0] [16, 256] [1, 1] : tensor<256x256xf32> to tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %3 = tensor.extract_slice %1[0, 0] [256, 16] [1, 1] : tensor<256x256xf32> to tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["true", "true", "true"]}
   %5 = linalg.matmul ins(%3, %4 : tensor<256x16xf32>, tensor<16x256xf32>) outs(%arg2 : tensor<256x256xf32>) -> tensor<256x256xf32>
   return %5 : tensor<256x256xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @fill_extract_matmul_
 func.func @fill_extract_matmul_1324(
-    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = true})
+    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = true})
   -> tensor<256x256xf32>
 {
   %c0 = arith.constant 0 : index
   %cst = arith.constant 0.000000e+00 : f32
   %cst_0 = arith.constant 1.000000e+00 : f32
   %0 = bufferization.alloc_tensor() : tensor<256x256xf32>
 
   // CHECK: {__inplace_operands_attr__ = ["none", "false"]}
   %1 = linalg.fill ins(%cst : f32) outs(%0 : tensor<256x256xf32>) -> tensor<256x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %3 = tensor.extract_slice %1[0, 0] [256, 16] [1, 1] : tensor<256x256xf32> to tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %2 = linalg.fill ins(%cst_0 : f32) outs(%0 : tensor<256x256xf32>) -> tensor<256x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %4 = tensor.extract_slice %2[0, 0] [16, 256] [1, 1] : tensor<256x256xf32> to tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true", "true", "true"]}
   %5 = linalg.matmul ins(%3, %4 : tensor<256x16xf32>, tensor<16x256xf32>) outs(%arg2 : tensor<256x256xf32>) -> tensor<256x256xf32>
   return %5 : tensor<256x256xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @fill_extract_matmul_
 func.func @fill_extract_matmul_1342(
-    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = true})
+    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = true})
   -> tensor<256x256xf32>
 {
   %c0 = arith.constant 0 : index
   %cst = arith.constant 0.000000e+00 : f32
   %cst_0 = arith.constant 1.000000e+00 : f32
   %0 = bufferization.alloc_tensor() : tensor<256x256xf32>
 
   // CHECK: {__inplace_operands_attr__ = ["none", "false"]}
   %1 = linalg.fill ins(%cst : f32) outs(%0 : tensor<256x256xf32>) -> tensor<256x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %3 = tensor.extract_slice %1[0, 0] [256, 16] [1, 1] : tensor<256x256xf32> to tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %4 = tensor.extract_slice %0[0, 0] [16, 256] [1, 1] : tensor<256x256xf32> to tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %2 = linalg.fill ins(%cst_0 : f32) outs(%4 : tensor<16x256xf32>) -> tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true", "true", "true"]}
   %5 = linalg.matmul ins(%3, %2 : tensor<256x16xf32>, tensor<16x256xf32>) outs(%arg2 : tensor<256x256xf32>) -> tensor<256x256xf32>
   return %5 : tensor<256x256xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @fill_extract_matmul_
 func.func @fill_extract_matmul_1423(
-    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = true})
+    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = true})
   -> tensor<256x256xf32>
 {
   %c0 = arith.constant 0 : index
   %cst = arith.constant 0.000000e+00 : f32
   %cst_0 = arith.constant 1.000000e+00 : f32
   %0 = bufferization.alloc_tensor() : tensor<256x256xf32>
 
   // CHECK: {__inplace_operands_attr__ = ["none", "false"]}
   %1 = linalg.fill ins(%cst : f32) outs(%0 : tensor<256x256xf32>) -> tensor<256x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %4 = tensor.extract_slice %0[0, 0] [16, 256] [1, 1] : tensor<256x256xf32> to tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %2 = linalg.fill ins(%cst_0 : f32) outs(%4 : tensor<16x256xf32>) -> tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %3 = tensor.extract_slice %1[0, 0] [256, 16] [1, 1] : tensor<256x256xf32> to tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["true", "true", "true"]}
   %5 = linalg.matmul ins(%3, %2 : tensor<256x16xf32>, tensor<16x256xf32>) outs(%arg2 : tensor<256x256xf32>) -> tensor<256x256xf32>
   return %5 : tensor<256x256xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @fill_extract_matmul_
 func.func @fill_extract_matmul_1432(
-    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = true})
+    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = true})
   -> tensor<256x256xf32>
 {
   %c0 = arith.constant 0 : index
   %cst = arith.constant 0.000000e+00 : f32
   %cst_0 = arith.constant 1.000000e+00 : f32
   %0 = bufferization.alloc_tensor() : tensor<256x256xf32>
 
   // CHECK: {__inplace_operands_attr__ = ["none", "false"]}
   %1 = linalg.fill ins(%cst : f32) outs(%0 : tensor<256x256xf32>) -> tensor<256x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %4 = tensor.extract_slice %0[0, 0] [16, 256] [1, 1] : tensor<256x256xf32> to tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %3 = tensor.extract_slice %1[0, 0] [256, 16] [1, 1] : tensor<256x256xf32> to tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %2 = linalg.fill ins(%cst_0 : f32) outs(%4 : tensor<16x256xf32>) -> tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true", "true", "true"]}
   %5 = linalg.matmul ins(%3, %2 : tensor<256x16xf32>, tensor<16x256xf32>) outs(%arg2 : tensor<256x256xf32>) -> tensor<256x256xf32>
   return %5 : tensor<256x256xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @fill_extract_matmul_
 func.func @fill_extract_matmul_2134(
-    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = true})
+    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = true})
   -> tensor<256x256xf32>
 {
   %c0 = arith.constant 0 : index
   %cst = arith.constant 0.000000e+00 : f32
   %cst_0 = arith.constant 1.000000e+00 : f32
   %0 = bufferization.alloc_tensor() : tensor<256x256xf32>
 
   // CHECK: {__inplace_operands_attr__ = ["none", "false"]}
   %2 = linalg.fill ins(%cst_0 : f32) outs(%0 : tensor<256x256xf32>) -> tensor<256x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %1 = linalg.fill ins(%cst : f32) outs(%0 : tensor<256x256xf32>) -> tensor<256x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %3 = tensor.extract_slice %1[0, 0] [256, 16] [1, 1] : tensor<256x256xf32> to tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %4 = tensor.extract_slice %2[0, 0] [16, 256] [1, 1] : tensor<256x256xf32> to tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true", "true", "true"]}
   %5 = linalg.matmul ins(%3, %4 : tensor<256x16xf32>, tensor<16x256xf32>) outs(%arg2 : tensor<256x256xf32>) -> tensor<256x256xf32>
   return %5 : tensor<256x256xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @fill_extract_matmul_
 func.func @fill_extract_matmul_2143(
-    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = true})
+    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = true})
   -> tensor<256x256xf32>
 {
   %c0 = arith.constant 0 : index
   %cst = arith.constant 0.000000e+00 : f32
   %cst_0 = arith.constant 1.000000e+00 : f32
   %0 = bufferization.alloc_tensor() : tensor<256x256xf32>
 
   // CHECK: {__inplace_operands_attr__ = ["none", "false"]}
   %2 = linalg.fill ins(%cst_0 : f32) outs(%0 : tensor<256x256xf32>) -> tensor<256x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %1 = linalg.fill ins(%cst : f32) outs(%0 : tensor<256x256xf32>) -> tensor<256x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %4 = tensor.extract_slice %2[0, 0] [16, 256] [1, 1] : tensor<256x256xf32> to tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %3 = tensor.extract_slice %1[0, 0] [256, 16] [1, 1] : tensor<256x256xf32> to tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["true", "true", "true"]}
   %5 = linalg.matmul ins(%3, %4 : tensor<256x16xf32>, tensor<16x256xf32>) outs(%arg2 : tensor<256x256xf32>) -> tensor<256x256xf32>
   return %5 : tensor<256x256xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @fill_extract_matmul_
 func.func @fill_extract_matmul_2314(
-    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = true})
+    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = true})
   -> tensor<256x256xf32>
 {
   %c0 = arith.constant 0 : index
   %cst = arith.constant 0.000000e+00 : f32
   %cst_0 = arith.constant 1.000000e+00 : f32
   %0 = bufferization.alloc_tensor() : tensor<256x256xf32>
 
   // CHECK: {__inplace_operands_attr__ = ["none", "false"]}
   %2 = linalg.fill ins(%cst_0 : f32) outs(%0 : tensor<256x256xf32>) -> tensor<256x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %3 = tensor.extract_slice %0[0, 0] [256, 16] [1, 1] : tensor<256x256xf32> to tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %1 = linalg.fill ins(%cst : f32) outs(%3 : tensor<256x16xf32>) -> tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %4 = tensor.extract_slice %2[0, 0] [16, 256] [1, 1] : tensor<256x256xf32> to tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true", "true", "true"]}
   %5 = linalg.matmul ins(%1, %4 : tensor<256x16xf32>, tensor<16x256xf32>) outs(%arg2 : tensor<256x256xf32>) -> tensor<256x256xf32>
   return %5 : tensor<256x256xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @fill_extract_matmul_
 func.func @fill_extract_matmul_2341(
-    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = true})
+    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = true})
   -> tensor<256x256xf32>
 {
   %c0 = arith.constant 0 : index
   %cst = arith.constant 0.000000e+00 : f32
   %cst_0 = arith.constant 1.000000e+00 : f32
   %0 = bufferization.alloc_tensor() : tensor<256x256xf32>
 
   // CHECK: {__inplace_operands_attr__ = ["none", "false"]}
   %2 = linalg.fill ins(%cst_0 : f32) outs(%0 : tensor<256x256xf32>) -> tensor<256x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %3 = tensor.extract_slice %0[0, 0] [256, 16] [1, 1] : tensor<256x256xf32> to tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %4 = tensor.extract_slice %2[0, 0] [16, 256] [1, 1] : tensor<256x256xf32> to tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %1 = linalg.fill ins(%cst : f32) outs(%3 : tensor<256x16xf32>) -> tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["true", "true", "true"]}
   %5 = linalg.matmul ins(%1, %4 : tensor<256x16xf32>, tensor<16x256xf32>) outs(%arg2 : tensor<256x256xf32>) -> tensor<256x256xf32>
   return %5 : tensor<256x256xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @fill_extract_matmul_
 func.func @fill_extract_matmul_2413(
-    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = true})
+    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = true})
   -> tensor<256x256xf32>
 {
   %c0 = arith.constant 0 : index
   %cst = arith.constant 0.000000e+00 : f32
   %cst_0 = arith.constant 1.000000e+00 : f32
   %0 = bufferization.alloc_tensor() : tensor<256x256xf32>
 
   // CHECK: {__inplace_operands_attr__ = ["none", "false"]}
   %2 = linalg.fill ins(%cst_0 : f32) outs(%0 : tensor<256x256xf32>) -> tensor<256x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %4 = tensor.extract_slice %2[0, 0] [16, 256] [1, 1] : tensor<256x256xf32> to tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %1 = linalg.fill ins(%cst : f32) outs(%0 : tensor<256x256xf32>) -> tensor<256x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %3 = tensor.extract_slice %1[0, 0] [256, 16] [1, 1] : tensor<256x256xf32> to tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["true", "true", "true"]}
   %5 = linalg.matmul ins(%3, %4 : tensor<256x16xf32>, tensor<16x256xf32>) outs(%arg2 : tensor<256x256xf32>) -> tensor<256x256xf32>
   return %5 : tensor<256x256xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @fill_extract_matmul_
 func.func @fill_extract_matmul_2431(
-    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = true})
+    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = true})
   -> tensor<256x256xf32>
 {
   %c0 = arith.constant 0 : index
   %cst = arith.constant 0.000000e+00 : f32
   %cst_0 = arith.constant 1.000000e+00 : f32
   %0 = bufferization.alloc_tensor() : tensor<256x256xf32>
 
   // CHECK: {__inplace_operands_attr__ = ["none", "false"]}
   %2 = linalg.fill ins(%cst_0 : f32) outs(%0 : tensor<256x256xf32>) -> tensor<256x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %4 = tensor.extract_slice %2[0, 0] [16, 256] [1, 1] : tensor<256x256xf32> to tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %3 = tensor.extract_slice %0[0, 0] [256, 16] [1, 1] : tensor<256x256xf32> to tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %1 = linalg.fill ins(%cst : f32) outs(%3 : tensor<256x16xf32>) -> tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["true", "true", "true"]}
   %5 = linalg.matmul ins(%1, %4 : tensor<256x16xf32>, tensor<16x256xf32>) outs(%arg2 : tensor<256x256xf32>) -> tensor<256x256xf32>
   return %5 : tensor<256x256xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @fill_extract_matmul_
 func.func @fill_extract_matmul_3124(
-    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = true})
+    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = true})
   -> tensor<256x256xf32>
 {
   %c0 = arith.constant 0 : index
   %cst = arith.constant 0.000000e+00 : f32
   %cst_0 = arith.constant 1.000000e+00 : f32
   %0 = bufferization.alloc_tensor() : tensor<256x256xf32>
 
   // CHECK: {__inplace_operands_attr__ = ["false"]}
   %3 = tensor.extract_slice %0[0, 0] [256, 16] [1, 1] : tensor<256x256xf32> to tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %1 = linalg.fill ins(%cst : f32) outs(%3 : tensor<256x16xf32>) -> tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %2 = linalg.fill ins(%cst_0 : f32) outs(%0 : tensor<256x256xf32>) -> tensor<256x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %4 = tensor.extract_slice %2[0, 0] [16, 256] [1, 1] : tensor<256x256xf32> to tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true", "true", "true"]}
   %5 = linalg.matmul ins(%1, %4 : tensor<256x16xf32>, tensor<16x256xf32>) outs(%arg2 : tensor<256x256xf32>) -> tensor<256x256xf32>
   return %5 : tensor<256x256xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @fill_extract_matmul_
 func.func @fill_extract_matmul_3142(
-    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = true})
+    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = true})
   -> tensor<256x256xf32>
 {
   %c0 = arith.constant 0 : index
   %cst = arith.constant 0.000000e+00 : f32
   %cst_0 = arith.constant 1.000000e+00 : f32
   %0 = bufferization.alloc_tensor() : tensor<256x256xf32>
 
   // CHECK: {__inplace_operands_attr__ = ["false"]}
   %3 = tensor.extract_slice %0[0, 0] [256, 16] [1, 1] : tensor<256x256xf32> to tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %1 = linalg.fill ins(%cst : f32) outs(%3 : tensor<256x16xf32>) -> tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %4 = tensor.extract_slice %0[0, 0] [16, 256] [1, 1] : tensor<256x256xf32> to tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %2 = linalg.fill ins(%cst_0 : f32) outs(%4 : tensor<16x256xf32>) -> tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true", "true", "true"]}
   %5 = linalg.matmul ins(%1, %2 : tensor<256x16xf32>, tensor<16x256xf32>) outs(%arg2 : tensor<256x256xf32>) -> tensor<256x256xf32>
   return %5 : tensor<256x256xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @fill_extract_matmul_
 func.func @fill_extract_matmul_3214(
-    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = true})  -> tensor<256x256xf32>
+    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = true})  -> tensor<256x256xf32>
 {
   %c0 = arith.constant 0 : index
   %cst = arith.constant 0.000000e+00 : f32
   %cst_0 = arith.constant 1.000000e+00 : f32
   %0 = bufferization.alloc_tensor() : tensor<256x256xf32>
 
   // CHECK: {__inplace_operands_attr__ = ["false"]}
   %3 = tensor.extract_slice %0[0, 0] [256, 16] [1, 1] : tensor<256x256xf32> to tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %2 = linalg.fill ins(%cst_0 : f32) outs(%0 : tensor<256x256xf32>) -> tensor<256x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %1 = linalg.fill ins(%cst : f32) outs(%3 : tensor<256x16xf32>) -> tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %4 = tensor.extract_slice %2[0, 0] [16, 256] [1, 1] : tensor<256x256xf32> to tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true", "true", "true"]}
   %5 = linalg.matmul ins(%1, %4 : tensor<256x16xf32>, tensor<16x256xf32>) outs(%arg2 : tensor<256x256xf32>) -> tensor<256x256xf32>
   return %5 : tensor<256x256xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @fill_extract_matmul_
 func.func @fill_extract_matmul_3241(
-    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = true})
+    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = true})
   -> tensor<256x256xf32>
 {
   %c0 = arith.constant 0 : index
   %cst = arith.constant 0.000000e+00 : f32
   %cst_0 = arith.constant 1.000000e+00 : f32
   %0 = bufferization.alloc_tensor() : tensor<256x256xf32>
 
   // CHECK: {__inplace_operands_attr__ = ["false"]}
   %3 = tensor.extract_slice %0[0, 0] [256, 16] [1, 1] : tensor<256x256xf32> to tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %2 = linalg.fill ins(%cst_0 : f32) outs(%0 : tensor<256x256xf32>) -> tensor<256x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %4 = tensor.extract_slice %2[0, 0] [16, 256] [1, 1] : tensor<256x256xf32> to tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %1 = linalg.fill ins(%cst : f32) outs(%3 : tensor<256x16xf32>) -> tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["true", "true", "true"]}
   %5 = linalg.matmul ins(%1, %4 : tensor<256x16xf32>, tensor<16x256xf32>) outs(%arg2 : tensor<256x256xf32>) -> tensor<256x256xf32>
   return %5 : tensor<256x256xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @fill_extract_matmul_
 func.func @fill_extract_matmul_3412(
-    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = true})
+    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = true})
   -> tensor<256x256xf32>
 {
   %c0 = arith.constant 0 : index
   %cst = arith.constant 0.000000e+00 : f32
   %cst_0 = arith.constant 1.000000e+00 : f32
   %0 = bufferization.alloc_tensor() : tensor<256x256xf32>
 
   // CHECK: {__inplace_operands_attr__ = ["false"]}
   %3 = tensor.extract_slice %0[0, 0] [256, 16] [1, 1] : tensor<256x256xf32> to tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %4 = tensor.extract_slice %0[0, 0] [16, 256] [1, 1] : tensor<256x256xf32> to tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %1 = linalg.fill ins(%cst : f32) outs(%3 : tensor<256x16xf32>) -> tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %2 = linalg.fill ins(%cst_0 : f32) outs(%4 : tensor<16x256xf32>) -> tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true", "true", "true"]}
   %5 = linalg.matmul ins(%1, %2 : tensor<256x16xf32>, tensor<16x256xf32>) outs(%arg2 : tensor<256x256xf32>) -> tensor<256x256xf32>
   return %5 : tensor<256x256xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @fill_extract_matmul_
 func.func @fill_extract_matmul_3421(
-    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = true})
+    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = true})
   -> tensor<256x256xf32>
 {
   %c0 = arith.constant 0 : index
   %cst = arith.constant 0.000000e+00 : f32
   %cst_0 = arith.constant 1.000000e+00 : f32
   %0 = bufferization.alloc_tensor() : tensor<256x256xf32>
 
   // CHECK: {__inplace_operands_attr__ = ["false"]}
   %3 = tensor.extract_slice %0[0, 0] [256, 16] [1, 1] : tensor<256x256xf32> to tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %4 = tensor.extract_slice %0[0, 0] [16, 256] [1, 1] : tensor<256x256xf32> to tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %2 = linalg.fill ins(%cst_0 : f32) outs(%4 : tensor<16x256xf32>) -> tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %1 = linalg.fill ins(%cst : f32) outs(%3 : tensor<256x16xf32>) -> tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["true", "true", "true"]}
   %5 = linalg.matmul ins(%1, %2 : tensor<256x16xf32>, tensor<16x256xf32>) outs(%arg2 : tensor<256x256xf32>) -> tensor<256x256xf32>
   return %5 : tensor<256x256xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @fill_extract_matmul_
 func.func @fill_extract_matmul_4123(
-    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = true})
+    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = true})
   -> tensor<256x256xf32>
 {
   %c0 = arith.constant 0 : index
   %cst = arith.constant 0.000000e+00 : f32
   %cst_0 = arith.constant 1.000000e+00 : f32
   %0 = bufferization.alloc_tensor() : tensor<256x256xf32>
 
   // CHECK: {__inplace_operands_attr__ = ["false"]}
   %4 = tensor.extract_slice %0[0, 0] [16, 256] [1, 1] : tensor<256x256xf32> to tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %1 = linalg.fill ins(%cst : f32) outs(%0 : tensor<256x256xf32>) -> tensor<256x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %2 = linalg.fill ins(%cst_0 : f32) outs(%4 : tensor<16x256xf32>) -> tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %3 = tensor.extract_slice %1[0, 0] [256, 16] [1, 1] : tensor<256x256xf32> to tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["true", "true", "true"]}
   %5 = linalg.matmul ins(%3, %2 : tensor<256x16xf32>, tensor<16x256xf32>) outs(%arg2 : tensor<256x256xf32>) -> tensor<256x256xf32>
   return %5 : tensor<256x256xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @fill_extract_matmul_
 func.func @fill_extract_matmul_4132(
-    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = true})
+    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = true})
   -> tensor<256x256xf32>
 {
   %c0 = arith.constant 0 : index
   %cst = arith.constant 0.000000e+00 : f32
   %cst_0 = arith.constant 1.000000e+00 : f32
   %0 = bufferization.alloc_tensor() : tensor<256x256xf32>
 
   // CHECK: {__inplace_operands_attr__ = ["false"]}
   %4 = tensor.extract_slice %0[0, 0] [16, 256] [1, 1] : tensor<256x256xf32> to tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %1 = linalg.fill ins(%cst : f32) outs(%0 : tensor<256x256xf32>) -> tensor<256x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %3 = tensor.extract_slice %1[0, 0] [256, 16] [1, 1] : tensor<256x256xf32> to tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %2 = linalg.fill ins(%cst_0 : f32) outs(%4 : tensor<16x256xf32>) -> tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true", "true", "true"]}
   %5 = linalg.matmul ins(%3, %2 : tensor<256x16xf32>, tensor<16x256xf32>) outs(%arg2 : tensor<256x256xf32>) -> tensor<256x256xf32>
   return %5 : tensor<256x256xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @fill_extract_matmul_
 func.func @fill_extract_matmul_4213(
-    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = true})
+    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = true})
   -> tensor<256x256xf32>
 {
   %c0 = arith.constant 0 : index
   %cst = arith.constant 0.000000e+00 : f32
   %cst_0 = arith.constant 1.000000e+00 : f32
   %0 = bufferization.alloc_tensor() : tensor<256x256xf32>
 
   // CHECK: {__inplace_operands_attr__ = ["false"]}
   %4 = tensor.extract_slice %0[0, 0] [16, 256] [1, 1] : tensor<256x256xf32> to tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %2 = linalg.fill ins(%cst_0 : f32) outs(%4 : tensor<16x256xf32>) -> tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %1 = linalg.fill ins(%cst : f32) outs(%0 : tensor<256x256xf32>) -> tensor<256x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %3 = tensor.extract_slice %1[0, 0] [256, 16] [1, 1] : tensor<256x256xf32> to tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["true", "true", "true"]}
   %5 = linalg.matmul ins(%3, %2 : tensor<256x16xf32>, tensor<16x256xf32>) outs(%arg2 : tensor<256x256xf32>) -> tensor<256x256xf32>
   return %5 : tensor<256x256xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @fill_extract_matmul_
 func.func @fill_extract_matmul_4231(
-    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = true})
+    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = true})
   -> tensor<256x256xf32>
 {
   %c0 = arith.constant 0 : index
   %cst = arith.constant 0.000000e+00 : f32
   %cst_0 = arith.constant 1.000000e+00 : f32
   %0 = bufferization.alloc_tensor() : tensor<256x256xf32>
 
   // CHECK: {__inplace_operands_attr__ = ["false"]}
   %4 = tensor.extract_slice %0[0, 0] [16, 256] [1, 1] : tensor<256x256xf32> to tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %2 = linalg.fill ins(%cst_0 : f32) outs(%4 : tensor<16x256xf32>) -> tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %3 = tensor.extract_slice %0[0, 0] [256, 16] [1, 1] : tensor<256x256xf32> to tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %1 = linalg.fill ins(%cst : f32) outs(%3 : tensor<256x16xf32>) -> tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["true", "true", "true"]}
   %5 = linalg.matmul ins(%1, %2 : tensor<256x16xf32>, tensor<16x256xf32>) outs(%arg2 : tensor<256x256xf32>) -> tensor<256x256xf32>
   return %5 : tensor<256x256xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @fill_extract_matmul_
 func.func @fill_extract_matmul_4312(
-    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = true})
+    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = true})
   -> tensor<256x256xf32>
 {
   %c0 = arith.constant 0 : index
   %cst = arith.constant 0.000000e+00 : f32
   %cst_0 = arith.constant 1.000000e+00 : f32
   %0 = bufferization.alloc_tensor() : tensor<256x256xf32>
 
   // CHECK: {__inplace_operands_attr__ = ["false"]}
   %4 = tensor.extract_slice %0[0, 0] [16, 256] [1, 1] : tensor<256x256xf32> to tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true"]}
   %3 = tensor.extract_slice %0[0, 0] [256, 16] [1, 1] : tensor<256x256xf32> to tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %1 = linalg.fill ins(%cst : f32) outs(%3 : tensor<256x16xf32>) -> tensor<256x16xf32>
   // CHECK: {__inplace_operands_attr__ = ["none", "true"]}
   %2 = linalg.fill ins(%cst_0 : f32) outs(%4 : tensor<16x256xf32>) -> tensor<16x256xf32>
   // CHECK: {__inplace_operands_attr__ = ["true", "true", "true"]}
   %5 = linalg.matmul ins(%1, %2 : tensor<256x16xf32>, tensor<16x256xf32>) outs(%arg2 : tensor<256x256xf32>) -> tensor<256x256xf32>
   return %5 : tensor<256x256xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @fill_extract_matmul_
 func.func @fill_extract_matmul_4321(
-    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = false},
-    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, linalg.inplaceable = true})
+    %arg0: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg1: tensor<518x518xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = false},
+    %arg2: tensor<256x256xf32> {bufferization.buffer_layout = affine_map<(d0, d1) -> (d0, d1)>, bufferization.writable = true})
   -> tensor<256x256xf32>
 {
   %c0 = arith.constant 0 : index
   %cst = arith.constant 0.000000e+00 : f32
   %cst_0 = arith.constant 1.000000e+00 : f32
   %0 = bufferization.alloc_tensor() : tensor<256x256xf32>
 
   // CHECK: {__inplace_operands_attr__ = ["false"]}

mlir/test/Dialect/SCF/one-shot-bufferize-analysis.mlir

func.func @scf_if_out_of_place3(%t1: tensor<?xf32> {bufferization.writable = true},
   %v2 = vector.transfer_read %t1_alias[%idx], %cst : tensor<?xf32>, vector<10xf32>
   return %r, %v2 : tensor<?xf32>, vector<10xf32>
 }
 
 // -----
 
 // CHECK-LABEL: func @write_to_same_tensor_in_loop_in_place(
 func.func @write_to_same_tensor_in_loop_in_place(
-    %A : tensor<?xf32> {linalg.inplaceable = true},
+    %A : tensor<?xf32> {bufferization.writable = true},
     %lb : index, %ub : index, %step : index, %sz: index)
   -> (tensor<?xf32>)
 {
   // CHECK: scf.for {{.*}} {
   %r0 = scf.for %i = %lb to %ub step %step iter_args(%t = %A) -> (tensor<?xf32>) {
     %B = bufferization.alloc_tensor(%sz) : tensor<?xf32>
     %i2 = arith.index_cast %i : index to i32
     %i3 = arith.sitofp %i2 : i32 to f32

mlir/test/Dialect/Transform/selective-targeting.mlir

 // RUN:  mlir-opt %s -test-transform-dialect-interpreter --split-input-file | FileCheck %s
 
 // CHECK-LABEL: func.func @matmul_tensors_1(
 func.func @matmul_tensors_1(
   %arg0: tensor<128x128xf32>, %arg1: tensor<128x128xf32>,
-  %arg2: tensor<128x128xf32> {linalg.inplaceable = true})
+  %arg2: tensor<128x128xf32>)
     -> tensor<128x128xf32> {
   // This operation is marked for tiling only.
   // CHECK-COUNT-3: scf.for
   // CHECK-COUNT-3: tensor.extract_slice
   // CHECK: linalg.matmul
   // CHECK-SAME: -> tensor<4x4xf32>
   %0 = linalg.matmul { test.attrA }
                       ins(%arg0, %arg1: tensor<128x128xf32>, tensor<128x128xf32>)
                      outs(%arg2: tensor<128x128xf32>)
     -> tensor<128x128xf32>
   func.return %0 : tensor<128x128xf32>
 }
 
 func.func @matmul_tensors_2(
   %arg0: tensor<128x128xf32>, %arg1: tensor<128x128xf32>,
-  %arg2: tensor<128x128xf32> {linalg.inplaceable = true})
+  %arg2: tensor<128x128xf32>)
     -> tensor<128x128xf32> {
   // This operation is marked f
   // This operation is marked for tiling and vectorization.
   // CHECK-COUNT-3: scf.for
   // CHECK-COUNT-3: vector.transfer_read
   // CHECK:       vector.contract
   // CHECK-NOT:   linalg.matmul
   // CHECK:       vector.transfer_write
   %0 = linalg.matmul { test.attrA, test.attrC }
                       ins(%arg0, %arg1: tensor<128x128xf32>, tensor<128x128xf32>)
                      outs(%arg2: tensor<128x128xf32>)
     -> tensor<128x128xf32>
   func.return %0 : tensor<128x128xf32>
 }
 
 func.func @matmul_tensors_3(
   %arg0: tensor<128x128xf32>, %arg1: tensor<128x128xf32>,
-  %arg2: tensor<128x128xf32> {linalg.inplaceable = true})
+  %arg2: tensor<128x128xf32>)
     -> tensor<128x128xf32> {
   // This operation is marked for vectorization only.
   // CHECK-NOT: scf.for
   // CHECK-COUNT-3: vector.transfer_read
   // CHECK: vector.contract
   // CHECK-SAME: into vector<128x128xf32>
   // CHECK: vector.transfer_write
   %0 = linalg.matmul { test.attrC }
^bb0(%arg0: !pdl.operation):
   }
 }
 
 // -----
 
 // CHECK-LABEL: @vectorize_one
 func.func @vectorize_one(
   %arg0: tensor<128x128xf32>, %arg1: tensor<128x128xf32>,
-  %arg2: tensor<128x128xf32> {linalg.inplaceable = true})
+  %arg2: tensor<128x128xf32>)
     -> tensor<128x128xf32> {
   // CHECK: vector.contract
   %0 = linalg.matmul {test.attrA}
                      ins(%arg0, %arg1: tensor<128x128xf32>, tensor<128x128xf32>)
                      outs(%arg2: tensor<128x128xf32>)
     -> tensor<128x128xf32>
   func.return %0 : tensor<128x128xf32>
 }
 
 func.func @vectorize_none(
   %arg0: tensor<128x128xf32>, %arg1: tensor<128x128xf32>,
-  %arg2: tensor<128x128xf32> {linalg.inplaceable = true})
+  %arg2: tensor<128x128xf32>)
     -> tensor<128x128xf32> {
   // CHECK: linalg.matmul
   %0 = linalg.matmul ins(%arg0, %arg1: tensor<128x128xf32>, tensor<128x128xf32>)
                      outs(%arg2: tensor<128x128xf32>)
     -> tensor<128x128xf32>
   func.return %0 : tensor<128x128xf32>
 }
 
^bb0(%arg0: !pdl.operation):
   }
 }
 
 // -----
 
 // CHECK-LABEL: @vectorize_all
 func.func @vectorize_all(
   %arg0: tensor<128x128xf32>, %arg1: tensor<128x128xf32>, %arg2: tensor<128x128xf32>,
-  %arg3: tensor<128x128xf32> {linalg.inplaceable = true})
+  %arg3: tensor<128x128xf32>)
     -> tensor<128x128xf32> {
   // CHECK: vector.contract
   %0 = linalg.matmul {test.attrA}
                      ins(%arg0, %arg1: tensor<128x128xf32>, tensor<128x128xf32>)
                      outs(%arg2: tensor<128x128xf32>)
     -> tensor<128x128xf32>
   // CHECK: vector.contract
   %1 = linalg.matmul ins(%arg0, %0: tensor<128x128xf32>, tensor<128x128xf32>)

mlir/test/Integration/Dialect/Linalg/CPU/test-one-shot-bufferize.mlir

 // RUN: mlir-opt %s -pass-pipeline="func.func(canonicalize,cse),one-shot-bufferize{bufferize-function-boundaries}" |\
 // RUN: mlir-opt -pass-pipeline="func.func(buffer-deallocation,convert-vector-to-scf,lower-affine,convert-linalg-to-loops)" |\
 // RUN: mlir-opt -pass-pipeline="func.func(canonicalize,convert-scf-to-cf),convert-vector-to-llvm,convert-memref-to-llvm,convert-func-to-llvm,reconcile-unrealized-casts" | \
 
 // RUN: mlir-cpu-runner -O3 -e main -entry-point-result=void \
 // RUN:   -shared-libs=%mlir_integration_test_dir/libmlir_runner_utils%shlibext,%mlir_integration_test_dir/libmlir_c_runner_utils%shlibext |\
 // RUN: FileCheck %s
 
 #map0 = affine_map<(d0, d1)[s0] -> ((d1 - d0) ceildiv s0)>
 #map1 = affine_map<(d0, d1)[s0] -> ((d0 - d1) ceildiv s0)>
 
-func.func @init_and_dot(%arg0: tensor<64xf32>, %arg1: tensor<64xf32>, %arg2: tensor<f32> {linalg.inplaceable = true}) -> tensor<f32> {
+func.func @init_and_dot(%arg0: tensor<64xf32>, %arg1: tensor<64xf32>, %arg2: tensor<f32>) -> tensor<f32> {
   %c64 = arith.constant 64 : index
   %cst = arith.constant 0.000000e+00 : f32
   %c2 = arith.constant 2 : index
   %c0 = arith.constant 0 : index
   %0 = linalg.fill ins(%cst : f32) outs(%arg2 : tensor<f32>) -> tensor<f32>
   %1 = affine.apply #map0(%c0, %c64)[%c2]
   %2 = bufferization.alloc_tensor(%1) : tensor<?x2xf32>
   %3 = scf.for %arg3 = %c0 to %c64 step %c2 iter_args(%arg4 = %2) -> (tensor<?x2xf32>) {