Skip to content

Emit nuw and nsw for mul and add when lowering to llvm.getelementptr #140966

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
May 22, 2025
Merged

Emit nuw and nsw for mul and add when lowering to llvm.getelementptr #140966

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
17 changes: 14 additions & 3 deletions mlir/lib/Conversion/LLVMCommon/Pattern.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -73,6 +73,15 @@ Value ConvertToLLVMPattern::getStridedElementPtr(
Value base =
memRefDescriptor.bufferPtr(rewriter, loc, *getTypeConverter(), type);

LLVM::IntegerOverflowFlags intOverflowFlags =
LLVM::IntegerOverflowFlags::none;
if (LLVM::bitEnumContainsAny(noWrapFlags, LLVM::GEPNoWrapFlags::nusw)) {
intOverflowFlags = intOverflowFlags | LLVM::IntegerOverflowFlags::nsw;
}
if (LLVM::bitEnumContainsAny(noWrapFlags, LLVM::GEPNoWrapFlags::nuw)) {
intOverflowFlags = intOverflowFlags | LLVM::IntegerOverflowFlags::nuw;
}

Type indexType = getIndexType();
Value index;
for (int i = 0, e = indices.size(); i < e; ++i) {
Expand All @@ -82,10 +91,12 @@ Value ConvertToLLVMPattern::getStridedElementPtr(
ShapedType::isDynamic(strides[i])
? memRefDescriptor.stride(rewriter, loc, i)
: createIndexAttrConstant(rewriter, loc, indexType, strides[i]);
increment = rewriter.create<LLVM::MulOp>(loc, increment, stride);
increment = rewriter.create<LLVM::MulOp>(loc, increment, stride,
intOverflowFlags);
}
index =
index ? rewriter.create<LLVM::AddOp>(loc, index, increment) : increment;
index = index ? rewriter.create<LLVM::AddOp>(loc, index, increment,
intOverflowFlags)
: increment;
}

Type elementPtrType = memRefDescriptor.getElementPtrType();
Expand Down
16 changes: 8 additions & 8 deletions mlir/test/Conversion/MemRefToLLVM/convert-dynamic-memref-ops.mlir
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -175,8 +175,8 @@ func.func @mixed_load(%mixed : memref<42x?xf32>, %i : index, %j : index) {
// CHECK-DAG: %[[J:.*]] = builtin.unrealized_conversion_cast %[[Jarg]]
// CHECK: %[[ptr:.*]] = llvm.extractvalue %[[ld:.*]][1] : !llvm.struct<(ptr, ptr, i64, array<2 x i64>, array<2 x i64>)>
// CHECK-NEXT: %[[st0:.*]] = llvm.extractvalue %[[ld]][4, 0] : !llvm.struct<(ptr, ptr, i64, array<2 x i64>, array<2 x i64>)>
// CHECK-NEXT: %[[offI:.*]] = llvm.mul %[[I]], %[[st0]] : i64
// CHECK-NEXT: %[[off1:.*]] = llvm.add %[[offI]], %[[J]] : i64
// CHECK-NEXT: %[[offI:.*]] = llvm.mul %[[I]], %[[st0]] overflow<nsw, nuw> : i64
// CHECK-NEXT: %[[off1:.*]] = llvm.add %[[offI]], %[[J]] overflow<nsw, nuw> : i64
// CHECK-NEXT: %[[addr:.*]] = llvm.getelementptr inbounds|nuw %[[ptr]][%[[off1]]] : (!llvm.ptr, i64) -> !llvm.ptr, f32
// CHECK-NEXT: llvm.load %[[addr]] : !llvm.ptr -> f32
%0 = memref.load %mixed[%i, %j] : memref<42x?xf32>
Expand All @@ -192,8 +192,8 @@ func.func @dynamic_load(%dynamic : memref<?x?xf32>, %i : index, %j : index) {
// CHECK-DAG: %[[J:.*]] = builtin.unrealized_conversion_cast %[[Jarg]]
// CHECK: %[[ptr:.*]] = llvm.extractvalue %[[ld:.*]][1] : !llvm.struct<(ptr, ptr, i64, array<2 x i64>, array<2 x i64>)>
// CHECK-NEXT: %[[st0:.*]] = llvm.extractvalue %[[ld]][4, 0] : !llvm.struct<(ptr, ptr, i64, array<2 x i64>, array<2 x i64>)>
// CHECK-NEXT: %[[offI:.*]] = llvm.mul %[[I]], %[[st0]] : i64
// CHECK-NEXT: %[[off1:.*]] = llvm.add %[[offI]], %[[J]] : i64
// CHECK-NEXT: %[[offI:.*]] = llvm.mul %[[I]], %[[st0]] overflow<nsw, nuw> : i64
// CHECK-NEXT: %[[off1:.*]] = llvm.add %[[offI]], %[[J]] overflow<nsw, nuw> : i64
// CHECK-NEXT: %[[addr:.*]] = llvm.getelementptr inbounds|nuw %[[ptr]][%[[off1]]] : (!llvm.ptr, i64) -> !llvm.ptr, f32
// CHECK-NEXT: llvm.load %[[addr]] : !llvm.ptr -> f32
%0 = memref.load %dynamic[%i, %j] : memref<?x?xf32>
Expand Down Expand Up @@ -230,8 +230,8 @@ func.func @dynamic_store(%dynamic : memref<?x?xf32>, %i : index, %j : index, %va
// CHECK-DAG: %[[J:.*]] = builtin.unrealized_conversion_cast %[[Jarg]]
// CHECK: %[[ptr:.*]] = llvm.extractvalue %[[ld:.*]][1] : !llvm.struct<(ptr, ptr, i64, array<2 x i64>, array<2 x i64>)>
// CHECK-NEXT: %[[st0:.*]] = llvm.extractvalue %[[ld]][4, 0] : !llvm.struct<(ptr, ptr, i64, array<2 x i64>, array<2 x i64>)>
// CHECK-NEXT: %[[offI:.*]] = llvm.mul %[[I]], %[[st0]] : i64
// CHECK-NEXT: %[[off1:.*]] = llvm.add %[[offI]], %[[J]] : i64
// CHECK-NEXT: %[[offI:.*]] = llvm.mul %[[I]], %[[st0]] overflow<nsw, nuw> : i64
// CHECK-NEXT: %[[off1:.*]] = llvm.add %[[offI]], %[[J]] overflow<nsw, nuw> : i64
// CHECK-NEXT: %[[addr:.*]] = llvm.getelementptr inbounds|nuw %[[ptr]][%[[off1]]] : (!llvm.ptr, i64) -> !llvm.ptr, f32
// CHECK-NEXT: llvm.store %{{.*}}, %[[addr]] : f32, !llvm.ptr
memref.store %val, %dynamic[%i, %j] : memref<?x?xf32>
Expand All @@ -247,8 +247,8 @@ func.func @mixed_store(%mixed : memref<42x?xf32>, %i : index, %j : index, %val :
// CHECK-DAG: %[[J:.*]] = builtin.unrealized_conversion_cast %[[Jarg]]
// CHECK: %[[ptr:.*]] = llvm.extractvalue %[[ld:.*]][1] : !llvm.struct<(ptr, ptr, i64, array<2 x i64>, array<2 x i64>)>
// CHECK-NEXT: %[[st0:.*]] = llvm.extractvalue %[[ld]][4, 0] : !llvm.struct<(ptr, ptr, i64, array<2 x i64>, array<2 x i64>)>
// CHECK-NEXT: %[[offI:.*]] = llvm.mul %[[I]], %[[st0]] : i64
// CHECK-NEXT: %[[off1:.*]] = llvm.add %[[offI]], %[[J]] : i64
// CHECK-NEXT: %[[offI:.*]] = llvm.mul %[[I]], %[[st0]] overflow<nsw, nuw> : i64
// CHECK-NEXT: %[[off1:.*]] = llvm.add %[[offI]], %[[J]] overflow<nsw, nuw> : i64
// CHECK-NEXT: %[[addr:.*]] = llvm.getelementptr inbounds|nuw %[[ptr]][%[[off1]]] : (!llvm.ptr, i64) -> !llvm.ptr, f32
// CHECK-NEXT: llvm.store %{{.*}}, %[[addr]] : f32, !llvm.ptr
memref.store %val, %mixed[%i, %j] : memref<42x?xf32>
Expand Down
8 changes: 4 additions & 4 deletions mlir/test/Conversion/MemRefToLLVM/convert-static-memref-ops.mlir
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -138,8 +138,8 @@ func.func @static_load(%static : memref<10x42xf32>, %i : index, %j : index) {
// CHECK-DAG: %[[JJ:.*]] = builtin.unrealized_conversion_cast %[[J]]
// CHECK: %[[ptr:.*]] = llvm.extractvalue %{{.*}}[1] : !llvm.struct<(ptr, ptr, i64, array<2 x i64>, array<2 x i64>)>
// CHECK: %[[st0:.*]] = llvm.mlir.constant(42 : index) : i64
// CHECK: %[[offI:.*]] = llvm.mul %[[II]], %[[st0]] : i64
// CHECK: %[[off1:.*]] = llvm.add %[[offI]], %[[JJ]] : i64
// CHECK: %[[offI:.*]] = llvm.mul %[[II]], %[[st0]] overflow<nsw, nuw> : i64
// CHECK: %[[off1:.*]] = llvm.add %[[offI]], %[[JJ]] overflow<nsw, nuw> : i64
// CHECK: %[[addr:.*]] = llvm.getelementptr inbounds|nuw %[[ptr]][%[[off1]]] : (!llvm.ptr, i64) -> !llvm.ptr, f32
// CHECK: llvm.load %[[addr]] : !llvm.ptr -> f32
%0 = memref.load %static[%i, %j] : memref<10x42xf32>
Expand All @@ -166,8 +166,8 @@ func.func @static_store(%static : memref<10x42xf32>, %i : index, %j : index, %va
// CHECK-DAG: %[[JJ:.*]] = builtin.unrealized_conversion_cast %[[J]]
// CHECK: %[[ptr:.*]] = llvm.extractvalue %{{.*}}[1] : !llvm.struct<(ptr, ptr, i64, array<2 x i64>, array<2 x i64>)>
// CHECK: %[[st0:.*]] = llvm.mlir.constant(42 : index) : i64
// CHECK: %[[offI:.*]] = llvm.mul %[[II]], %[[st0]] : i64
// CHECK: %[[off1:.*]] = llvm.add %[[offI]], %[[JJ]] : i64
// CHECK: %[[offI:.*]] = llvm.mul %[[II]], %[[st0]] overflow<nsw, nuw> : i64
// CHECK: %[[off1:.*]] = llvm.add %[[offI]], %[[JJ]] overflow<nsw, nuw> : i64
// CHECK: %[[addr:.*]] = llvm.getelementptr inbounds|nuw %[[ptr]][%[[off1]]] : (!llvm.ptr, i64) -> !llvm.ptr, f32
// CHECK: llvm.store %{{.*}}, %[[addr]] : f32, !llvm.ptr

Expand Down
Loading