[flang] Support non-index shape/shift/slice for CG box operations. #124625
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
That is another problem uncovered during hlfir.reshape inlining, where the shape bits could be any integer type. This patch adds explicit convertions to `index` type where needed.
llvmbot
added
flang
Member
|
@llvm/pr-subscribers-flang-fir-hlfir @llvm/pr-subscribers-flang-codegen Author: Slava Zakharin (vzakhari) ChangesThat is another problem uncovered during hlfir.reshape inlining, Full diff: https://github.com/llvm/llvm-project/pull/124625.diff 2 Files Affected:
diff --git a/flang/lib/Optimizer/CodeGen/CodeGen.cpp b/flang/lib/Optimizer/CodeGen/CodeGen.cpp
index 6ff2c20d744537..17f76faa9b0025 100644
--- a/flang/lib/Optimizer/CodeGen/CodeGen.cpp
+++ b/flang/lib/Optimizer/CodeGen/CodeGen.cpp
@@ -1675,22 +1675,26 @@ struct EmboxCommonConversion : public fir::FIROpConversion<OP> {
this->attachTBAATag(storeOp, boxTy, boxTy, nullptr);
return storage;
}
-};
-/// Compute the extent of a triplet slice (lb:ub:step).
-static mlir::Value
-computeTripletExtent(mlir::ConversionPatternRewriter &rewriter,
- mlir::Location loc, mlir::Value lb, mlir::Value ub,
- mlir::Value step, mlir::Value zero, mlir::Type type) {
- mlir::Value extent = rewriter.create<mlir::LLVM::SubOp>(loc, type, ub, lb);
- extent = rewriter.create<mlir::LLVM::AddOp>(loc, type, extent, step);
- extent = rewriter.create<mlir::LLVM::SDivOp>(loc, type, extent, step);
- // If the resulting extent is negative (`ub-lb` and `step` have different
- // signs), zero must be returned instead.
- auto cmp = rewriter.create<mlir::LLVM::ICmpOp>(
- loc, mlir::LLVM::ICmpPredicate::sgt, extent, zero);
- return rewriter.create<mlir::LLVM::SelectOp>(loc, cmp, extent, zero);
-}
+ /// Compute the extent of a triplet slice (lb:ub:step).
+ mlir::Value computeTripletExtent(mlir::ConversionPatternRewriter &rewriter,
+ mlir::Location loc, mlir::Value lb,
+ mlir::Value ub, mlir::Value step,
+ mlir::Value zero, mlir::Type type) const {
+ lb = this->integerCast(loc, rewriter, type, lb);
+ ub = this->integerCast(loc, rewriter, type, ub);
+ step = this->integerCast(loc, rewriter, type, step);
+ zero = this->integerCast(loc, rewriter, type, zero);
+ mlir::Value extent = rewriter.create<mlir::LLVM::SubOp>(loc, type, ub, lb);
+ extent = rewriter.create<mlir::LLVM::AddOp>(loc, type, extent, step);
+ extent = rewriter.create<mlir::LLVM::SDivOp>(loc, type, extent, step);
+ // If the resulting extent is negative (`ub-lb` and `step` have different
+ // signs), zero must be returned instead.
+ auto cmp = rewriter.create<mlir::LLVM::ICmpOp>(
+ loc, mlir::LLVM::ICmpPredicate::sgt, extent, zero);
+ return rewriter.create<mlir::LLVM::SelectOp>(loc, cmp, extent, zero);
+ }
+};
/// Create a generic box on a memory reference. This conversions lowers the
/// abstract box to the appropriate, initialized descriptor.
@@ -1851,14 +1855,16 @@ struct XEmboxOpConversion : public EmboxCommonConversion<fir::cg::XEmboxOp> {
// translating everything to values in the descriptor wherever the entity
// has a dynamic array dimension.
for (unsigned di = 0, descIdx = 0; di < rank; ++di) {
- mlir::Value extent = operands[shapeOffset];
+ mlir::Value extent =
+ integerCast(loc, rewriter, i64Ty, operands[shapeOffset]);
mlir::Value outerExtent = extent;
bool skipNext = false;
if (hasSlice) {
- mlir::Value off = operands[sliceOffset];
+ mlir::Value off =
+ integerCast(loc, rewriter, i64Ty, operands[sliceOffset]);
mlir::Value adj = one;
if (hasShift)
- adj = operands[shiftOffset];
+ adj = integerCast(loc, rewriter, i64Ty, operands[shiftOffset]);
auto ao = rewriter.create<mlir::LLVM::SubOp>(loc, i64Ty, off, adj);
if (constRows > 0) {
cstInteriorIndices.push_back(ao);
@@ -1895,7 +1901,7 @@ struct XEmboxOpConversion : public EmboxCommonConversion<fir::cg::XEmboxOp> {
// the lower bound.
if (hasShift && !(hasSlice || hasSubcomp || hasSubstr) &&
(isaPointerOrAllocatable || !normalizedLowerBound(xbox))) {
- lb = operands[shiftOffset];
+ lb = integerCast(loc, rewriter, i64Ty, operands[shiftOffset]);
auto extentIsEmpty = rewriter.create<mlir::LLVM::ICmpOp>(
loc, mlir::LLVM::ICmpPredicate::eq, extent, zero);
lb = rewriter.create<mlir::LLVM::SelectOp>(loc, extentIsEmpty, one,
@@ -1907,9 +1913,12 @@ struct XEmboxOpConversion : public EmboxCommonConversion<fir::cg::XEmboxOp> {
// store step (scaled by shaped extent)
mlir::Value step = prevDimByteStride;
- if (hasSlice)
- step = rewriter.create<mlir::LLVM::MulOp>(loc, i64Ty, step,
- operands[sliceOffset + 2]);
+ if (hasSlice) {
+ mlir::Value sliceStep =
+ integerCast(loc, rewriter, i64Ty, operands[sliceOffset + 2]);
+ step =
+ rewriter.create<mlir::LLVM::MulOp>(loc, i64Ty, step, sliceStep);
+ }
dest = insertStride(rewriter, loc, dest, descIdx, step);
++descIdx;
}
diff --git a/flang/test/Fir/convert-to-llvm.fir b/flang/test/Fir/convert-to-llvm.fir
index 4c9f965e1241a0..6d7a4a09918e5a 100644
--- a/flang/test/Fir/convert-to-llvm.fir
+++ b/flang/test/Fir/convert-to-llvm.fir
@@ -1909,6 +1909,67 @@ func.func @xembox0(%arg0: !fir.ref<!fir.array<?xi32>>) {
// CHECK: %[[BOX10:.*]] = llvm.insertvalue %[[BASE_PTR]], %[[BOX9]][0] : !llvm.struct<(ptr, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, array<1 x array<3 x i64>>)>
// CHECK: llvm.store %[[BOX10]], %[[ALLOCA]] : !llvm.struct<(ptr, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, array<1 x array<3 x i64>>)>, !llvm.ptr
+// Test i32 shape/shift/slice:
+func.func @xembox0_i32(%arg0: !fir.ref<!fir.array<?xi32>>) {
+ %c0 = arith.constant 0 : i32
+ %c0_i64 = arith.constant 0 : i64
+ %0 = fircg.ext_embox %arg0(%c0) origin %c0[%c0, %c0, %c0] : (!fir.ref<!fir.array<?xi32>>, i32, i32, i32, i32, i32) -> !fir.box<!fir.array<?xi32>>
+ return
+}
+// CHECK-LABEL: llvm.func @xembox0_i32(
+// CHECK-SAME: %[[ARG0:.*]]: !llvm.ptr
+// CHECK: %[[ALLOCA_SIZE:.*]] = llvm.mlir.constant(1 : i32) : i32
+// GENERIC: %[[ALLOCA:.*]] = llvm.alloca %[[ALLOCA_SIZE]] x !llvm.struct<(ptr, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, array<1 x array<3 x i64>>)> {alignment = 8 : i64} : (i32) -> !llvm.ptr
+// AMDGPU: %[[AA:.*]] = llvm.alloca %[[ALLOCA_SIZE]] x !llvm.struct<(ptr, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, array<1 x array<3 x i64>>)> {alignment = 8 : i64} : (i32) -> !llvm.ptr<5>
+// AMDGPU: %[[ALLOCA:.*]] = llvm.addrspacecast %[[AA]] : !llvm.ptr<5> to !llvm.ptr
+// CHECK: %[[C0_I32:.*]] = llvm.mlir.constant(0 : i32) : i32
+// CHECK: %[[C0:.*]] = llvm.mlir.constant(0 : i64) : i64
+// CHECK: %[[TYPE:.*]] = llvm.mlir.constant(9 : i32) : i32
+// CHECK: %[[NULL:.*]] = llvm.mlir.zero : !llvm.ptr
+// CHECK: %[[GEP:.*]] = llvm.getelementptr %[[NULL]][1]
+// CHECK: %[[ELEM_LEN_I64:.*]] = llvm.ptrtoint %[[GEP]] : !llvm.ptr to i64
+// CHECK: %[[BOX0:.*]] = llvm.mlir.undef : !llvm.struct<(ptr, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, array<1 x array<3 x i64>>)>
+// CHECK: %[[BOX1:.*]] = llvm.insertvalue %[[ELEM_LEN_I64]], %[[BOX0]][1] : !llvm.struct<(ptr, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, array<1 x array<3 x i64>>)>
+// CHECK: %[[VERSION:.*]] = llvm.mlir.constant(20240719 : i32) : i32
+// CHECK: %[[BOX2:.*]] = llvm.insertvalue %[[VERSION]], %[[BOX1]][2] : !llvm.struct<(ptr, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, array<1 x array<3 x i64>>)>
+// CHECK: %[[RANK:.*]] = llvm.mlir.constant(1 : i32) : i32
+// CHECK: %[[RANK_I8:.*]] = llvm.trunc %[[RANK]] : i32 to i8
+// CHECK: %[[BOX3:.*]] = llvm.insertvalue %[[RANK_I8]], %[[BOX2]][3] : !llvm.struct<(ptr, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, array<1 x array<3 x i64>>)>
+// CHECK: %[[TYPE_I8:.*]] = llvm.trunc %[[TYPE]] : i32 to i8
+// CHECK: %[[BOX4:.*]] = llvm.insertvalue %[[TYPE_I8]], %[[BOX3]][4] : !llvm.struct<(ptr, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, array<1 x array<3 x i64>>)>
+// CHECK: %[[ATTR:.*]] = llvm.mlir.constant(0 : i32) : i32
+// CHECK: %[[ATTR_I8:.*]] = llvm.trunc %[[ATTR]] : i32 to i8
+// CHECK: %[[BOX5:.*]] = llvm.insertvalue %[[ATTR_I8]], %[[BOX4]][5] : !llvm.struct<(ptr, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, array<1 x array<3 x i64>>)>
+// CHECK: %[[F18ADDENDUM:.*]] = llvm.mlir.constant(0 : i32) : i32
+// CHECK: %[[F18ADDENDUM_I8:.*]] = llvm.trunc %[[F18ADDENDUM]] : i32 to i8
+// CHECK: %[[BOX6:.*]] = llvm.insertvalue %[[F18ADDENDUM_I8]], %[[BOX5]][6] : !llvm.struct<(ptr, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, array<1 x array<3 x i64>>)>
+// CHECK: %[[ZERO:.*]] = llvm.mlir.constant(0 : i64) : i64
+// CHECK: %[[ONE:.*]] = llvm.mlir.constant(1 : i64) : i64
+// CHECK: %[[C0_1:.*]] = llvm.sext %[[C0_I32]] : i32 to i64
+// CHECK: %[[C0_2:.*]] = llvm.sext %[[C0_I32]] : i32 to i64
+// CHECK: %[[C0_3:.*]] = llvm.sext %[[C0_I32]] : i32 to i64
+// CHECK: %[[ADJUSTED_OFFSET:.*]] = llvm.sub %[[C0_2]], %[[C0_3]] : i64
+// CHECK: %[[DIM_OFFSET:.*]] = llvm.mul %[[ADJUSTED_OFFSET]], %[[ONE]] : i64
+// CHECK: %[[PTR_OFFSET:.*]] = llvm.add %[[DIM_OFFSET]], %[[ZERO]] : i64
+// CHECK: %[[C0_4:.*]] = llvm.sext %[[C0_I32]] : i32 to i64
+// CHECK: %[[C0_5:.*]] = llvm.sext %[[C0_I32]] : i32 to i64
+// CHECK: %[[C0_6:.*]] = llvm.sext %[[C0_I32]] : i32 to i64
+// CHECK: %[[EXTENT0:.*]] = llvm.sub %[[C0_5]], %[[C0_4]] : i64
+// CHECK: %[[EXTENT1:.*]] = llvm.add %[[EXTENT0]], %[[C0_6]] : i64
+// CHECK: %[[EXTENT2:.*]] = llvm.sdiv %[[EXTENT1]], %[[C0_6]] : i64
+// CHECK: %[[EXTENT_CMP:.*]] = llvm.icmp "sgt" %[[EXTENT2]], %[[ZERO]] : i64
+// CHECK: %[[EXTENT:.*]] = llvm.select %[[EXTENT_CMP]], %[[EXTENT2]], %[[ZERO]] : i1, i64
+// CHECK: %[[BOX7:.*]] = llvm.insertvalue %[[ONE]], %[[BOX6]][7, 0, 0] : !llvm.struct<(ptr, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, array<1 x array<3 x i64>>)>
+// CHECK: %[[BOX8:.*]] = llvm.insertvalue %[[EXTENT]], %[[BOX7]][7, 0, 1] : !llvm.struct<(ptr, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, array<1 x array<3 x i64>>)>
+// CHECK: %[[C0_7:.*]] = llvm.sext %[[C0_I32]] : i32 to i64
+// CHECK: %[[STRIDE:.*]] = llvm.mul %[[ELEM_LEN_I64]], %[[C0_7]] : i64
+// CHECK: %[[BOX9:.*]] = llvm.insertvalue %[[STRIDE]], %[[BOX8]][7, 0, 2] : !llvm.struct<(ptr, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, array<1 x array<3 x i64>>)>
+// CHECK: %[[PREV_DIM:.*]] = llvm.mul %[[ELEM_LEN_I64]], %[[C0_1]] : i64
+// CHECK: %[[PREV_PTROFF:.*]] = llvm.mul %[[ONE]], %[[C0_1]] : i64
+// CHECK: %[[BASE_PTR:.*]] = llvm.getelementptr %[[ARG0]][%[[PTR_OFFSET]]] : (!llvm.ptr, i64) -> !llvm.ptr, i32
+// CHECK: %[[BOX10:.*]] = llvm.insertvalue %[[BASE_PTR]], %[[BOX9]][0] : !llvm.struct<(ptr, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, array<1 x array<3 x i64>>)>
+// CHECK: llvm.store %[[BOX10]], %[[ALLOCA]] : !llvm.struct<(ptr, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, i{{.*}}, array<1 x array<3 x i64>>)>, !llvm.ptr
+
// Check adjustment of element scaling factor.
func.func @xembox1(%arg0: !fir.ref<!fir.array<?x!fir.char<1, 10>>>) {
|
tblah
approved these changes
Jan 28, 2025
Contributor
tblah
left a comment
tblah
left a comment
There was a problem hiding this comment.
A minor nit, otherwise LGTM
| mlir::Location loc, mlir::Value lb, | ||
| mlir::Value ub, mlir::Value step, | ||
| mlir::Value zero, mlir::Type type) const { | ||
| lb = this->integerCast(loc, rewriter, type, lb); |
Contributor
There was a problem hiding this comment.
why are the this-> pointers required?
Contributor
Author
There was a problem hiding this comment.
integerCast is protected within ConvertFIRToLLVMPattern, so either this-> or FIROpConversion:: should be used. I chose the former :)
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
That is another problem uncovered during hlfir.reshape inlining,
where the shape bits could be any integer type.
This patch adds explicit convertions to
indextype where needed.