[llvm] 4f7f71b - [VPlan] Compare APInt instead of getSExtValue to fix crash in unroll.
Florian Hahn via llvm-commits
llvm-commits at lists.llvm.org
Fri Dec 6 08:29:37 PST 2024
Author: Florian Hahn
Date: 2024-12-06T16:28:49Z
New Revision: 4f7f71b7bccdc38f37b82981e8fa9ceb536a7016
URL: https://github.com/llvm/llvm-project/commit/4f7f71b7bccdc38f37b82981e8fa9ceb536a7016
DIFF: https://github.com/llvm/llvm-project/commit/4f7f71b7bccdc38f37b82981e8fa9ceb536a7016.diff
LOG: [VPlan] Compare APInt instead of getSExtValue to fix crash in unroll.
getSExtValue assumes the result fits in 64 bits, but this may not be the
case for indcutions with wider types. Instead, directly perform the
compare on the APInt for the ConstantInt.
Fixes https://github.com/llvm/llvm-project/issues/118850.
Added:
llvm/test/Transforms/LoopVectorize/interleave-with-i65-induction.ll
Modified:
llvm/lib/Transforms/Vectorize/VPlanUnroll.cpp
Removed:
################################################################################
diff --git a/llvm/lib/Transforms/Vectorize/VPlanUnroll.cpp b/llvm/lib/Transforms/Vectorize/VPlanUnroll.cpp
index f653269713b30b..ff6c9295ee2057 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanUnroll.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanUnroll.cpp
@@ -168,7 +168,7 @@ void UnrollState::unrollWidenInductionByUF(
auto *ConstStep = ScalarStep->isLiveIn()
? dyn_cast<ConstantInt>(ScalarStep->getLiveInIRValue())
: nullptr;
- if (!ConstStep || ConstStep->getZExtValue() != 1) {
+ if (!ConstStep || ConstStep->getValue() != 1) {
if (TypeInfo.inferScalarType(ScalarStep) != IVTy) {
ScalarStep =
Builder.createWidenCast(Instruction::Trunc, ScalarStep, IVTy);
diff --git a/llvm/test/Transforms/LoopVectorize/interleave-with-i65-induction.ll b/llvm/test/Transforms/LoopVectorize/interleave-with-i65-induction.ll
new file mode 100644
index 00000000000000..c1384a1dc474fc
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/interleave-with-i65-induction.ll
@@ -0,0 +1,80 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 5
+; RUN: opt -p loop-vectorize -force-vector-width=4 -force-vector-interleave=2 -S %s | FileCheck %s
+
+; Test case for https://github.com/llvm/llvm-project/issues/118850.
+define void @i65_induction_with_negative_step(ptr %dst) {
+; CHECK-LABEL: define void @i65_induction_with_negative_step(
+; CHECK-SAME: ptr [[DST:%.*]]) {
+; CHECK-NEXT: [[ENTRY:.*]]:
+; CHECK-NEXT: br i1 false, label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
+; CHECK: [[VECTOR_PH]]:
+; CHECK-NEXT: br label %[[VECTOR_BODY:.*]]
+; CHECK: [[VECTOR_BODY]]:
+; CHECK-NEXT: [[INDEX:%.*]] = phi i65 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
+; CHECK-NEXT: [[VECTOR_RECUR:%.*]] = phi <4 x i64> [ <i64 poison, i64 poison, i64 poison, i64 0>, %[[VECTOR_PH]] ], [ [[STEP_ADD:%.*]], %[[VECTOR_BODY]] ]
+; CHECK-NEXT: [[VEC_IND:%.*]] = phi <4 x i64> [ <i64 0, i64 -1, i64 -2, i64 -3>, %[[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], %[[VECTOR_BODY]] ]
+; CHECK-NEXT: [[STEP_ADD]] = add <4 x i64> [[VEC_IND]], splat (i64 -4)
+; CHECK-NEXT: [[OFFSET_IDX:%.*]] = sub i65 0, [[INDEX]]
+; CHECK-NEXT: [[TMP0:%.*]] = trunc i65 [[OFFSET_IDX]] to i64
+; CHECK-NEXT: [[TMP1:%.*]] = add i64 [[TMP0]], 0
+; CHECK-NEXT: [[TMP2:%.*]] = shufflevector <4 x i64> [[VECTOR_RECUR]], <4 x i64> [[VEC_IND]], <4 x i32> <i32 3, i32 4, i32 5, i32 6>
+; CHECK-NEXT: [[TMP3:%.*]] = shufflevector <4 x i64> [[VEC_IND]], <4 x i64> [[STEP_ADD]], <4 x i32> <i32 3, i32 4, i32 5, i32 6>
+; CHECK-NEXT: [[TMP4:%.*]] = getelementptr inbounds i64, ptr [[DST]], i64 [[TMP1]]
+; CHECK-NEXT: [[TMP5:%.*]] = getelementptr inbounds i64, ptr [[TMP4]], i32 0
+; CHECK-NEXT: [[TMP6:%.*]] = getelementptr inbounds i64, ptr [[TMP5]], i32 -3
+; CHECK-NEXT: [[TMP7:%.*]] = getelementptr inbounds i64, ptr [[TMP4]], i32 -4
+; CHECK-NEXT: [[TMP8:%.*]] = getelementptr inbounds i64, ptr [[TMP7]], i32 -3
+; CHECK-NEXT: [[REVERSE:%.*]] = shufflevector <4 x i64> [[TMP2]], <4 x i64> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
+; CHECK-NEXT: store <4 x i64> [[REVERSE]], ptr [[TMP6]], align 8
+; CHECK-NEXT: [[REVERSE1:%.*]] = shufflevector <4 x i64> [[TMP3]], <4 x i64> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
+; CHECK-NEXT: store <4 x i64> [[REVERSE1]], ptr [[TMP8]], align 8
+; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i65 [[INDEX]], 8
+; CHECK-NEXT: [[VEC_IND_NEXT]] = add <4 x i64> [[STEP_ADD]], splat (i64 -4)
+; CHECK-NEXT: [[TMP9:%.*]] = icmp eq i65 [[INDEX_NEXT]], 1000
+; CHECK-NEXT: br i1 [[TMP9]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; CHECK: [[MIDDLE_BLOCK]]:
+; CHECK-NEXT: [[VECTOR_RECUR_EXTRACT:%.*]] = extractelement <4 x i64> [[STEP_ADD]], i32 3
+; CHECK-NEXT: br i1 true, label %[[EXIT:.*]], label %[[SCALAR_PH]]
+; CHECK: [[SCALAR_PH]]:
+; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ 1000, %[[MIDDLE_BLOCK]] ], [ 0, %[[ENTRY]] ]
+; CHECK-NEXT: [[BC_RESUME_VAL2:%.*]] = phi i65 [ -1000, %[[MIDDLE_BLOCK]] ], [ 0, %[[ENTRY]] ]
+; CHECK-NEXT: [[SCALAR_RECUR_INIT:%.*]] = phi i64 [ [[VECTOR_RECUR_EXTRACT]], %[[MIDDLE_BLOCK]] ], [ 0, %[[ENTRY]] ]
+; CHECK-NEXT: br label %[[LOOP:.*]]
+; CHECK: [[LOOP]]:
+; CHECK-NEXT: [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], %[[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], %[[LOOP]] ]
+; CHECK-NEXT: [[IV_I65:%.*]] = phi i65 [ [[BC_RESUME_VAL2]], %[[SCALAR_PH]] ], [ [[IV_I65_NEXT:%.*]], %[[LOOP]] ]
+; CHECK-NEXT: [[FOR:%.*]] = phi i64 [ [[SCALAR_RECUR_INIT]], %[[SCALAR_PH]] ], [ [[TRUNC:%.*]], %[[LOOP]] ]
+; CHECK-NEXT: [[TRUNC]] = trunc i65 [[IV_I65]] to i64
+; CHECK-NEXT: [[GEP:%.*]] = getelementptr inbounds i64, ptr [[DST]], i64 [[TRUNC]]
+; CHECK-NEXT: store i64 [[FOR]], ptr [[GEP]], align 8
+; CHECK-NEXT: [[IV_NEXT]] = add i64 [[IV]], 1
+; CHECK-NEXT: [[ICMP:%.*]] = icmp eq i64 [[IV_NEXT]], 1000
+; CHECK-NEXT: [[IV_I65_NEXT]] = add i65 [[IV_I65]], -1
+; CHECK-NEXT: br i1 [[ICMP]], label %[[EXIT]], label %[[LOOP]], !llvm.loop [[LOOP3:![0-9]+]]
+; CHECK: [[EXIT]]:
+; CHECK-NEXT: ret void
+;
+entry:
+ br label %loop
+
+loop:
+ %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
+ %iv.i65 = phi i65 [ 0, %entry ], [ %iv.i65.next, %loop ]
+ %for = phi i64 [ 0, %entry ], [ %trunc, %loop ]
+ %trunc = trunc i65 %iv.i65 to i64
+ %gep = getelementptr inbounds i64, ptr %dst, i64 %trunc
+ store i64 %for, ptr %gep, align 8
+ %iv.next = add i64 %iv, 1
+ %icmp = icmp eq i64 %iv.next, 1000
+ %iv.i65.next = add i65 %iv.i65, -1
+ br i1 %icmp, label %exit, label %loop
+
+exit:
+ ret void
+}
+;.
+; CHECK: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]}
+; CHECK: [[META1]] = !{!"llvm.loop.isvectorized", i32 1}
+; CHECK: [[META2]] = !{!"llvm.loop.unroll.runtime.disable"}
+; CHECK: [[LOOP3]] = distinct !{[[LOOP3]], [[META2]], [[META1]]}
+;.
More information about the llvm-commits
mailing list