[llvm] [SCEV] Fix NSW flag propagation in getAddExpr (PR #155267)
Sebastian Pop via llvm-commits
llvm-commits at lists.llvm.org
Mon Aug 25 09:25:29 PDT 2025
https://github.com/sebpop created https://github.com/llvm/llvm-project/pull/155267
This patch depends on https://github.com/llvm/llvm-project/pull/155266
SCEV was losing NSW flags during AddRec operations, causing Dependence
Analysis to add unnecessary runtime assumptions for inbounds GEPs.
This patch fixes getAddExpr: when combining AddRecs from the same loop, preserve
compatible NSW/NUW flags from input AddRecs instead of always using FlagAnyWrap.
>From a2aec677e2e3ec3177737a25f4068f23f3e9e998 Mon Sep 17 00:00:00 2001
From: Sebastian Pop <spop at nvidia.com>
Date: Sun, 24 Aug 2025 01:11:32 -0500
Subject: [PATCH 1/2] [LAA] Fix WAW dependency analysis with negative distances
Previously, LAA would incorrectly classify Write-After-Write dependencies
with negative distances as safe Forward dependencies, allowing inappropriate
vectorization of loops with bidirectional WAW dependencies.
The issue occurred in loops like:
for(int i = 0; i < n; ++i) {
A[(i+1)*4] = 10; // First store
A[i] = 100; // Second store
}
The dependence distance from first store to second store is negative:
{-16,+,-12}. However, this represents a bidirectional WAW dependency
that DependenceAnalysis would report as 'output [<>]!', indicating
dependency in both directions.
This patch fixes LAA to properly detect WAW dependencies with negative
distances as unsafe, preventing incorrect vectorization.
The fix adds a check specifically for Write-After-Write dependencies
before the general negative distance handling, ensuring they are
classified as Unknown (unsafe) rather than Forward (safe).
A proper fix, not implemented here because it would be a major rework of LAA,
is to implement bidirectional dependence checking that computes distances
in both directions and detects inconsistent direction vectors.
---
llvm/lib/Analysis/LoopAccessAnalysis.cpp | 15 +++
.../LoopAccessAnalysis/depend_diff_types.ll | 7 +-
.../forward-loop-carried.ll | 2 +-
.../forward-loop-independent.ll | 2 +-
.../waw-negative-dependence.ll | 109 ++++++++++++++++++
.../AArch64/sve-interleaved-accesses.ll | 75 ++++--------
.../LoopVectorize/interleaved-accesses.ll | 84 ++------------
7 files changed, 163 insertions(+), 131 deletions(-)
create mode 100644 llvm/test/Analysis/LoopAccessAnalysis/waw-negative-dependence.ll
diff --git a/llvm/lib/Analysis/LoopAccessAnalysis.cpp b/llvm/lib/Analysis/LoopAccessAnalysis.cpp
index bceddd0325276..e522d1392f7f9 100644
--- a/llvm/lib/Analysis/LoopAccessAnalysis.cpp
+++ b/llvm/lib/Analysis/LoopAccessAnalysis.cpp
@@ -2196,6 +2196,21 @@ MemoryDepChecker::isDependent(const MemAccessInfo &A, unsigned AIdx,
return Dependence::Unknown;
}
+ // For WAW (Write-After-Write) dependencies, negative distances in one
+ // direction can still represent unsafe dependencies. Since we only check
+ // dependencies in program order (AIdx < BIdx), a negative distance means
+ // the later write accesses memory locations before the earlier write.
+ // However, in a vectorized loop, both writes could execute simultaneously,
+ // potentially causing incorrect behavior. Therefore, WAW with negative
+ // distances should be treated as unsafe.
+ bool IsWriteAfterWrite = (AIsWrite && BIsWrite);
+ if (IsWriteAfterWrite) {
+ LLVM_DEBUG(
+ dbgs() << "LAA: WAW dependence with negative distance is unsafe\n");
+ return CheckCompletelyBeforeOrAfter() ? Dependence::NoDep
+ : Dependence::Unknown;
+ }
+
bool IsTrueDataDependence = (AIsWrite && !BIsWrite);
// Check if the first access writes to a location that is read in a later
// iteration, where the distance between them is not a multiple of a vector
diff --git a/llvm/test/Analysis/LoopAccessAnalysis/depend_diff_types.ll b/llvm/test/Analysis/LoopAccessAnalysis/depend_diff_types.ll
index 023a8c056968f..00a2ce7337d0d 100644
--- a/llvm/test/Analysis/LoopAccessAnalysis/depend_diff_types.ll
+++ b/llvm/test/Analysis/LoopAccessAnalysis/depend_diff_types.ll
@@ -1,4 +1,4 @@
-; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py UTC_ARGS: --version 4
+; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py UTC_ARGS: --version 5
; RUN: opt -S -disable-output -passes='print<access-info>' < %s 2>&1 | FileCheck %s
@@ -449,9 +449,10 @@ exit:
define void @different_type_sizes_forward(ptr %dst) {
; CHECK-LABEL: 'different_type_sizes_forward'
; CHECK-NEXT: loop:
-; CHECK-NEXT: Memory dependences are safe
+; CHECK-NEXT: Report: unsafe dependent memory operations in loop. Use #pragma clang loop distribute(enable) to allow loop distribution to attempt to isolate the offending operations into a separate loop
+; CHECK-NEXT: Unknown data dependence.
; CHECK-NEXT: Dependences:
-; CHECK-NEXT: Forward:
+; CHECK-NEXT: Unknown:
; CHECK-NEXT: store i32 0, ptr %gep.10.iv, align 4 ->
; CHECK-NEXT: store i16 1, ptr %gep.iv, align 2
; CHECK-EMPTY:
diff --git a/llvm/test/Analysis/LoopAccessAnalysis/forward-loop-carried.ll b/llvm/test/Analysis/LoopAccessAnalysis/forward-loop-carried.ll
index adfd19923e921..8b00ff0ad2a59 100644
--- a/llvm/test/Analysis/LoopAccessAnalysis/forward-loop-carried.ll
+++ b/llvm/test/Analysis/LoopAccessAnalysis/forward-loop-carried.ll
@@ -70,7 +70,7 @@ define void @forward_different_access_sizes(ptr readnone %end, ptr %start) {
; CHECK-NEXT: store i32 0, ptr %gep.2, align 4 ->
; CHECK-NEXT: %l = load i24, ptr %gep.1, align 1
; CHECK-EMPTY:
-; CHECK-NEXT: Forward:
+; CHECK-NEXT: Unknown:
; CHECK-NEXT: store i32 0, ptr %gep.2, align 4 ->
; CHECK-NEXT: store i24 %l, ptr %ptr.iv, align 1
; CHECK-EMPTY:
diff --git a/llvm/test/Analysis/LoopAccessAnalysis/forward-loop-independent.ll b/llvm/test/Analysis/LoopAccessAnalysis/forward-loop-independent.ll
index 7fc9958dba552..218166526d7c0 100644
--- a/llvm/test/Analysis/LoopAccessAnalysis/forward-loop-independent.ll
+++ b/llvm/test/Analysis/LoopAccessAnalysis/forward-loop-independent.ll
@@ -35,7 +35,7 @@ define void @f(ptr noalias %A, ptr noalias %B, ptr noalias %C, i64 %N) {
; CHECK-NEXT: store i32 %b_p2, ptr %Aidx_next, align 4 ->
; CHECK-NEXT: %a = load i32, ptr %Aidx, align 4
; CHECK-EMPTY:
-; CHECK-NEXT: Forward:
+; CHECK-NEXT: Unknown:
; CHECK-NEXT: store i32 %b_p2, ptr %Aidx_next, align 4 ->
; CHECK-NEXT: store i32 %b_p1, ptr %Aidx, align 4
; CHECK-EMPTY:
diff --git a/llvm/test/Analysis/LoopAccessAnalysis/waw-negative-dependence.ll b/llvm/test/Analysis/LoopAccessAnalysis/waw-negative-dependence.ll
new file mode 100644
index 0000000000000..be49af7d75460
--- /dev/null
+++ b/llvm/test/Analysis/LoopAccessAnalysis/waw-negative-dependence.ll
@@ -0,0 +1,109 @@
+; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py UTC_ARGS: --version 5
+; RUN: opt -passes='print<access-info>' -disable-output %s 2>&1 | FileCheck %s
+
+; Test that LAA correctly identifies Write-After-Write dependencies with negative
+; distances as unsafe. Previously, LAA would incorrectly classify negative distance
+; WAW dependencies as safe Forward dependencies, allowing inappropriate vectorization.
+;
+; This corresponds to the loop:
+; for(int i = 0; i < n; ++i) {
+; A[(i+1)*4] = 10; // First store: A[4, 8, 12, 16, ...]
+; A[i] = 100; // Second store: A[0, 1, 2, 3, 4, ...]
+; }
+;
+; The dependence distance from first store to second store is negative:
+; A[i] - A[(i+1)*4] = {0,+,4} - {16,+,16} = {-16,+,-12}
+; However, the dependence from second store to first store in the next iteration
+; would be positive: A[(i+1)*4] - A[i] = {16,+,16} - {0,+,4} = {16,+,12}
+;
+; This bidirectional dependence pattern (negative in one direction, positive in the
+; other) creates a Write-After-Write dependency that is unsafe for vectorization.
+; DependenceAnalysis would report this as "output [<>]!" indicating the complex
+; dependence direction. LAA must detect this as unsafe even when only checking
+; the negative distance direction.
+
+define void @test_waw_negative_dependence(i64 %n, ptr nocapture %A) {
+; CHECK-LABEL: 'test_waw_negative_dependence'
+; CHECK-NEXT: loop:
+; CHECK-NEXT: Report: unsafe dependent memory operations in loop. Use #pragma clang loop distribute(enable) to allow loop distribution to attempt to isolate the offending operations into a separate loop
+; CHECK-NEXT: Unknown data dependence.
+; CHECK-NEXT: Dependences:
+; CHECK-NEXT: Unknown:
+; CHECK-NEXT: store i32 10, ptr %arrayidx1, align 4 ->
+; CHECK-NEXT: store i32 100, ptr %arrayidx2, align 4
+; CHECK-EMPTY:
+; CHECK-NEXT: Run-time memory checks:
+; CHECK-NEXT: Grouped accesses:
+; CHECK-EMPTY:
+; CHECK-NEXT: Non vectorizable stores to invariant address were not found in loop.
+; CHECK-NEXT: SCEV assumptions:
+; CHECK-EMPTY:
+; CHECK-NEXT: Expressions re-written:
+;
+entry:
+ %cmp8 = icmp sgt i64 %n, 0
+ br i1 %cmp8, label %loop, label %exit
+
+loop:
+ %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %loop ]
+ %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+
+ ; First store: A[(i+1)*4] = 10
+ %0 = shl nsw i64 %indvars.iv.next, 2 ; (i+1)*4
+ %arrayidx1 = getelementptr inbounds i32, ptr %A, i64 %0
+ store i32 10, ptr %arrayidx1, align 4
+
+ ; Second store: A[i] = 100
+ %arrayidx2 = getelementptr inbounds i32, ptr %A, i64 %indvars.iv
+ store i32 100, ptr %arrayidx2, align 4
+
+ %exitcond.not = icmp eq i64 %indvars.iv.next, %n
+ br i1 %exitcond.not, label %exit, label %loop
+
+exit:
+ ret void
+}
+
+; Test a similar case but with different stride to ensure the fix is general.
+define void @test_waw_negative_dependence_different_stride(i64 %n, ptr nocapture %A) {
+; CHECK-LABEL: 'test_waw_negative_dependence_different_stride'
+; CHECK-NEXT: loop:
+; CHECK-NEXT: Report: unsafe dependent memory operations in loop. Use #pragma clang loop distribute(enable) to allow loop distribution to attempt to isolate the offending operations into a separate loop
+; CHECK-NEXT: Unknown data dependence.
+; CHECK-NEXT: Dependences:
+; CHECK-NEXT: Unknown:
+; CHECK-NEXT: store i32 10, ptr %arrayidx1, align 4 ->
+; CHECK-NEXT: store i32 100, ptr %arrayidx2, align 4
+; CHECK-EMPTY:
+; CHECK-NEXT: Run-time memory checks:
+; CHECK-NEXT: Grouped accesses:
+; CHECK-EMPTY:
+; CHECK-NEXT: Non vectorizable stores to invariant address were not found in loop.
+; CHECK-NEXT: SCEV assumptions:
+; CHECK-EMPTY:
+; CHECK-NEXT: Expressions re-written:
+;
+entry:
+ %cmp8 = icmp sgt i64 %n, 0
+ br i1 %cmp8, label %loop, label %exit
+
+loop:
+ %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %loop ]
+ %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+
+ ; First store: A[(i+2)*2] = 10
+ %0 = add nsw i64 %indvars.iv, 2 ; i+2
+ %1 = shl nsw i64 %0, 1 ; (i+2)*2
+ %arrayidx1 = getelementptr inbounds i32, ptr %A, i64 %1
+ store i32 10, ptr %arrayidx1, align 4
+
+ ; Second store: A[i] = 100
+ %arrayidx2 = getelementptr inbounds i32, ptr %A, i64 %indvars.iv
+ store i32 100, ptr %arrayidx2, align 4
+
+ %exitcond.not = icmp eq i64 %indvars.iv.next, %n
+ br i1 %exitcond.not, label %exit, label %loop
+
+exit:
+ ret void
+}
diff --git a/llvm/test/Transforms/LoopVectorize/AArch64/sve-interleaved-accesses.ll b/llvm/test/Transforms/LoopVectorize/AArch64/sve-interleaved-accesses.ll
index fd0bc0b6c20ef..e68744b8a6f37 100644
--- a/llvm/test/Transforms/LoopVectorize/AArch64/sve-interleaved-accesses.ll
+++ b/llvm/test/Transforms/LoopVectorize/AArch64/sve-interleaved-accesses.ll
@@ -1158,53 +1158,22 @@ for.end:
define void @PR27626_5(ptr %a, i32 %x, i32 %y, i32 %z, i64 %n) #1 {
; CHECK-LABEL: @PR27626_5(
; CHECK-NEXT: entry:
-; CHECK-NEXT: [[SMAX:%.*]] = call i64 @llvm.smax.i64(i64 [[N:%.*]], i64 5)
-; CHECK-NEXT: [[TMP0:%.*]] = add nsw i64 [[SMAX]], -4
-; CHECK-NEXT: [[TMP1:%.*]] = lshr i64 [[TMP0]], 1
-; CHECK-NEXT: [[TMP2:%.*]] = add nuw nsw i64 [[TMP1]], 1
-; CHECK-NEXT: [[TMP3:%.*]] = call i64 @llvm.vscale.i64()
-; CHECK-NEXT: [[TMP4:%.*]] = shl nuw nsw i64 [[TMP3]], 2
-; CHECK-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp samesign ult i64 [[TMP2]], [[TMP4]]
-; CHECK-NEXT: br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
-; CHECK: vector.ph:
-; CHECK-NEXT: [[TMP7:%.*]] = call i64 @llvm.vscale.i64()
-; CHECK-NEXT: [[TMP8:%.*]] = shl nuw nsw i64 [[TMP7]], 2
-; CHECK-NEXT: [[DOTNOT:%.*]] = sub nsw i64 0, [[TMP8]]
-; CHECK-NEXT: [[N_VEC:%.*]] = and i64 [[TMP2]], [[DOTNOT]]
-; CHECK-NEXT: [[TMP11:%.*]] = shl nuw i64 [[N_VEC]], 1
-; CHECK-NEXT: [[IND_END:%.*]] = or disjoint i64 [[TMP11]], 3
-; CHECK-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[X:%.*]], i64 0
-; CHECK-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[BROADCAST_SPLATINSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
-; CHECK-NEXT: [[BROADCAST_SPLATINSERT1:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[Y:%.*]], i64 0
-; CHECK-NEXT: [[BROADCAST_SPLAT2:%.*]] = shufflevector <vscale x 4 x i32> [[BROADCAST_SPLATINSERT1]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
-; CHECK-NEXT: [[BROADCAST_SPLATINSERT3:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[Z:%.*]], i64 0
-; CHECK-NEXT: [[BROADCAST_SPLAT4:%.*]] = shufflevector <vscale x 4 x i32> [[BROADCAST_SPLATINSERT3]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
-; CHECK-NEXT: [[TMP10:%.*]] = call <vscale x 4 x i64> @llvm.stepvector.nxv4i64()
-; CHECK-NEXT: [[TMP21:%.*]] = shl <vscale x 4 x i64> [[TMP10]], splat (i64 1)
-; CHECK-NEXT: [[INDUCTION:%.*]] = add <vscale x 4 x i64> [[TMP21]], splat (i64 3)
-; CHECK-NEXT: [[TMP12:%.*]] = shl nuw nsw i64 [[TMP7]], 3
-; CHECK-NEXT: [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 4 x i64> poison, i64 [[TMP12]], i64 0
-; CHECK-NEXT: [[DOTSPLAT:%.*]] = shufflevector <vscale x 4 x i64> [[DOTSPLATINSERT]], <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer
; CHECK-NEXT: br label [[VECTOR_BODY:%.*]]
-; CHECK: vector.body:
-; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; CHECK-NEXT: [[VEC_IND:%.*]] = phi <vscale x 4 x i64> [ [[INDUCTION]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
-; CHECK-NEXT: [[TMP13:%.*]] = add <vscale x 4 x i64> [[VEC_IND]], splat (i64 -1)
-; CHECK-NEXT: [[TMP14:%.*]] = add <vscale x 4 x i64> [[VEC_IND]], splat (i64 -3)
-; CHECK-NEXT: [[TMP15:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], <vscale x 4 x i64> [[VEC_IND]]
-; CHECK-NEXT: [[TMP16:%.*]] = getelementptr inbounds i32, ptr [[A]], <vscale x 4 x i64> [[TMP13]]
-; CHECK-NEXT: [[TMP17:%.*]] = getelementptr inbounds i32, ptr [[A]], <vscale x 4 x i64> [[TMP14]]
-; CHECK-NEXT: call void @llvm.masked.scatter.nxv4i32.nxv4p0(<vscale x 4 x i32> [[BROADCAST_SPLAT]], <vscale x 4 x ptr> [[TMP16]], i32 4, <vscale x 4 x i1> splat (i1 true))
-; CHECK-NEXT: call void @llvm.masked.scatter.nxv4i32.nxv4p0(<vscale x 4 x i32> [[BROADCAST_SPLAT2]], <vscale x 4 x ptr> [[TMP17]], i32 4, <vscale x 4 x i1> splat (i1 true))
-; CHECK-NEXT: call void @llvm.masked.scatter.nxv4i32.nxv4p0(<vscale x 4 x i32> [[BROADCAST_SPLAT4]], <vscale x 4 x ptr> [[TMP15]], i32 4, <vscale x 4 x i1> splat (i1 true))
-; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP8]]
-; CHECK-NEXT: [[VEC_IND_NEXT]] = add <vscale x 4 x i64> [[VEC_IND]], [[DOTSPLAT]]
-; CHECK-NEXT: [[TMP18:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; CHECK-NEXT: br i1 [[TMP18]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP32:![0-9]+]]
-; CHECK: middle.block:
-; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[TMP2]], [[N_VEC]]
-; CHECK-NEXT: br i1 [[CMP_N]], label [[FOR_END:%.*]], label [[SCALAR_PH]]
-; CHECK: scalar.ph:
+; CHECK: for.body:
+; CHECK-NEXT: [[I:%.*]] = phi i64 [ [[I_NEXT:%.*]], [[VECTOR_BODY]] ], [ 3, [[ENTRY:%.*]] ]
+; CHECK-NEXT: [[A_I:%.*]] = getelementptr inbounds nuw i32, ptr [[A:%.*]], i64 [[I]]
+; CHECK-NEXT: [[TMP0:%.*]] = getelementptr i32, ptr [[A]], i64 [[I]]
+; CHECK-NEXT: [[A_I_MINUS_1:%.*]] = getelementptr i8, ptr [[TMP0]], i64 -4
+; CHECK-NEXT: [[TMP1:%.*]] = getelementptr i32, ptr [[A]], i64 [[I]]
+; CHECK-NEXT: [[A_I_MINUS_3:%.*]] = getelementptr i8, ptr [[TMP1]], i64 -12
+; CHECK-NEXT: store i32 [[X:%.*]], ptr [[A_I_MINUS_1]], align 4
+; CHECK-NEXT: store i32 [[Y:%.*]], ptr [[A_I_MINUS_3]], align 4
+; CHECK-NEXT: store i32 [[Z:%.*]], ptr [[A_I]], align 4
+; CHECK-NEXT: [[I_NEXT]] = add nuw nsw i64 [[I]], 2
+; CHECK-NEXT: [[COND:%.*]] = icmp slt i64 [[I_NEXT]], [[N:%.*]]
+; CHECK-NEXT: br i1 [[COND]], label [[VECTOR_BODY]], label [[FOR_END:%.*]]
+; CHECK: for.end:
+; CHECK-NEXT: ret void
;
entry:
br label %for.body
@@ -1281,21 +1250,21 @@ define void @PR34743(ptr %a, ptr %b, i64 %n) #1 {
; CHECK-NEXT: [[TMP18:%.*]] = add nuw nsw <vscale x 4 x i64> [[VEC_IND]], splat (i64 1)
; CHECK-NEXT: [[TMP19:%.*]] = add nuw nsw <vscale x 4 x i64> [[VEC_IND]], splat (i64 2)
; CHECK-NEXT: [[TMP20:%.*]] = getelementptr inbounds i16, ptr [[A]], <vscale x 4 x i64> [[TMP18]]
-; CHECK-NEXT: [[WIDE_MASKED_GATHER:%.*]] = call <vscale x 4 x i16> @llvm.masked.gather.nxv4i16.nxv4p0(<vscale x 4 x ptr> [[TMP20]], i32 4, <vscale x 4 x i1> splat (i1 true), <vscale x 4 x i16> poison), !alias.scope [[META34:![0-9]+]]
+; CHECK-NEXT: [[WIDE_MASKED_GATHER:%.*]] = call <vscale x 4 x i16> @llvm.masked.gather.nxv4i16.nxv4p0(<vscale x 4 x ptr> [[TMP20]], i32 4, <vscale x 4 x i1> splat (i1 true), <vscale x 4 x i16> poison), !alias.scope [[META32:![0-9]+]]
; CHECK-NEXT: [[TMP21:%.*]] = sext <vscale x 4 x i16> [[WIDE_MASKED_GATHER]] to <vscale x 4 x i32>
; CHECK-NEXT: [[TMP22:%.*]] = getelementptr inbounds i16, ptr [[A]], <vscale x 4 x i64> [[TMP19]]
-; CHECK-NEXT: [[WIDE_MASKED_GATHER4]] = call <vscale x 4 x i16> @llvm.masked.gather.nxv4i16.nxv4p0(<vscale x 4 x ptr> [[TMP22]], i32 4, <vscale x 4 x i1> splat (i1 true), <vscale x 4 x i16> poison), !alias.scope [[META34]]
+; CHECK-NEXT: [[WIDE_MASKED_GATHER4]] = call <vscale x 4 x i16> @llvm.masked.gather.nxv4i16.nxv4p0(<vscale x 4 x ptr> [[TMP22]], i32 4, <vscale x 4 x i1> splat (i1 true), <vscale x 4 x i16> poison), !alias.scope [[META32]]
; CHECK-NEXT: [[TMP23:%.*]] = call <vscale x 4 x i16> @llvm.vector.splice.nxv4i16(<vscale x 4 x i16> [[VECTOR_RECUR]], <vscale x 4 x i16> [[WIDE_MASKED_GATHER4]], i32 -1)
; CHECK-NEXT: [[TMP24:%.*]] = sext <vscale x 4 x i16> [[TMP23]] to <vscale x 4 x i32>
; CHECK-NEXT: [[TMP25:%.*]] = sext <vscale x 4 x i16> [[WIDE_MASKED_GATHER4]] to <vscale x 4 x i32>
; CHECK-NEXT: [[TMP26:%.*]] = mul nsw <vscale x 4 x i32> [[TMP24]], [[TMP21]]
; CHECK-NEXT: [[TMP27:%.*]] = mul nsw <vscale x 4 x i32> [[TMP26]], [[TMP25]]
; CHECK-NEXT: [[TMP28:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[INDEX]]
-; CHECK-NEXT: store <vscale x 4 x i32> [[TMP27]], ptr [[TMP28]], align 4, !alias.scope [[META37:![0-9]+]], !noalias [[META34]]
+; CHECK-NEXT: store <vscale x 4 x i32> [[TMP27]], ptr [[TMP28]], align 4, !alias.scope [[META35:![0-9]+]], !noalias [[META32]]
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP10]]
; CHECK-NEXT: [[VEC_IND_NEXT]] = add <vscale x 4 x i64> [[VEC_IND]], [[DOTSPLAT]]
; CHECK-NEXT: [[TMP29:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; CHECK-NEXT: br i1 [[TMP29]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP39:![0-9]+]]
+; CHECK-NEXT: br i1 [[TMP29]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP37:![0-9]+]]
; CHECK: middle.block:
; CHECK-NEXT: [[TMP30:%.*]] = call i32 @llvm.vscale.i32()
; CHECK-NEXT: [[TMP31:%.*]] = shl nuw nsw i32 [[TMP30]], 2
@@ -1388,7 +1357,7 @@ define void @interleave_deinterleave_factor3(ptr writeonly noalias %dst, ptr rea
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP3]]
; CHECK-NEXT: [[VEC_IND_NEXT]] = add <vscale x 4 x i64> [[VEC_IND]], [[DOTSPLAT]]
; CHECK-NEXT: [[TMP18:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; CHECK-NEXT: br i1 [[TMP18]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP41:![0-9]+]]
+; CHECK-NEXT: br i1 [[TMP18]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP39:![0-9]+]]
; CHECK: middle.block:
; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[N_MOD_VF]], 0
; CHECK-NEXT: br i1 [[CMP_N]], label [[FOR_END:%.*]], label [[SCALAR_PH]]
@@ -1475,7 +1444,7 @@ define void @interleave_deinterleave(ptr writeonly noalias %dst, ptr readonly %a
; CHECK-NEXT: store <vscale x 16 x i32> [[INTERLEAVED_VEC13]], ptr [[TMP21]], align 4
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP3]]
; CHECK-NEXT: [[TMP25:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; CHECK-NEXT: br i1 [[TMP25]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP43:![0-9]+]]
+; CHECK-NEXT: br i1 [[TMP25]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP41:![0-9]+]]
; CHECK: middle.block:
; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[N_MOD_VF]], 0
; CHECK-NEXT: br i1 [[CMP_N]], label [[FOR_COND_CLEANUP:%.*]], label [[SCALAR_PH]]
@@ -1584,7 +1553,7 @@ define void @interleave_deinterleave_reverse(ptr noalias nocapture readonly %A,
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP1]]
; CHECK-NEXT: [[VEC_IND_NEXT]] = add <vscale x 4 x i32> [[VEC_IND]], [[DOTSPLAT]]
; CHECK-NEXT: [[TMP27:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1024
-; CHECK-NEXT: br i1 [[TMP27]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP45:![0-9]+]]
+; CHECK-NEXT: br i1 [[TMP27]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP43:![0-9]+]]
; CHECK: middle.block:
; CHECK-NEXT: br i1 true, label [[FOR_COND_CLEANUP:%.*]], label [[SCALAR_PH]]
; CHECK: scalar.ph:
diff --git a/llvm/test/Transforms/LoopVectorize/interleaved-accesses.ll b/llvm/test/Transforms/LoopVectorize/interleaved-accesses.ll
index add58758788f9..aab21965d6877 100644
--- a/llvm/test/Transforms/LoopVectorize/interleaved-accesses.ll
+++ b/llvm/test/Transforms/LoopVectorize/interleaved-accesses.ll
@@ -1348,82 +1348,20 @@ for.end:
define void @PR27626_5(ptr %a, i32 %x, i32 %y, i32 %z, i64 %n) {
; CHECK-LABEL: @PR27626_5(
; CHECK-NEXT: entry:
-; CHECK-NEXT: [[SMAX:%.*]] = call i64 @llvm.smax.i64(i64 [[N:%.*]], i64 5)
-; CHECK-NEXT: [[TMP0:%.*]] = add nsw i64 [[SMAX]], -4
-; CHECK-NEXT: [[TMP1:%.*]] = lshr i64 [[TMP0]], 1
-; CHECK-NEXT: [[TMP2:%.*]] = add nuw nsw i64 [[TMP1]], 1
-; CHECK-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp slt i64 [[N]], 10
-; CHECK-NEXT: br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
-; CHECK: vector.ph:
-; CHECK-NEXT: [[N_VEC:%.*]] = and i64 [[TMP2]], 9223372036854775804
-; CHECK-NEXT: [[TMP3:%.*]] = shl nuw i64 [[N_VEC]], 1
-; CHECK-NEXT: [[IND_END:%.*]] = or disjoint i64 [[TMP3]], 3
-; CHECK-NEXT: br label [[VECTOR_BODY:%.*]]
-; CHECK: vector.body:
-; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; CHECK-NEXT: [[VEC_IND:%.*]] = phi <4 x i64> [ <i64 3, i64 5, i64 7, i64 9>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
-; CHECK-NEXT: [[TMP4:%.*]] = shl i64 [[INDEX]], 1
-; CHECK-NEXT: [[TMP7:%.*]] = add <4 x i64> [[VEC_IND]], splat (i64 -1)
-; CHECK-NEXT: [[TMP8:%.*]] = add <4 x i64> [[VEC_IND]], splat (i64 -3)
-; CHECK-NEXT: [[TMP12:%.*]] = getelementptr i32, ptr [[A:%.*]], i64 [[TMP4]]
-; CHECK-NEXT: [[TMP9:%.*]] = getelementptr i8, ptr [[TMP12]], i64 12
-; CHECK-NEXT: [[TMP33:%.*]] = getelementptr i32, ptr [[A]], i64 [[TMP4]]
-; CHECK-NEXT: [[TMP10:%.*]] = getelementptr i8, ptr [[TMP33]], i64 20
-; CHECK-NEXT: [[TMP34:%.*]] = getelementptr i32, ptr [[A]], i64 [[TMP4]]
-; CHECK-NEXT: [[TMP11:%.*]] = getelementptr i8, ptr [[TMP34]], i64 28
-; CHECK-NEXT: [[TMP35:%.*]] = getelementptr i32, ptr [[A]], i64 [[TMP4]]
-; CHECK-NEXT: [[TMP13:%.*]] = getelementptr i8, ptr [[TMP35]], i64 36
-; CHECK-NEXT: [[TMP14:%.*]] = extractelement <4 x i64> [[TMP7]], i64 0
-; CHECK-NEXT: [[TMP15:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[TMP14]]
-; CHECK-NEXT: [[TMP16:%.*]] = extractelement <4 x i64> [[TMP7]], i64 1
-; CHECK-NEXT: [[TMP17:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[TMP16]]
-; CHECK-NEXT: [[TMP18:%.*]] = extractelement <4 x i64> [[TMP7]], i64 2
-; CHECK-NEXT: [[TMP19:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[TMP18]]
-; CHECK-NEXT: [[TMP20:%.*]] = extractelement <4 x i64> [[TMP7]], i64 3
-; CHECK-NEXT: [[TMP21:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[TMP20]]
-; CHECK-NEXT: [[TMP22:%.*]] = extractelement <4 x i64> [[TMP8]], i64 0
-; CHECK-NEXT: [[TMP23:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[TMP22]]
-; CHECK-NEXT: [[TMP24:%.*]] = extractelement <4 x i64> [[TMP8]], i64 1
-; CHECK-NEXT: [[TMP25:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[TMP24]]
-; CHECK-NEXT: [[TMP26:%.*]] = extractelement <4 x i64> [[TMP8]], i64 2
-; CHECK-NEXT: [[TMP27:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[TMP26]]
-; CHECK-NEXT: [[TMP28:%.*]] = extractelement <4 x i64> [[TMP8]], i64 3
-; CHECK-NEXT: [[TMP29:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[TMP28]]
-; CHECK-NEXT: store i32 [[X:%.*]], ptr [[TMP15]], align 4
-; CHECK-NEXT: store i32 [[X]], ptr [[TMP17]], align 4
-; CHECK-NEXT: store i32 [[X]], ptr [[TMP19]], align 4
-; CHECK-NEXT: store i32 [[X]], ptr [[TMP21]], align 4
-; CHECK-NEXT: store i32 [[Y:%.*]], ptr [[TMP23]], align 4
-; CHECK-NEXT: store i32 [[Y]], ptr [[TMP25]], align 4
-; CHECK-NEXT: store i32 [[Y]], ptr [[TMP27]], align 4
-; CHECK-NEXT: store i32 [[Y]], ptr [[TMP29]], align 4
-; CHECK-NEXT: store i32 [[Z:%.*]], ptr [[TMP9]], align 4
-; CHECK-NEXT: store i32 [[Z]], ptr [[TMP10]], align 4
-; CHECK-NEXT: store i32 [[Z]], ptr [[TMP11]], align 4
-; CHECK-NEXT: store i32 [[Z]], ptr [[TMP13]], align 4
-; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
-; CHECK-NEXT: [[VEC_IND_NEXT]] = add <4 x i64> [[VEC_IND]], splat (i64 8)
-; CHECK-NEXT: [[TMP30:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; CHECK-NEXT: br i1 [[TMP30]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP34:![0-9]+]]
-; CHECK: middle.block:
-; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[TMP2]], [[N_VEC]]
-; CHECK-NEXT: br i1 [[CMP_N]], label [[FOR_END:%.*]], label [[SCALAR_PH]]
-; CHECK: scalar.ph:
-; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ 3, [[ENTRY:%.*]] ]
; CHECK-NEXT: br label [[FOR_BODY:%.*]]
; CHECK: for.body:
-; CHECK-NEXT: [[I:%.*]] = phi i64 [ [[I_NEXT:%.*]], [[FOR_BODY]] ], [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ]
-; CHECK-NEXT: [[A_I:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[I]]
+; CHECK-NEXT: [[I:%.*]] = phi i64 [ [[I_NEXT:%.*]], [[FOR_BODY]] ], [ 3, [[ENTRY:%.*]] ]
+; CHECK-NEXT: [[A_I:%.*]] = getelementptr inbounds nuw i32, ptr [[A:%.*]], i64 [[I]]
; CHECK-NEXT: [[TMP31:%.*]] = getelementptr i32, ptr [[A]], i64 [[I]]
; CHECK-NEXT: [[A_I_MINUS_1:%.*]] = getelementptr i8, ptr [[TMP31]], i64 -4
; CHECK-NEXT: [[TMP32:%.*]] = getelementptr i32, ptr [[A]], i64 [[I]]
; CHECK-NEXT: [[A_I_MINUS_3:%.*]] = getelementptr i8, ptr [[TMP32]], i64 -12
-; CHECK-NEXT: store i32 [[X]], ptr [[A_I_MINUS_1]], align 4
-; CHECK-NEXT: store i32 [[Y]], ptr [[A_I_MINUS_3]], align 4
-; CHECK-NEXT: store i32 [[Z]], ptr [[A_I]], align 4
+; CHECK-NEXT: store i32 [[X:%.*]], ptr [[A_I_MINUS_1]], align 4
+; CHECK-NEXT: store i32 [[Y:%.*]], ptr [[A_I_MINUS_3]], align 4
+; CHECK-NEXT: store i32 [[Z:%.*]], ptr [[A_I]], align 4
; CHECK-NEXT: [[I_NEXT]] = add nuw nsw i64 [[I]], 2
-; CHECK-NEXT: [[COND:%.*]] = icmp slt i64 [[I_NEXT]], [[N]]
-; CHECK-NEXT: br i1 [[COND]], label [[FOR_BODY]], label [[FOR_END]], !llvm.loop [[LOOP35:![0-9]+]]
+; CHECK-NEXT: [[COND:%.*]] = icmp slt i64 [[I_NEXT]], [[N:%.*]]
+; CHECK-NEXT: br i1 [[COND]], label [[FOR_BODY]], label [[FOR_END:%.*]]
; CHECK: for.end:
; CHECK-NEXT: ret void
;
@@ -1489,7 +1427,7 @@ define void @PR34743(ptr %a, ptr %b, i64 %n) {
; CHECK-NEXT: [[DOTIDX:%.*]] = shl i64 [[INDEX]], 2
; CHECK-NEXT: [[TMP17:%.*]] = getelementptr i8, ptr [[A]], i64 [[DOTIDX]]
; CHECK-NEXT: [[TMP7:%.*]] = getelementptr i8, ptr [[TMP17]], i64 2
-; CHECK-NEXT: [[WIDE_VEC:%.*]] = load <8 x i16>, ptr [[TMP7]], align 4
+; CHECK-NEXT: [[WIDE_VEC:%.*]] = load <8 x i16>, ptr [[TMP7]], align 4, !alias.scope [[META34:![0-9]+]]
; CHECK-NEXT: [[STRIDED_VEC:%.*]] = shufflevector <8 x i16> [[WIDE_VEC]], <8 x i16> poison, <4 x i32> <i32 0, i32 2, i32 4, i32 6>
; CHECK-NEXT: [[STRIDED_VEC4]] = shufflevector <8 x i16> [[WIDE_VEC]], <8 x i16> poison, <4 x i32> <i32 1, i32 3, i32 5, i32 7>
; CHECK-NEXT: [[TMP8:%.*]] = shufflevector <4 x i16> [[VECTOR_RECUR]], <4 x i16> [[STRIDED_VEC4]], <4 x i32> <i32 3, i32 4, i32 5, i32 6>
@@ -1499,10 +1437,10 @@ define void @PR34743(ptr %a, ptr %b, i64 %n) {
; CHECK-NEXT: [[TMP12:%.*]] = mul nsw <4 x i32> [[TMP9]], [[TMP10]]
; CHECK-NEXT: [[TMP13:%.*]] = mul nsw <4 x i32> [[TMP12]], [[TMP11]]
; CHECK-NEXT: [[TMP14:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[INDEX]]
-; CHECK-NEXT: store <4 x i32> [[TMP13]], ptr [[TMP14]], align 4, !alias.scope [[META36:![0-9]+]], !noalias [[META39:![0-9]+]]
+; CHECK-NEXT: store <4 x i32> [[TMP13]], ptr [[TMP14]], align 4, !alias.scope [[META37:![0-9]+]], !noalias [[META34]]
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; CHECK-NEXT: [[TMP15:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; CHECK-NEXT: br i1 [[TMP15]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP41:![0-9]+]]
+; CHECK-NEXT: br i1 [[TMP15]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP39:![0-9]+]]
; CHECK: middle.block:
; CHECK-NEXT: [[VECTOR_RECUR_EXTRACT:%.*]] = extractelement <8 x i16> [[WIDE_VEC]], i64 7
; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[TMP1]], [[N_VEC]]
@@ -1531,7 +1469,7 @@ define void @PR34743(ptr %a, ptr %b, i64 %n) {
; CHECK-NEXT: [[ARRAYIDX5:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[I]]
; CHECK-NEXT: store i32 [[MUL012]], ptr [[ARRAYIDX5]], align 4
; CHECK-NEXT: [[EXITCOND:%.*]] = icmp eq i64 [[IV]], [[N]]
-; CHECK-NEXT: br i1 [[EXITCOND]], label [[END]], label [[LOOP]], !llvm.loop [[LOOP42:![0-9]+]]
+; CHECK-NEXT: br i1 [[EXITCOND]], label [[END]], label [[LOOP]], !llvm.loop [[LOOP40:![0-9]+]]
; CHECK: end:
; CHECK-NEXT: ret void
;
>From 696f52ca02136edac799bb72a1d39603aa34975f Mon Sep 17 00:00:00 2001
From: Sebastian Pop <spop at nvidia.com>
Date: Sat, 23 Aug 2025 23:16:38 -0500
Subject: [PATCH 2/2] [SCEV] Fix NSW flag propagation in getAddExpr
SCEV was losing NSW flags during AddRec operations, causing Dependence
Analysis to add unnecessary runtime assumptions for inbounds GEPs.
This patch fixes getAddExpr: when combining AddRecs from the same loop, preserve
compatible NSW/NUW flags from input AddRecs instead of always using FlagAnyWrap.
---
llvm/lib/Analysis/ScalarEvolution.cpp | 31 +++++-
.../Delinearization/fixed_size_array.ll | 4 +-
.../scev-nsw-flags-enable-analysis.ll | 45 ++++++++
.../ScalarEvolution/different-loops-recs.ll | 30 +++---
.../gep-nsw-flag-preservation.ll | 100 ++++++++++++++++++
5 files changed, 191 insertions(+), 19 deletions(-)
create mode 100644 llvm/test/Analysis/DependenceAnalysis/scev-nsw-flags-enable-analysis.ll
create mode 100644 llvm/test/Analysis/ScalarEvolution/gep-nsw-flag-preservation.ll
diff --git a/llvm/lib/Analysis/ScalarEvolution.cpp b/llvm/lib/Analysis/ScalarEvolution.cpp
index d2c445f1ffaa0..9867994527c45 100644
--- a/llvm/lib/Analysis/ScalarEvolution.cpp
+++ b/llvm/lib/Analysis/ScalarEvolution.cpp
@@ -2951,25 +2951,52 @@ const SCEV *ScalarEvolution::getAddExpr(SmallVectorImpl<const SCEV *> &Ops,
if (AddRecLoop == cast<SCEVAddRecExpr>(Ops[OtherIdx])->getLoop()) {
// Other + {A,+,B}<L> + {C,+,D}<L> --> Other + {A+C,+,B+D}<L>
SmallVector<const SCEV *, 4> AddRecOps(AddRec->operands());
+
+ // Track flags: start with the flags from the first AddRec.
+ bool AllHaveNSW = AddRec->hasNoSignedWrap();
+ bool AllHaveNUW = AddRec->hasNoUnsignedWrap();
+
for (; OtherIdx != Ops.size() && isa<SCEVAddRecExpr>(Ops[OtherIdx]);
++OtherIdx) {
const auto *OtherAddRec = cast<SCEVAddRecExpr>(Ops[OtherIdx]);
if (OtherAddRec->getLoop() == AddRecLoop) {
+ // Update flags based on this AddRec
+ if (!OtherAddRec->hasNoSignedWrap())
+ AllHaveNSW = false;
+ if (!OtherAddRec->hasNoUnsignedWrap())
+ AllHaveNUW = false;
for (unsigned i = 0, e = OtherAddRec->getNumOperands();
i != e; ++i) {
if (i >= AddRecOps.size()) {
append_range(AddRecOps, OtherAddRec->operands().drop_front(i));
break;
}
+ // Preserve no-wrap flags when combining AddRec operands.
+ SCEV::NoWrapFlags CombineFlags = SCEV::FlagAnyWrap;
+ if (auto *AR1 = dyn_cast<SCEVAddRecExpr>(AddRecOps[i]))
+ if (auto *AR2 =
+ dyn_cast<SCEVAddRecExpr>(OtherAddRec->getOperand(i))) {
+ if (AR1->hasNoSignedWrap() && AR2->hasNoSignedWrap())
+ CombineFlags = setFlags(CombineFlags, SCEV::FlagNSW);
+ if (AR1->hasNoUnsignedWrap() && AR2->hasNoUnsignedWrap())
+ CombineFlags = setFlags(CombineFlags, SCEV::FlagNUW);
+ }
SmallVector<const SCEV *, 2> TwoOps = {
AddRecOps[i], OtherAddRec->getOperand(i)};
- AddRecOps[i] = getAddExpr(TwoOps, SCEV::FlagAnyWrap, Depth + 1);
+ AddRecOps[i] = getAddExpr(TwoOps, CombineFlags, Depth + 1);
}
Ops.erase(Ops.begin() + OtherIdx); --OtherIdx;
}
}
// Step size has changed, so we cannot guarantee no self-wraparound.
- Ops[Idx] = getAddRecExpr(AddRecOps, AddRecLoop, SCEV::FlagAnyWrap);
+ // However, preserve NSW/NUW flags if all combined AddRecs had them.
+ SCEV::NoWrapFlags FinalFlags = SCEV::FlagAnyWrap;
+ if (AllHaveNSW)
+ FinalFlags = setFlags(FinalFlags, SCEV::FlagNSW);
+ if (AllHaveNUW)
+ FinalFlags = setFlags(FinalFlags, SCEV::FlagNUW);
+
+ Ops[Idx] = getAddRecExpr(AddRecOps, AddRecLoop, FinalFlags);
return getAddExpr(Ops, SCEV::FlagAnyWrap, Depth + 1);
}
}
diff --git a/llvm/test/Analysis/Delinearization/fixed_size_array.ll b/llvm/test/Analysis/Delinearization/fixed_size_array.ll
index 634850bb4a5a2..ffd0202e205ce 100644
--- a/llvm/test/Analysis/Delinearization/fixed_size_array.ll
+++ b/llvm/test/Analysis/Delinearization/fixed_size_array.ll
@@ -339,7 +339,7 @@ define void @a_i_j2k_i(ptr %a) {
; CHECK-LABEL: 'a_i_j2k_i'
; CHECK-NEXT: Inst: store i32 1, ptr %idx, align 4
; CHECK-NEXT: In Loop with Header: for.k
-; CHECK-NEXT: AccessFunction: {{\{\{\{}}0,+,1028}<%for.i.header>,+,256}<nw><%for.j.header>,+,128}<nw><%for.k>
+; CHECK-NEXT: AccessFunction: {{\{\{\{}}0,+,1028}<nuw><nsw><%for.i.header>,+,256}<nw><%for.j.header>,+,128}<nw><%for.k>
; CHECK-NEXT: failed to delinearize
;
entry:
@@ -391,7 +391,7 @@ define void @a_i_i_jk(ptr %a) {
; CHECK-LABEL: 'a_i_i_jk'
; CHECK-NEXT: Inst: store i32 1, ptr %idx, align 4
; CHECK-NEXT: In Loop with Header: for.k
-; CHECK-NEXT: AccessFunction: {{\{\{\{}}0,+,1152}<%for.i.header>,+,4}<nw><%for.j.header>,+,4}<nw><%for.k>
+; CHECK-NEXT: AccessFunction: {{\{\{\{}}0,+,1152}<nuw><nsw><%for.i.header>,+,4}<nw><%for.j.header>,+,4}<nw><%for.k>
; CHECK-NEXT: Base offset: %a
; CHECK-NEXT: ArrayDecl[UnknownSize][288] with elements of 4 bytes.
; CHECK-NEXT: ArrayRef[{0,+,1}<nuw><nsw><%for.i.header>][{{\{\{}}0,+,1}<nuw><nsw><%for.j.header>,+,1}<nuw><nsw><%for.k>]
diff --git a/llvm/test/Analysis/DependenceAnalysis/scev-nsw-flags-enable-analysis.ll b/llvm/test/Analysis/DependenceAnalysis/scev-nsw-flags-enable-analysis.ll
new file mode 100644
index 0000000000000..b717ce9815341
--- /dev/null
+++ b/llvm/test/Analysis/DependenceAnalysis/scev-nsw-flags-enable-analysis.ll
@@ -0,0 +1,45 @@
+; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py UTC_ARGS: --version 5
+; RUN: opt < %s -disable-output "-passes=print<da>" -aa-pipeline=basic-aa 2>&1 | FileCheck %s
+
+; Test that SCEV NSW flag preservation enables dependence analysis to work
+; correctly. Previously, SCEV would lose NSW flags when combining AddRec
+; expressions from GEP operations, causing dependence analysis to incorrectly
+; classify expressions as "wrapping" and fail analysis.
+
+define void @test_da_with_scev_flags(ptr %A) {
+; This test verifies that dependence analysis now correctly identifies
+; self-dependences when SCEV preserves NSW flags from GEP index computations.
+; CHECK-LABEL: 'test_da_with_scev_flags'
+; CHECK-NEXT: Src: %val = load i32, ptr %gep, align 4 --> Dst: %val = load i32, ptr %gep, align 4
+; CHECK-NEXT: da analyze - none!
+; CHECK-NEXT: Src: %val = load i32, ptr %gep, align 4 --> Dst: store i32 %val, ptr %gep, align 4
+; CHECK-NEXT: da analyze - consistent anti [0|<]!
+; CHECK-NEXT: Src: store i32 %val, ptr %gep, align 4 --> Dst: store i32 %val, ptr %gep, align 4
+; CHECK-NEXT: da analyze - none!
+;
+
+entry:
+ br label %loop
+
+loop:
+ %i = phi i64 [ 0, %entry ], [ %i.next, %loop ]
+
+ ; Create NSW-flagged index computation
+ %mul = mul nsw i64 %i, 3
+ %sub = add nsw i64 %mul, -6
+
+ ; GEP that should result in SCEV: {(-2424 + %A),+,1212}<nw>
+ ; The <nw> flag should prevent false "wrapping" detection in DA
+ %gep = getelementptr inbounds [100 x i32], ptr %A, i64 %sub, i64 %sub
+
+ ; Self-dependence: should be detected as "none" (no dependence)
+ %val = load i32, ptr %gep
+ store i32 %val, ptr %gep
+
+ %i.next = add nsw i64 %i, 1
+ %cond = icmp ult i64 %i.next, 50
+ br i1 %cond, label %loop, label %exit
+
+exit:
+ ret void
+}
diff --git a/llvm/test/Analysis/ScalarEvolution/different-loops-recs.ll b/llvm/test/Analysis/ScalarEvolution/different-loops-recs.ll
index 44bff5638bc85..b700f6cd0b432 100644
--- a/llvm/test/Analysis/ScalarEvolution/different-loops-recs.ll
+++ b/llvm/test/Analysis/ScalarEvolution/different-loops-recs.ll
@@ -23,9 +23,9 @@ define void @test_00(i1 %arg) {
; CHECK-NEXT: %phi3.inc = add i32 %phi3, 3
; CHECK-NEXT: --> {9,+,3}<nuw><nsw><%loop1> U: [9,502) S: [9,502) Exits: 501 LoopDispositions: { %loop1: Computable }
; CHECK-NEXT: %sum1 = add i32 %phi1, %phi2
-; CHECK-NEXT: --> {14,+,3}<%loop1> U: [14,507) S: [14,507) Exits: 506 LoopDispositions: { %loop1: Computable }
+; CHECK-NEXT: --> {14,+,3}<nuw><nsw><%loop1> U: [14,507) S: [14,507) Exits: 506 LoopDispositions: { %loop1: Computable }
; CHECK-NEXT: %sum2 = add i32 %sum1, %phi3
-; CHECK-NEXT: --> {20,+,6}<%loop1> U: [20,1005) S: [20,1005) Exits: 1004 LoopDispositions: { %loop1: Computable }
+; CHECK-NEXT: --> {20,+,6}<nuw><nsw><%loop1> U: [20,1005) S: [20,1005) Exits: 1004 LoopDispositions: { %loop1: Computable }
; CHECK-NEXT: %phi4 = phi i32 [ 63, %loop1 ], [ %phi4.inc, %loop2 ]
; CHECK-NEXT: --> {63,+,1}<nuw><nsw><%loop2> U: [63,205) S: [63,205) Exits: 204 LoopDispositions: { %loop2: Computable }
; CHECK-NEXT: %phi5 = phi i32 [ 53, %loop1 ], [ %phi5.inc, %loop2 ]
@@ -39,21 +39,21 @@ define void @test_00(i1 %arg) {
; CHECK-NEXT: %phi6.inc = add i32 %phi6, 3
; CHECK-NEXT: --> {46,+,3}<nuw><nsw><%loop2> U: [46,470) S: [46,470) Exits: 469 LoopDispositions: { %loop2: Computable }
; CHECK-NEXT: %sum3 = add i32 %phi4, %phi5
-; CHECK-NEXT: --> {116,+,3}<%loop2> U: [116,540) S: [116,540) Exits: 539 LoopDispositions: { %loop2: Computable }
+; CHECK-NEXT: --> {116,+,3}<nuw><nsw><%loop2> U: [116,540) S: [116,540) Exits: 539 LoopDispositions: { %loop2: Computable }
; CHECK-NEXT: %sum4 = add i32 %sum3, %phi6
-; CHECK-NEXT: --> {159,+,6}<%loop2> U: [159,1006) S: [159,1006) Exits: 1005 LoopDispositions: { %loop2: Computable }
+; CHECK-NEXT: --> {159,+,6}<nuw><nsw><%loop2> U: [159,1006) S: [159,1006) Exits: 1005 LoopDispositions: { %loop2: Computable }
; CHECK-NEXT: %s1 = add i32 %phi1, %phi4
; CHECK-NEXT: --> {{\{\{}}73,+,1}<nuw><nsw><%loop1>,+,1}<nw><%loop2> U: [73,379) S: [73,379) --> 378 U: [378,379) S: [378,379)
; CHECK-NEXT: %s2 = add i32 %phi5, %phi2
; CHECK-NEXT: --> {{\{\{}}57,+,2}<nuw><nsw><%loop1>,+,2}<nw><%loop2> U: [57,668) S: [57,668) --> 667 U: [667,668) S: [667,668)
; CHECK-NEXT: %s3 = add i32 %sum1, %sum3
-; CHECK-NEXT: --> {{\{\{}}130,+,3}<%loop1>,+,3}<%loop2> U: [130,1046) S: [130,1046) --> 1045 U: [1045,1046) S: [1045,1046)
+; CHECK-NEXT: --> {{\{\{}}130,+,3}<nuw><nsw><%loop1>,+,3}<nw><%loop2> U: [130,1046) S: [130,1046) --> 1045 U: [1045,1046) S: [1045,1046)
; CHECK-NEXT: %s4 = add i32 %sum4, %sum2
-; CHECK-NEXT: --> {{\{\{}}179,+,6}<%loop1>,+,6}<%loop2> U: [179,2010) S: [179,2010) --> 2009 U: [2009,2010) S: [2009,2010)
+; CHECK-NEXT: --> {{\{\{}}179,+,6}<nuw><nsw><%loop1>,+,6}<nw><%loop2> U: [179,2010) S: [179,2010) --> 2009 U: [2009,2010) S: [2009,2010)
; CHECK-NEXT: %s5 = add i32 %phi3, %sum3
-; CHECK-NEXT: --> {{\{\{}}122,+,3}<nuw><nsw><%loop1>,+,3}<%loop2> U: [122,1038) S: [122,1038) --> 1037 U: [1037,1038) S: [1037,1038)
+; CHECK-NEXT: --> {{\{\{}}122,+,3}<nuw><nsw><%loop1>,+,3}<nw><%loop2> U: [122,1038) S: [122,1038) --> 1037 U: [1037,1038) S: [1037,1038)
; CHECK-NEXT: %s6 = add i32 %sum2, %phi6
-; CHECK-NEXT: --> {{\{\{}}63,+,6}<%loop1>,+,3}<nw><%loop2> U: [63,1471) S: [63,1471) --> 1470 U: [1470,1471) S: [1470,1471)
+; CHECK-NEXT: --> {{\{\{}}63,+,6}<nuw><nsw><%loop1>,+,3}<nw><%loop2> U: [63,1471) S: [63,1471) --> 1470 U: [1470,1471) S: [1470,1471)
; CHECK-NEXT: Determining loop execution counts for: @test_00
; CHECK-NEXT: Loop %loop2: backedge-taken count is i32 141
; CHECK-NEXT: Loop %loop2: constant max backedge-taken count is i32 141
@@ -139,13 +139,13 @@ define void @test_01(i32 %a, i32 %b) {
; CHECK-NEXT: %phi6.inc = add i32 %phi6, 3
; CHECK-NEXT: --> {46,+,3}<nuw><nsw><%loop2> U: [46,548) S: [46,548) Exits: (46 + (3 * (((-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (-1 * (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))))))))<nuw><nsw> + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))) /u 6) + (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))))))<nuw>)<nuw> LoopDispositions: { %loop2: Computable }
; CHECK-NEXT: %sum3 = add i32 %phi4, %phi5
-; CHECK-NEXT: --> {116,+,3}<%loop2> U: [116,618) S: [116,618) Exits: (116 + (3 * (((-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (-1 * (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))))))))<nuw><nsw> + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))) /u 6) + (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))))))<nuw>)<nuw> LoopDispositions: { %loop2: Computable }
+; CHECK-NEXT: --> {116,+,3}<nuw><nsw><%loop2> U: [116,618) S: [116,618) Exits: (116 + (3 * (((-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (-1 * (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))))))))<nuw><nsw> + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))) /u 6) + (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))))))<nuw>)<nuw> LoopDispositions: { %loop2: Computable }
; CHECK-NEXT: %sum4 = add i32 %sum3, %phi6
-; CHECK-NEXT: --> {159,+,6}<%loop2> U: [159,1162) S: [159,1162) Exits: (159 + (6 * (((-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (-1 * (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))))))))<nuw><nsw> + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))) /u 6) + (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))))))) LoopDispositions: { %loop2: Computable }
+; CHECK-NEXT: --> {159,+,6}<nuw><nsw><%loop2> U: [159,1162) S: [159,1162) Exits: (159 + (6 * (((-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (-1 * (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))))))))<nuw><nsw> + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))) /u 6) + (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))))))) LoopDispositions: { %loop2: Computable }
; CHECK-NEXT: %is2 = add i32 %sum4, %b
-; CHECK-NEXT: --> {(159 + %b),+,6}<%loop2> U: full-set S: full-set Exits: (159 + (6 * (((-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (-1 * (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))))))))<nuw><nsw> + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))) /u 6) + (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))))))))) + %b) LoopDispositions: { %loop2: Computable }
+; CHECK-NEXT: --> {(159 + %b),+,6}<nw><%loop2> U: full-set S: full-set Exits: (159 + (6 * (((-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (-1 * (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))))))))<nuw><nsw> + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))) /u 6) + (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))))))))) + %b) LoopDispositions: { %loop2: Computable }
; CHECK-NEXT: %ec2 = add i32 %is1, %is2
-; CHECK-NEXT: --> {{\{\{}}(165 + (2 * %a) + (2 * %b)),+,6}<%loop1>,+,6}<%loop2> U: full-set S: full-set --> {(165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))),+,6}<%loop2> U: full-set S: full-set Exits: (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (6 * (((-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (-1 * (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))))))))<nuw><nsw> + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))) /u 6) + (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))))))) LoopDispositions: { %loop2: Computable }
+; CHECK-NEXT: --> {{\{\{}}(165 + (2 * %a) + (2 * %b)),+,6}<%loop1>,+,6}<nw><%loop2> U: full-set S: full-set --> {(165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))),+,6}<nw><%loop2> U: full-set S: full-set Exits: (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (6 * (((-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (-1 * (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))))))))<nuw><nsw> + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))) /u 6) + (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))))))) LoopDispositions: { %loop2: Computable }
; CHECK-NEXT: %s1 = add i32 %phi1, %is1
; CHECK-NEXT: --> {(6 + (3 * %a) + %b),+,7}<%loop1> U: full-set S: full-set --> (6 + (3 * %a) + (7 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + %b) U: full-set S: full-set
; CHECK-NEXT: %s2 = add i32 %is2, %phi4
@@ -153,11 +153,11 @@ define void @test_01(i32 %a, i32 %b) {
; CHECK-NEXT: %s3 = add i32 %is1, %phi5
; CHECK-NEXT: --> {{\{\{}}(59 + (2 * %a) + %b),+,6}<%loop1>,+,2}<nw><%loop2> U: full-set S: full-set --> (59 + (2 * (((-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (-1 * (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))))))))<nuw><nsw> + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))) /u 6) + (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))))))<nuw><nsw> + (2 * %a) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + %b) U: full-set S: full-set
; CHECK-NEXT: %s4 = add i32 %phi2, %is2
-; CHECK-NEXT: --> {{\{\{}}(159 + (2 * %b)),+,2}<nw><%loop1>,+,6}<%loop2> U: full-set S: full-set --> (159 + (2 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))<nuw><nsw> + (2 * %b) + (6 * (((-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (-1 * (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))))))))<nuw><nsw> + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))) /u 6) + (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))))))) U: full-set S: full-set
+; CHECK-NEXT: --> {{\{\{}}(159 + (2 * %b)),+,2}<nw><%loop1>,+,6}<nw><%loop2> U: full-set S: full-set --> (159 + (2 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))<nuw><nsw> + (2 * %b) + (6 * (((-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (-1 * (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))))))))<nuw><nsw> + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))) /u 6) + (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))))))) U: full-set S: full-set
; CHECK-NEXT: %s5 = add i32 %is1, %is2
-; CHECK-NEXT: --> {{\{\{}}(165 + (2 * %a) + (2 * %b)),+,6}<%loop1>,+,6}<%loop2> U: full-set S: full-set --> (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (6 * (((-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (-1 * (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))))))))<nuw><nsw> + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))) /u 6) + (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))))))) U: full-set S: full-set
+; CHECK-NEXT: --> {{\{\{}}(165 + (2 * %a) + (2 * %b)),+,6}<%loop1>,+,6}<nw><%loop2> U: full-set S: full-set --> (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (6 * (((-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (-1 * (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))))))))<nuw><nsw> + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))) /u 6) + (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))))))) U: full-set S: full-set
; CHECK-NEXT: %s6 = add i32 %is2, %is1
-; CHECK-NEXT: --> {{\{\{}}(165 + (2 * %a) + (2 * %b)),+,6}<%loop1>,+,6}<%loop2> U: full-set S: full-set --> (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (6 * (((-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (-1 * (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))))))))<nuw><nsw> + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))) /u 6) + (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))))))) U: full-set S: full-set
+; CHECK-NEXT: --> {{\{\{}}(165 + (2 * %a) + (2 * %b)),+,6}<%loop1>,+,6}<nw><%loop2> U: full-set S: full-set --> (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (6 * (((-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (-1 * (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))))))))<nuw><nsw> + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))) /u 6) + (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))))))) U: full-set S: full-set
; CHECK-NEXT: Determining loop execution counts for: @test_01
; CHECK-NEXT: Loop %loop2: backedge-taken count is (((-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (-1 * (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))))))))<nuw><nsw> + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))))))))) /u 6) + (1 umin (-165 + (-6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))) + (-2 * %a) + (-2 * %b) + (1000 umax (165 + (2 * %a) + (2 * %b) + (6 * (((-6 + (-2 * %a) + (-1 * (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))<nuw><nsw> + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b))) /u 6) + (1 umin (-6 + (-2 * %a) + (-1 * %b) + (1000 umax (6 + (2 * %a) + %b)))))))))))
; CHECK-NEXT: Loop %loop2: constant max backedge-taken count is i32 167
diff --git a/llvm/test/Analysis/ScalarEvolution/gep-nsw-flag-preservation.ll b/llvm/test/Analysis/ScalarEvolution/gep-nsw-flag-preservation.ll
new file mode 100644
index 0000000000000..3d1865540c78f
--- /dev/null
+++ b/llvm/test/Analysis/ScalarEvolution/gep-nsw-flag-preservation.ll
@@ -0,0 +1,100 @@
+; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
+; RUN: opt < %s -disable-output "-passes=print<scalar-evolution>" 2>&1 | FileCheck %s
+
+; Test that SCEV preserves NSW flags when combining AddRec expressions
+; from GEP operations. Previously, the NSW flags would be lost when
+; combining offset calculations like:
+; {-2400,+,1200}<nsw> + {-24,+,12}<nsw> = {-2424,+,1212}<nsw>
+; This test ensures the fix in getAddExpr properly preserves flags.
+
+define void @test_gep_nsw_preservation(ptr %A) {
+; CHECK-LABEL: 'test_gep_nsw_preservation'
+; CHECK-NEXT: Classifying expressions for: @test_gep_nsw_preservation
+; CHECK-NEXT: %i = phi i64 [ 0, %entry ], [ %i.next, %loop ]
+; CHECK-NEXT: --> {0,+,1}<nuw><nsw><%loop> U: [0,50) S: [0,50) Exits: 49 LoopDispositions: { %loop: Computable }
+; CHECK-NEXT: %mul = mul nsw i64 %i, 3
+; CHECK-NEXT: --> {0,+,3}<nuw><nsw><%loop> U: [0,148) S: [0,148) Exits: 147 LoopDispositions: { %loop: Computable }
+; CHECK-NEXT: %sub = add nsw i64 %mul, -6
+; CHECK-NEXT: --> {-6,+,3}<nsw><%loop> U: [-6,142) S: [-6,142) Exits: 141 LoopDispositions: { %loop: Computable }
+; CHECK-NEXT: %gep = getelementptr inbounds [100 x i32], ptr %A, i64 %sub, i64 %sub
+; CHECK-NEXT: --> {(-2424 + %A),+,1212}<nw><%loop> U: full-set S: full-set Exits: (56964 + %A) LoopDispositions: { %loop: Computable }
+; CHECK-NEXT: %i.next = add nsw i64 %i, 1
+; CHECK-NEXT: --> {1,+,1}<nuw><nsw><%loop> U: [1,51) S: [1,51) Exits: 50 LoopDispositions: { %loop: Computable }
+; CHECK-NEXT: Determining loop execution counts for: @test_gep_nsw_preservation
+; CHECK-NEXT: Loop %loop: backedge-taken count is i64 49
+; CHECK-NEXT: Loop %loop: constant max backedge-taken count is i64 49
+; CHECK-NEXT: Loop %loop: symbolic max backedge-taken count is i64 49
+; CHECK-NEXT: Loop %loop: Trip multiple is 50
+;
+entry:
+ br label %loop
+
+loop:
+ %i = phi i64 [ 0, %entry ], [ %i.next, %loop ]
+
+ ; Create NSW-flagged arithmetic that gets used in GEP indices
+ %mul = mul nsw i64 %i, 3
+ %sub = add nsw i64 %mul, -6
+
+ ; GEP with inbounds using the NSW-flagged expressions as indices
+ ; This should result in SCEV: {(-2424 + %A),+,1212}<nw>
+ ; The <nw> flag should be preserved from the NSW flags on the indices
+ %gep = getelementptr inbounds [100 x i32], ptr %A, i64 %sub, i64 %sub
+
+ %i.next = add nsw i64 %i, 1
+ %cond = icmp ult i64 %i.next, 50
+ br i1 %cond, label %loop, label %exit
+
+exit:
+ ret void
+}
+
+; Test that the preserved flags enable dependence analysis
+define void @test_dependence_analysis(ptr %A) {
+; This test should show that dependence analysis works correctly
+; when SCEV preserves NSW flags, preventing false "wrapping" detection
+; CHECK-LABEL: 'test_dependence_analysis'
+; CHECK-NEXT: Classifying expressions for: @test_dependence_analysis
+; CHECK-NEXT: %i = phi i64 [ 0, %entry ], [ %i.next, %loop ]
+; CHECK-NEXT: --> {0,+,1}<nuw><nsw><%loop> U: [0,50) S: [0,50) Exits: 49 LoopDispositions: { %loop: Computable }
+; CHECK-NEXT: %mul = mul nsw i64 %i, 3
+; CHECK-NEXT: --> {0,+,3}<nuw><nsw><%loop> U: [0,148) S: [0,148) Exits: 147 LoopDispositions: { %loop: Computable }
+; CHECK-NEXT: %sub = add nsw i64 %mul, -6
+; CHECK-NEXT: --> {-6,+,3}<nsw><%loop> U: [-6,142) S: [-6,142) Exits: 141 LoopDispositions: { %loop: Computable }
+; CHECK-NEXT: %gep1 = getelementptr inbounds [100 x i32], ptr %A, i64 %sub, i64 %sub
+; CHECK-NEXT: --> {(-2424 + %A),+,1212}<nw><%loop> U: full-set S: full-set Exits: (56964 + %A) LoopDispositions: { %loop: Computable }
+; CHECK-NEXT: %gep2 = getelementptr inbounds [100 x i32], ptr %A, i64 %sub, i64 %sub
+; CHECK-NEXT: --> {(-2424 + %A),+,1212}<nw><%loop> U: full-set S: full-set Exits: (56964 + %A) LoopDispositions: { %loop: Computable }
+; CHECK-NEXT: %val = load i32, ptr %gep2, align 4
+; CHECK-NEXT: --> %val U: full-set S: full-set Exits: <<Unknown>> LoopDispositions: { %loop: Variant }
+; CHECK-NEXT: %i.next = add nsw i64 %i, 1
+; CHECK-NEXT: --> {1,+,1}<nuw><nsw><%loop> U: [1,51) S: [1,51) Exits: 50 LoopDispositions: { %loop: Computable }
+; CHECK-NEXT: Determining loop execution counts for: @test_dependence_analysis
+; CHECK-NEXT: Loop %loop: backedge-taken count is i64 49
+; CHECK-NEXT: Loop %loop: constant max backedge-taken count is i64 49
+; CHECK-NEXT: Loop %loop: symbolic max backedge-taken count is i64 49
+; CHECK-NEXT: Loop %loop: Trip multiple is 50
+;
+entry:
+ br label %loop
+
+loop:
+ %i = phi i64 [ 0, %entry ], [ %i.next, %loop ]
+
+ %mul = mul nsw i64 %i, 3
+ %sub = add nsw i64 %mul, -6
+
+ ; Two identical GEPs - should be detected as same location (no dependence)
+ %gep1 = getelementptr inbounds [100 x i32], ptr %A, i64 %sub, i64 %sub
+ %gep2 = getelementptr inbounds [100 x i32], ptr %A, i64 %sub, i64 %sub
+
+ %val = load i32, ptr %gep2
+ store i32 %val, ptr %gep1
+
+ %i.next = add nsw i64 %i, 1
+ %cond = icmp ult i64 %i.next, 50
+ br i1 %cond, label %loop, label %exit
+
+exit:
+ ret void
+}
More information about the llvm-commits
mailing list