[MemDep] DBG intrinsics don't impact abort limit for call site dependence analysis

Summary:
Memory dependence analysis no longer counts DbgInfoIntrinsics towards the
limit where to abort the analysis. Before, a bunch of calls to dbg.value
could affect the generated code, meaning that with -g we could generate
different code than without.

Reviewers: chandlerc, Prazek, davide, efriedma

Reviewed By: efriedma

Subscribers: llvm-commits

Differential Revision: https://reviews.llvm.org/D39181

llvm-svn: 316551

lib/Analysis/MemoryDependenceAnalysis.cpp

@@ -196,14 +196,17 @@ MemDepResult MemoryDependenceResults::getCallSiteDependencyFrom(
 
   // Walk backwards through the block, looking for dependencies.
   while (ScanIt != BB->begin()) {
+    Instruction *Inst = &*--ScanIt;
+    // Debug intrinsics don't cause dependences and should not affect Limit
+    if (isa<DbgInfoIntrinsic>(Inst))
+      continue;
+
     // Limit the amount of scanning we do so we don't end up with quadratic
     // running time on extreme testcases.
     --Limit;
     if (!Limit)
       return MemDepResult::getUnknown();
 
-    Instruction *Inst = &*--ScanIt;
-
     // If this inst is a memory op, get the pointer it accessed
     MemoryLocation Loc;
     ModRefInfo MR = GetLocation(Inst, Loc, TLI);
@@ -215,9 +218,6 @@ MemDepResult MemoryDependenceResults::getCallSiteDependencyFrom(
     }
 
     if (auto InstCS = CallSite(Inst)) {
-      // Debug intrinsics don't cause dependences.
-      if (isa<DbgInfoIntrinsic>(Inst))
-        continue;
       // If these two calls do not interfere, look past it.
       switch (AA.getModRefInfo(CS, InstCS)) {
       case MRI_NoModRef:

test/Transforms/DeadStoreElimination/mda-with-dbg-values.ll

@@ -0,0 +1,72 @@
+; RUN: opt -S -dse -memdep-block-scan-limit=3 < %s | FileCheck %s
+; RUN: opt -S -strip-debug -dse -memdep-block-scan-limit=3 < %s | FileCheck %s
+
+; Test case to check that the memory dependency analysis gets the same
+; result even if we have a dbg value between the memcpy and
+; store. The memory dependency is then used by DSE to remove the store.
+
+; We use -memdep-block-scan-limit=3 to be able to create a small test case.
+; Without it, we would need to squeeze in 100 instructions since the default
+; limit is 100.
+
+target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-unknown-linux-gnu"
+
+@g = common global [1 x i8] zeroinitializer, align 1, !dbg !0
+
+; Function Attrs: noinline nounwind uwtable
+define void @foo() #0 !dbg !14 {
+entry:
+  %i = alloca i8, align 1
+  store i8 1, i8* %i, align 1, !dbg !19
+  call void @llvm.dbg.value(metadata i32 0, i64 0, metadata !17, metadata !DIExpression()), !dbg !18
+  %0 = bitcast [1 x i8]* @g to i8*
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %i, i8* %0, i64 1, i32 1, i1 false), !dbg !20
+  br label %bb2
+
+bb2:                                              ; preds = %0
+  ret void, !dbg !21
+}
+
+; Function Attrs: nounwind readnone speculatable
+declare void @llvm.dbg.value(metadata, i64, metadata, metadata) #1
+
+; Function Attrs: argmemonly nounwind
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture writeonly, i8* nocapture readonly, i64, i32, i1) #2
+
+attributes #0 = { noinline nounwind uwtable "correctly-rounded-divide-sqrt-fp-math"="false" "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "no-trapping-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="x86-64" "target-features"="+fxsr,+mmx,+sse,+sse2,+x87" "unsafe-fp-math"="false" "use-soft-float"="false" }
+attributes #1 = { nounwind readnone speculatable }
+attributes #2 = { argmemonly nounwind }
+
+!llvm.dbg.cu = !{!2}
+!llvm.module.flags = !{!10, !11, !12}
+!llvm.ident = !{!13}
+
+!0 = !DIGlobalVariableExpression(var: !1, expr: !DIExpression())
+!1 = distinct !DIGlobalVariable(name: "g", scope: !2, file: !3, line: 3, type: !6, isLocal: false, isDefinition: true)
+!2 = distinct !DICompileUnit(language: DW_LANG_C99, file: !3, producer: "clang version 6.0.0", isOptimized: false, runtimeVersion: 0, emissionKind: FullDebug, enums: !4, globals: !5)
+!3 = !DIFile(filename: "foo.c", directory: "/bar")
+!4 = !{}
+!5 = !{!0}
+!6 = !DICompositeType(tag: DW_TAG_array_type, baseType: !7, size: 8, elements: !8)
+!7 = !DIBasicType(name: "char", size: 8, encoding: DW_ATE_signed_char)
+!8 = !{!9}
+!9 = !DISubrange(count: 1)
+!10 = !{i32 2, !"Dwarf Version", i32 4}
+!11 = !{i32 2, !"Debug Info Version", i32 3}
+!12 = !{i32 1, !"wchar_size", i32 4}
+!13 = !{!"clang version 6.0.0"}
+!14 = distinct !DISubprogram(name: "foo", scope: !3, file: !3, line: 5, type: !15, isLocal: false, isDefinition: true, scopeLine: 6, isOptimized: false, unit: !2, variables: !4)
+!15 = !DISubroutineType(types: !16)
+!16 = !{null}
+!17 = !DILocalVariable(name: "i", scope: !14, file: !3, line: 7, type: !7)
+!18 = !DILocation(line: 7, column: 10, scope: !14)
+!19 = !DILocation(line: 8, column: 7, scope: !14)
+!20 = !DILocation(line: 9, column: 5, scope: !14)
+!21 = !DILocation(line: 10, column: 1, scope: !14)
+
+; Check that the store is removed and that the memcpy is still there
+; CHECK-LABEL: foo
+; CHECK-NOT:   store i8
+; CHECK:       call void @llvm.memcpy
+; CHECK:       ret void