Refuse to (even try to) vectorize loops which have uniform writes,

even if erroneously annotated with the parallel loop metadata.

Fixes Bug 15794: 
"Loop Vectorizer: Crashes with the use of llvm.loop.parallel metadata"



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@180081 91177308-0d34-0410-b5e6-96231b3b80d8

2 files changed, 67 insertions, 9 deletions

diff --git a/lib/Transforms/Vectorize/LoopVectorize.cpp b/lib/Transforms/Vectorize/LoopVectorize.cpp
index 0c88ba7835..2e42585e35 100644
--- a/lib/Transforms/Vectorize/LoopVectorize.cpp
+++ b/lib/Transforms/Vectorize/LoopVectorize.cpp
@@ -2548,13 +2548,6 @@ LoopVectorizationLegality::hasPossibleGlobalWriteReorder(
 
 bool LoopVectorizationLegality::canVectorizeMemory() {
 
-  if (TheLoop->isAnnotatedParallel()) {
-    DEBUG(dbgs()
-          << "LV: A loop annotated parallel, ignore memory dependency "
-          << "checks.\n");
-    return true;
-  }
-
   typedef SmallVector<Value*, 16> ValueVector;
   typedef SmallPtrSet<Value*, 16> ValueSet;
   // Holds the Load and Store *instructions*.
@@ -2577,7 +2570,7 @@ bool LoopVectorizationLegality::canVectorizeMemory() {
       if (it->mayReadFromMemory()) {
         LoadInst *Ld = dyn_cast<LoadInst>(it);
         if (!Ld) return false;
-        if (!Ld->isSimple()) {
+        if (!Ld->isSimple() && !TheLoop->isAnnotatedParallel()) {
           DEBUG(dbgs() << "LV: Found a non-simple load.\n");
           return false;
         }
@@ -2589,7 +2582,7 @@ bool LoopVectorizationLegality::canVectorizeMemory() {
       if (it->mayWriteToMemory()) {
         StoreInst *St = dyn_cast<StoreInst>(it);
         if (!St) return false;
-        if (!St->isSimple()) {
+        if (!St->isSimple() && !TheLoop->isAnnotatedParallel()) {
           DEBUG(dbgs() << "LV: Found a non-simple store.\n");
           return false;
         }
@@ -2636,6 +2629,13 @@ bool LoopVectorizationLegality::canVectorizeMemory() {
       ReadWrites.insert(std::make_pair(Ptr, ST));
   }
 
+  if (TheLoop->isAnnotatedParallel()) {
+    DEBUG(dbgs()
+          << "LV: A loop annotated parallel, ignore memory dependency "
+          << "checks.\n");
+    return true;
+  }
+
   for (I = Loads.begin(), IE = Loads.end(); I != IE; ++I) {
     LoadInst *LD = cast<LoadInst>(*I);
     Value* Ptr = LD->getPointerOperand();
diff --git a/test/Transforms/LoopVectorize/X86/illegal-parallel-loop-uniform-write.ll b/test/Transforms/LoopVectorize/X86/illegal-parallel-loop-uniform-write.ll
new file mode 100644
index 0000000000..bc606f3ebd
--- /dev/null
+++ b/test/Transforms/LoopVectorize/X86/illegal-parallel-loop-uniform-write.ll
@@ -0,0 +1,58 @@
+; RUN: opt < %s  -loop-vectorize -force-vector-unroll=1 -force-vector-width=4 -dce -instcombine -S | FileCheck %s
+
+target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
+target triple = "x86_64-unknown-linux-gnu"
+
+;CHECK: @foo
+;CHECK-NOT: <4 x 
+
+; Function Attrs: nounwind uwtable 
+define void @foo(i32* nocapture %a, i32* nocapture %b, i32 %k, i32 %m) #0 {
+entry:
+  %cmp27 = icmp sgt i32 %m, 0
+  br i1 %cmp27, label %for.body3.lr.ph.us, label %for.end15
+
+for.end.us:                                       ; preds = %for.body3.us
+  %arrayidx9.us = getelementptr inbounds i32* %b, i64 %indvars.iv33
+  %0 = load i32* %arrayidx9.us, align 4, !tbaa !0, !llvm.mem.parallel_loop_access !3
+  %add10.us = add nsw i32 %0, 3
+  store i32 %add10.us, i32* %arrayidx9.us, align 4, !tbaa !0, !llvm.mem.parallel_loop_access !3
+  %indvars.iv.next34 = add i64 %indvars.iv33, 1
+  %lftr.wideiv35 = trunc i64 %indvars.iv.next34 to i32
+  %exitcond36 = icmp eq i32 %lftr.wideiv35, %m
+  br i1 %exitcond36, label %for.end15, label %for.body3.lr.ph.us, !llvm.loop.parallel !5
+
+for.body3.us:                                     ; preds = %for.body3.us, %for.body3.lr.ph.us
+  %indvars.iv29 = phi i64 [ 0, %for.body3.lr.ph.us ], [ %indvars.iv.next30, %for.body3.us ]
+  %1 = trunc i64 %indvars.iv29 to i32
+  %add4.us = add i32 %add.us, %1
+  %idxprom.us = sext i32 %add4.us to i64
+  %arrayidx.us = getelementptr inbounds i32* %a, i64 %idxprom.us
+  %2 = load i32* %arrayidx.us, align 4, !tbaa !0, !llvm.mem.parallel_loop_access !3
+  %add5.us = add nsw i32 %2, 1
+  store i32 %add5.us, i32* %arrayidx7.us, align 4, !tbaa !0, !llvm.mem.parallel_loop_access !3
+  %indvars.iv.next30 = add i64 %indvars.iv29, 1
+  %lftr.wideiv31 = trunc i64 %indvars.iv.next30 to i32
+  %exitcond32 = icmp eq i32 %lftr.wideiv31, %m
+  br i1 %exitcond32, label %for.end.us, label %for.body3.us, !llvm.loop.parallel !4
+
+for.body3.lr.ph.us:                               ; preds = %for.end.us, %entry
+  %indvars.iv33 = phi i64 [ %indvars.iv.next34, %for.end.us ], [ 0, %entry ]
+  %3 = trunc i64 %indvars.iv33 to i32
+  %add.us = add i32 %3, %k
+  %arrayidx7.us = getelementptr inbounds i32* %a, i64 %indvars.iv33
+  br label %for.body3.us
+
+for.end15:                                        ; preds = %for.end.us, %entry
+  ret void
+}
+
+attributes #0 = { nounwind uwtable "less-precise-fpmad"="false" "no-frame-pointer-elim"="false" "no-frame-pointer-elim-non-leaf"="false" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "unsafe-fp-math"="false" "use-soft-float"="false" }
+
+!0 = metadata !{metadata !"int", metadata !1}
+!1 = metadata !{metadata !"omnipotent char", metadata !2}
+!2 = metadata !{metadata !"Simple C/C++ TBAA"}
+!3 = metadata !{metadata !4, metadata !5}
+!4 = metadata !{metadata !4}
+!5 = metadata !{metadata !5}
+