Merging r183035:

------------------------------------------------------------------------
r183035 | arnolds | 2013-05-31 12:53:50 -0700 (Fri, 31 May 2013) | 7 lines

LoopVectorize: PHIs with only outside users should prevent vectorization

We check that instructions in the loop don't have outside users (except if
they are reduction values). Unfortunately, we skipped this check for
if-convertable PHIs.

Fixes PR16184.
------------------------------------------------------------------------

llvm-svn: 183189

2 files changed, 71 insertions, 13 deletions

diff --git a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
index bc420bd..08d3725 100644
--- a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
+++ b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
@@ -2378,6 +2378,26 @@ bool LoopVectorizationLegality::canVectorize() {
   return true;
 }
 
+/// \brief Check that the instruction has outside loop users and is not an
+/// identified reduction variable.
+static bool hasOutsideLoopUser(const Loop *TheLoop, Instruction *Inst,
+                               SmallPtrSet<Value *, 4> &Reductions) {
+  // Reduction instructions are allowed to have exit users. All other
+  // instructions must not have external users.
+  if (!Reductions.count(Inst))
+    //Check that all of the users of the loop are inside the BB.
+    for (Value::use_iterator I = Inst->use_begin(), E = Inst->use_end();
+         I != E; ++I) {
+      Instruction *U = cast<Instruction>(*I);
+      // This user may be a reduction exit value.
+      if (!TheLoop->contains(U)) {
+        DEBUG(dbgs() << "LV: Found an outside user for : "<< *U << "\n");
+        return true;
+      }
+    }
+  return false;
+}
+
 bool LoopVectorizationLegality::canVectorizeInstrs() {
   BasicBlock *PreHeader = TheLoop->getLoopPreheader();
   BasicBlock *Header = TheLoop->getHeader();
@@ -2416,8 +2436,13 @@ bool LoopVectorizationLegality::canVectorizeInstrs() {
         // If this PHINode is not in the header block, then we know that we
         // can convert it to select during if-conversion. No need to check if
         // the PHIs in this block are induction or reduction variables.
-        if (*bb != Header)
-          continue;
+        if (*bb != Header) {
+          // Check that this instruction has no outside users or is an
+          // identified reduction value with an outside user.
+          if(!hasOutsideLoopUser(TheLoop, it, AllowedExit))
+            continue;
+          return false;
+        }
 
         // We only allow if-converted PHIs with more than two incoming values.
         if (Phi->getNumIncomingValues() != 2) {
@@ -2510,17 +2535,9 @@ bool LoopVectorizationLegality::canVectorizeInstrs() {
 
       // Reduction instructions are allowed to have exit users.
       // All other instructions must not have external users.
-      if (!AllowedExit.count(it))
-        //Check that all of the users of the loop are inside the BB.
-        for (Value::use_iterator I = it->use_begin(), E = it->use_end();
-             I != E; ++I) {
-          Instruction *U = cast<Instruction>(*I);
-          // This user may be a reduction exit value.
-          if (!TheLoop->contains(U)) {
-            DEBUG(dbgs() << "LV: Found an outside user for : "<< *U << "\n");
-            return false;
-          }
-        }
+      if (hasOutsideLoopUser(TheLoop, it, AllowedExit))
+        return false;
+
     } // next instr.
 
   }
diff --git a/llvm/test/Transforms/LoopVectorize/no_outside_user.ll b/llvm/test/Transforms/LoopVectorize/no_outside_user.ll
new file mode 100644
index 0000000..6f0357c
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/no_outside_user.ll
@@ -0,0 +1,41 @@
+; RUN: opt -S -loop-vectorize -force-vector-unroll=1 -force-vector-width=2 < %s | FileCheck %s
+
+target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:32:32-n8:16:32-S128"
+
+@f = common global i32 0, align 4
+@.str = private unnamed_addr constant [4 x i8] c"%d\0A\00", align 1
+@c = common global i32 0, align 4
+@a = common global i32 0, align 4
+@b = common global i32 0, align 4
+@e = common global i32 0, align 4
+
+; We used to vectorize this loop. But it has a value that is used outside of the
+; and is not a recognized reduction variable "tmp17".
+
+; CHECK-NOT: <2 x i32>
+
+define i32 @main()  {
+bb:
+  %b.promoted = load i32* @b, align 4
+  br label %.lr.ph.i
+
+.lr.ph.i:
+  %tmp8 = phi i32 [ %tmp18, %bb16 ], [ %b.promoted, %bb ]
+  %tmp2 = icmp sgt i32 %tmp8, 10
+  br i1 %tmp2, label %bb16, label %bb10
+
+bb10:
+  br label %bb16
+
+bb16:
+  %tmp17 = phi i32 [ 0, %bb10 ], [ 1, %.lr.ph.i ]
+  %tmp18 = add nsw i32 %tmp8, 1
+  %tmp19 = icmp slt i32 %tmp18, 4
+  br i1 %tmp19, label %.lr.ph.i, label %f1.exit.loopexit
+
+f1.exit.loopexit:
+  %.lcssa = phi i32 [ %tmp17, %bb16 ]
+  ret i32 %.lcssa
+}
+
+