[llvm] r193434 - LoopVectorizer: Don't attempt to vectorize extractelement instructions

[Hal Finkel]

Author: hfinkel
Date: Fri Oct 25 15:40:15 2013
New Revision: 193434

URL: http://llvm.org/viewvc/llvm-project?rev=193434&view=rev
Log:
LoopVectorizer: Don't attempt to vectorize extractelement instructions

The loop vectorizer does not currently understand how to vectorize
extractelement instructions. The existing check, which excluded all
vector-valued instructions, did not catch extractelement instructions because
it checked only the return value. As a result, vectorization would proceed,
producing illegal instructions like this:

  %58 = extractelement <2 x i32> %15, i32 0
  %59 = extractelement i32 %58, i32 0

where the second extractelement is illegal because its first operand is not a vector.

Added:
    llvm/trunk/test/Transforms/LoopVectorize/ee-crash.ll
Modified:
    llvm/trunk/lib/Transforms/Vectorize/LoopVectorize.cpp

Modified: llvm/trunk/lib/Transforms/Vectorize/LoopVectorize.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Transforms/Vectorize/LoopVectorize.cpp?rev=193434&r1=193433&r2=193434&view=diff
==============================================================================
--- llvm/trunk/lib/Transforms/Vectorize/LoopVectorize.cpp (original)
+++ llvm/trunk/lib/Transforms/Vectorize/LoopVectorize.cpp Fri Oct 25 15:40:15 2013
@@ -2965,8 +2965,9 @@ bool LoopVectorizationLegality::canVecto
       }
 
       // Check that the instruction return type is vectorizable.
-      if (!VectorType::isValidElementType(it->getType()) &&
-          !it->getType()->isVoidTy()) {
+      // Also, we can't vectorize extractelement instructions.
+      if ((!VectorType::isValidElementType(it->getType()) &&
+           !it->getType()->isVoidTy()) || isa<ExtractElementInst>(it)) {
         DEBUG(dbgs() << "LV: Found unvectorizable type.\n");
         return false;
       }

Added: llvm/trunk/test/Transforms/LoopVectorize/ee-crash.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/Transforms/LoopVectorize/ee-crash.ll?rev=193434&view=auto
==============================================================================
--- llvm/trunk/test/Transforms/LoopVectorize/ee-crash.ll (added)
+++ llvm/trunk/test/Transforms/LoopVectorize/ee-crash.ll Fri Oct 25 15:40:15 2013
@@ -0,0 +1,35 @@
+; RUN: opt < %s -loop-vectorize -force-vector-unroll=1 -force-vector-width=4 -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-apple-macosx10.8.0"
+
+; This test checks that we deal with an in-loop extractelement (for now, this
+; means not crashing by not vectorizing).
+; CHECK-LABEL: @_Z4foo1Pii(
+; CHECK-NOT: <4 x i32>
+; CHECK: ret
+define i32 @_Z4foo1Pii(i32* %A, i32 %n, <2 x i32> %q) #0 {
+entry:
+  %idx.ext = sext i32 %n to i64
+  %add.ptr = getelementptr inbounds i32* %A, i64 %idx.ext
+  %cmp3.i = icmp eq i32 %n, 0
+  br i1 %cmp3.i, label %_ZSt10accumulateIPiiET0_T_S2_S1_.exit, label %for.body.i
+
+for.body.i:                                       ; preds = %entry, %for.body.i
+  %__init.addr.05.i = phi i32 [ %add.i, %for.body.i ], [ 0, %entry ]
+  %__first.addr.04.i = phi i32* [ %incdec.ptr.i, %for.body.i ], [ %A, %entry ]
+  %0 = load i32* %__first.addr.04.i, align 4
+  %q1 = extractelement <2 x i32> %q, i32 %n
+  %q2 = add nsw i32 %0, %q1
+  %add.i = add nsw i32 %q2, %__init.addr.05.i
+  %incdec.ptr.i = getelementptr inbounds i32* %__first.addr.04.i, i64 1
+  %cmp.i = icmp eq i32* %incdec.ptr.i, %add.ptr
+  br i1 %cmp.i, label %_ZSt10accumulateIPiiET0_T_S2_S1_.exit, label %for.body.i
+
+_ZSt10accumulateIPiiET0_T_S2_S1_.exit:            ; preds = %for.body.i, %entry
+  %__init.addr.0.lcssa.i = phi i32 [ 0, %entry ], [ %add.i, %for.body.i ]
+  ret i32 %__init.addr.0.lcssa.i
+}
+
+attributes #0 = { nounwind readonly ssp uwtable }
+