diff options
| author | Nadav Rotem <nrotem@apple.com> | 2012-10-21 06:49:10 +0000 |
|---|---|---|
| committer | Nadav Rotem <nrotem@apple.com> | 2012-10-21 06:49:10 +0000 |
| commit | bb950854acbb5966875763eaae7ab58e48e4f5a9 (patch) | |
| tree | 09bf75778dbc3d93744730114b6a769b0e3f0d6f /test/Transforms | |
| parent | c84787262942a400a8de6539ee79f4de72d3f872 (diff) | |
| download | llvm-bb950854acbb5966875763eaae7ab58e48e4f5a9.tar.gz llvm-bb950854acbb5966875763eaae7ab58e48e4f5a9.tar.bz2 llvm-bb950854acbb5966875763eaae7ab58e48e4f5a9.tar.xz | |
Fix a bug in the vectorization of wide load/store operations.
We used a SCEV to detect that A[X] is consecutive. We assumed that X was
the induction variable. But X can be any expression that uses the induction
for example: X = i + 2;
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@166388 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'test/Transforms')
| -rw-r--r-- | test/Transforms/LoopVectorize/induction_plus.ll | 30 |
1 files changed, 30 insertions, 0 deletions
diff --git a/test/Transforms/LoopVectorize/induction_plus.ll b/test/Transforms/LoopVectorize/induction_plus.ll new file mode 100644 index 0000000000..bd90113e52 --- /dev/null +++ b/test/Transforms/LoopVectorize/induction_plus.ll @@ -0,0 +1,30 @@ +; RUN: opt < %s -loop-vectorize -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-apple-macosx10.8.0" + +@array = common global [1024 x i32] zeroinitializer, align 16 + +;CHECK: @array_at_plus_one +;CHECK: add <4 x i64> +;CHECK: trunc <4 x i64> +;CHECK: add i64 %index, 12 +;CHECK: ret i32 +define i32 @array_at_plus_one(i32 %n) nounwind uwtable ssp { + %1 = icmp sgt i32 %n, 0 + br i1 %1, label %.lr.ph, label %._crit_edge + +.lr.ph: ; preds = %0, %.lr.ph + %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 0, %0 ] + %2 = add nsw i64 %indvars.iv, 12 + %3 = getelementptr inbounds [1024 x i32]* @array, i64 0, i64 %2 + %4 = trunc i64 %indvars.iv to i32 + store i32 %4, i32* %3, align 4 + %indvars.iv.next = add i64 %indvars.iv, 1 + %lftr.wideiv = trunc i64 %indvars.iv.next to i32 + %exitcond = icmp eq i32 %lftr.wideiv, %n + br i1 %exitcond, label %._crit_edge, label %.lr.ph + +._crit_edge: ; preds = %.lr.ph, %0 + ret i32 undef +} |