Commit 1e21181a authored by Matt Arsenault's avatar Matt Arsenault
Browse files

LoopDistribute/LAA: Add tests to catch regressions 

I broke 2 of these with a patch, but were not covered by existing
tests.

https://reviews.llvm.org/D63035

llvm-svn: 363158
parent 52d7326f

llvm/test/Transforms/LoopDistribute/basic-with-memchecks.ll

+65 −0
Original line number Diff line number Diff line
@@ -23,6 +23,7 @@ target triple = "x86_64-apple-macosx10.10.0" 
@D = common global i32* null, align 8

@E = common global i32* null, align 8



; CHECK-LABEL: @f(

define void @f() {

entry:

  %a = load i32*, i32** @A, align 8
@@ -108,3 +109,67 @@ for.body: ; preds = %for.body, %entry 
for.end:                                          ; preds = %for.body

  ret void

}



; Make sure there's no "Multiple reports generated" assert with a

; volatile load, and no distribution



; TODO: Distribution of volatile may be possible under some

; circumstance, but the current implementation does not touch them.



; CHECK-LABEL: @f_volatile_load(

; CHECK: br label %for.body{{$}}



; CHECK-NOT: load



; CHECK: {{^}}for.body:

; CHECK: load i32

; CHECK: load i32

; CHECK: load volatile i32

; CHECK: load i32

; CHECK: br i1 %exitcond, label %for.end, label %for.body{{$}}



; CHECK-NOT: load



; VECTORIZE-NOT: load <4 x i32>

; VECTORIZE-NOT: mul <4 x i32>

define void @f_volatile_load() {

entry:

  %a = load i32*, i32** @A, align 8

  %b = load i32*, i32** @B, align 8

  %c = load i32*, i32** @C, align 8

  %d = load i32*, i32** @D, align 8

  %e = load i32*, i32** @E, align 8

  br label %for.body



for.body:

  %ind = phi i64 [ 0, %entry ], [ %add, %for.body ]



  %arrayidxA = getelementptr inbounds i32, i32* %a, i64 %ind

  %loadA = load i32, i32* %arrayidxA, align 4



  %arrayidxB = getelementptr inbounds i32, i32* %b, i64 %ind

  %loadB = load i32, i32* %arrayidxB, align 4



  %mulA = mul i32 %loadB, %loadA



  %add = add nuw nsw i64 %ind, 1

  %arrayidxA_plus_4 = getelementptr inbounds i32, i32* %a, i64 %add

  store i32 %mulA, i32* %arrayidxA_plus_4, align 4



  %arrayidxD = getelementptr inbounds i32, i32* %d, i64 %ind

  %loadD = load volatile i32, i32* %arrayidxD, align 4



  %arrayidxE = getelementptr inbounds i32, i32* %e, i64 %ind

  %loadE = load i32, i32* %arrayidxE, align 4



  %mulC = mul i32 %loadD, %loadE



  %arrayidxC = getelementptr inbounds i32, i32* %c, i64 %ind

  store i32 %mulC, i32* %arrayidxC, align 4



  %exitcond = icmp eq i64 %add, 20

  br i1 %exitcond, label %for.end, label %for.body



for.end:

  ret void

}

llvm/test/Transforms/LoopVectorize/read-only.ll

+30 −0
Original line number Diff line number Diff line
@@ -29,3 +29,33 @@ define i32 @read_only_func(i32* nocapture %A, i32* nocapture %B, i32 %n) nounwin 
  %sum.0.lcssa = phi i32 [ 0, %0 ], [ %9, %.lr.ph ]

  ret i32 %sum.0.lcssa

}



; Ensure that volatile loads are not vectorized in a read-only loop.



;CHECK-LABEL: @read_only_func_volatile(

;CHECK-NOT: load <4 x i32>

;CHECK: ret i32

define i32 @read_only_func_volatile(i32* nocapture %A, i32* nocapture %B, i32 %n) nounwind uwtable readonly 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 ]

  %sum.02 = phi i32 [ %9, %.lr.ph ], [ 0, %0 ]

  %2 = getelementptr inbounds i32, i32* %A, i64 %indvars.iv

  %3 = load volatile i32, i32* %2, align 4

  %4 = add nsw i64 %indvars.iv, 13

  %5 = getelementptr inbounds i32, i32* %B, i64 %4

  %6 = load i32, i32* %5, align 4

  %7 = shl i32 %6, 1

  %8 = add i32 %3, %sum.02

  %9 = add i32 %8, %7

  %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

  %sum.0.lcssa = phi i32 [ 0, %0 ], [ %9, %.lr.ph ]

  ret i32 %sum.0.lcssa

}

llvm/test/Transforms/LoopVectorize/write-only.ll

+23 −0
Original line number Diff line number Diff line
@@ -23,3 +23,26 @@ define i32 @read_mod_write_single_ptr(float* nocapture %a, i32 %n) nounwind uwta 
._crit_edge:                                      ; preds = %.lr.ph, %0

  ret i32 undef

}



; Ensure that volatile stores are not vectorized.

; CHECK-LABEL: @read_mod_write_single_ptr_volatile_store(

; CHECK-NOT: store <4 x float>

; CHECK: ret i32

define i32 @read_mod_write_single_ptr_volatile_store(float* nocapture %a, 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 = getelementptr inbounds float, float* %a, i64 %indvars.iv

  %3 = load float, float* %2, align 4

  %4 = fmul float %3, 3.000000e+00

  store volatile float %4, float* %2, 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

}