1 ; RUN: opt -S %loadPolly -basicaa -polly-opt-isl \
2 ; RUN: -polly-pattern-matching-based-opts=false -polly-vectorizer=polly \
3 ; RUN: -polly-ast -analyze < %s | FileCheck %s
4 ; RUN: opt -S %loadPolly -basicaa -polly-opt-isl \
5 ; RUN: -polly-pattern-matching-based-opts=false -polly-vectorizer=stripmine \
6 ; RUN: -polly-ast -analyze < %s | FileCheck %s
8 ; RUN: opt -S %loadPolly -basicaa -polly-opt-isl \
9 ; RUN: -polly-vectorizer=polly -polly-pattern-matching-based-opts=false \
10 ; RUN: -polly-ast -analyze -polly-prevect-width=16 < %s | \
11 ; RUN: FileCheck %s -check-prefix=VEC16
13 target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
15 @C = common global [1536 x [1536 x float]] zeroinitializer, align 16
16 @A = common global [1536 x [1536 x float]] zeroinitializer, align 16
17 @B = common global [1536 x [1536 x float]] zeroinitializer, align 16
19 ; Function Attrs: nounwind uwtable
20 define void @foo() #0 {
24 entry.split: ; preds = %entry
25 br label %for.cond1.preheader
27 for.cond1.preheader: ; preds = %entry.split, %for.inc28
28 %indvar4 = phi i64 [ 0, %entry.split ], [ %indvar.next5, %for.inc28 ]
31 for.body3: ; preds = %for.cond1.preheader, %for.inc25
32 %indvar6 = phi i64 [ 0, %for.cond1.preheader ], [ %indvar.next7, %for.inc25 ]
33 %arrayidx24 = getelementptr [1536 x [1536 x float]], [1536 x [1536 x float]]* @C, i64 0, i64 %indvar4, i64 %indvar6
34 store float 0.000000e+00, float* %arrayidx24, align 4
37 for.body8: ; preds = %for.body3, %for.body8
38 %indvar = phi i64 [ 0, %for.body3 ], [ %indvar.next, %for.body8 ]
39 %arrayidx16 = getelementptr [1536 x [1536 x float]], [1536 x [1536 x float]]* @A, i64 0, i64 %indvar4, i64 %indvar
40 %arrayidx20 = getelementptr [1536 x [1536 x float]], [1536 x [1536 x float]]* @B, i64 0, i64 %indvar, i64 %indvar6
41 %0 = load float, float* %arrayidx24, align 4
42 %1 = load float, float* %arrayidx16, align 4
43 %2 = load float, float* %arrayidx20, align 4
44 %mul = fmul float %1, %2
45 %add = fadd float %0, %mul
46 store float %add, float* %arrayidx24, align 4
47 %indvar.next = add i64 %indvar, 1
48 %exitcond = icmp ne i64 %indvar.next, 1536
49 br i1 %exitcond, label %for.body8, label %for.inc25
51 for.inc25: ; preds = %for.body8
52 %indvar.next7 = add i64 %indvar6, 1
53 %exitcond8 = icmp ne i64 %indvar.next7, 1536
54 br i1 %exitcond8, label %for.body3, label %for.inc28
56 for.inc28: ; preds = %for.inc25
57 %indvar.next5 = add i64 %indvar4, 1
58 %exitcond9 = icmp ne i64 %indvar.next5, 1536
59 br i1 %exitcond9, label %for.cond1.preheader, label %for.end30
61 for.end30: ; preds = %for.inc28
65 attributes #0 = { nounwind uwtable "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "unsafe-fp-math"="false" "use-soft-float"="false" }
67 ; CHECK: #pragma known-parallel
68 ; CHECK: for (int c0 = 0; c0 <= 47; c0 += 1)
69 ; CHECK: for (int c1 = 0; c1 <= 47; c1 += 1)
70 ; CHECK: for (int c2 = 0; c2 <= 31; c2 += 1)
71 ; CHECK: for (int c3 = 0; c3 <= 7; c3 += 1)
73 ; CHECK: for (int c4 = 0; c4 <= 3; c4 += 1)
74 ; CHECK: Stmt_for_body3(32 * c0 + c2, 32 * c1 + 4 * c3 + c4);
75 ; CHECK: #pragma known-parallel
76 ; CHECK: for (int c0 = 0; c0 <= 47; c0 += 1)
77 ; CHECK: for (int c1 = 0; c1 <= 47; c1 += 1)
78 ; CHECK: for (int c2 = 0; c2 <= 47; c2 += 1)
79 ; CHECK: for (int c3 = 0; c3 <= 31; c3 += 1)
80 ; CHECK: for (int c4 = 0; c4 <= 7; c4 += 1)
81 ; CHECK: for (int c5 = 0; c5 <= 31; c5 += 1)
83 ; CHECK: for (int c6 = 0; c6 <= 3; c6 += 1)
84 ; CHECK: Stmt_for_body8(32 * c0 + c3, 32 * c1 + 4 * c4 + c6, 32 * c2 + c5);
87 ; VEC16: #pragma known-parallel
88 ; VEC16: for (int c0 = 0; c0 <= 47; c0 += 1)
89 ; VEC16: for (int c1 = 0; c1 <= 47; c1 += 1)
90 ; VEC16: for (int c2 = 0; c2 <= 31; c2 += 1)
91 ; VEC16: for (int c3 = 0; c3 <= 1; c3 += 1)
93 ; VEC16: for (int c4 = 0; c4 <= 15; c4 += 1)
94 ; VEC16: Stmt_for_body3(32 * c0 + c2, 32 * c1 + 16 * c3 + c4);
95 ; VEC16: #pragma known-parallel
96 ; VEC16: for (int c0 = 0; c0 <= 47; c0 += 1)
97 ; VEC16: for (int c1 = 0; c1 <= 47; c1 += 1)
98 ; VEC16: for (int c2 = 0; c2 <= 47; c2 += 1)
99 ; VEC16: for (int c3 = 0; c3 <= 31; c3 += 1)
100 ; VEC16: for (int c4 = 0; c4 <= 1; c4 += 1)
101 ; VEC16: for (int c5 = 0; c5 <= 31; c5 += 1)
103 ; VEC16: for (int c6 = 0; c6 <= 15; c6 += 1)
104 ; VEC16: Stmt_for_body8(32 * c0 + c3, 32 * c1 + 16 * c4 + c6, 32 * c2 + c5);
110 !0 = !{!"clang version 3.5.0 "}