1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
|
; RUN: opt %loadPolly -polly-codegen -polly-parallel -S < %s | FileCheck %s --check-prefix=AUTO
; RUN: opt %loadPolly -polly-codegen -polly-parallel -polly-num-threads=1 -S < %s | FileCheck %s --check-prefix=ONE
; RUN: opt %loadPolly -polly-codegen -polly-parallel -polly-num-threads=4 -S < %s | FileCheck %s --check-prefix=FOUR
; RUN: opt %loadPolly -polly-codegen -polly-parallel -polly-omp-backend=LLVM -S < %s | FileCheck %s --check-prefix=LIBOMP-AUTO
; RUN: opt %loadPolly -polly-codegen -polly-parallel -polly-omp-backend=LLVM -polly-num-threads=1 -S < %s | FileCheck %s --check-prefix=LIBOMP-ONE
; RUN: opt %loadPolly -polly-codegen -polly-parallel -polly-omp-backend=LLVM -polly-num-threads=4 -S < %s | FileCheck %s --check-prefix=LIBOMP-FOUR
; Ensure that the provided thread numbers are forwarded to the OpenMP calls.
;
; void storePosition(int *A) {
; for (int i = 0; i < 1024; i++)
; for (int j = 0; j < 1024; j++)
; A[i + j * 1024] = 0;
; }
; AUTO: call void @GOMP_parallel_loop_runtime_start(ptr @storePosition_polly_subfn, ptr %polly.par.userContext{{[0-9]*}}, i32 0, i64 0, i64 1024, i64 1)
; ONE: call void @GOMP_parallel_loop_runtime_start(ptr @storePosition_polly_subfn, ptr %polly.par.userContext{{[0-9]*}}, i32 1, i64 0, i64 1024, i64 1)
; FOUR: call void @GOMP_parallel_loop_runtime_start(ptr @storePosition_polly_subfn, ptr %polly.par.userContext{{[0-9]*}}, i32 4, i64 0, i64 1024, i64 1)
; In automatic mode, no threads are pushed explicitly.
; LIBOMP-AUTO-NOT: call void @__kmpc_push_num_threads
; LIBOMP-ONE: call void @__kmpc_push_num_threads(ptr @.loc.dummy{{[.0-9]*}}, i32 %{{[0-9]+}}, i32 1)
; LIBOMP-FOUR: call void @__kmpc_push_num_threads(ptr @.loc.dummy{{[.0-9]*}}, i32 %{{[0-9]+}}, i32 4)
target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
define void @storePosition(ptr %A) {
entry:
br label %for.cond
for.cond: ; preds = %for.inc4, %entry
%indvars.iv3 = phi i64 [ %indvars.iv.next4, %for.inc4 ], [ 0, %entry ]
%exitcond5 = icmp ne i64 %indvars.iv3, 1024
br i1 %exitcond5, label %for.body, label %for.end6
for.body: ; preds = %for.cond
br label %for.cond1
for.cond1: ; preds = %for.inc, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.inc ], [ 0, %for.body ]
%exitcond = icmp ne i64 %indvars.iv, 1024
br i1 %exitcond, label %for.body3, label %for.end
for.body3: ; preds = %for.cond1
%tmp = shl nsw i64 %indvars.iv, 10
%tmp6 = add nsw i64 %indvars.iv3, %tmp
%arrayidx = getelementptr inbounds i32, ptr %A, i64 %tmp6
store i32 0, ptr %arrayidx, align 4
br label %for.inc
for.inc: ; preds = %for.body3
%indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
br label %for.cond1
for.end: ; preds = %for.cond1
br label %for.inc4
for.inc4: ; preds = %for.end
%indvars.iv.next4 = add nuw nsw i64 %indvars.iv3, 1
br label %for.cond
for.end6: ; preds = %for.cond
ret void
}
|
