File: loop_dispatch.c

package info (click to toggle)
swiftlang 6.0.3-2
  • links: PTS, VCS
  • area: main
  • in suites: forky, sid, trixie
  • size: 2,519,992 kB
  • sloc: cpp: 9,107,863; ansic: 2,040,022; asm: 1,135,751; python: 296,500; objc: 82,456; f90: 60,502; lisp: 34,951; pascal: 19,946; sh: 18,133; perl: 7,482; ml: 4,937; javascript: 4,117; makefile: 3,840; awk: 3,535; xml: 914; fortran: 619; cs: 573; ruby: 573
file content (127 lines) | stat: -rw-r--r-- 5,788 bytes parent folder | download | duplicates (6)
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
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
// RUN: %libomp-compile-and-run | %sort-threads | FileCheck %s
// REQUIRES: ompt
// XFAIL: gcc
// GCC doesn't call runtime for static schedule

#include "callback.h"

#define WORK_SIZE 64

int main() {
  int i;
  int wait_s = 0;

#pragma omp parallel num_threads(4)
  {
    int wait_id = 0;
    int team_size = omp_get_num_threads();
#pragma omp for schedule(static, WORK_SIZE / 4)
    for (i = 0; i < WORK_SIZE; i++) {}

#pragma omp for schedule(dynamic)
    for (i = 0; i < WORK_SIZE; i++) {
      if (wait_id == 0) {
        // Wait until every thread has at least one iteration assigned
        OMPT_SIGNAL(wait_s);
        OMPT_WAIT(wait_s, team_size);
        wait_id++;
      }
    }

#pragma omp for schedule(guided)
    for (i = 0; i < WORK_SIZE; i++) {
      if (wait_id == 1) {
        // Wait until every thread has at least one iteration assigned
        OMPT_SIGNAL(wait_s);
        OMPT_WAIT(wait_s, 2 * team_size);
        wait_id++;
      }
    }
  }

  return 0;
}

// CHECK-NOT: {{^}}0: Could not register callback 'ompt_callback_parallel_begin'
// CHECK-NOT: {{^}}0: Could not register callback 'ompt_callback_implicit_task'
// CHECK-NOT: {{^}}0: Could not register callback 'ompt_callback_work'
// CHECK-NOT: {{^}}0: Could not register callback 'ompt_callback_dispatch'

// CHECK: 0: NULL_POINTER=[[NULL:.*$]]
// CHECK: {{^}}[[THREAD_ID0:[0-9]+]]: ompt_event_parallel_begin:
// CHECK-SAME: parallel_id=[[PARALLEL_ID:[0-9]+]]

// Each thread should have at least one ws-loop-chunk-begin event for each
// for loop.

// CHECK: {{^}}[[THREAD_ID0]]: ompt_event_implicit_task_begin:
// CHECK-SAME: task_id=[[TASK_ID0:[0-9]+]]
// CHECK: {{^}}[[THREAD_ID0]]: ompt_event_loop_begin:
// CHECK-SAME: parallel_id=[[PARALLEL_ID]], parent_task_id=[[TASK_ID0]]
// CHECK: {{^}}[[THREAD_ID0]]: ompt_event_ws_loop_chunk_begin:
// CHECK-SAME: parallel_id=[[PARALLEL_ID]], task_id=[[TASK_ID0]]
// CHECK-SAME: chunk_start={{[0-9]+}}, chunk_iterations=16
// CHECK: {{^}}[[THREAD_ID0]]: ompt_event_loop_begin:
// CHECK-SAME: parallel_id=[[PARALLEL_ID]], parent_task_id=[[TASK_ID0]]
// CHECK: {{^}}[[THREAD_ID0]]: ompt_event_ws_loop_chunk_begin:
// CHECK-SAME: parallel_id=[[PARALLEL_ID]], task_id=[[TASK_ID0]]
// CHECK-SAME: chunk_start={{[0-9]+}}, chunk_iterations=1
// CHECK: {{^}}[[THREAD_ID0]]: ompt_event_loop_begin:
// CHECK-SAME: parallel_id=[[PARALLEL_ID]], parent_task_id=[[TASK_ID0]]
// CHECK: {{^}}[[THREAD_ID0]]: ompt_event_ws_loop_chunk_begin:
// CHECK-SAME: parallel_id=[[PARALLEL_ID]], task_id=[[TASK_ID0]]
// CHECK-SAME: chunk_start={{[0-9]+}}, chunk_iterations={{[1-9][0-9]*}}

// CHECK: {{^}}[[THREAD_ID1:[0-9]+]]: ompt_event_implicit_task_begin:
// CHECK-SAME: task_id=[[TASK_ID1:[0-9]+]]
// CHECK: {{^}}[[THREAD_ID1]]: ompt_event_loop_begin:
// CHECK-SAME: parallel_id=[[PARALLEL_ID]], parent_task_id=[[TASK_ID1]]
// CHECK: {{^}}[[THREAD_ID1]]: ompt_event_ws_loop_chunk_begin:
// CHECK-SAME: parallel_id=[[PARALLEL_ID]], task_id=[[TASK_ID1]]
// CHECK-SAME: chunk_start={{[0-9]+}}, chunk_iterations=16
// CHECK: {{^}}[[THREAD_ID1]]: ompt_event_loop_begin:
// CHECK-SAME: parallel_id=[[PARALLEL_ID]], parent_task_id=[[TASK_ID1]]
// CHECK: {{^}}[[THREAD_ID1]]: ompt_event_ws_loop_chunk_begin:
// CHECK-SAME: parallel_id=[[PARALLEL_ID]], task_id=[[TASK_ID1]]
// CHECK-SAME: chunk_start={{[0-9]+}}, chunk_iterations=1
// CHECK: {{^}}[[THREAD_ID1]]: ompt_event_loop_begin:
// CHECK-SAME: parallel_id=[[PARALLEL_ID]], parent_task_id=[[TASK_ID1]]
// CHECK: {{^}}[[THREAD_ID1]]: ompt_event_ws_loop_chunk_begin:
// CHECK-SAME: parallel_id=[[PARALLEL_ID]], task_id=[[TASK_ID1]]
// CHECK-SAME: chunk_start={{[0-9]+}}, chunk_iterations={{[1-9][0-9]*}}

// CHECK: {{^}}[[THREAD_ID2:[0-9]+]]: ompt_event_implicit_task_begin:
// CHECK-SAME: task_id=[[TASK_ID2:[0-9]+]]
// CHECK: {{^}}[[THREAD_ID2]]: ompt_event_loop_begin:
// CHECK-SAME: parallel_id=[[PARALLEL_ID]], parent_task_id=[[TASK_ID2]]
// CHECK: {{^}}[[THREAD_ID2]]: ompt_event_ws_loop_chunk_begin:
// CHECK-SAME: parallel_id=[[PARALLEL_ID]], task_id=[[TASK_ID2]]
// CHECK-SAME: chunk_start={{[0-9]+}}, chunk_iterations=16
// CHECK: {{^}}[[THREAD_ID2]]: ompt_event_loop_begin:
// CHECK-SAME: parallel_id=[[PARALLEL_ID]], parent_task_id=[[TASK_ID2]]
// CHECK: {{^}}[[THREAD_ID2]]: ompt_event_ws_loop_chunk_begin:
// CHECK-SAME: parallel_id=[[PARALLEL_ID]], task_id=[[TASK_ID2]]
// CHECK-SAME: chunk_start={{[0-9]+}}, chunk_iterations=1
// CHECK: {{^}}[[THREAD_ID2]]: ompt_event_loop_begin:
// CHECK-SAME: parallel_id=[[PARALLEL_ID]], parent_task_id=[[TASK_ID2]]
// CHECK: {{^}}[[THREAD_ID2]]: ompt_event_ws_loop_chunk_begin:
// CHECK-SAME: parallel_id=[[PARALLEL_ID]], task_id=[[TASK_ID2]]
// CHECK-SAME: chunk_start={{[0-9]+}}, chunk_iterations={{[1-9][0-9]*}}

// CHECK: {{^}}[[THREAD_ID3:[0-9]+]]: ompt_event_implicit_task_begin:
// CHECK-SAME: task_id=[[TASK_ID3:[0-9]+]]
// CHECK: {{^}}[[THREAD_ID3]]: ompt_event_loop_begin:
// CHECK-SAME: parallel_id=[[PARALLEL_ID]], parent_task_id=[[TASK_ID3]]
// CHECK: {{^}}[[THREAD_ID3]]: ompt_event_ws_loop_chunk_begin:
// CHECK-SAME: parallel_id=[[PARALLEL_ID]], task_id=[[TASK_ID3]]
// CHECK-SAME: chunk_start={{[0-9]+}}, chunk_iterations=16
// CHECK: {{^}}[[THREAD_ID3]]: ompt_event_loop_begin:
// CHECK-SAME: parallel_id=[[PARALLEL_ID]], parent_task_id=[[TASK_ID3]]
// CHECK: {{^}}[[THREAD_ID3]]: ompt_event_ws_loop_chunk_begin:
// CHECK-SAME: parallel_id=[[PARALLEL_ID]], task_id=[[TASK_ID3]]
// CHECK-SAME: chunk_start={{[0-9]+}}, chunk_iterations=1
// CHECK: {{^}}[[THREAD_ID3]]: ompt_event_loop_begin:
// CHECK-SAME: parallel_id=[[PARALLEL_ID]], parent_task_id=[[TASK_ID3]]
// CHECK: {{^}}[[THREAD_ID3]]: ompt_event_ws_loop_chunk_begin:
// CHECK-SAME: parallel_id=[[PARALLEL_ID]], task_id=[[TASK_ID3]]
// CHECK-SAME: chunk_start={{[0-9]+}}, chunk_iterations={{[1-9][0-9]*}}