darshan-parser.c 13.8 KB
Newer Older
1
2
3
4
5
/*
 *  (C) 2009 by Argonne National Laboratory.
 *      See COPYRIGHT in top-level directory.
 */

6
7
8
9
10
11
12
13
14
#include <stdio.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
#include <unistd.h>
#include <string.h>
#include <zlib.h>
#include <time.h>

15
#include "darshan-logutils.h"
16
17
18
19
20
21
22
23
24

int main(int argc, char **argv)
{
    int ret;
    struct darshan_job job;
    struct darshan_file cp_file;
    char tmp_string[1024];
    int no_files_flag = 0;
    time_t tmp_time = 0;
25
    darshan_fd file;
26
27
28
29
30
31
32

    if(argc != 2)
    {
        fprintf(stderr, "Usage: %s <filename>\n", argv[0]);
        return(-1);
    }

33
    file = darshan_log_open(argv[1]);
34
35
    if(!file)
    {
36
        perror("darshan_log_open");
37
38
39
40
        return(-1);
    }
   
    /* read job info */
41
    ret = darshan_log_getjob(file, &job);
42
    if(ret < 0)
43
    {
44
        fprintf(stderr, "Error: unable to read job information from log file.\n");
45
        darshan_log_close(file);
46
47
48
        return(-1);
    }

49
    ret = darshan_log_getexe(file, tmp_string, &no_files_flag);
50
    if(ret < 0)
51
    {
52
        fprintf(stderr, "Error: unable to read trailing job information.\n");
53
        darshan_log_close(file);
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
        return(-1);
    }

    printf("# darshan log version: %s\n", CP_VERSION);
    printf("# size of file statistics: %d bytes\n", sizeof(cp_file));
    printf("# size of job statistics: %d bytes\n", sizeof(job));
    printf("# exe: %s\n", tmp_string);
    printf("# uid: %d\n", job.uid);
    printf("# start_time: %ld\n", (long)job.start_time);
    tmp_time = (time_t)job.start_time;
    printf("# start_time_asci: %s", ctime(&tmp_time));
    printf("# end_time: %ld\n", (long)job.end_time);
    tmp_time = (time_t)job.end_time;
    printf("# end_time_asci: %s", ctime(&tmp_time));
    printf("# nprocs: %d\n", job.nprocs);
    printf("# run time: %ld\n", (long)(job.end_time - job.start_time + 1));
   
    if(no_files_flag)
    {
        /* it looks like the app didn't open any files */
        printf("# no files opened.\n");
75
        darshan_log_close(file);
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
        return(0);
    }

    printf("\n# description of columns:\n");
    printf("#   <rank>: MPI rank.  -1 indicates that the file is shared\n");
    printf("#      across all processes and statistics are aggregated.\n");
    printf("#   <file>: hash of file path.  0 indicates that statistics\n");
    printf("#      are condensed to refer to all files opened at the given\n");
    printf("#      process.\n");
    printf("#   <counter> and <value>: statistical counters.\n");
    printf("#   <name suffix>: last %d characters of file name.\n", CP_NAME_SUFFIX_LEN);
    printf("\n# description of counters:\n");
    printf("#   CP_POSIX_*: posix operation counts.\n");
    printf("#   CP_COLL_*: MPI collective operation counts.\n");
    printf("#   CP_INDEP_*: MPI independent operation counts.\n");
    printf("#   CP_SPIT_*: MPI split collective operation counts.\n");
    printf("#   CP_NB_*: MPI non blocking operation counts.\n");
    printf("#   READS,WRITES,OPENS,SEEKS,STATS, and MMAPS are types of operations.\n");
94
    printf("#   CP_*_NC_OPENS: number of indep. and collective pnetcdf opens.\n");
95
    printf("#   CP_HDF5_OPENS: number of hdf5 opens.\n");
96
97
98
99
100
101
102
103
104
105
106
    printf("#   CP_COMBINER_*: combiner counts for MPI mem and file datatypes.\n");
    printf("#   CP_HINTS: number of times MPI hints were used.\n");
    printf("#   CP_VIEWS: number of times MPI file views were used.\n");
    printf("#   CP_MODE: mode that file was opened in.\n");
    printf("#   CP_BYTES_*: total bytes read and written.\n");
    printf("#   CP_MAX_BYTE_*: highest offset byte read and written.\n");
    printf("#   CP_CONSEC_*: number of exactly adjacent reads and writes.\n");
    printf("#   CP_SEQ_*: number of reads and writes from increasing offsets.\n");
    printf("#   CP_RW_SWITCHES: number of times access alternated between read and write.\n");
    printf("#   CP_*_ALIGNMENT: memory and file alignment.\n");
    printf("#   CP_*_NOT_ALIGNED: number of reads and writes that were not aligned.\n");
107
    printf("#   CP_MAX_*_TIME_SIZE: size of the slowest read and write operations.\n");
108
109
110
111
112
113
114
115
116
117
118
    printf("#   CP_SIZE_READ_*: histogram of read access sizes.\n");
    printf("#   CP_SIZE_READ_AGG_*: histogram of MPI datatype total sizes.\n");
    printf("#   CP_EXTENT_READ_*: histogram of MPI datatype extents.\n");
    printf("#   CP_STRIDE*_STRIDE: the four most common strides detected.\n");
    printf("#   CP_STRIDE*_COUNT: count of the four most common strides.\n");
    printf("#   CP_ACCESS*_ACCESS: the four most common access sizes.\n");
    printf("#   CP_ACCESS*_COUNT: count of the four most common access sizes.\n");
    printf("#   CP_F_OPEN_TIMESTAMP: timestamp of first open (mpi or posix).\n");
    printf("#   CP_F_*_START_TIMESTAMP: timestamp of first read/write (mpi or posix).\n");
    printf("#   CP_F_*_END_TIMESTAMP: timestamp of last read/write (mpi or posix).\n");
    printf("#   CP_F_CLOSE_TIMESTAMP: timestamp of last close (mpi or posix).\n");
119
120
121
122
    printf("#   CP_F_POSIX_READ/WRITE_TIME: cumulative time spent in posix reads or writes.\n");
    printf("#   CP_F_MPI_READ/WRITE_TIME: cumulative time spent in mpi-io reads or writes.\n");
    printf("#   CP_F_POSIX_META_TIME: cumulative time spent in posix open, close, fsync, stat and seek, .\n");
    printf("#   CP_F_MPI_META_TIME: cumulative time spent in mpi-io open, close, set_view, and sync.\n");
123
    printf("#   CP_MAX_*_TIME: duration of the slowest read and write operations.\n");
124
125
126
127
128

    printf("\n");

    CP_PRINT_HEADER();

129
    while((ret = darshan_log_getfile(file, &cp_file)) == 1)
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
    {
        CP_PRINT(&job, &cp_file, CP_POSIX_READS);
        CP_PRINT(&job, &cp_file, CP_POSIX_WRITES);
        CP_PRINT(&job, &cp_file, CP_POSIX_OPENS);
        CP_PRINT(&job, &cp_file, CP_POSIX_SEEKS);
        CP_PRINT(&job, &cp_file, CP_POSIX_STATS);
        CP_PRINT(&job, &cp_file, CP_POSIX_MMAPS);
        CP_PRINT(&job, &cp_file, CP_POSIX_FREADS);
        CP_PRINT(&job, &cp_file, CP_POSIX_FWRITES);
        CP_PRINT(&job, &cp_file, CP_POSIX_FOPENS);
        CP_PRINT(&job, &cp_file, CP_POSIX_FSEEKS);
        CP_PRINT(&job, &cp_file, CP_POSIX_FSYNCS);
        CP_PRINT(&job, &cp_file, CP_POSIX_FDSYNCS);
        CP_PRINT(&job, &cp_file, CP_INDEP_OPENS);
        CP_PRINT(&job, &cp_file, CP_COLL_OPENS);
        CP_PRINT(&job, &cp_file, CP_INDEP_READS);
        CP_PRINT(&job, &cp_file, CP_INDEP_WRITES);
        CP_PRINT(&job, &cp_file, CP_COLL_READS);
        CP_PRINT(&job, &cp_file, CP_COLL_WRITES);
        CP_PRINT(&job, &cp_file, CP_SPLIT_READS);
        CP_PRINT(&job, &cp_file, CP_SPLIT_WRITES);
        CP_PRINT(&job, &cp_file, CP_NB_READS);
        CP_PRINT(&job, &cp_file, CP_NB_WRITES);
        CP_PRINT(&job, &cp_file, CP_SYNCS);
        CP_PRINT(&job, &cp_file, CP_COMBINER_NAMED);
        CP_PRINT(&job, &cp_file, CP_COMBINER_DUP);
        CP_PRINT(&job, &cp_file, CP_COMBINER_CONTIGUOUS);
        CP_PRINT(&job, &cp_file, CP_COMBINER_VECTOR);
        CP_PRINT(&job, &cp_file, CP_COMBINER_HVECTOR_INTEGER);
        CP_PRINT(&job, &cp_file, CP_COMBINER_HVECTOR);
        CP_PRINT(&job, &cp_file, CP_COMBINER_INDEXED);
        CP_PRINT(&job, &cp_file, CP_COMBINER_HINDEXED_INTEGER);
        CP_PRINT(&job, &cp_file, CP_COMBINER_HINDEXED);
        CP_PRINT(&job, &cp_file, CP_COMBINER_INDEXED_BLOCK);
        CP_PRINT(&job, &cp_file, CP_COMBINER_STRUCT_INTEGER);
        CP_PRINT(&job, &cp_file, CP_COMBINER_STRUCT);
        CP_PRINT(&job, &cp_file, CP_COMBINER_SUBARRAY);
        CP_PRINT(&job, &cp_file, CP_COMBINER_DARRAY);
        CP_PRINT(&job, &cp_file, CP_COMBINER_F90_REAL);
        CP_PRINT(&job, &cp_file, CP_COMBINER_F90_COMPLEX);
        CP_PRINT(&job, &cp_file, CP_COMBINER_F90_INTEGER);
        CP_PRINT(&job, &cp_file, CP_COMBINER_RESIZED);
172
173
        CP_PRINT(&job, &cp_file, CP_INDEP_NC_OPENS);
        CP_PRINT(&job, &cp_file, CP_COLL_NC_OPENS);
174
        CP_PRINT(&job, &cp_file, CP_HDF5_OPENS);
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
        CP_PRINT(&job, &cp_file, CP_HINTS);
        CP_PRINT(&job, &cp_file, CP_VIEWS);
        CP_PRINT(&job, &cp_file, CP_MODE);
        CP_PRINT(&job, &cp_file, CP_BYTES_READ);
        CP_PRINT(&job, &cp_file, CP_BYTES_WRITTEN);
        CP_PRINT(&job, &cp_file, CP_MAX_BYTE_READ);
        CP_PRINT(&job, &cp_file, CP_MAX_BYTE_WRITTEN);
        CP_PRINT(&job, &cp_file, CP_CONSEC_READS);
        CP_PRINT(&job, &cp_file, CP_CONSEC_WRITES);
        CP_PRINT(&job, &cp_file, CP_SEQ_READS);
        CP_PRINT(&job, &cp_file, CP_SEQ_WRITES);
        CP_PRINT(&job, &cp_file, CP_RW_SWITCHES);
        CP_PRINT(&job, &cp_file, CP_MEM_NOT_ALIGNED);
        CP_PRINT(&job, &cp_file, CP_MEM_ALIGNMENT);
        CP_PRINT(&job, &cp_file, CP_FILE_NOT_ALIGNED);
        CP_PRINT(&job, &cp_file, CP_FILE_ALIGNMENT);
191
192
        CP_PRINT(&job, &cp_file, CP_MAX_READ_TIME_SIZE);
        CP_PRINT(&job, &cp_file, CP_MAX_WRITE_TIME_SIZE);
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
        CP_PRINT(&job, &cp_file, CP_SIZE_READ_0_100);
        CP_PRINT(&job, &cp_file, CP_SIZE_READ_100_1K);
        CP_PRINT(&job, &cp_file, CP_SIZE_READ_1K_10K);
        CP_PRINT(&job, &cp_file, CP_SIZE_READ_10K_100K);
        CP_PRINT(&job, &cp_file, CP_SIZE_READ_100K_1M);
        CP_PRINT(&job, &cp_file, CP_SIZE_READ_1M_4M);
        CP_PRINT(&job, &cp_file, CP_SIZE_READ_4M_10M);
        CP_PRINT(&job, &cp_file, CP_SIZE_READ_10M_100M);
        CP_PRINT(&job, &cp_file, CP_SIZE_READ_100M_1G);
        CP_PRINT(&job, &cp_file, CP_SIZE_READ_1G_PLUS);
        CP_PRINT(&job, &cp_file, CP_SIZE_READ_AGG_0_100);
        CP_PRINT(&job, &cp_file, CP_SIZE_READ_AGG_100_1K);
        CP_PRINT(&job, &cp_file, CP_SIZE_READ_AGG_1K_10K);
        CP_PRINT(&job, &cp_file, CP_SIZE_READ_AGG_10K_100K);
        CP_PRINT(&job, &cp_file, CP_SIZE_READ_AGG_100K_1M);
        CP_PRINT(&job, &cp_file, CP_SIZE_READ_AGG_1M_4M);
        CP_PRINT(&job, &cp_file, CP_SIZE_READ_AGG_4M_10M);
        CP_PRINT(&job, &cp_file, CP_SIZE_READ_AGG_10M_100M);
        CP_PRINT(&job, &cp_file, CP_SIZE_READ_AGG_100M_1G);
        CP_PRINT(&job, &cp_file, CP_SIZE_READ_AGG_1G_PLUS);
        CP_PRINT(&job, &cp_file, CP_EXTENT_READ_0_100);
        CP_PRINT(&job, &cp_file, CP_EXTENT_READ_100_1K);
        CP_PRINT(&job, &cp_file, CP_EXTENT_READ_1K_10K);
        CP_PRINT(&job, &cp_file, CP_EXTENT_READ_10K_100K);
        CP_PRINT(&job, &cp_file, CP_EXTENT_READ_100K_1M);
        CP_PRINT(&job, &cp_file, CP_EXTENT_READ_1M_4M);
        CP_PRINT(&job, &cp_file, CP_EXTENT_READ_4M_10M);
        CP_PRINT(&job, &cp_file, CP_EXTENT_READ_10M_100M);
        CP_PRINT(&job, &cp_file, CP_EXTENT_READ_100M_1G);
        CP_PRINT(&job, &cp_file, CP_EXTENT_READ_1G_PLUS);
        CP_PRINT(&job, &cp_file, CP_SIZE_WRITE_0_100);
        CP_PRINT(&job, &cp_file, CP_SIZE_WRITE_100_1K);
        CP_PRINT(&job, &cp_file, CP_SIZE_WRITE_1K_10K);
        CP_PRINT(&job, &cp_file, CP_SIZE_WRITE_10K_100K);
        CP_PRINT(&job, &cp_file, CP_SIZE_WRITE_100K_1M);
        CP_PRINT(&job, &cp_file, CP_SIZE_WRITE_1M_4M);
        CP_PRINT(&job, &cp_file, CP_SIZE_WRITE_4M_10M);
        CP_PRINT(&job, &cp_file, CP_SIZE_WRITE_10M_100M);
        CP_PRINT(&job, &cp_file, CP_SIZE_WRITE_100M_1G);
        CP_PRINT(&job, &cp_file, CP_SIZE_WRITE_1G_PLUS);
        CP_PRINT(&job, &cp_file, CP_SIZE_WRITE_AGG_0_100);
        CP_PRINT(&job, &cp_file, CP_SIZE_WRITE_AGG_100_1K);
        CP_PRINT(&job, &cp_file, CP_SIZE_WRITE_AGG_1K_10K);
        CP_PRINT(&job, &cp_file, CP_SIZE_WRITE_AGG_10K_100K);
        CP_PRINT(&job, &cp_file, CP_SIZE_WRITE_AGG_100K_1M);
        CP_PRINT(&job, &cp_file, CP_SIZE_WRITE_AGG_1M_4M);
        CP_PRINT(&job, &cp_file, CP_SIZE_WRITE_AGG_4M_10M);
        CP_PRINT(&job, &cp_file, CP_SIZE_WRITE_AGG_10M_100M);
        CP_PRINT(&job, &cp_file, CP_SIZE_WRITE_AGG_100M_1G);
        CP_PRINT(&job, &cp_file, CP_SIZE_WRITE_AGG_1G_PLUS);
        CP_PRINT(&job, &cp_file, CP_EXTENT_WRITE_0_100);
        CP_PRINT(&job, &cp_file, CP_EXTENT_WRITE_100_1K);
        CP_PRINT(&job, &cp_file, CP_EXTENT_WRITE_1K_10K);
        CP_PRINT(&job, &cp_file, CP_EXTENT_WRITE_10K_100K);
        CP_PRINT(&job, &cp_file, CP_EXTENT_WRITE_100K_1M);
        CP_PRINT(&job, &cp_file, CP_EXTENT_WRITE_1M_4M);
        CP_PRINT(&job, &cp_file, CP_EXTENT_WRITE_4M_10M);
        CP_PRINT(&job, &cp_file, CP_EXTENT_WRITE_10M_100M);
        CP_PRINT(&job, &cp_file, CP_EXTENT_WRITE_100M_1G);
        CP_PRINT(&job, &cp_file, CP_EXTENT_WRITE_1G_PLUS);
        CP_PRINT(&job, &cp_file, CP_STRIDE1_STRIDE);
        CP_PRINT(&job, &cp_file, CP_STRIDE1_COUNT);
        CP_PRINT(&job, &cp_file, CP_STRIDE2_STRIDE);
        CP_PRINT(&job, &cp_file, CP_STRIDE2_COUNT);
        CP_PRINT(&job, &cp_file, CP_STRIDE3_STRIDE);
        CP_PRINT(&job, &cp_file, CP_STRIDE3_COUNT);
        CP_PRINT(&job, &cp_file, CP_STRIDE4_STRIDE);
        CP_PRINT(&job, &cp_file, CP_STRIDE4_COUNT);
        CP_PRINT(&job, &cp_file, CP_ACCESS1_ACCESS);
        CP_PRINT(&job, &cp_file, CP_ACCESS1_COUNT);
        CP_PRINT(&job, &cp_file, CP_ACCESS2_ACCESS);
        CP_PRINT(&job, &cp_file, CP_ACCESS2_COUNT);
        CP_PRINT(&job, &cp_file, CP_ACCESS3_ACCESS);
        CP_PRINT(&job, &cp_file, CP_ACCESS3_COUNT);
        CP_PRINT(&job, &cp_file, CP_ACCESS4_ACCESS);
        CP_PRINT(&job, &cp_file, CP_ACCESS4_COUNT);

        CP_F_PRINT(&job, &cp_file, CP_F_OPEN_TIMESTAMP);
        CP_F_PRINT(&job, &cp_file, CP_F_CLOSE_TIMESTAMP);
        CP_F_PRINT(&job, &cp_file, CP_F_READ_START_TIMESTAMP);
        CP_F_PRINT(&job, &cp_file, CP_F_READ_END_TIMESTAMP);
        CP_F_PRINT(&job, &cp_file, CP_F_WRITE_START_TIMESTAMP);
        CP_F_PRINT(&job, &cp_file, CP_F_WRITE_END_TIMESTAMP);
        CP_F_PRINT(&job, &cp_file, CP_F_POSIX_READ_TIME);
        CP_F_PRINT(&job, &cp_file, CP_F_POSIX_WRITE_TIME);
        CP_F_PRINT(&job, &cp_file, CP_F_POSIX_META_TIME);
        CP_F_PRINT(&job, &cp_file, CP_F_MPI_META_TIME);
        CP_F_PRINT(&job, &cp_file, CP_F_MPI_READ_TIME);
        CP_F_PRINT(&job, &cp_file, CP_F_MPI_WRITE_TIME);
282
283
        CP_F_PRINT(&job, &cp_file, CP_F_MAX_READ_TIME);
        CP_F_PRINT(&job, &cp_file, CP_F_MAX_WRITE_TIME);
284
285
    }

286
    if(ret < 0)
287
    {
288
        fprintf(stderr, "Error: failed to parse log file.\n");
289
290
291
        return(-1);
    }

292
    darshan_log_close(file);
293
294
    return(0);
}