darshan-logutils.c 31.8 KB
Newer Older
1
/*
Shane Snyder's avatar
Shane Snyder committed
2 3 4
 * Copyright (C) 2015 University of Chicago.
 * See COPYRIGHT notice in top-level directory.
 *
5 6
 */

7
#define _GNU_SOURCE
8
#include "darshan-util-config.h"
9 10
#include <stdio.h>
#include <string.h>
11
#include <assert.h>
12
#include <stdlib.h>
13
#include <unistd.h>
14
#include <inttypes.h>
15 16 17
#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
18
#include <errno.h>
19

20 21
#include "darshan-logutils.h"

22 23
static int darshan_log_seek(darshan_fd fd, off_t offset);
static int darshan_log_read(darshan_fd fd, void *buf, int len);
24
static int darshan_log_write(darshan_fd fd, void *buf, int len);
25
static int darshan_decompress_buf(char* comp_buf, int comp_buf_sz,
26 27
    char* decomp_buf, int* inout_decomp_buf_sz,
    enum darshan_comp_type comp_type);
28
static int darshan_compress_buf(char* decomp_buf, int decomp_buf_sz,
29 30 31 32 33 34 35 36 37 38
    char* comp_buf, int* inout_comp_buf_sz,
    enum darshan_comp_type comp_type);
static int darshan_zlib_decomp(char* comp_buf, int comp_buf_sz,
    char* decomp_buf, int* inout_decomp_buf_sz);
static int darshan_zlib_comp(char* decomp_buf, int decomp_buf_sz,
    char* comp_buf, int* inout_comp_buf_sz);
#ifdef HAVE_LIBBZ2
static int darshan_bzip2_decomp(char* comp_buf, int comp_buf_sz,
    char* decomp_buf, int* inout_decomp_buf_sz);
static int darshan_bzip2_comp(char* decomp_buf, int decomp_buf_sz,
39
    char* comp_buf, int* inout_comp_buf_sz);
40 41
#endif

42 43 44 45

/* TODO: can we make this s.t. we don't care about ordering (i.e., X macro it ) */
struct darshan_mod_logutil_funcs *mod_logutils[DARSHAN_MAX_MODS] =
{
46
    NULL,               /* NULL */
47
    &posix_logutils,    /* POSIX */
48 49 50
    &mpiio_logutils,    /* MPI-IO */
    &hdf5_logutils,     /* HDF5 */
    &pnetcdf_logutils,  /* PNETCDF */
51 52 53 54 55 56 57 58 59 60 61 62
    NULL,
    NULL,
    NULL,
    NULL,
    NULL,
    NULL,
    NULL,
    NULL,
    NULL,
    NULL,
    NULL
};
63 64 65

/* darshan_log_open()
 *
66
 * open an existing darshan log file for reading only
67 68 69
 *
 * returns 0 on success, -1 on failure
 */
70
darshan_fd darshan_log_open(const char *name)
71
{
72
    darshan_fd tmp_fd;
73

74 75 76 77 78
    tmp_fd = malloc(sizeof(*tmp_fd));
    if(!tmp_fd)
        return(NULL);
    memset(tmp_fd, 0, sizeof(*tmp_fd));

79 80
    tmp_fd->fildes = open(name, O_RDONLY);
    if(tmp_fd->fildes < 0)
81
    {
82 83 84
        perror("darshan_log_open: ");
        free(tmp_fd);
        tmp_fd = NULL;
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
    return(tmp_fd);
}

/* darshan_log_create()
 *
 * create a darshan log file for writing with the given compression method
 *
 * returns 0 on success, -1 on failure
 */
darshan_fd darshan_log_create(const char *name, enum darshan_comp_type comp_type)
{
    darshan_fd tmp_fd;

    tmp_fd = malloc(sizeof(*tmp_fd));
    if(!tmp_fd)
        return(NULL);
    memset(tmp_fd, 0, sizeof(*tmp_fd));

    /* TODO: permissions when creating?  umask */
    /* when writing, we create the log file making sure not to overwrite
     * an existing log
     */
    tmp_fd->comp_type = comp_type;
    tmp_fd->fildes = open(name, O_WRONLY | O_CREAT | O_EXCL, 0400);
111
    if(tmp_fd->fildes < 0)
112
    {
113
        perror("darshan_log_create: ");
114
        free(tmp_fd);
115
        tmp_fd = NULL;
116
    }
117 118

    return(tmp_fd);
119 120
}

121 122 123 124
/* darshan_log_getheader()
 *
 * read the header of the darshan log and set internal data structures
 * NOTE: this function must be called before reading other portions of the log
125
 * NOTE: this is the only portion of the darshan log which is uncompressed
126 127 128 129
 *
 * returns 0 on success, -1 on failure
 */
int darshan_log_getheader(darshan_fd fd, struct darshan_header *header)
130
{
131
    int i;
132
    int ret;
133

134
    ret = darshan_log_seek(fd, 0);
135
    if(ret < 0)
136 137
    {
        fprintf(stderr, "Error: unable to seek in darshan log file.\n");
138
        return(-1);
139
    }
140

141
    /* read uncompressed header from log file */
142
    ret = darshan_log_read(fd, header, sizeof(*header));
143
    if(ret != sizeof(*header))
144
    {
145
        fprintf(stderr, "Error: failed to read darshan log file header.\n");
146 147 148
        return(-1);
    }

149 150
    /* save the version string */
    strncpy(fd->version, header->version_string, 8);
151

152
    if(header->magic_nr == DARSHAN_MAGIC_NR)
153
    {
154
        /* no byte swapping needed, this file is in host format already */
155 156 157 158 159 160
        fd->swap_flag = 0;
    }
    else
    {
        /* try byte swapping */
        DARSHAN_BSWAP64(&header->magic_nr);
161
        if(header->magic_nr == DARSHAN_MAGIC_NR)
162 163
        {
            fd->swap_flag = 1;
164 165 166 167

            /* swap the log map variables in the header */
            DARSHAN_BSWAP64(&header->rec_map.off);
            DARSHAN_BSWAP64(&header->rec_map.len);
168
            for(i = 0; i < DARSHAN_MAX_MODS; i++)
169 170 171 172
            {
                DARSHAN_BSWAP64(&header->mod_map[i].off);
                DARSHAN_BSWAP64(&header->mod_map[i].len);
            }
173 174 175 176 177 178 179
        }
        else
        {
            /* otherwise this file is just broken */
            fprintf(stderr, "Error: bad magic number in darshan log file.\n");
            return(-1);
        }
180 181
    }

182 183
    fd->comp_type = header->comp_type;

184
    /* save the mapping of data within log file to this file descriptor */
185 186
    fd->job_map.off = sizeof(struct darshan_header);
    fd->job_map.len = header->rec_map.off - fd->job_map.off;
187 188
    memcpy(&fd->rec_map, &header->rec_map, sizeof(struct darshan_log_map));
    memcpy(&fd->mod_map, &header->mod_map, DARSHAN_MAX_MODS * sizeof(struct darshan_log_map));
189 190

    return(0);
191
}
192

193 194 195
/* darshan_log_putheader()
 *
 * write a darshan header to log file
196 197
 * NOTE: the header is not passed in, and is instead built using
 * contents of the given file descriptor
198 199 200
 *
 * returns 0 on success, -1 on failure
 */
201
int darshan_log_putheader(darshan_fd fd)
202
{
203
    struct darshan_header header;
204 205 206 207 208 209 210 211 212
    int ret;

    ret = darshan_log_seek(fd, 0);
    if(ret < 0)
    {
        fprintf(stderr, "Error: unable to seek in darshan log file.\n");
        return(-1);
    }

213 214 215
    memset(&header, 0, sizeof(header));
    strcpy(header.version_string, DARSHAN_LOG_VERSION);
    header.magic_nr = DARSHAN_MAGIC_NR;
216
    header.comp_type = fd->comp_type;
217 218 219 220 221

    /* copy the mapping information to the header */
    memcpy(&header.rec_map, &fd->rec_map, sizeof(struct darshan_log_map));
    memcpy(&header.mod_map, &fd->mod_map, DARSHAN_MAX_MODS * sizeof(struct darshan_log_map));

222
    /* write header to file */
223 224
    ret = darshan_log_write(fd, &header, sizeof(header));
    if(ret != sizeof(header))
225 226 227 228 229 230 231 232
    {
        fprintf(stderr, "Error: failed to write Darshan log file header.\n");
        return(-1);
    }

    return(0);
}

233
/* darshan_log_getjob()
234 235
 *
 * read job level metadata from the darshan log file
236
 *
237
 * returns 0 on success, -1 on failure
238
 */
239
int darshan_log_getjob(darshan_fd fd, struct darshan_job *job)
240
{
241 242 243
    char *comp_buf;
    char job_buf[DARSHAN_JOB_RECORD_SIZE] = {0};
    int job_buf_sz = DARSHAN_JOB_RECORD_SIZE;
244
    int ret;
245

246 247
    assert(fd->job_map.len > 0 && fd->job_map.off > 0);

248 249 250 251 252 253
    /* allocate buffer to store compressed job info */
    comp_buf = malloc(fd->job_map.len);
    if(!comp_buf)
        return(-1);

    ret = darshan_log_seek(fd, fd->job_map.off);
254
    if(ret < 0)
255
    {
256
        fprintf(stderr, "Error: unable to seek in darshan log file.\n");
257
        free(comp_buf);
258
        return(-1);
259 260
    }

261 262 263
    /* read the compressed job data from the log file */
    ret = darshan_log_read(fd, comp_buf, fd->job_map.len);
    if(ret != fd->job_map.len)
264
    {
265
        fprintf(stderr, "Error: failed to read darshan log file job data.\n");
266 267 268 269 270
        free(comp_buf);
        return(-1);
    }

    /* decompress the job data */
271 272
    ret = darshan_decompress_buf(comp_buf, fd->job_map.len,
        job_buf, &job_buf_sz, fd->comp_type);
273 274 275 276
    if(ret < 0)
    {
        fprintf(stderr, "Error: failed to decompress darshan job data.\n");
        free(comp_buf);
277 278
        return(-1);
    }
279 280 281
    free(comp_buf);

    memcpy(job, job_buf, sizeof(*job));
282

283
    if(fd->swap_flag)
284
    {
285 286 287 288 289 290
        /* swap bytes if necessary */
        DARSHAN_BSWAP64(&job->uid);
        DARSHAN_BSWAP64(&job->start_time);
        DARSHAN_BSWAP64(&job->end_time);
        DARSHAN_BSWAP64(&job->nprocs);
        DARSHAN_BSWAP64(&job->jobid);
291 292
    }

293 294 295 296 297 298 299
    /* save trailing exe & mount information, so it can be retrieved later */
    if(!fd->exe_mnt_data)
        fd->exe_mnt_data = malloc(DARSHAN_EXE_LEN+1);
    if(!fd->exe_mnt_data)
        return(-1);
    memcpy(fd->exe_mnt_data, &job_buf[sizeof(*job)], DARSHAN_EXE_LEN+1);

300 301 302
    return(0);
}

303 304 305 306 307 308 309 310 311
/* darshan_log_putjob()
 *
 * write job level metadat to darshan log file
 *
 * returns 0 on success, -1 on failure
 */
int darshan_log_putjob(darshan_fd fd, struct darshan_job *job)
{
    struct darshan_job job_copy;
312 313
    char *comp_buf;
    int comp_buf_sz;
314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337
    int len;
    int ret;

    ret = darshan_log_seek(fd, sizeof(struct darshan_header));
    if(ret < 0)
    {
        fprintf(stderr, "Error: unable to seek in darshan log file.\n");
        return(-1);
    }

    memset(&job_copy, 0, sizeof(*job));
    memcpy(&job_copy, job, sizeof(*job));

    /* check for newline in existing metadata, add if needed */
    len = strlen(job_copy.metadata);
    if(len > 0 && len < DARSHAN_JOB_METADATA_LEN)
    {
        if(job_copy.metadata[len-1] != '\n')
        {
            job_copy.metadata[len] = '\n';
            job_copy.metadata[len+1] = '\0';
        }
    }

338 339 340 341 342 343
    comp_buf = malloc(sizeof(*job));
    if(!comp_buf)
        return(-1);
    comp_buf_sz = sizeof(*job);

    /* compress the job data */
344 345
    ret = darshan_compress_buf((char*)&job_copy, sizeof(*job),
        comp_buf, &comp_buf_sz, fd->comp_type);
346 347 348 349 350 351 352
    if(ret < 0)
    {
        fprintf(stderr, "Error: failed to decompress darshan job data.\n");
        free(comp_buf);
        return(-1);
    }    

353
    /* write job data to file */
354 355
    ret = darshan_log_write(fd, comp_buf, comp_buf_sz);
    if(ret != comp_buf_sz)
356 357
    {
        fprintf(stderr, "Error: failed to write darshan log file job data.\n");
Shane Snyder's avatar
Shane Snyder committed
358
        free(comp_buf);
359 360 361
        return(-1);
    }

Shane Snyder's avatar
Shane Snyder committed
362
    free(comp_buf);
363 364 365 366 367 368 369 370 371
    return(0);
}

/* darshan_log_getexe()
 *
 * reads the application exe name from darshan log file
 * 
 * returns 0 on success, -1 on failure 
 */
372 373 374
int darshan_log_getexe(darshan_fd fd, char *buf)
{
    char *newline;
375
    int ret;
376

377 378
    /* if the exe/mount data has not been saved yet, read in the job info */
    if(!fd->exe_mnt_data)
379
    {
380 381
        struct darshan_job job;
        ret = darshan_log_getjob(fd, &job);
382

383 384
        if(ret < 0 || !fd->exe_mnt_data)
            return(-1);
385
    }
386

387 388 389 390
    /* exe string is located before the first line break */
    newline = strchr(fd->exe_mnt_data, '\n');

    /* copy over the exe string */
391
    if(newline)
392
        memcpy(buf, fd->exe_mnt_data, (newline - fd->exe_mnt_data));
393 394 395 396

    return (0);
}

397 398 399
/* darshan_log_putexe()
 *
 * wrties the application exe name to darshan log file
400 401 402
 * NOTE: this needs to be called immediately following put_job as it
 * expects the final pointer to be positioned immediately following
 * the darshan job information
403 404 405 406 407 408 409
 *
 * returns 0 on success, -1 on failure 
 */
int darshan_log_putexe(darshan_fd fd, char *buf)
{
    int len;
    int ret;
410 411 412 413
    char comp_buf[DARSHAN_EXE_LEN] = {0};
    int comp_buf_sz = DARSHAN_EXE_LEN;

    len = strlen(buf);
414

415
    /* compress the input exe string */
416
    ret = darshan_compress_buf(buf, len, comp_buf, &comp_buf_sz, fd->comp_type);
417 418
    if(ret < 0)
    {
419
        fprintf(stderr, "Error: unable to compress exe string.\n");
420 421 422
        return(-1);
    }

423 424
    ret = darshan_log_write(fd, comp_buf, comp_buf_sz);
    if(ret != comp_buf_sz)
425 426 427 428 429 430 431 432
    {
        fprintf(stderr, "Error: failed to write exe string to darshan log file.\n");
        return(-1);
    }

    return(0);
}

433 434
/* darshan_log_getmounts()
 * 
435
 * retrieves mount table information from the log. Note that mnt_pts and
436
 * fs_types are arrays that will be allocated by the function and must be
437 438 439
 * freed by the caller. count will indicate the size of the arrays
 *
 * returns 0 on success, -1 on failure
440
 */
441
int darshan_log_getmounts(darshan_fd fd, char*** mnt_pts,
442 443 444 445
    char*** fs_types, int* count)
{
    char *pos;
    int array_index = 0;
446
    int ret;
447

448 449
    /* if the exe/mount data has not been saved yet, read in the job info */
    if(!fd->exe_mnt_data)
450
    {
451 452
        struct darshan_job job;
        ret = darshan_log_getjob(fd, &job);
453

454 455
        if(ret < 0 || !fd->exe_mnt_data)
            return(-1);
456
    }
457

458
    /* count entries */
459
    *count = 0;
460
    pos = fd->exe_mnt_data;
461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481
    while((pos = strchr(pos, '\n')) != NULL)
    {
        pos++;
        (*count)++;
    }

    if(*count == 0)
    {
        /* no mount entries present */
        return(0);
    }

    /* allocate output arrays */
    *mnt_pts = malloc((*count)*sizeof(char*));
    assert(*mnt_pts);
    *fs_types = malloc((*count)*sizeof(char*));
    assert(*fs_types);

    /* work backwards through the table and parse each line (except for
     * first, which holds command line information)
     */
482
    while((pos = strrchr(fd->exe_mnt_data, '\n')) != NULL)
483 484
    {
        /* overestimate string lengths */
485
        (*mnt_pts)[array_index] = malloc(DARSHAN_EXE_LEN);
486
        assert((*mnt_pts)[array_index]);
487
        (*fs_types)[array_index] = malloc(DARSHAN_EXE_LEN);
488 489
        assert((*fs_types)[array_index]);

490 491 492
        ret = sscanf(++pos, "%s\t%s", (*fs_types)[array_index],
            (*mnt_pts)[array_index]);
        if(ret != 2)
493
        {
494
            fprintf(stderr, "Error: poorly formatted mount table in darshan log file.\n");
495 496 497 498 499 500 501
            return(-1);
        }
        pos--;
        *pos = '\0';
        array_index++;
    }

502 503 504
    return(0);
}

505 506 507 508 509 510 511 512 513 514 515 516 517
/* darshan_log_putmounts()
 *
 * writes mount information to the darshan log file
 * NOTE: this function call should follow immediately after the call
 * to darshan_log_putexe(), as it assumes the darshan log file pointer
 * is pointing to the offset immediately following the exe string
 *
 * returns 0 on success, -1 on failure
 */
int darshan_log_putmounts(darshan_fd fd, char** mnt_pts, char** fs_types, int count)
{
    int i;
    char line[1024];
518 519 520 521 522
    char mnt_dat[DARSHAN_EXE_LEN] = {0};
    int mnt_dat_sz = 0;
    char comp_buf[DARSHAN_EXE_LEN] = {0};
    int comp_buf_sz = DARSHAN_EXE_LEN;
    char *tmp;
523 524 525
    int ret;

    /* write each mount entry to file */
526
    tmp = mnt_dat;
527 528 529
    for(i=count-1; i>=0; i--)
    {
        sprintf(line, "\n%s\t%s", fs_types[i], mnt_pts[i]);
530 531 532 533 534 535

        memcpy(tmp, line, strlen(line));
        tmp += strlen(line);
        mnt_dat_sz += strlen(line);
    }

536
    ret = darshan_compress_buf(mnt_dat, mnt_dat_sz, comp_buf, &comp_buf_sz, fd->comp_type);
537 538 539 540
    if(ret < 0)
    {
        fprintf(stderr, "Error: unable to compress mount data.\n");
        return(-1);
541 542
    }

543 544
    ret = darshan_log_write(fd, comp_buf, comp_buf_sz);
    if (ret != comp_buf_sz)
545
    {
546
        fprintf(stderr, "Error: failed to write darshan log mount data.\n");
547 548 549 550 551 552
        return(-1);
    }

    return(0);
}

553
/* darshan_log_gethash()
554
 *
555
 * read the hash of records from the darshan log file
556 557 558
 *
 * returns 0 on success, -1 on failure
 */
559
int darshan_log_gethash(darshan_fd fd, struct darshan_record_ref **hash)
560
{
561
    char *comp_buf;
562
    char *hash_buf;
563
    int hash_buf_sz;
564
    char *buf_ptr;
565 566 567 568
    darshan_record_id *rec_id_ptr;
    uint32_t *path_len_ptr;
    char *path_ptr;
    struct darshan_record_ref *ref;
569 570
    int ret;

571 572 573 574 575 576 577 578 579 580
    /* allocate buffers to store compressed and decompressed record
     * hash data
     */
    comp_buf = malloc(fd->rec_map.len);
    hash_buf = malloc(DARSHAN_DEF_COMP_BUF_SZ);
    if(!comp_buf || !hash_buf)
        return(-1);
    hash_buf_sz = DARSHAN_DEF_COMP_BUF_SZ;
    memset(hash_buf, 0, DARSHAN_DEF_COMP_BUF_SZ);

581
    ret = darshan_log_seek(fd, fd->rec_map.off);
582
    if(ret < 0)
583 584
    {
        fprintf(stderr, "Error: unable to seek in darshan log file.\n");
585
        free(comp_buf);
Shane Snyder's avatar
Shane Snyder committed
586
        free(hash_buf);
587
        return(-1);
588
    }
589

590
    /* read the record hash from the log file */
591
    ret = darshan_log_read(fd, comp_buf, fd->rec_map.len);
592
    if(ret != fd->rec_map.len)
593
    {
594
        fprintf(stderr, "Error: failed to read record hash from darshan log file.\n");
595
        free(comp_buf);
Shane Snyder's avatar
Shane Snyder committed
596
        free(hash_buf);
597 598 599 600 601
        return(-1);
    }

    /* decompress the record hash buffer */
    ret = darshan_decompress_buf(comp_buf, fd->rec_map.len,
602
        hash_buf, &hash_buf_sz, fd->comp_type);
603 604 605 606
    if(ret < 0)
    {
        fprintf(stderr, "Error: unable to decompress darshan job data.\n");
        free(comp_buf);
Shane Snyder's avatar
Shane Snyder committed
607
        free(hash_buf);
608 609
        return(-1);
    }
610
    free(comp_buf);
611

612
    buf_ptr = hash_buf;
613
    while(buf_ptr < (hash_buf + hash_buf_sz))
614
    {
615
        /* get pointers for each field of this darshan record */
616
        /* NOTE: darshan record hash serialization method: 
617 618
         *          ... darshan_record_id | (uint32_t) path_len | path ...
         */
619 620 621 622 623 624 625
        rec_id_ptr = (darshan_record_id *)buf_ptr;
        buf_ptr += sizeof(darshan_record_id);
        path_len_ptr = (uint32_t *)buf_ptr;
        buf_ptr += sizeof(uint32_t);
        path_ptr = (char *)buf_ptr;

        if(fd->swap_flag)
626
        {
627 628 629
            /* we need to sort out endianness issues before deserializing */
            DARSHAN_BSWAP64(rec_id_ptr);
            DARSHAN_BSWAP32(path_len_ptr);
630 631
        }

632 633 634 635 636 637
        HASH_FIND(hlink, *hash, rec_id_ptr, sizeof(darshan_record_id), ref);
        if(!ref)
        {
            ref = malloc(sizeof(*ref));
            if(!ref)
            {
Shane Snyder's avatar
Shane Snyder committed
638
                free(hash_buf);
639 640 641 642 643 644
                return(-1);
            }
            ref->rec.name = malloc(*path_len_ptr + 1);
            if(!ref->rec.name)
            {
                free(ref);
Shane Snyder's avatar
Shane Snyder committed
645
                free(hash_buf);
646 647 648 649 650 651 652
                return(-1);
            }

            /* set the fields for this record */
            ref->rec.id = *rec_id_ptr;
            memcpy(ref->rec.name, path_ptr, *path_len_ptr);
            ref->rec.name[*path_len_ptr] = '\0';
653

654 655 656
            /* add this record to the hash */
            HASH_ADD(hlink, *hash, rec.id, sizeof(darshan_record_id), ref);
        }
657 658

        buf_ptr += *path_len_ptr;
659 660
    }

661
    free(hash_buf);
662 663 664
    return(0);
}

665 666 667
/* darshan_log_puthash()
 *
 * writes the hash table of records to the darshan log file
668 669 670
 * NOTE: this function call should follow immediately after the call
 * to darshan_log_putmounts(), as it assumes the darshan log file pointer
 * is pointing to the offset immediately following the mount information
671 672 673 674 675 676 677 678 679 680 681
 *
 * returns 0 on success, -1 on failure
 */
int darshan_log_puthash(darshan_fd fd, struct darshan_record_ref *hash)
{
    size_t hash_buf_sz;
    char *hash_buf;
    char *hash_buf_off;
    struct darshan_record_ref *ref, *tmp;
    uint32_t name_len;
    size_t record_sz;
682
    char *comp_buf;
Shane Snyder's avatar
Shane Snyder committed
683
    int comp_buf_sz;
684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735
    int ret;

    /* allocate a buffer to store 2 MiB worth of record data */
    /* NOTE: this buffer may be reallocated if estimate is too small */
    hash_buf_sz = 2 * 1024 * 1024;
    hash_buf = malloc(hash_buf_sz);
    if(!hash_buf)
    {
        return(-1);
    }

    /* serialize the record hash into a buffer for writing */
    hash_buf_off = hash_buf;
    HASH_ITER(hlink, hash, ref, tmp)
    {
        name_len = strlen(ref->rec.name);
        record_sz = sizeof(darshan_record_id) + sizeof(uint32_t) + name_len;
        /* make sure there is room in the buffer for this record */
        if((hash_buf_off + record_sz) > (hash_buf + hash_buf_sz))
        {
            char *tmp_buf;
            size_t old_buf_sz;

            /* if no room, reallocate the hash buffer at twice the current size */
            old_buf_sz = hash_buf_off - hash_buf;
            hash_buf_sz *= 2;
            tmp_buf = malloc(hash_buf_sz);
            if(!tmp_buf)
            {
                free(hash_buf);
                return(-1);
            }

            memcpy(tmp_buf, hash_buf, old_buf_sz);
            free(hash_buf);
            hash_buf = tmp_buf;
            hash_buf_off = hash_buf + old_buf_sz;
        }

        /* now serialize the record into the hash buffer.
         * NOTE: darshan record hash serialization method: 
         *          ... darshan_record_id | (uint32_t) path_len | path ...
         */
        *((darshan_record_id *)hash_buf_off) = ref->rec.id;
        hash_buf_off += sizeof(darshan_record_id);
        *((uint32_t *)hash_buf_off) = name_len;
        hash_buf_off += sizeof(uint32_t);
        memcpy(hash_buf_off, ref->rec.name, name_len);
        hash_buf_off += name_len;
    }
    hash_buf_sz = hash_buf_off - hash_buf;

Shane Snyder's avatar
Shane Snyder committed
736
    comp_buf = malloc(DARSHAN_DEF_COMP_BUF_SZ);
737 738
    if(!comp_buf)
        return(-1);
Shane Snyder's avatar
Shane Snyder committed
739
    comp_buf_sz = DARSHAN_DEF_COMP_BUF_SZ;
740 741

    /* compress the record hash */
742 743
    ret = darshan_compress_buf(hash_buf, hash_buf_sz,
        comp_buf, &comp_buf_sz, fd->comp_type);
744 745 746
    if(ret < 0)
    {
        fprintf(stderr, "Error: unable to compress darshan record hash.\n");
Shane Snyder's avatar
Shane Snyder committed
747 748
        free(comp_buf);
        free(hash_buf);
749 750 751 752 753 754 755
        return(-1);
    }

    /* set the appropriate mapping info for the record hash in the file descriptor */
    fd->rec_map.off = fd->pos;
    fd->rec_map.len = comp_buf_sz;

756
    /* write the record hash to file */
757 758
    ret = darshan_log_write(fd, comp_buf, comp_buf_sz);
    if(ret != comp_buf_sz)
759 760
    {
        fprintf(stderr, "Error: failed to write record hash to darshan log file.\n");
Shane Snyder's avatar
Shane Snyder committed
761 762
        free(comp_buf);
        free(hash_buf);
763 764 765
        return(-1);
    }

Shane Snyder's avatar
Shane Snyder committed
766
    free(comp_buf);
767 768 769 770 771 772 773
    free(hash_buf);

    return(0);
}

/* darshan_log_getmod()
 *
774
 * returns 1 on successful read of module data, 0 on no data, -1 on failure
775 776
 */
int darshan_log_getmod(darshan_fd fd, darshan_module_id mod_id,
777
    void *mod_buf, int *mod_buf_sz)
778
{
779
    char *comp_buf;
780
    int ret;
781

782
    if(mod_id < 0 || mod_id >= DARSHAN_MAX_MODS)
783
    {
784 785
        fprintf(stderr, "Error: invalid Darshan module id.\n");
        return(-1);
786 787
    }

788 789 790 791 792
    if(fd->mod_map[mod_id].len == 0)
        return(0); /* no data corresponding to this mod_id */

    comp_buf = malloc(fd->mod_map[mod_id].len);
    if(!comp_buf)
793
        return(-1);
794 795 796 797 798 799 800

    ret = darshan_log_seek(fd, fd->mod_map[mod_id].off);
    if(ret < 0)
    {
        fprintf(stderr, "Error: unable to seek in darshan log file.\n");
        free(comp_buf);
        return(ret);
801 802
    }

803 804 805
    /* read this module's data from the log file */
    ret = darshan_log_read(fd, comp_buf, fd->mod_map[mod_id].len);
    if(ret != fd->mod_map[mod_id].len)
806
    {
807
        fprintf(stderr,
808
            "Error: failed to read module %s data from darshan log file.\n",
809
            darshan_module_names[mod_id]);
810
        free(comp_buf);
811 812 813
        return(-1);
    }

814 815
    /* decompress this module's data */
    ret = darshan_decompress_buf(comp_buf, fd->mod_map[mod_id].len,
816
        mod_buf, mod_buf_sz, fd->comp_type);
817 818 819 820 821 822 823 824 825
    if(ret < 0)
    {
        fprintf(stderr, "Error: unable to decompress module %s data.\n",
            darshan_module_names[mod_id]);
        free(comp_buf);
        return(-1);
    }
    free(comp_buf);

826
    return(1);
827 828
}

829 830 831
/* darshan_log_putmod()
 *
 * write a chunk of module data to the darshan log file
Shane Snyder's avatar
Shane Snyder committed
832 833 834 835 836 837
 * NOTE: this function call should be called directly after the
 * put_hash() function, as it expects the file pointer to be
 * positioned directly past the record hash location. Also,
 * for a set of modules with data to write to file, this function
 * should be called in order of increasing module identifiers,
 * as the darshan log file format expects this ordering.
838 839 840 841 842 843
 *
 * returns 0 on success, -1 on failure
 */
int darshan_log_putmod(darshan_fd fd, darshan_module_id mod_id,
    void *mod_buf, int mod_buf_sz)
{
Shane Snyder's avatar
Shane Snyder committed
844 845
    char *comp_buf;
    int comp_buf_sz;
846 847
    int ret;

Shane Snyder's avatar
Shane Snyder committed
848
    if(mod_id < 0 || mod_id >= DARSHAN_MAX_MODS)
849
    {
Shane Snyder's avatar
Shane Snyder committed
850 851
        fprintf(stderr, "Error: invalid Darshan module id.\n");
        return(-1);
852 853
    }

Shane Snyder's avatar
Shane Snyder committed
854
    comp_buf = malloc(DARSHAN_DEF_COMP_BUF_SZ);
Shane Snyder's avatar
Shane Snyder committed
855 856
    if(!comp_buf)
        return(-1);
Shane Snyder's avatar
Shane Snyder committed
857
    comp_buf_sz = DARSHAN_DEF_COMP_BUF_SZ;
Shane Snyder's avatar
Shane Snyder committed
858 859

    /* compress the module's data */
860 861
    ret = darshan_compress_buf(mod_buf, mod_buf_sz,
        comp_buf, &comp_buf_sz, fd->comp_type);
Shane Snyder's avatar
Shane Snyder committed
862
    if(ret < 0)
863
    {
Shane Snyder's avatar
Shane Snyder committed
864 865 866
        fprintf(stderr, "Error: unable to compress module %s data.\n",
            darshan_module_names[mod_id]);
        free(comp_buf);
867 868 869
        return(-1);
    }

Shane Snyder's avatar
Shane Snyder committed
870 871 872 873
    /* set the appropriate mapping information for this module */
    fd->mod_map[mod_id].off = fd->pos;
    fd->mod_map[mod_id].len = comp_buf_sz;

874
    /* write the module chunk to the log file */
Shane Snyder's avatar
Shane Snyder committed
875 876
    ret = darshan_log_write(fd, comp_buf, comp_buf_sz);
    if(ret != comp_buf_sz)
877 878 879 880
    {
        fprintf(stderr,
            "Error: failed to write module %s data to darshan log file.\n",
            darshan_module_names[mod_id]);
Shane Snyder's avatar
Shane Snyder committed
881
        free(comp_buf);
882 883 884
        return(-1);
    }

Shane Snyder's avatar
Shane Snyder committed
885
    free(comp_buf);
886 887 888
    return(0);
}

889 890 891 892 893 894 895
/* darshan_log_close()
 *
 * close an open darshan file descriptor
 *
 * returns 0 on success, -1 on failure
 */
void darshan_log_close(darshan_fd fd)
896
{
897 898 899 900 901
    if(fd->fildes)
        close(fd->fildes);

    if(fd->exe_mnt_data)
        free(fd->exe_mnt_data);
902

903
    free(fd);
904 905

    return;
906 907
}

908
/* **************************************************** */
909

910 911 912 913
/* return 0 on successful seek to offset, -1 on failure.
 */
static int darshan_log_seek(darshan_fd fd, off_t offset)
{
914
    off_t ret_off;
915 916 917 918

    if(fd->pos == offset)
        return(0);

919 920
    ret_off = lseek(fd->fildes, offset, SEEK_SET);
    if(ret_off == offset)
921
    {
922 923
        fd->pos = offset;
        return(0);
924 925 926 927 928
    }

    return(-1);
}

929
/* return amount read on success, 0 on EOF, -1 on failure.
930
 */
931
static int darshan_log_read(darshan_fd fd, void* buf, int len)
932 933 934
{
    int ret;

935 936 937 938
    /* read data from the log file using the given map */
    ret = read(fd->fildes, buf, len);
    if(ret > 0)
        fd->pos += ret;
939

940
    return(ret);
941 942
}

943
/* return amount written on success, -1 on failure.
944
 */
945
static int darshan_log_write(darshan_fd fd, void* buf, int len)
946 947 948
{
    int ret;

949 950 951 952 953 954 955 956 957 958
    ret = write(fd->fildes, buf, len);
    if(ret > 0)
        fd->pos += ret;

    return(ret);
}

/* return 0 on successful decompression, -1 on failure
 */
static int darshan_decompress_buf(char* comp_buf, int comp_buf_sz,
959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983
    char* decomp_buf, int* inout_decomp_buf_sz,
    enum darshan_comp_type comp_type)
{
    int ret;

    switch(comp_type)
    {
        case DARSHAN_ZLIB_COMP:
            ret = darshan_zlib_decomp(comp_buf, comp_buf_sz,
                decomp_buf, inout_decomp_buf_sz);
            break;
#ifdef HAVE_LIBBZ2
        case DARSHAN_BZIP2_COMP:
            ret = darshan_bzip2_decomp(comp_buf, comp_buf_sz,
                decomp_buf, inout_decomp_buf_sz);
            break;
#endif
        default:
            fprintf(stderr, "Error: invalid decompression method.\n");
            return(-1);
    }

    return(ret);
}

984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010
static int darshan_compress_buf(char* decomp_buf, int decomp_buf_sz,
    char* comp_buf, int* inout_comp_buf_sz,
    enum darshan_comp_type comp_type)
{
    int ret;

    switch(comp_type)
    {
        case DARSHAN_ZLIB_COMP:
            ret = darshan_zlib_comp(decomp_buf, decomp_buf_sz,
                comp_buf, inout_comp_buf_sz);
            break;
#ifdef HAVE_LIBBZ2
        case DARSHAN_BZIP2_COMP:
            ret = darshan_bzip2_comp(decomp_buf, decomp_buf_sz,
                comp_buf, inout_comp_buf_sz);
            break;
#endif
        default:
            fprintf(stderr, "Error: invalid compression method.\n");
            return(-1);
    }

    return(ret);

}

1011
static int darshan_zlib_decomp(char* comp_buf, int comp_buf_sz,
1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031
    char* decomp_buf, int* inout_decomp_buf_sz)
{
    int ret;
    int total_out = 0;
    z_stream tmp_stream;

    memset(&tmp_stream, 0, sizeof(tmp_stream));
    tmp_stream.zalloc = Z_NULL;
    tmp_stream.zfree = Z_NULL;
    tmp_stream.opaque = Z_NULL;
    tmp_stream.next_in = (unsigned char*)comp_buf;
    tmp_stream.avail_in = comp_buf_sz;
    tmp_stream.next_out = (unsigned char*)decomp_buf;
    tmp_stream.avail_out = *inout_decomp_buf_sz;

    /* initialize the zlib decompression parameters */
    /* TODO: check these parameters? */
    //ret = inflateInit2(&tmp_stream, 31);
    ret = inflateInit(&tmp_stream);
    if(ret != Z_OK)
1032
    {
1033
        return(-1);
1034 1035
    }

1036 1037 1038 1039 1040
    /* while we have not finished consuming all of the compressed input data */
    while(tmp_stream.avail_in)
    {
        if(tmp_stream.avail_out == 0)
        {
1041
            /* We ran out of buffer space for decompression.  In theory,
1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066
             * we could just alloc more space, but probably just easier
             * to bump up the default size of the output buffer.
             */
            inflateEnd(&tmp_stream);
            return(-1);
        }

        /* decompress data */
        ret = inflate(&tmp_stream, Z_FINISH);
        if(ret != Z_STREAM_END)
        {
            inflateEnd(&tmp_stream);
            return(-1);
        }

        total_out += tmp_stream.total_out;
        if(tmp_stream.avail_in)
            inflateReset(&tmp_stream);
    }
    inflateEnd(&tmp_stream);

    *inout_decomp_buf_sz = total_out;
    return(0);
}

1067
static int darshan_zlib_comp(char* decomp_buf, int decomp_buf_sz,
1068 1069
    char* comp_buf, int* inout_comp_buf_sz)
{
1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087
    int ret;
    z_stream tmp_stream;

    memset(&tmp_stream, 0, sizeof(tmp_stream));
    tmp_stream.zalloc = Z_NULL;
    tmp_stream.zfree = Z_NULL;
    tmp_stream.opaque = Z_NULL;
    tmp_stream.next_in = (unsigned char*)decomp_buf;
    tmp_stream.avail_in = decomp_buf_sz;
    tmp_stream.next_out = (unsigned char*)comp_buf;
    tmp_stream.avail_out = *inout_comp_buf_sz;

    ret = deflateInit(&tmp_stream, Z_DEFAULT_COMPRESSION);
    if(ret != Z_OK)
    {
        return(-1);
    }

1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123
    /* compress data */
    ret = deflate(&tmp_stream, Z_FINISH);
    if(ret != Z_STREAM_END)
    {
        deflateEnd(&tmp_stream);
        return(-1);
    }
    deflateEnd(&tmp_stream);

    *inout_comp_buf_sz = tmp_stream.total_out;
    return(0);
}

#ifdef HAVE_LIBBZ2
static int darshan_bzip2_decomp(char* comp_buf, int comp_buf_sz,
    char* decomp_buf, int* inout_decomp_buf_sz)
{
    int ret;
    int total_out = 0;
    bz_stream tmp_stream;

    memset(&tmp_stream, 0, sizeof(tmp_stream));
    tmp_stream.bzalloc = NULL;
    tmp_stream.bzfree = NULL;
    tmp_stream.opaque = NULL;
    tmp_stream.next_in = comp_buf;
    tmp_stream.avail_in = comp_buf_sz;
    tmp_stream.next_out = decomp_buf;
    tmp_stream.avail_out = *inout_decomp_buf_sz;

    ret = BZ2_bzDecompressInit(&tmp_stream, 1, 0);
    if(ret != BZ_OK)
    {
        return(-1);
    }

1124 1125 1126 1127 1128
    /* while we have not finished consuming all of the uncompressed input data */
    while(tmp_stream.avail_in)
    {
        if(tmp_stream.avail_out == 0)
        {
1129
            /* We ran out of buffer space for decompression.  In theory,
1130 1131 1132
             * we could just alloc more space, but probably just easier
             * to bump up the default size of the output buffer.
             */
1133
            BZ2_bzDecompressEnd(&tmp_stream);
1134 1135 1136
            return(-1);
        }

1137 1138 1139
        /* decompress data */
        ret = BZ2_bzDecompress(&tmp_stream);
        if(ret != BZ_STREAM_END)
1140
        {
1141
            BZ2_bzDecompressEnd(&tmp_stream);
1142 1143 1144
            return(-1);
        }

1145 1146
        assert(tmp_stream.total_out_hi32 == 0);
        total_out += tmp_stream.total_out_lo32;
1147
        if(tmp_stream.avail_in)
1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158
        {
            /* reinitialize bzip2 stream, we have more data to
             * decompress
             */
            BZ2_bzDecompressEnd(&tmp_stream);
            ret = BZ2_bzDecompressInit(&tmp_stream, 1, 0);
            if(ret != BZ_OK)
            {
                return(-1);
            }
        }
1159
    }
1160
    BZ2_bzDecompressEnd(&tmp_stream);
1161

1162
    *inout_decomp_buf_sz = total_out;
1163
    return(0);
1164
}
1165

1166 1167 1168
static int darshan_bzip2_comp(char* decomp_buf, int decomp_buf_sz,
    char* comp_buf, int* inout_comp_buf_sz)
{
1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187
    int ret;
    bz_stream tmp_stream;

    memset(&tmp_stream, 0, sizeof(tmp_stream));
    tmp_stream.bzalloc = NULL;
    tmp_stream.bzfree = NULL;
    tmp_stream.opaque = NULL;
    tmp_stream.next_in = decomp_buf;
    tmp_stream.avail_in = decomp_buf_sz;
    tmp_stream.next_out = comp_buf;
    tmp_stream.avail_out = *inout_comp_buf_sz;

    ret = BZ2_bzCompressInit(&tmp_stream, 9, 1, 30);
    if(ret != BZ_OK)
    {
        return(-1);
    }

    /* compress data */
Shane Snyder's avatar
Shane Snyder committed
1188
    do
1189
    {
Shane Snyder's avatar
Shane Snyder committed
1190 1191 1192 1193 1194 1195 1196
        ret = BZ2_bzCompress(&tmp_stream, BZ_FINISH);
        if(ret < 0)
        {
            BZ2_bzCompressEnd(&tmp_stream);
            return(-1);
        }
    } while (ret != BZ_STREAM_END);
1197 1198 1199 1200 1201
    BZ2_bzCompressEnd(&tmp_stream);

    assert(tmp_stream.total_out_hi32 == 0);
    *inout_comp_buf_sz = tmp_stream.total_out_lo32;
    return(0);
1202 1203 1204
}
#endif

1205 1206 1207 1208 1209 1210 1211 1212
/*
 * Local variables:
 *  c-indent-level: 4
 *  c-basic-offset: 4
 * End:
 *
 * vim: ts=8 sts=4 sw=4 expandtab
 */