-
Notifications
You must be signed in to change notification settings - Fork 113
/
atop.c
1387 lines (1180 loc) · 34.8 KB
/
atop.c
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
128
129
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
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
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
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
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
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
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
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
/*
** ATOP - System & Process Monitor
**
** The program 'atop' offers the possibility to view the activity of
** the system on system-level as well as process-level.
**
** This source-file contains the main-function, which verifies the
** calling-parameters and takes care of initialization.
** The engine-function drives the main sample-loop in which after the
** indicated interval-time a snapshot is taken of the system-level and
** process-level counters and the deviations are calculated and
** visualized for the user.
** ==========================================================================
** Author: Gerlof Langeveld
** E-mail: [email protected]
** Date: November 1996
** Linux-port: June 2000
** Modified: May 2001 - Ported to kernel 2.4
** --------------------------------------------------------------------------
** Copyright (C) 2000-2024 Gerlof Langeveld
**
** This program is free software; you can redistribute it and/or modify it
** under the terms of the GNU General Public License as published by the
** Free Software Foundation; either version 2, or (at your option) any
** later version.
**
** This program is distributed in the hope that it will be useful, but
** WITHOUT ANY WARRANTY; without even the implied warranty of
** MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
** See the GNU General Public License for more details.
**
** You should have received a copy of the GNU General Public License
** along with this program; if not, write to the Free Software
** Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
** --------------------------------------------------------------------------
**
** After initialization, the main-function calls the ENGINE.
** For every cycle (so after another interval) the ENGINE calls various
** functions as shown below:
**
** +------------------------------------------------------------------------+
** | E N G I N E |
** | |
** | |
** | _____________________await interval-timer________________________ |
** | | ^ |
** | | _______ _______ _______ ________ ________ | |
** | | ^ | ^ | ^ | ^ | ^ | | |
** +---|---|-------|--|-------|--|-------|-----|--------|----|--------|--|--+
** | | | | | | | | | | | |
** +--V-----+ +--V----+ +--V----+ +--V--------+ +--V-------+ +--V-----+
** | | | | | | | | | deviate | | |
** | photo | | photo | | photo | | acct | | ..cgroup | | print |
** | cgroup | | syst | | proc | | photoproc | | ..syst | | |
** | | | | | | | | | ..proc | | |
** +--------+ +-------+ +-------+ +-----------+ +----------+ +--------+
** ^ ^ ^ ^ | |
** | | | | | |
** | | | V V V
** _______ _____ _____ __________ ________ _________
** / \ / \ / \ / \ / \ / \
** /sys/fs /proc /proc accounting task screen or
** /cgroup file database file
** \_______/ \_____/ \_____/ \__________/ \________/ \_________/
**
** - photocgroup()
** Takes a snapshot of the counters related to resource usage on
** cgroup-level v2 (cpu, disk, memory).
**
** - photosyst()
** Takes a snapshot of the counters related to resource-usage on
** system-level (cpu, disk, memory, network).
**
** - photoproc()
** Takes a snapshot of the counters related to resource-usage of
** tasks which are currently active. For this purpose the whole
** task-list is read.
**
** - acctphotoproc()
** Takes a snapshot of the counters related to resource-usage of
** tasks which have been finished during the last interval.
** For this purpose all new records in the accounting-file are read.
**
** When all counters have been gathered, functions are called to calculate
** the difference between the current counter values and the counter values
** of the previous cycle. These functions operate on cgroup level, system level
** as well as on task level.
** These differences are stored in a new structure (table).
**
** - deviatcgroup()
** Calculates the differences between the current cgroup-level
** counters and the corresponding counters of the previous cycle.
**
** - deviatsyst()
** Calculates the differences between the current system-level
** counters and the corresponding counters of the previous cycle.
**
** - deviattask()
** Calculates the differences between the current task-level
** counters and the corresponding counters of the previous cycle.
** The per-task counters of the previous cycle are stored in the
** task-database; this "database" is implemented as a linked list
** of taskinfo structures in memory (so no disk-accesses needed).
** Within this linked list hash-buckets are maintained for fast searches.
** The entire task-database is handled via a set of well-defined
** functions from which the name starts with "pdb_..." (see the
** source-file procdbase.c).
** The processes which have been finished during the last cycle
** are also treated by deviattask() in order to calculate what their
** resource-usage was before they finished.
**
** All information is ready to be visualized now.
** There is a structure which holds the start-address of the
** visualization-function to be called. Initially this structure contains
** the address of the generic visualization-function ("generic_samp"), but
** these addresses can be modified in the main-function depending on particular
** flags. In this way various representation-layers (ASCII, graphical, ...)
** can be linked with 'atop'; the one to use can eventually be chosen
** at runtime.
*/
#include <sys/types.h>
#include <sys/param.h>
#include <sys/mman.h>
#include <sys/stat.h>
#include <time.h>
#include <stdio.h>
#include <errno.h>
#include <fcntl.h>
#include <unistd.h>
#include <stdlib.h>
#include <signal.h>
#include <sys/utsname.h>
#include <string.h>
#include <sys/time.h>
#include <sys/resource.h>
#include <regex.h>
#include <glib.h>
#include <sys/inotify.h>
#include "atop.h"
#include "acctproc.h"
#include "ifprop.h"
#include "photoproc.h"
#include "photosyst.h"
#include "cgroups.h"
#include "showgeneric.h"
#include "showlinux.h"
#include "parseable.h"
#include "json.h"
#include "gpucom.h"
#include "netatop.h"
#define allflags "ab:cde:fghijklmnopqrstuvwxyz:123456789ABCDEFGHIJ:KL:MNOP:QRSTUVWXYZ"
#define MAXFL 84 /* maximum number of command-line flags */
/*
** declaration of global variables
*/
struct utsname utsname;
int utsnodenamelen;
time_t pretime; /* timing info */
time_t curtime; /* timing info */
unsigned long interval = 10;
unsigned long sampcnt;
char screen;
int fdinotify = -1; /* inotify fd for twin mode */
pid_t twinpid; /* PID of lower half for twin mode */
char twindir[RAWNAMESZ] = "/tmp";
int linelen = 80;
char acctreason; /* accounting not active (return val) */
char irawname[RAWNAMESZ];
char orawname[RAWNAMESZ];
char rawreadflag;
char idnamesuppress; /* suppress UID/GID to name translation */
time_t begintime, endtime, cursortime; // epoch or time in day
char flaglist[MAXFL];
char deviatonly = 1;
char usecolors = 1; /* boolean: colors for high occupation */
char threadview = 0; /* boolean: show individual threads */
char calcpss = 0; /* boolean: read/calculate process PSS */
char getwchan = 0; /* boolean: obtain wchan string */
char rmspaces = 0; /* boolean: remove spaces from command */
/* name in case of parsable output */
char displaymode = 'T'; /* 'T' = text, 'D' = draw */
char barmono = 0; /* boolean: bar without categories? */
/* name in case of parseable output */
char prependenv = 0; /* boolean: prepend selected */
/* environment variables to cmdline */
regex_t envregex;
unsigned short hertz;
unsigned int pidwidth;
unsigned int pagesize;
unsigned int nrgpus;
int osrel;
int osvers;
int ossub;
extern GHashTable *ghash_net;
int supportflags; /* supported features */
char **argvp;
struct handler handlers[MAXHANDLERS];
int numhandlers;
/*
** argument values
*/
static char awaittrigger; /* boolean: awaiting trigger */
static unsigned int nsamples = 0xffffffff;
static char midnightflag;
static char rawwriteflag;
static char parseoutflag;
static char jsonoutflag;
static char screenoutflag;
char twinmodeflag;
/*
** interpretation of defaults-file /etc/atoprc and $HOME/.atop
*/
static void readrc(char *, int);
static void do_interval(char *, char *);
static void do_linelength(char *, char *);
static struct {
char *tag;
void (*func)(char *, char *);
int sysonly;
} manrc[] = {
{ "flags", do_flags, 0, },
{ "twindir", do_twindir, 0, },
{ "interval", do_interval, 0, },
{ "linelen", do_linelength, 0, },
{ "username", do_username, 0, },
{ "procname", do_procname, 0, },
{ "maxlinecpu", do_maxcpu, 0, },
{ "maxlinegpu", do_maxgpu, 0, },
{ "maxlinedisk", do_maxdisk, 0, },
{ "maxlinemdd", do_maxmdd, 0, },
{ "maxlinelvm", do_maxlvm, 0, },
{ "maxlineintf", do_maxintf, 0, },
{ "maxlineifb", do_maxifb, 0, },
{ "maxlinenfsm", do_maxnfsm, 0, },
{ "maxlinecont", do_maxcont, 0, },
{ "maxlinenuma", do_maxnuma, 0, },
{ "maxlinellc", do_maxllc, 0, },
{ "colorinfo", do_colinfo, 0, },
{ "coloralmost", do_colalmost, 0, },
{ "colorcritical", do_colcrit, 0, },
{ "colorthread", do_colthread, 0, },
{ "ownallcpuline", do_ownallcpuline, 0, },
{ "ownonecpuline", do_ownindivcpuline, 0, },
{ "owncplline", do_owncplline, 0, },
{ "ownmemline", do_ownmemline, 0, },
{ "ownswpline", do_ownswpline, 0, },
{ "ownpagline", do_ownpagline, 0, },
{ "ownmemnumaline", do_ownmemnumaline, 0, },
{ "ownnumacpuline", do_owncpunumaline, 0, },
{ "ownllcline", do_ownllcline, 0, },
{ "owndskline", do_owndskline, 0, },
{ "ownnettrline", do_ownnettransportline, 0, },
{ "ownnetnetline", do_ownnetnetline, 0, },
{ "ownnetifline", do_ownnetinterfaceline, 0, },
{ "ownifbline", do_owninfinibandline, 0, },
{ "ownprocline", do_ownprocline, 0, },
{ "ownsysprcline", do_ownsysprcline, 0, },
{ "owndskline", do_owndskline, 0, },
{ "cpucritperc", do_cpucritperc, 0, },
{ "gpucritperc", do_gpucritperc, 0, },
{ "memcritperc", do_memcritperc, 0, },
{ "swpcritperc", do_swpcritperc, 0, },
{ "dskcritperc", do_dskcritperc, 0, },
{ "netcritperc", do_netcritperc, 0, },
{ "swoutcritsec", do_swoutcritsec, 0, },
{ "almostcrit", do_almostcrit, 0, },
{ "atopsarflags", do_atopsarflags, 0, },
{ "perfevents", do_perfevents, 0, },
{ "pacctdir", do_pacctdir, 1, },
};
/*
** internal prototypes
*/
static void engine(void);
static void twinprepare(void);
static void twinclean(void);
int
main(int argc, char *argv[])
{
register int i;
int c;
char *p;
struct rlimit rlim;
/*
** since privileged actions will be done later on, at this stage
** the root-privileges are dropped by switching effective user-id
** to real user-id (security reasons)
*/
if (! droprootprivs() )
{
fprintf(stderr, "not possible to drop root privs\n");
exit(42);
}
/*
** preserve command arguments to allow restart of other version
*/
argvp = argv;
/*
** read defaults-files /etc/atoprc en $HOME/.atoprc (if any)
*/
readrc("/etc/atoprc", 1);
if ( (p = getenv("HOME")) )
{
char path[1024];
snprintf(path, sizeof path, "%s/.atoprc", p);
readrc(path, 0);
}
/*
** check if we are supposed to behave as 'atopsar'
** i.e. system statistics only
*/
if ( (p = strrchr(argv[0], '/')))
p++;
else
p = argv[0];
if ( memcmp(p, "atopsar", 7) == 0)
return atopsar(argc, argv);
/*
** interpret command-line arguments & flags
*/
if (argc > 1)
{
/*
** gather all flags for visualization-functions
**
** generic flags will be handled here;
** unrecognized flags are passed to the print-routines
*/
i = 0;
while (i < MAXFL-1 && (c=getopt(argc, argv, allflags)) != EOF)
{
switch (c)
{
case '?': /* usage wanted ? */
prusage(argv[0]);
break;
case 'V': /* version wanted ? */
printf("%s\n", getstrvers());
exit(0);
case 'w': /* writing of raw data ? */
if (optind >= argc)
prusage(argv[0]);
strncpy(orawname, argv[optind++], RAWNAMESZ-1);
if (!rawwriteflag)
{
rawwriteflag++;
handlers[numhandlers++].handle_sample = rawwrite;
}
break;
case 'r': /* reading of raw data ? */
if (optind < argc)
{
if (*(argv[optind]) == '-')
{
if (strlen(argv[optind]) == 1)
{
strcpy(irawname, "/dev/stdin");
optind++;
}
}
else
{
strncpy(irawname, argv[optind],
RAWNAMESZ-1);
optind++;
}
}
rawreadflag++;
break;
case 't': /* twin mode ? */
// optional absolute path name of directory?
if (optind < argc)
{
if (*(argv[optind]) == '/')
{
strncpy(twindir, argv[optind],
RAWNAMESZ-1);
optind++;
}
}
twinmodeflag++;
break;
case 'B': /* bar graphs ? */
displaymode = 'D';
break;
case 'H': /* bar graphs ? */
barmono = 1;
break;
case 'S': /* midnight limit ? */
midnightflag++;
break;
case 'I': /* suppress ID translation ? */
idnamesuppress++;
break;
case 'b': /* begin time ? */
if ( !getbranchtime(optarg, &begintime) )
prusage(argv[0]);
break;
case 'e': /* end time ? */
if ( !getbranchtime(optarg, &endtime) )
prusage(argv[0]);
break;
case 'P': /* parsable output? */
if ( !parsedef(optarg) )
prusage(argv[0]);
if (!parseoutflag)
{
parseoutflag++;
handlers[numhandlers++].handle_sample = parseout;
}
break;
case 'J': /* json output? */
if ( !jsondef(optarg) )
prusage(argv[0]);
if (!jsonoutflag)
{
jsonoutflag++;
handlers[numhandlers++].handle_sample = jsonout;
}
break;
case 'L': /* line length */
if ( !numeric(optarg) )
prusage(argv[0]);
linelen = atoi(optarg);
break;
case MALLACTIVE: /* all processes/cgroups ? */
deviatonly = 0;
break;
case MCALCPSS: /* calculate PSS per sample ? */
if (rawreadflag)
{
fprintf(stderr,
"PSIZE gathering depends on rawfile\n");
sleep(3);
break;
}
calcpss = 1;
if (!rootprivs())
{
fprintf(stderr,
"PSIZE gathering only for own "
"processes\n");
sleep(3);
}
break;
case MGETWCHAN: /* obtain wchan string? */
getwchan = 1;
break;
case MRMSPACES: /* remove spaces from command */
rmspaces = 1;
break;
case 'z': /* prepend regex matching environment variables */
if (regcomp(&envregex, optarg, REG_NOSUB|REG_EXTENDED))
{
fprintf(stderr, "Invalid environment regular expression!");
prusage(argv[0]);
}
prependenv = 1;
break;
default: /* gather other flags */
flaglist[i++] = c;
}
/*
** check if this flag explicitly refers to
** generic (screen) output
*/
if (strchr("gmdnsevcoBGaCMDNEAupjSf", c))
screenoutflag++;
}
/*
** get optional interval-value and optional number of samples
*/
if (optind < argc && optind < MAXFL)
{
if (!numeric(argv[optind]))
prusage(argv[0]);
interval = atoi(argv[optind]);
optind++;
if (optind < argc)
{
if (!numeric(argv[optind]) )
prusage(argv[0]);
if ( (nsamples = atoi(argv[optind])) < 1)
prusage(argv[0]);
}
}
}
/*
** verify if the generic handler has to be installed as default
** (no other handler choosen) or if the generic screen handler
** has to be added due to an explicit flag
*/
if (numhandlers == 0 || screenoutflag)
handlers[numhandlers++].handle_sample = generic_samp;
/*
** determine the name of this node (without domain-name)
** and the kernel-version
*/
(void) uname(&utsname);
if ( (p = strchr(utsname.nodename, '.')) )
*p = '\0';
utsnodenamelen = strlen(utsname.nodename);
sscanf(utsname.release, "%d.%d.%d", &osrel, &osvers, &ossub);
/*
** determine the clock rate and memory page size for this machine
*/
hertz = sysconf(_SC_CLK_TCK);
pagesize = sysconf(_SC_PAGESIZE);
pidwidth = getpidwidth();
/*
** check if twin mode wanted with two atop processes:
**
** - lower half: gather statistics and write to raw file
** - upper half: read statistics and present to user
**
** consistency checks
*/
if (twinmodeflag)
twinprepare();
/*
** check if raw data from a file must be viewed
*/
if (rawreadflag)
{
rawread();
cleanstop(0);
}
/*
** be sure to be leader of an own process group when
** running as a daemon (or at least: when not interactive);
** needed for systemd
*/
if (rawwriteflag)
(void) setpgid(0, 0);
/*
** determine start-time for gathering current statistics
*/
curtime = getboot() / hertz;
/*
** catch signals for proper close-down
*/
signal(SIGHUP, cleanstop);
signal(SIGTERM, cleanstop);
/*
** regain the root-privileges that we dropped at the beginning
** to do some privileged work
*/
regainrootprivs();
/*
** lock ATOP in memory to get reliable samples (also when
** memory is low and swapping is going on);
** ignored if not running under superuser privileges!
*/
rlim.rlim_cur = RLIM_INFINITY;
rlim.rlim_max = RLIM_INFINITY;
if (setrlimit(RLIMIT_MEMLOCK, &rlim) == 0)
(void) mlockall(MCL_CURRENT|MCL_FUTURE);
/*
** increment CPU scheduling-priority to get reliable samples (also
** during heavy CPU load);
** ignored if not running under superuser privileges!
*/
if ( nice(-20) == -1)
;
set_oom_score_adj();
/*
** switch-on the process-accounting mechanism to register the
** (remaining) resource-usage by processes which have finished
*/
acctreason = acctswon();
/*
** determine properties (like speed) of all interfaces
*/
initifprop();
/*
** open socket to the IP layer to issue getsockopt() calls later on
*/
netatop_ipopen();
/*
** since privileged activities are finished now, there is no
** need to keep running under root-privileges, so switch
** effective user-id to real user-id
*/
if (! droprootprivs() )
mcleanstop(42, "failed to drop root privs\n");
/*
** determine if cgroups v2 is supported
*/
cgroupv2support();
/*
** start the engine now .....
*/
engine();
cleanstop(0);
return 0; /* never reached */
}
/*
** The engine() drives the main-loop of the program
*/
static void
engine(void)
{
struct sigaction sigact;
static time_t timelimit;
/*
** reserve space for cgroup-level statistics
*/
static struct cgchainer *devcstat;
int ncgroups = 0;
int npids = 0;
int i;
/*
** reserve space for system-level statistics
*/
static struct sstat *cursstat; /* current */
static struct sstat *presstat; /* previous */
static struct sstat *devsstat; /* deviation */
static struct sstat *hlpsstat;
/*
** reserve space for task-level statistics
*/
static struct tstat *curtpres; /* current present list */
static unsigned long curtlen; /* size of present list */
struct tstat *curpexit; /* exited process list */
static struct devtstat devtstat; /* deviation info */
unsigned long ntaskpres; /* number of tasks present */
unsigned long nprocexit; /* number of exited procs */
unsigned long nprocexitnet; /* number of exited procs */
/* via netatopd daemon */
unsigned long noverflow;
int nrgpuproc=0, /* number of GPU processes */
gpupending=0; /* boolean: request sent */
struct gpupidstat *gp = NULL;
/*
** initialization: allocate required memory dynamically
*/
cursstat = calloc(1, sizeof(struct sstat));
presstat = calloc(1, sizeof(struct sstat));
devsstat = calloc(1, sizeof(struct sstat));
ptrverify(cursstat, "Malloc failed for current sysstats\n");
ptrverify(presstat, "Malloc failed for prev sysstats\n");
ptrverify(devsstat, "Malloc failed for deviate sysstats\n");
/*
** install the signal-handler for ALARM, USR1 and USR2 (triggers
* for the next sample)
*/
memset(&sigact, 0, sizeof sigact);
sigact.sa_handler = getusr1;
sigaction(SIGUSR1, &sigact, (struct sigaction *)0);
memset(&sigact, 0, sizeof sigact);
sigact.sa_handler = getusr2;
sigaction(SIGUSR2, &sigact, (struct sigaction *)0);
memset(&sigact, 0, sizeof sigact);
sigact.sa_handler = getalarm;
sigaction(SIGALRM, &sigact, (struct sigaction *)0);
if (interval > 0)
alarm(interval);
if (midnightflag)
{
time_t timenow = time(0);
struct tm *tp = localtime(&timenow);
tp->tm_hour = 23;
tp->tm_min = 59;
tp->tm_sec = 59;
timelimit = mktime(tp);
}
/*
** open socket to the atopgpud daemon for GPU statistics
*/
nrgpus = gpud_init();
if (nrgpus)
supportflags |= GPUSTAT;
/*
** MAIN-LOOP:
** - Wait for the requested number of seconds or for other trigger
**
** - System-level counters
** get current counters
** calculate the differences with the previous sample
**
** - Process-level counters
** get current counters from running & exited processes
** calculate the differences with the previous sample
**
** - Call the print-function to visualize the differences
*/
for (sampcnt=0; sampcnt < nsamples; sampcnt++)
{
char lastcmd = ' ';
/*
** if the limit-flag is specified:
** check if the next sample is expected before midnight;
** if not, stop atop now
*/
if (midnightflag && (curtime+interval) > timelimit)
break;
/*
** wait for alarm-signal to arrive (except first sample)
** or wait for SIGUSR1/SIGUSR2
*/
if (sampcnt > 0 && awaittrigger)
pause();
awaittrigger = 1;
/*
** gather time info for this sample
*/
pretime = curtime;
curtime = time(0); /* seconds since 1-1-1970 */
/*
** send request for statistics to atopgpud
*/
if (nrgpus)
gpupending = gpud_statrequest();
/*
** take a snapshot of the current system-level metrics
** and calculate the deviations (i.e. calculate the activity
** during the last sample)
*/
hlpsstat = cursstat; /* swap current/prev. stats */
cursstat = presstat;
presstat = hlpsstat;
photosyst(cursstat); /* obtain new system-level counters */
/*
** take a snapshot of the current cgroup-level metrics
** when cgroups v2 supported
*/
if ( (supportflags&CGROUPV2) )
photocgroup();
/*
** receive and parse response from atopgpud
*/
if (nrgpus && gpupending)
{
nrgpuproc = gpud_statresponse(nrgpus, cursstat->gpu.gpu, &gp);
gpupending = 0;
// connection lost or timeout on receive?
if (nrgpuproc == -1)
{
int ng;
// try to reconnect
ng = gpud_init();
if (ng != nrgpus) // no success
nrgpus = 0;
if (nrgpus)
{
// request for stats again
if (gpud_statrequest())
{
// receive stats response
nrgpuproc = gpud_statresponse(nrgpus,
cursstat->gpu.gpu, &gp);
// persistent failure?
if (nrgpuproc == -1)
nrgpus = 0;
}
}
}
cursstat->gpu.nrgpus = nrgpus;
}
deviatsyst(cursstat, presstat, devsstat,
curtime-pretime > 0 ? curtime-pretime : 1);
/*
** take a snapshot of the current task-level statistics
** and calculate the deviations (i.e. calculate the activity
** during the last sample)
**
** first register active tasks
*/
curtpres = NULL;
do
{
curtlen = counttasks(); // worst-case value
curtpres = realloc(curtpres,
curtlen * sizeof(struct tstat));
ptrverify(curtpres, "Malloc failed for %lu tstats\n",
curtlen);
memset(curtpres, 0, curtlen * sizeof(struct tstat));
}
while ( (ntaskpres = photoproc(curtpres, curtlen)) == curtlen);
/*
** register processes that exited during last sample;
** first determine how many processes exited
**
** the number of exited processes is limited to avoid
** that atop explodes in memory and introduces OOM killing
*/
nprocexit = acctprocnt(); /* number of exited processes */
if (nprocexit > MAXACCTPROCS)
{
noverflow = nprocexit - MAXACCTPROCS;
nprocexit = MAXACCTPROCS;
}
else
noverflow = 0;
/*
** determine how many processes have been exited
** for the netatop module (only processes that have
** used the network)
*/
if (nprocexit > 0 && (supportflags & NETATOPD))
nprocexitnet = netatop_exitstore();
else
nprocexitnet = 0;
/*
** reserve space for the exited processes and read them
*/
if (nprocexit > 0)
{
curpexit = malloc(nprocexit * sizeof(struct tstat));
ptrverify(curpexit,
"Malloc failed for %lu exited processes\n",
nprocexit);
memset(curpexit, 0, nprocexit * sizeof(struct tstat));
nprocexit = acctphotoproc(curpexit, nprocexit);
/*
** reposition offset in accounting file when not
** all exited processes have been read (i.e. skip
** those processes)
*/
if (noverflow)
acctrepos(noverflow);
}
else
{
curpexit = NULL;
}
/*
** merge GPU per-process stats with other per-process stats
*/
if (nrgpus && nrgpuproc)
gpumergeproc(curtpres, ntaskpres,
curpexit, nprocexit,
gp, nrgpuproc);
/*
** calculate process-level deviations
*/
deviattask(curtpres, ntaskpres, curpexit, nprocexit,
&devtstat, devsstat);
if (supportflags & NETATOPBPF)
{
g_hash_table_destroy(ghash_net);
ghash_net = NULL;
}
/*
** calculate cgroup-level v2 deviations
**
** allocation and deallocation of structs
** is arranged at a lower level
*/
if ( (supportflags&CGROUPV2) )
ncgroups = deviatcgroup(&devcstat, &npids);
/*
** activate the installed print function to visualize
** the deviations
*/
for (i=0; handlers[i].handle_sample; i++)
{