procps/vmstat.c

963 lines
32 KiB
C
Raw Normal View History

/*
* old: "Copyright 1994 by Henry Ware <al172@yfn.ysu.edu>. Copyleft same year."
* most code copyright 2002 Albert Cahalan
*
* 27/05/2003 (Fabian Frederick) : Add unit conversion + interface
* Export proc/stat access to libproc
* Adapt vmstat helpfile
* 31/05/2003 (Fabian) : Add diskstat support (/libproc)
* June 2003 (Fabian) : -S <x> -s & -s -S <x> patch
* June 2003 (Fabian) : Adding diskstat against 3.1.9, slabinfo
* patching 'header' in disk & slab
* July 2003 (Fabian) : Adding disk partition output
* Adding disk table
* Syncing help / usage
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
* This library 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
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with this library; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
2002-11-28 17:07:59 +08:00
2002-02-02 06:47:29 +08:00
#include <assert.h>
#include <ctype.h>
#include <dirent.h>
#include <errno.h>
2002-02-02 06:47:29 +08:00
#include <fcntl.h>
#include <getopt.h>
2002-02-02 06:47:29 +08:00
#include <limits.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <sys/ioctl.h>
#include <sys/types.h>
#include <termios.h>
2002-02-02 06:47:29 +08:00
#include <unistd.h>
#include <time.h>
2002-02-02 06:47:29 +08:00
#include "c.h"
#include "fileutils.h"
#include "nls.h"
#include "strutils.h"
2002-11-26 05:12:25 +08:00
#include "proc/sysinfo.h"
#include <proc/vmstat.h>
#include <proc/readstat.h>
#include <proc/meminfo.h>
#include <proc/diskstat.h>
#include <proc/slab.h>
2003-06-09 01:28:06 +08:00
2003-08-11 07:40:41 +08:00
#define UNIT_B 1
#define UNIT_k 1000
#define UNIT_K 1024
#define UNIT_m 1000000
#define UNIT_M 1048576
static unsigned long dataUnit = UNIT_K;
static char szDataUnit[3] = "K";
2003-08-11 07:40:41 +08:00
#define VMSTAT 0
#define DISKSTAT 0x00000001
#define VMSUMSTAT 0x00000002
#define SLABSTAT 0x00000004
#define PARTITIONSTAT 0x00000008
#define DISKSUMSTAT 0x00000010
2003-06-09 01:28:06 +08:00
static int statMode = VMSTAT;
2002-11-26 05:12:25 +08:00
/* "-a" means "show active/inactive" */
static int a_option;
2002-11-09 07:29:34 +08:00
/* "-w" means "wide output" */
static int w_option;
/* "-t" means "show timestamp" */
static int t_option;
2002-11-08 15:42:54 +08:00
static unsigned sleep_time = 1;
static int infinite_updates = 0;
static unsigned long num_updates =1;
/* window height */
static unsigned int height;
static unsigned int moreheaders = TRUE;
2002-02-02 06:47:29 +08:00
static void __attribute__ ((__noreturn__))
usage(FILE * out)
{
fputs(USAGE_HEADER, out);
fprintf(out,
_(" %s [options] [delay [count]]\n"),
program_invocation_short_name);
fputs(USAGE_OPTIONS, out);
fputs(_(" -a, --active active/inactive memory\n"), out);
fputs(_(" -f, --forks number of forks since boot\n"), out);
fputs(_(" -m, --slabs slabinfo\n"), out);
fputs(_(" -n, --one-header do not redisplay header\n"), out);
fputs(_(" -s, --stats event counter statistics\n"), out);
fputs(_(" -d, --disk disk statistics\n"), out);
fputs(_(" -D, --disk-sum summarize disk statistics\n"), out);
fputs(_(" -p, --partition <dev> partition specific statistics\n"), out);
fputs(_(" -S, --unit <char> define display unit\n"), out);
fputs(_(" -w, --wide wide output\n"), out);
fputs(_(" -t, --timestamp show timestamp\n"), out);
fputs(USAGE_SEPARATOR, out);
fputs(USAGE_HELP, out);
fputs(USAGE_VERSION, out);
fprintf(out, USAGE_MAN_TAIL("vmstat(8)"));
exit(out == stderr ? EXIT_FAILURE : EXIT_SUCCESS);
2002-02-02 06:47:29 +08:00
}
static void new_header(void)
{
struct tm *tm_ptr;
time_t the_time;
char timebuf[32];
/* Translation Hint: Translating folloging header & fields
* that follow (marked with max x chars) might not work,
* unless manual page is translated as well. */
const char *header =
_("procs -----------memory---------- ---swap-- -----io---- -system-- ------cpu-----");
const char *wide_header =
_("procs -----------------------memory---------------------- ---swap-- -----io---- -system-- --------cpu--------");
const char *timestamp_header = _(" -----timestamp-----");
const char format[] =
"%2s %2s %6s %6s %6s %6s %4s %4s %5s %5s %4s %4s %2s %2s %2s %2s %2s";
const char wide_format[] =
"%2s %2s %12s %12s %12s %12s %4s %4s %5s %5s %4s %4s %3s %3s %3s %3s %3s";
printf("%s", w_option ? wide_header : header);
if (t_option) {
printf("%s", timestamp_header);
}
printf("\n");
printf(
w_option ? wide_format : format,
/* Translation Hint: max 2 chars */
_("r"),
/* Translation Hint: max 2 chars */
_("b"),
/* Translation Hint: max 6 chars */
_("swpd"),
/* Translation Hint: max 6 chars */
_("free"),
/* Translation Hint: max 6 chars */
a_option ? _("inact") :
/* Translation Hint: max 6 chars */
_("buff"),
/* Translation Hint: max 6 chars */
a_option ? _("active") :
/* Translation Hint: max 6 chars */
_("cache"),
/* Translation Hint: max 4 chars */
_("si"),
/* Translation Hint: max 4 chars */
_("so"),
/* Translation Hint: max 5 chars */
_("bi"),
/* Translation Hint: max 5 chars */
_("bo"),
/* Translation Hint: max 4 chars */
_("in"),
/* Translation Hint: max 4 chars */
_("cs"),
/* Translation Hint: max 2 chars */
_("us"),
/* Translation Hint: max 2 chars */
_("sy"),
/* Translation Hint: max 2 chars */
_("id"),
/* Translation Hint: max 2 chars */
_("wa"),
/* Translation Hint: max 2 chars */
_("st"));
if (t_option) {
(void) time( &the_time );
tm_ptr = localtime( &the_time );
if (strftime(timebuf, sizeof(timebuf), "%Z", tm_ptr)) {
timebuf[strlen(timestamp_header) - 1] = '\0';
} else {
timebuf[0] = '\0';
}
printf(" %*s", (int)(strlen(timestamp_header) - 1), timebuf);
}
printf("\n");
2002-02-02 06:47:29 +08:00
}
2002-11-08 15:42:54 +08:00
static unsigned long unitConvert(unsigned long size)
{
float cvSize;
cvSize = (float)size / dataUnit * ((statMode == SLABSTAT) ? 1 : 1024);
return ((unsigned long)cvSize);
}
2002-11-09 07:29:34 +08:00
static void new_format(void)
{
const char format[] =
"%2u %2u %6lu %6lu %6lu %6lu %4u %4u %5u %5u %4u %4u %2u %2u %2u %2u %2u";
const char wide_format[] =
"%2u %2u %12lu %12lu %12lu %12lu %4u %4u %5u %5u %4u %4u %3u %3u %3u %3u %3u";
unsigned int tog = 0; /* toggle switch for cleaner code */
unsigned int i;
long hz;
jiff cpu_use[2], cpu_sys[2], cpu_idl[2], cpu_iow[2], cpu_sto[2];
jiff duse, dsys, didl, diow, dstl, Div, divo2;
unsigned long pgpgin[2], pgpgout[2], pswpin[2] = {0,0}, pswpout[2];
unsigned int intr[2], ctxt[2];
unsigned int sleep_half;
unsigned long kb_per_page = sysconf(_SC_PAGESIZE) / 1024ul;
int debt = 0; /* handle idle ticks running backwards */
struct tm *tm_ptr;
time_t the_time;
char timebuf[32];
struct procps_vmstat *vm_info;
struct procps_stat *sys_info;
struct procps_meminfo *mem_info;
sleep_half = (sleep_time / 2);
hz = procps_hertz_get();
new_header();
if (procps_vmstat_new(&vm_info) < 0)
xerrx(EXIT_FAILURE,
_("Unable to create vmstat structure"));
if (procps_vmstat_read(vm_info) < 0)
xerrx(EXIT_FAILURE,
_("Unable to read vmstat information"));
if (procps_stat_new(&sys_info) < 0)
xerrx(EXIT_FAILURE,
_("Unable to create system stat structure"));
if (procps_stat_read(sys_info, 0) < 0)
xerrx(EXIT_FAILURE,
_("Unable to read system stat information"));
if (procps_meminfo_new(&mem_info) < 0)
xerrx(EXIT_FAILURE, _("Unable to create meminfo structure"));
if (procps_meminfo_read(mem_info) < 0)
xerrx(EXIT_FAILURE, _("Unable to read meminfo information"));
if (t_option) {
(void) time( &the_time );
tm_ptr = localtime( &the_time );
strftime(timebuf, sizeof(timebuf), "%Y-%m-%d %H:%M:%S", tm_ptr);
}
/* Do the initial fill */
library: readstat redesigned using 'stack' vs. 'chain' In addition to that text shown below the line which is common to several commit messages, this patch contains several minor changes with lessor impact upon the API: . A call to procps_stat_read_jiffs() has been added to those jiffs functions carrying the 'fill' nomenclature to parallel like functions in some of our other files. . The #include header files are ordered alphabetically now, with all those <sys/??> types separately grouped. . Standard copyright boilerplate was added in .c file. . The header file follows the conventions of indenting (by 4 spaces) those parameters too lengthy for 1 line. ------------------------------------------------------ . The former 'chains' have now become 'stacks' without the 'next' pointer in each result struct. The pointers initially seemed to offer some flexibility with memory allocations and benefits for the library access logic. However, user access was always via displacement and a a statically allocated chain was cumbersome to define. . An enumerator ending in '_noop' will no longer serve as a fencepost delimiter. Rather, it has become a much more important and flexible user oriented tool. Adding one or more such 'items' in any items list passed into the library becomes the means of extending the 'stack' to also include user (not just library) data. Any such data is guaranteed to never be altered by the library. . Anticipating PID support, where many different types must be represented in a result structure, we'll adopt a common naming standard. And, while not every results structure currently needs to reflect disparate types a union will be employed so the same dot qualifier ('.') can be used consistently when accessing all such data. Signed-off-by: Jim Warner <james.warner@comcast.net>
2015-07-21 13:00:00 +08:00
cpu_use[tog] = procps_stat_cpu_get(sys_info, PROCPS_CPU_USER) +
procps_stat_cpu_get(sys_info, PROCPS_CPU_NICE);
cpu_sys[tog] = procps_stat_cpu_get(sys_info, PROCPS_CPU_SYSTEM);
procps_stat_cpu_get(sys_info, PROCPS_CPU_IRQ) +
procps_stat_cpu_get(sys_info, PROCPS_CPU_SIRQ);
cpu_idl[tog] = procps_stat_cpu_get(sys_info, PROCPS_CPU_IDLE);
cpu_iow[tog] = procps_stat_cpu_get(sys_info, PROCPS_CPU_IOWAIT);
cpu_sto[tog] = procps_stat_cpu_get(sys_info, PROCPS_CPU_STOLEN);
intr[tog] = procps_stat_sys_get(sys_info, PROCPS_STAT_INTR);
ctxt[tog] = procps_stat_sys_get(sys_info, PROCPS_STAT_CTXT);
pgpgin[tog] = procps_vmstat_get(vm_info, PROCPS_VMSTAT_PGPGIN);
pgpgout[tog] = procps_vmstat_get(vm_info, PROCPS_VMSTAT_PGPGOUT);
pswpin[tog] = procps_vmstat_get(vm_info, PROCPS_VMSTAT_PSWPIN);
pswpout[tog] = procps_vmstat_get(vm_info, PROCPS_VMSTAT_PSWPOUT);
Div = cpu_use[tog] + cpu_sys[tog] + cpu_idl[tog] + cpu_iow[tog] + cpu_sto[tog];
if (!Div) {
Div = 1;
cpu_idl[tog] = 1;
}
divo2 = Div / 2UL;
printf(w_option ? wide_format : format,
library: readstat redesigned using 'stack' vs. 'chain' In addition to that text shown below the line which is common to several commit messages, this patch contains several minor changes with lessor impact upon the API: . A call to procps_stat_read_jiffs() has been added to those jiffs functions carrying the 'fill' nomenclature to parallel like functions in some of our other files. . The #include header files are ordered alphabetically now, with all those <sys/??> types separately grouped. . Standard copyright boilerplate was added in .c file. . The header file follows the conventions of indenting (by 4 spaces) those parameters too lengthy for 1 line. ------------------------------------------------------ . The former 'chains' have now become 'stacks' without the 'next' pointer in each result struct. The pointers initially seemed to offer some flexibility with memory allocations and benefits for the library access logic. However, user access was always via displacement and a a statically allocated chain was cumbersome to define. . An enumerator ending in '_noop' will no longer serve as a fencepost delimiter. Rather, it has become a much more important and flexible user oriented tool. Adding one or more such 'items' in any items list passed into the library becomes the means of extending the 'stack' to also include user (not just library) data. Any such data is guaranteed to never be altered by the library. . Anticipating PID support, where many different types must be represented in a result structure, we'll adopt a common naming standard. And, while not every results structure currently needs to reflect disparate types a union will be employed so the same dot qualifier ('.') can be used consistently when accessing all such data. Signed-off-by: Jim Warner <james.warner@comcast.net>
2015-07-21 13:00:00 +08:00
procps_stat_sys_get(sys_info, PROCPS_STAT_PROCS_RUN),
procps_stat_sys_get(sys_info, PROCPS_STAT_PROCS_BLK),
unitConvert(procps_meminfo_get(mem_info, PROCPS_SWAP_USED)),
unitConvert(procps_meminfo_get(mem_info, PROCPS_MEM_FREE)),
unitConvert(procps_meminfo_get(mem_info, (a_option?PROCPS_MEM_INACTIVE:PROCPS_MEM_BUFFERS))),
unitConvert(procps_meminfo_get(mem_info, a_option?PROCPS_MEM_ACTIVE:PROCPS_MEM_CACHED)),
(unsigned)( (unitConvert(procps_vmstat_get(vm_info, PROCPS_VMSTAT_PSWPIN) * kb_per_page) * hz + divo2) / Div ),
(unsigned)( (unitConvert(procps_vmstat_get(vm_info, PROCPS_VMSTAT_PSWPOUT) * kb_per_page) * hz + divo2) / Div ),
(unsigned)( (procps_vmstat_get(vm_info, PROCPS_VMSTAT_PGPGIN) * hz + divo2) / Div ),
(unsigned)( (procps_vmstat_get(vm_info, PROCPS_VMSTAT_PGPGOUT) * hz + divo2) / Div ),
(unsigned)( (intr[tog] * hz + divo2) / Div ),
(unsigned)( (ctxt[tog] * hz + divo2) / Div ),
(unsigned)( (100*cpu_use[tog] + divo2) / Div ),
(unsigned)( (100*cpu_sys[tog] + divo2) / Div ),
(unsigned)( (100*cpu_idl[tog] + divo2) / Div ),
(unsigned)( (100*cpu_iow[tog] + divo2) / Div ),
(unsigned)( (100*cpu_sto[tog] + divo2) / Div )
);
if (t_option) {
printf(" %s", timebuf);
}
printf("\n");
/* main loop */
for (i = 1; infinite_updates || i < num_updates; i++) {
sleep(sleep_time);
if (moreheaders && ((i % height) == 0))
new_header();
tog = !tog;
if (procps_stat_read(sys_info,0) < 0)
xerrx(EXIT_FAILURE,
_("Unable to read system stat information"));
if (procps_vmstat_read(vm_info) < 0)
xerrx(EXIT_FAILURE,
_("Unable to read vmstat information"));
library: readstat redesigned using 'stack' vs. 'chain' In addition to that text shown below the line which is common to several commit messages, this patch contains several minor changes with lessor impact upon the API: . A call to procps_stat_read_jiffs() has been added to those jiffs functions carrying the 'fill' nomenclature to parallel like functions in some of our other files. . The #include header files are ordered alphabetically now, with all those <sys/??> types separately grouped. . Standard copyright boilerplate was added in .c file. . The header file follows the conventions of indenting (by 4 spaces) those parameters too lengthy for 1 line. ------------------------------------------------------ . The former 'chains' have now become 'stacks' without the 'next' pointer in each result struct. The pointers initially seemed to offer some flexibility with memory allocations and benefits for the library access logic. However, user access was always via displacement and a a statically allocated chain was cumbersome to define. . An enumerator ending in '_noop' will no longer serve as a fencepost delimiter. Rather, it has become a much more important and flexible user oriented tool. Adding one or more such 'items' in any items list passed into the library becomes the means of extending the 'stack' to also include user (not just library) data. Any such data is guaranteed to never be altered by the library. . Anticipating PID support, where many different types must be represented in a result structure, we'll adopt a common naming standard. And, while not every results structure currently needs to reflect disparate types a union will be employed so the same dot qualifier ('.') can be used consistently when accessing all such data. Signed-off-by: Jim Warner <james.warner@comcast.net>
2015-07-21 13:00:00 +08:00
cpu_use[tog] = procps_stat_cpu_get(sys_info, PROCPS_CPU_USER) +
procps_stat_cpu_get(sys_info, PROCPS_CPU_NICE);
cpu_sys[tog] = procps_stat_cpu_get(sys_info, PROCPS_CPU_SYSTEM);
procps_stat_cpu_get(sys_info, PROCPS_CPU_IRQ) +
procps_stat_cpu_get(sys_info, PROCPS_CPU_SIRQ);
cpu_idl[tog] = procps_stat_cpu_get(sys_info, PROCPS_CPU_IDLE);
cpu_iow[tog] = procps_stat_cpu_get(sys_info, PROCPS_CPU_IOWAIT);
cpu_sto[tog] = procps_stat_cpu_get(sys_info, PROCPS_CPU_STOLEN);
intr[tog] = procps_stat_sys_get(sys_info, PROCPS_STAT_INTR);
ctxt[tog] = procps_stat_sys_get(sys_info, PROCPS_STAT_CTXT);
pgpgin[tog] = procps_vmstat_get(vm_info, PROCPS_VMSTAT_PGPGIN);
pgpgout[tog] = procps_vmstat_get(vm_info, PROCPS_VMSTAT_PGPGOUT);
pswpin[tog] = procps_vmstat_get(vm_info, PROCPS_VMSTAT_PSWPIN);
pswpout[tog] = procps_vmstat_get(vm_info, PROCPS_VMSTAT_PSWPOUT);
if (procps_meminfo_read(mem_info) < 0)
xerrx(EXIT_FAILURE,
_("Unable to read memory information"));
if (t_option) {
(void) time( &the_time );
tm_ptr = localtime( &the_time );
strftime(timebuf, sizeof(timebuf), "%Y-%m-%d %H:%M:%S", tm_ptr);
}
duse = cpu_use[tog] - cpu_use[!tog];
dsys = cpu_sys[tog] - cpu_sys[!tog];
didl = cpu_idl[tog] - cpu_idl[!tog];
diow = cpu_iow[tog] - cpu_iow[!tog];
dstl = cpu_sto[tog] - cpu_sto[!tog];
/* idle can run backwards for a moment -- kernel "feature" */
if (debt) {
didl = (int)didl + debt;
debt = 0;
}
if ((int)didl < 0) {
debt = (int)didl;
didl = 0;
}
Div = duse + dsys + didl + diow + dstl;
if (!Div) Div = 1, didl = 1;
divo2 = Div / 2UL;
printf(w_option ? wide_format : format,
library: readstat redesigned using 'stack' vs. 'chain' In addition to that text shown below the line which is common to several commit messages, this patch contains several minor changes with lessor impact upon the API: . A call to procps_stat_read_jiffs() has been added to those jiffs functions carrying the 'fill' nomenclature to parallel like functions in some of our other files. . The #include header files are ordered alphabetically now, with all those <sys/??> types separately grouped. . Standard copyright boilerplate was added in .c file. . The header file follows the conventions of indenting (by 4 spaces) those parameters too lengthy for 1 line. ------------------------------------------------------ . The former 'chains' have now become 'stacks' without the 'next' pointer in each result struct. The pointers initially seemed to offer some flexibility with memory allocations and benefits for the library access logic. However, user access was always via displacement and a a statically allocated chain was cumbersome to define. . An enumerator ending in '_noop' will no longer serve as a fencepost delimiter. Rather, it has become a much more important and flexible user oriented tool. Adding one or more such 'items' in any items list passed into the library becomes the means of extending the 'stack' to also include user (not just library) data. Any such data is guaranteed to never be altered by the library. . Anticipating PID support, where many different types must be represented in a result structure, we'll adopt a common naming standard. And, while not every results structure currently needs to reflect disparate types a union will be employed so the same dot qualifier ('.') can be used consistently when accessing all such data. Signed-off-by: Jim Warner <james.warner@comcast.net>
2015-07-21 13:00:00 +08:00
procps_stat_sys_get(sys_info, PROCPS_STAT_PROCS_RUN),
procps_stat_sys_get(sys_info, PROCPS_STAT_PROCS_BLK),
unitConvert(procps_meminfo_get(mem_info, PROCPS_SWAP_USED)),
unitConvert(procps_meminfo_get(mem_info, PROCPS_MEM_FREE)),
unitConvert(procps_meminfo_get(mem_info,
(a_option?PROCPS_MEM_INACTIVE:PROCPS_MEM_BUFFERS))),
unitConvert(procps_meminfo_get(mem_info,
(a_option?PROCPS_MEM_ACTIVE:PROCPS_MEM_CACHED))),
/*si */
(unsigned)( ( unitConvert((pswpin [tog] - pswpin [!tog])*kb_per_page)+sleep_half )/sleep_time ),
/* so */
(unsigned)( ( unitConvert((pswpout[tog] - pswpout[!tog])*kb_per_page)+sleep_half )/sleep_time ),
/* bi */
(unsigned)( ( pgpgin [tog] - pgpgin [!tog] +sleep_half )/sleep_time ),
/* bo */
(unsigned)( ( pgpgout[tog] - pgpgout[!tog] +sleep_half )/sleep_time ),
/* in */
(unsigned)( ( intr [tog] - intr [!tog] +sleep_half )/sleep_time ),
/* cs */
(unsigned)( ( ctxt [tog] - ctxt [!tog] +sleep_half )/sleep_time ),
/* us */
(unsigned)( (100*duse+divo2)/Div ),
/* sy */
(unsigned)( (100*dsys+divo2)/Div ),
/* id */
(unsigned)( (100*didl+divo2)/Div ),
/* wa */
(unsigned)( (100*diow+divo2)/Div ),
/* st */
(unsigned)( (100*dstl+divo2)/Div )
);
if (t_option) {
printf(" %s", timebuf);
}
printf("\n");
}
2004-05-05 07:29:40 +08:00
}
static void diskpartition_header(const char *partition_name)
{
printf("%-10s %10s %10s %10s %10s\n",
partition_name,
/* Translation Hint: Translating folloging disk partition
* header fields that follow (marked with max x chars) might
* not work, unless manual page is translated as well. */
/* Translation Hint: max 10 chars. The word is
* expected to be centralized, use spaces at the end
* to do that. */
_("reads "),
/* Translation Hint: max 10 chars */
_("read sectors"),
/* Translation Hint: max 10 chars. The word is
* expected to be centralized, use spaces at the end
* to do that. */
_("writes "),
/* Translation Hint: max 10 chars */
_("requested writes"));
2003-08-11 07:40:41 +08:00
}
static int diskpartition_format(const char *partition_name)
{
#define PARTGET(x) procps_diskstat_dev_get(disk_stat, (x), partid)
struct procps_diskstat *disk_stat;
const char format[] = "%20lu %10lu %10lu %10lu\n";
int i, partid;
if (procps_diskstat_new(&disk_stat) < 0)
xerr(EXIT_FAILURE,
_("Unable to create diskstat structure"));
if (procps_diskstat_read(disk_stat) < 0)
xerr(EXIT_FAILURE,
_("Unable to read diskstat"));
if ((partid = procps_diskstat_dev_getbyname(disk_stat, partition_name))
< 0)
xerrx(EXIT_FAILURE, _("Partition %s not found"), partition_name);
diskpartition_header(partition_name);
for (i=0; infinite_updates || i < num_updates ; i++) {
if (procps_diskstat_read(disk_stat) < 0)
xerr(EXIT_FAILURE,
_("Unable to read diskstat"));
if ((partid = procps_diskstat_dev_getbyname(disk_stat, partition_name))
< 0)
xerrx(EXIT_FAILURE,
_("Partition %s not found"), partition_name);
printf(format,
PARTGET(PROCPS_DISKSTAT_READS),
PARTGET(PROCPS_DISKSTAT_READ_SECTORS),
PARTGET(PROCPS_DISKSTAT_WRITES),
PARTGET(PROCPS_DISKSTAT_WRITE_SECTORS)
);
if (infinite_updates || i+1 < num_updates)
sleep(sleep_time);
}
return 0;
}
2003-08-11 07:40:41 +08:00
static void diskheader(void)
{
struct tm *tm_ptr;
time_t the_time;
char timebuf[32];
/* Translation Hint: Translating folloging header & fields
* that follow (marked with max x chars) might not work,
* unless manual page is translated as well. */
const char *header =
_("disk- ------------reads------------ ------------writes----------- -----IO------");
const char *wide_header =
_("disk- -------------------reads------------------- -------------------writes------------------ ------IO-------");
const char *timestamp_header = _(" -----timestamp-----");
const char format[] =
"%5s %6s %6s %7s %7s %6s %6s %7s %7s %6s %6s";
const char wide_format[] =
"%5s %9s %9s %11s %11s %9s %9s %11s %11s %7s %7s";
printf("%s", w_option ? wide_header : header);
if (t_option) {
printf("%s", timestamp_header);
}
printf("\n");
printf(w_option ? wide_format : format,
" ",
/* Translation Hint: max 6 chars */
_("total"),
/* Translation Hint: max 6 chars */
_("merged"),
/* Translation Hint: max 7 chars */
_("sectors"),
/* Translation Hint: max 7 chars */
_("ms"),
/* Translation Hint: max 6 chars */
_("total"),
/* Translation Hint: max 6 chars */
_("merged"),
/* Translation Hint: max 7 chars */
_("sectors"),
/* Translation Hint: max 7 chars */
_("ms"),
/* Translation Hint: max 6 chars */
_("cur"),
/* Translation Hint: max 6 chars */
_("sec"));
if (t_option) {
(void) time( &the_time );
tm_ptr = localtime( &the_time );
if (strftime(timebuf, sizeof(timebuf), "%Z", tm_ptr)) {
timebuf[strlen(timestamp_header) - 1] = '\0';
} else {
timebuf[0] = '\0';
}
printf(" %*s", (int)(strlen(timestamp_header) - 1), timebuf);
}
printf("\n");
}
2004-05-05 07:29:40 +08:00
static void diskformat(void)
{
#define DSTAT(x) procps_diskstat_dev_get(disk_stat, (x), diskid)
struct procps_diskstat *disk_stat;
int i,diskid, disk_count;
time_t the_time;
struct tm *tm_ptr;
char timebuf[32];
const char format[] = "%-5s %6lu %6lu %7lu %7lu %6lu %6lu %7lu %7lu %6lu %6lu";
miscellaneous: silence a whole bunch of clang warnings [ but most definitely not all of them by a long shot ] Reference(s): proc/diskstat.c:186:17: warning: unused variable 'is_disk' [-Wunused-variable] int retval, is_disk; ^ proc/namespace.c:110:1: warning: control may reach end of non-void function [-Wreturn-type] } ^ proc/readproc.c:1131:50: warning: address of array 'ent->d_name' will always evaluate to 'true' [-Wpointer-bo if(unlikely(unlikely(!ent) || unlikely(!ent->d_name))) return 0; ~~~~~~^~~~~~ proc/readproc.c:1158:50: warning: address of array 'ent->d_name' will always evaluate to 'true' [-Wpointer-bo if(unlikely(unlikely(!ent) || unlikely(!ent->d_name))) return 0; ~~~~~~^~~~~~ proc/sysinfo.c:45:12: warning: unused variable 'stat_fd' [-Wunused-variable] static int stat_fd = -1; ^ proc/sysinfo.c:49:12: warning: unused variable 'meminfo_fd' [-Wunused-variable] static int meminfo_fd = -1; ^ proc/sysinfo.c:51:12: warning: unused variable 'vminfo_fd' [-Wunused-variable] static int vminfo_fd = -1; ^ proc/sysinfo.c:53:12: warning: unused variable 'vm_min_free_fd' [-Wunused-variable] static int vm_min_free_fd = -1; ^ proc/uptime.c:157:12: warning: unused variable 'realseconds' [-Wunused-variable] time_t realseconds; ^ proc/uptime.c:158:16: warning: unused variable 'realtime' [-Wunused-variable] struct tm *realtime; ^ vmstat.c:574:20: warning: format specifies type 'unsigned int' but the argument has type 'unsigned long' [-Wformat] DSTAT(PROCPS_DISKSTAT_READ_TIME), ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ vmstat.c:578:20: warning: format specifies type 'unsigned int' but the argument has type 'unsigned long' [-Wformat] DSTAT(PROCPS_DISKSTAT_WRITE_TIME), ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ watch.c:230:7: warning: variable 'endptr' is uninitialized when used here [-Wuninitialized] if (*endptr == '\0') set_ansi_attribute(0); /* [m treated as [0m */ ^~~~~~ Signed-off-by: Jim Warner <james.warner@comcast.net>
2015-09-07 13:00:00 +08:00
const char wide_format[] = "%-5s %9lu %9lu %lu %lu %9lu %9lu %lu %lu %7lu %7lu";
if (procps_diskstat_new(&disk_stat) < 0)
xerr(EXIT_FAILURE,
_("Unable to create diskstat structure"));
if (!moreheaders)
diskheader();
for (i=0; infinite_updates || i < num_updates ; i++) {
if (procps_diskstat_read(disk_stat) < 0)
xerr(EXIT_FAILURE,
_("Unable to read diskstat data"));
if (t_option) {
(void) time( &the_time );
tm_ptr = localtime( &the_time );
strftime(timebuf, sizeof(timebuf), "%Y-%m-%d %H:%M:%S", tm_ptr);
}
disk_count = procps_diskstat_dev_count(disk_stat);
for (diskid = 0; diskid < disk_count; diskid++) {
if (procps_diskstat_dev_isdisk(disk_stat, diskid) != 1)
continue; /* not a disk */
if (moreheaders && ((diskid % height) == 0))
diskheader();
printf(w_option ? wide_format : format,
procps_diskstat_dev_getname(disk_stat, diskid),
DSTAT(PROCPS_DISKSTAT_READS),
DSTAT(PROCPS_DISKSTAT_READS_MERGED),
DSTAT(PROCPS_DISKSTAT_READ_SECTORS),
DSTAT(PROCPS_DISKSTAT_READ_TIME),
DSTAT(PROCPS_DISKSTAT_WRITES),
DSTAT(PROCPS_DISKSTAT_WRITES_MERGED),
DSTAT(PROCPS_DISKSTAT_WRITE_SECTORS),
DSTAT(PROCPS_DISKSTAT_WRITE_TIME),
DSTAT(PROCPS_DISKSTAT_IO_INPROGRESS) / 1000,
DSTAT(PROCPS_DISKSTAT_IO_TIME) / 1000);
if (t_option)
printf(" %s\n", timebuf);
else
printf("\n");
fflush(stdout);
}
if (infinite_updates || i+1 < num_updates)
sleep(sleep_time);
}
#undef DSTAT
2004-05-05 07:29:40 +08:00
}
static void slabheader(void)
{
printf("%-24s %6s %6s %6s %6s\n",
/* Translation Hint: Translating folloging slab fields that
* follow (marked with max x chars) might not work, unless
* manual page is translated as well. */
/* Translation Hint: max 24 chars */
_("Cache"),
/* Translation Hint: max 6 chars */
_("Num"),
/* Translation Hint: max 6 chars */
_("Total"),
/* Translation Hint: max 6 chars */
_("Size"),
/* Translation Hint: max 6 chars */
_("Pages"));
2003-06-09 01:28:06 +08:00
}
static void slabformat (void)
{
#define CHAINS_ALLOC 150
#define MAX_ITEMS (int)(sizeof(node_items) / sizeof(node_items[0]))
2015-07-04 12:59:59 +08:00
struct procps_slabinfo *slab_info;
library: slab is redesigned to use 'stack' vs. 'chain' In addition to that text shown below the line which is common to several commit messages, this patch contains several minor changes with lessor impact upon the API: . A 'read' was added to function procps_slabnode_count (but only when necessary, i.e. info->nodes_used == 0). . The #include header files are ordered alphabetically now, with all those <sys/??> types separately grouped. ------------------------------------------------------ . The former 'chains' have now become 'stacks' without the 'next' pointer in each result struct. The pointers initially seemed to offer some flexibility with memory allocations and benefits for the library access logic. However, user access was always via displacement and a a statically allocated chain was cumbersome to define. . An enumerator ending in '_noop' will no longer serve as a fencepost delimiter. Rather, it has become a much more important and flexible user oriented tool. Adding one or more such 'items' in any items list passed into the library becomes the means of extending the 'stack' to also include user (not just library) data. Any such data is guaranteed to never be altered by the library. . Anticipating PID support, where many different types must be represented in a result structure, we'll adopt a common naming standard. And, while not every results structure currently needs to reflect disparate types a union will be employed so the same dot qualifier ('.') can be used consistently when accessing all such data. Signed-off-by: Jim Warner <james.warner@comcast.net>
2015-07-21 13:00:00 +08:00
struct slabnode_stack **v, *p;
int i, j, nr_slabs;
enum slabnode_item node_items[] = {
PROCPS_SLABNODE_AOBJS, PROCPS_SLABNODE_OBJS,
PROCPS_SLABNODE_OBJ_SIZE, PROCPS_SLABNODE_OBJS_PER_SLAB,
library: slab is redesigned to use 'stack' vs. 'chain' In addition to that text shown below the line which is common to several commit messages, this patch contains several minor changes with lessor impact upon the API: . A 'read' was added to function procps_slabnode_count (but only when necessary, i.e. info->nodes_used == 0). . The #include header files are ordered alphabetically now, with all those <sys/??> types separately grouped. ------------------------------------------------------ . The former 'chains' have now become 'stacks' without the 'next' pointer in each result struct. The pointers initially seemed to offer some flexibility with memory allocations and benefits for the library access logic. However, user access was always via displacement and a a statically allocated chain was cumbersome to define. . An enumerator ending in '_noop' will no longer serve as a fencepost delimiter. Rather, it has become a much more important and flexible user oriented tool. Adding one or more such 'items' in any items list passed into the library becomes the means of extending the 'stack' to also include user (not just library) data. Any such data is guaranteed to never be altered by the library. . Anticipating PID support, where many different types must be represented in a result structure, we'll adopt a common naming standard. And, while not every results structure currently needs to reflect disparate types a union will be employed so the same dot qualifier ('.') can be used consistently when accessing all such data. Signed-off-by: Jim Warner <james.warner@comcast.net>
2015-07-21 13:00:00 +08:00
PROCPS_SLABNODE_NAME, PROCPS_SLABNODE_stack_end };
enum rel_enums {
slab_AOBJS, slab_OBJS, slab_OSIZE, slab_OPS, slab_NAME };
2015-07-04 12:59:59 +08:00
if (procps_slabinfo_new(&slab_info) < 0)
xerrx(EXIT_FAILURE, _("Unable to create slabinfo structure"));
if (!(v = procps_slabnode_stacks_alloc(slab_info, CHAINS_ALLOC, MAX_ITEMS, node_items)))
xerrx(EXIT_FAILURE, _("Unable to allocate slabinfo nodes"));
2015-07-04 12:59:59 +08:00
if (!moreheaders)
slabheader();
for (i = 0; infinite_updates || i < num_updates; i++) {
// this next guy also performs the procps_slabnode_read() call
library: slab is redesigned to use 'stack' vs. 'chain' In addition to that text shown below the line which is common to several commit messages, this patch contains several minor changes with lessor impact upon the API: . A 'read' was added to function procps_slabnode_count (but only when necessary, i.e. info->nodes_used == 0). . The #include header files are ordered alphabetically now, with all those <sys/??> types separately grouped. ------------------------------------------------------ . The former 'chains' have now become 'stacks' without the 'next' pointer in each result struct. The pointers initially seemed to offer some flexibility with memory allocations and benefits for the library access logic. However, user access was always via displacement and a a statically allocated chain was cumbersome to define. . An enumerator ending in '_noop' will no longer serve as a fencepost delimiter. Rather, it has become a much more important and flexible user oriented tool. Adding one or more such 'items' in any items list passed into the library becomes the means of extending the 'stack' to also include user (not just library) data. Any such data is guaranteed to never be altered by the library. . Anticipating PID support, where many different types must be represented in a result structure, we'll adopt a common naming standard. And, while not every results structure currently needs to reflect disparate types a union will be employed so the same dot qualifier ('.') can be used consistently when accessing all such data. Signed-off-by: Jim Warner <james.warner@comcast.net>
2015-07-21 13:00:00 +08:00
if ((nr_slabs = procps_slabnode_stacks_fill(slab_info, v, CHAINS_ALLOC)) < 0)
xerrx(EXIT_FAILURE, _("Unable to get slabinfo node data, requires root permission"));
library: slab is redesigned to use 'stack' vs. 'chain' In addition to that text shown below the line which is common to several commit messages, this patch contains several minor changes with lessor impact upon the API: . A 'read' was added to function procps_slabnode_count (but only when necessary, i.e. info->nodes_used == 0). . The #include header files are ordered alphabetically now, with all those <sys/??> types separately grouped. ------------------------------------------------------ . The former 'chains' have now become 'stacks' without the 'next' pointer in each result struct. The pointers initially seemed to offer some flexibility with memory allocations and benefits for the library access logic. However, user access was always via displacement and a a statically allocated chain was cumbersome to define. . An enumerator ending in '_noop' will no longer serve as a fencepost delimiter. Rather, it has become a much more important and flexible user oriented tool. Adding one or more such 'items' in any items list passed into the library becomes the means of extending the 'stack' to also include user (not just library) data. Any such data is guaranteed to never be altered by the library. . Anticipating PID support, where many different types must be represented in a result structure, we'll adopt a common naming standard. And, while not every results structure currently needs to reflect disparate types a union will be employed so the same dot qualifier ('.') can be used consistently when accessing all such data. Signed-off-by: Jim Warner <james.warner@comcast.net>
2015-07-21 13:00:00 +08:00
if (!(v = procps_slabnode_stacks_sort(slab_info, v, nr_slabs, PROCPS_SLABNODE_NAME)))
xerrx(EXIT_FAILURE, _("Unable to sort slab nodes"));
for (j = 0; j < nr_slabs; j++) {
library: slab is redesigned to use 'stack' vs. 'chain' In addition to that text shown below the line which is common to several commit messages, this patch contains several minor changes with lessor impact upon the API: . A 'read' was added to function procps_slabnode_count (but only when necessary, i.e. info->nodes_used == 0). . The #include header files are ordered alphabetically now, with all those <sys/??> types separately grouped. ------------------------------------------------------ . The former 'chains' have now become 'stacks' without the 'next' pointer in each result struct. The pointers initially seemed to offer some flexibility with memory allocations and benefits for the library access logic. However, user access was always via displacement and a a statically allocated chain was cumbersome to define. . An enumerator ending in '_noop' will no longer serve as a fencepost delimiter. Rather, it has become a much more important and flexible user oriented tool. Adding one or more such 'items' in any items list passed into the library becomes the means of extending the 'stack' to also include user (not just library) data. Any such data is guaranteed to never be altered by the library. . Anticipating PID support, where many different types must be represented in a result structure, we'll adopt a common naming standard. And, while not every results structure currently needs to reflect disparate types a union will be employed so the same dot qualifier ('.') can be used consistently when accessing all such data. Signed-off-by: Jim Warner <james.warner@comcast.net>
2015-07-21 13:00:00 +08:00
p = v[j];
if (moreheaders && ((j % height) == 0))
2015-07-04 12:59:59 +08:00
slabheader();
library: slab is redesigned to use 'stack' vs. 'chain' In addition to that text shown below the line which is common to several commit messages, this patch contains several minor changes with lessor impact upon the API: . A 'read' was added to function procps_slabnode_count (but only when necessary, i.e. info->nodes_used == 0). . The #include header files are ordered alphabetically now, with all those <sys/??> types separately grouped. ------------------------------------------------------ . The former 'chains' have now become 'stacks' without the 'next' pointer in each result struct. The pointers initially seemed to offer some flexibility with memory allocations and benefits for the library access logic. However, user access was always via displacement and a a statically allocated chain was cumbersome to define. . An enumerator ending in '_noop' will no longer serve as a fencepost delimiter. Rather, it has become a much more important and flexible user oriented tool. Adding one or more such 'items' in any items list passed into the library becomes the means of extending the 'stack' to also include user (not just library) data. Any such data is guaranteed to never be altered by the library. . Anticipating PID support, where many different types must be represented in a result structure, we'll adopt a common naming standard. And, while not every results structure currently needs to reflect disparate types a union will be employed so the same dot qualifier ('.') can be used consistently when accessing all such data. Signed-off-by: Jim Warner <james.warner@comcast.net>
2015-07-21 13:00:00 +08:00
printf("%-24.24s %6u %6u %6u %6u\n",
p->head[slab_NAME ].result.str,
p->head[slab_AOBJS].result.u_int,
p->head[slab_OBJS ].result.u_int,
p->head[slab_OSIZE].result.u_int,
p->head[slab_OPS ].result.u_int);
2015-07-04 12:59:59 +08:00
}
if (infinite_updates || i+1 < num_updates)
sleep(sleep_time);
}
procps_slabinfo_unref(&slab_info);
#undef CHAINS_ALLOC
#undef MAX_ITEMS
2004-05-05 07:29:40 +08:00
}
static void disksum_format(void)
{
#define DSTAT(x) procps_diskstat_dev_get(disk_stat, (x), devid)
struct procps_diskstat *disk_stat;
if (procps_diskstat_new(&disk_stat) < 0)
xerr(EXIT_FAILURE,
_("Unable to create diskstat structure"));
if (procps_diskstat_read(disk_stat) < 0)
xerr(EXIT_FAILURE,
_("Unable to read diskstat"));
int devid, dev_count, disk_count, part_count ;
unsigned long reads, merged_reads, read_sectors, milli_reading, writes,
merged_writes, written_sectors, milli_writing, inprogress_IO,
milli_spent_IO, weighted_milli_spent_IO;
reads = merged_reads = read_sectors = milli_reading = writes =
merged_writes = written_sectors = milli_writing = inprogress_IO =
milli_spent_IO = weighted_milli_spent_IO = 0;
disk_count = part_count = 0;
if ((dev_count = procps_diskstat_dev_count(disk_stat)) < 0)
xerr(EXIT_FAILURE,
_("Unable to count diskstat devices"));
for (devid=0; devid < dev_count; devid++) {
if (procps_diskstat_dev_isdisk(disk_stat, devid) != 1) {
part_count++;
continue; /* not a disk */
}
disk_count++;
reads += DSTAT(PROCPS_DISKSTAT_READS);
merged_reads += DSTAT(PROCPS_DISKSTAT_READS_MERGED);
read_sectors += DSTAT(PROCPS_DISKSTAT_READ_SECTORS);
milli_reading += DSTAT(PROCPS_DISKSTAT_READ_TIME);
writes += DSTAT(PROCPS_DISKSTAT_WRITES);
merged_writes += DSTAT(PROCPS_DISKSTAT_WRITES_MERGED);
written_sectors += DSTAT(PROCPS_DISKSTAT_WRITE_SECTORS);
milli_writing += DSTAT(PROCPS_DISKSTAT_WRITE_TIME);
inprogress_IO += DSTAT(PROCPS_DISKSTAT_IO_INPROGRESS) / 1000;
milli_spent_IO += DSTAT(PROCPS_DISKSTAT_IO_TIME) / 1000;
weighted_milli_spent_IO += DSTAT(PROCPS_DISKSTAT_IO_TIME) / 1000;
}
printf(_("%13d disks\n"), disk_count);
printf(_("%13d partitions\n"), part_count);
printf(_("%13lu reads\n"), reads);
printf(_("%13lu merged reads\n"), merged_reads);
printf(_("%13lu read sectors\n"), read_sectors);
printf(_("%13lu milli reading\n"), milli_reading);
printf(_("%13lu writes\n"), writes);
printf(_("%13lu merged writes\n"), merged_writes);
printf(_("%13lu written sectors\n"), written_sectors);
printf(_("%13lu milli writing\n"), milli_writing);
printf(_("%13lu inprogress IO\n"), inprogress_IO);
printf(_("%13lu milli spent IO\n"), milli_spent_IO);
printf(_("%13lu milli weighted IO\n"), weighted_milli_spent_IO);
#undef DSTAT
2003-08-11 07:40:41 +08:00
}
static void sum_format(void)
{
struct procps_stat *sys_info;
struct procps_vmstat *vm_info;
struct procps_meminfo *mem_info;
if (procps_stat_new(&sys_info) < 0)
xerrx(EXIT_FAILURE,
_("Unable to create system stat structure"));
if (procps_stat_read(sys_info,0) < 0)
xerrx(EXIT_FAILURE,
_("Unable to read system stat information"));
if (procps_vmstat_new(&vm_info) < 0)
xerrx(EXIT_FAILURE,
_("Unable to create vmstat structure"));
if (procps_vmstat_read(vm_info) < 0)
xerrx(EXIT_FAILURE,
_("Unable to read vmstat information"));
if (procps_meminfo_new(&mem_info) < 0)
xerrx(EXIT_FAILURE, _("Unable to create meminfo structure"));
if (procps_meminfo_read(mem_info) < 0)
xerrx(EXIT_FAILURE, _("Unable to read meminfo information"));
printf(_("%13lu %s total memory\n"), unitConvert(procps_meminfo_get(
mem_info, PROCPS_MEM_TOTAL)), szDataUnit);
printf(_("%13lu %s used memory\n"), unitConvert(procps_meminfo_get(
mem_info, PROCPS_MEM_USED)), szDataUnit);
printf(_("%13lu %s active memory\n"), unitConvert(procps_meminfo_get(
mem_info, PROCPS_MEM_ACTIVE)), szDataUnit);
printf(_("%13lu %s inactive memory\n"), unitConvert(
procps_meminfo_get(mem_info, PROCPS_MEM_INACTIVE)), szDataUnit);
printf(_("%13lu %s free memory\n"), unitConvert(procps_meminfo_get(
mem_info, PROCPS_MEM_FREE)), szDataUnit);
printf(_("%13lu %s buffer memory\n"), unitConvert(procps_meminfo_get(
mem_info, PROCPS_MEM_BUFFERS)), szDataUnit);
printf(_("%13lu %s swap cache\n"), unitConvert(procps_meminfo_get(
mem_info, PROCPS_MEM_CACHED)), szDataUnit);
printf(_("%13lu %s total swap\n"), unitConvert(procps_meminfo_get(
mem_info, PROCPS_SWAP_TOTAL)), szDataUnit);
printf(_("%13lu %s used swap\n"), unitConvert(procps_meminfo_get(
mem_info, PROCPS_SWAP_USED)), szDataUnit);
printf(_("%13lu %s free swap\n"), unitConvert(procps_meminfo_get(
mem_info, PROCPS_SWAP_FREE)), szDataUnit);
library: readstat redesigned using 'stack' vs. 'chain' In addition to that text shown below the line which is common to several commit messages, this patch contains several minor changes with lessor impact upon the API: . A call to procps_stat_read_jiffs() has been added to those jiffs functions carrying the 'fill' nomenclature to parallel like functions in some of our other files. . The #include header files are ordered alphabetically now, with all those <sys/??> types separately grouped. . Standard copyright boilerplate was added in .c file. . The header file follows the conventions of indenting (by 4 spaces) those parameters too lengthy for 1 line. ------------------------------------------------------ . The former 'chains' have now become 'stacks' without the 'next' pointer in each result struct. The pointers initially seemed to offer some flexibility with memory allocations and benefits for the library access logic. However, user access was always via displacement and a a statically allocated chain was cumbersome to define. . An enumerator ending in '_noop' will no longer serve as a fencepost delimiter. Rather, it has become a much more important and flexible user oriented tool. Adding one or more such 'items' in any items list passed into the library becomes the means of extending the 'stack' to also include user (not just library) data. Any such data is guaranteed to never be altered by the library. . Anticipating PID support, where many different types must be represented in a result structure, we'll adopt a common naming standard. And, while not every results structure currently needs to reflect disparate types a union will be employed so the same dot qualifier ('.') can be used consistently when accessing all such data. Signed-off-by: Jim Warner <james.warner@comcast.net>
2015-07-21 13:00:00 +08:00
printf(_("%13lld non-nice user cpu ticks\n"), procps_stat_cpu_get(sys_info, PROCPS_CPU_USER));
printf(_("%13lld nice user cpu ticks\n"), procps_stat_cpu_get(sys_info, PROCPS_CPU_NICE));
printf(_("%13lld system cpu ticks\n"), procps_stat_cpu_get(sys_info, PROCPS_CPU_SYSTEM));
printf(_("%13lld idle cpu ticks\n"), procps_stat_cpu_get(sys_info, PROCPS_CPU_IDLE));
printf(_("%13lld IO-wait cpu ticks\n"), procps_stat_cpu_get(sys_info, PROCPS_CPU_IOWAIT));
printf(_("%13lld IRQ cpu ticks\n"), procps_stat_cpu_get(sys_info, PROCPS_CPU_IRQ));
printf(_("%13lld softirq cpu ticks\n"), procps_stat_cpu_get(sys_info, PROCPS_CPU_SIRQ));
printf(_("%13lld stolen cpu ticks\n"), procps_stat_cpu_get(sys_info, PROCPS_CPU_STOLEN));
printf(_("%13lld non-nice guest cpu ticks\n"), procps_stat_cpu_get(sys_info, PROCPS_CPU_GUEST));
printf(_("%13lld nice guest cpu ticks\n"), procps_stat_cpu_get(sys_info, PROCPS_CPU_GNICE));
printf(_("%13lu pages paged in\n"), procps_vmstat_get(vm_info, PROCPS_VMSTAT_PGPGIN));
printf(_("%13lu pages paged out\n"), procps_vmstat_get(vm_info, PROCPS_VMSTAT_PGPGOUT));
printf(_("%13lu pages swapped in\n"), procps_vmstat_get(vm_info, PROCPS_VMSTAT_PSWPIN));
printf(_("%13lu pages swapped out\n"), procps_vmstat_get(vm_info, PROCPS_VMSTAT_PSWPOUT));
library: readstat redesigned using 'stack' vs. 'chain' In addition to that text shown below the line which is common to several commit messages, this patch contains several minor changes with lessor impact upon the API: . A call to procps_stat_read_jiffs() has been added to those jiffs functions carrying the 'fill' nomenclature to parallel like functions in some of our other files. . The #include header files are ordered alphabetically now, with all those <sys/??> types separately grouped. . Standard copyright boilerplate was added in .c file. . The header file follows the conventions of indenting (by 4 spaces) those parameters too lengthy for 1 line. ------------------------------------------------------ . The former 'chains' have now become 'stacks' without the 'next' pointer in each result struct. The pointers initially seemed to offer some flexibility with memory allocations and benefits for the library access logic. However, user access was always via displacement and a a statically allocated chain was cumbersome to define. . An enumerator ending in '_noop' will no longer serve as a fencepost delimiter. Rather, it has become a much more important and flexible user oriented tool. Adding one or more such 'items' in any items list passed into the library becomes the means of extending the 'stack' to also include user (not just library) data. Any such data is guaranteed to never be altered by the library. . Anticipating PID support, where many different types must be represented in a result structure, we'll adopt a common naming standard. And, while not every results structure currently needs to reflect disparate types a union will be employed so the same dot qualifier ('.') can be used consistently when accessing all such data. Signed-off-by: Jim Warner <james.warner@comcast.net>
2015-07-21 13:00:00 +08:00
printf(_("%13u interrupts\n"), procps_stat_sys_get(sys_info, PROCPS_STAT_INTR));
printf(_("%13u CPU context switches\n"), procps_stat_sys_get(sys_info, PROCPS_STAT_CTXT));
printf(_("%13u boot time\n"), procps_stat_sys_get(sys_info, PROCPS_STAT_BTIME));
printf(_("%13u forks\n"), procps_stat_sys_get(sys_info, PROCPS_STAT_PROCS));
2002-11-28 17:07:59 +08:00
}
2003-06-09 01:28:06 +08:00
static void fork_format(void)
{
struct procps_stat *sys_info;
if (procps_stat_new(&sys_info) < 0)
xerrx(EXIT_FAILURE,
_("Unable to create system stat structure"));
if (procps_stat_read(sys_info,0) < 0)
xerrx(EXIT_FAILURE,
_("Unable to read system stat information"));
library: readstat redesigned using 'stack' vs. 'chain' In addition to that text shown below the line which is common to several commit messages, this patch contains several minor changes with lessor impact upon the API: . A call to procps_stat_read_jiffs() has been added to those jiffs functions carrying the 'fill' nomenclature to parallel like functions in some of our other files. . The #include header files are ordered alphabetically now, with all those <sys/??> types separately grouped. . Standard copyright boilerplate was added in .c file. . The header file follows the conventions of indenting (by 4 spaces) those parameters too lengthy for 1 line. ------------------------------------------------------ . The former 'chains' have now become 'stacks' without the 'next' pointer in each result struct. The pointers initially seemed to offer some flexibility with memory allocations and benefits for the library access logic. However, user access was always via displacement and a a statically allocated chain was cumbersome to define. . An enumerator ending in '_noop' will no longer serve as a fencepost delimiter. Rather, it has become a much more important and flexible user oriented tool. Adding one or more such 'items' in any items list passed into the library becomes the means of extending the 'stack' to also include user (not just library) data. Any such data is guaranteed to never be altered by the library. . Anticipating PID support, where many different types must be represented in a result structure, we'll adopt a common naming standard. And, while not every results structure currently needs to reflect disparate types a union will be employed so the same dot qualifier ('.') can be used consistently when accessing all such data. Signed-off-by: Jim Warner <james.warner@comcast.net>
2015-07-21 13:00:00 +08:00
printf(_("%13u forks\n"), procps_stat_sys_get(sys_info, PROCPS_STAT_PROCS));
2002-11-28 17:07:59 +08:00
}
static int winhi(void)
{
struct winsize win;
int rows = 24;
if (ioctl(STDOUT_FILENO, TIOCGWINSZ, &win) != -1 && 0 < win.ws_row)
rows = win.ws_row;
2002-11-28 17:07:59 +08:00
return rows;
2002-11-09 07:29:34 +08:00
}
int main(int argc, char *argv[])
{
char *partition = NULL;
int c;
long tmp;
static const struct option longopts[] = {
{"active", no_argument, NULL, 'a'},
{"forks", no_argument, NULL, 'f'},
{"slabs", no_argument, NULL, 'm'},
{"one-header", no_argument, NULL, 'n'},
{"stats", no_argument, NULL, 's'},
{"disk", no_argument, NULL, 'd'},
{"disk-sum", no_argument, NULL, 'D'},
{"partition", required_argument, NULL, 'p'},
{"unit", required_argument, NULL, 'S'},
{"wide", no_argument, NULL, 'w'},
{"timestamp", no_argument, NULL, 't'},
{"help", no_argument, NULL, 'h'},
{"version", no_argument, NULL, 'V'},
{NULL, 0, NULL, 0}
};
#ifdef HAVE_PROGRAM_INVOCATION_NAME
program_invocation_name = program_invocation_short_name;
#endif
setlocale (LC_ALL, "");
bindtextdomain(PACKAGE, LOCALEDIR);
textdomain(PACKAGE);
atexit(close_stdout);
while ((c =
getopt_long(argc, argv, "afmnsdDp:S:wthV", longopts,
NULL)) != EOF)
switch (c) {
case 'V':
printf(PROCPS_NG_VERSION);
return EXIT_SUCCESS;
case 'h':
usage(stdout);
case 'd':
statMode |= DISKSTAT;
break;
case 'a':
/* active/inactive mode */
a_option = 1;
break;
case 'f':
/* FIXME: check for conflicting args */
fork_format();
exit(0);
case 'm':
statMode |= SLABSTAT;
break;
case 'D':
statMode |= DISKSUMSTAT;
break;
case 'n':
/* print only one header */
moreheaders = FALSE;
break;
case 'p':
statMode |= PARTITIONSTAT;
partition = optarg;
if (memcmp(partition, "/dev/", 5) == 0)
partition += 5;
break;
case 'S':
switch (optarg[0]) {
case 'b':
case 'B':
dataUnit = UNIT_B;
break;
case 'k':
dataUnit = UNIT_k;
break;
case 'K':
dataUnit = UNIT_K;
break;
case 'm':
dataUnit = UNIT_m;
break;
case 'M':
dataUnit = UNIT_M;
break;
default:
xerrx(EXIT_FAILURE,
/* Translation Hint: do not change argument characters */
_("-S requires k, K, m or M (default is KiB)"));
}
szDataUnit[0] = optarg[0];
break;
case 's':
statMode |= VMSUMSTAT;
break;
case 'w':
w_option = 1;
break;
case 't':
t_option = 1;
break;
default:
/* no other aguments defined yet. */
usage(stderr);
}
if (optind < argc) {
tmp = strtol_or_err(argv[optind++], _("failed to parse argument"));
if (tmp < 1)
xerrx(EXIT_FAILURE, _("delay must be positive integer"));
else if (UINT_MAX < tmp)
xerrx(EXIT_FAILURE, _("too large delay value"));
sleep_time = tmp;
infinite_updates = 1;
}
2015-07-04 12:59:59 +08:00
num_updates = 1;
if (optind < argc) {
num_updates = strtol_or_err(argv[optind++], _("failed to parse argument"));
infinite_updates = 0;
}
if (optind < argc)
usage(stderr);
if (moreheaders) {
int wheight = winhi() - 3;
height = ((wheight > 0) ? wheight : 22);
}
setlinebuf(stdout);
switch (statMode) {
case (VMSTAT):
new_format();
break;
case (VMSUMSTAT):
sum_format();
break;
case (DISKSTAT):
diskformat();
break;
case (PARTITIONSTAT):
if (diskpartition_format(partition) == -1)
printf(_("partition was not found\n"));
break;
case (SLABSTAT):
slabformat();
break;
case (DISKSUMSTAT):
disksum_format();
break;
default:
usage(stderr);
break;
}
return 0;
2002-11-08 15:42:54 +08:00
}