Show dirty buffers and the percentage of time a disk was busy.

PR:		12858
Submitted by:	Arjan de Vet <Arjan.deVet@adv.iae.nl>
This commit is contained in:
Dag-Erling Smørgrav 1999-07-30 07:44:25 +00:00
parent 55293461db
commit a4a026b4c0
Notes: svn2git 2020-12-20 02:59:44 +00:00
svn path=/head/; revision=49245
2 changed files with 37 additions and 14 deletions

View File

@ -30,7 +30,7 @@
.\" SUCH DAMAGE.
.\"
.\" @(#)systat.1 8.2 (Berkeley) 12/30/93
.\" $Id$
.\" $Id: systat.1,v 1.18 1999/07/12 20:24:00 nik Exp $
.\"
.Dd September 9, 1997
.Dt SYSTAT 1
@ -306,9 +306,9 @@ handled by the per process name translation cache.
.Pp
At the bottom left is the disk usage display.
It reports the number of
kilobytes per transaction, transactions per second and megabytes
per second averaged over the
refresh period of the display (by default, five seconds).
kilobytes per transaction, transactions per second, megabytes
per second and the percentage of the time the disk was busy averaged
over the refresh period of the display (by default, five seconds).
The system keeps statistics on most every storage device. In general, up
to seven devices are displayed. The devices displayed by default are the
first devices in the kernel's device list. See
@ -351,6 +351,7 @@ per second over the refresh interval.
.Pp
At the bottom of this column are lines showing the
amount of memory, in kilobytes, used for the buffer cache (`buf'),
the number of dirty buffers in the buffer cache (`dirtybuf'),
desired maximum size of vnode cache (`desiredvnodes') (mostly unused,
except to size the name cache),
number of vnodes actually allocated (`numvnodes'),

View File

@ -36,7 +36,7 @@
static char sccsid[] = "@(#)vmstat.c 8.2 (Berkeley) 1/12/94";
#endif
static const char rcsid[] =
"$Id: vmstat.c,v 1.34 1999/02/08 02:39:45 dillon Exp $";
"$Id: vmstat.c,v 1.35 1999/03/22 03:44:01 bde Exp $";
#endif /* not lint */
/*
@ -82,6 +82,7 @@ static struct Info {
int desiredvnodes;
long numvnodes;
long freevnodes;
long numdirtybuffers;
} s, s1, s2, z;
struct statinfo cur, last, run;
@ -164,6 +165,8 @@ static struct nlist namelist[] = {
{ "_numvnodes" },
#define X_FREEVNODES 10
{ "_freevnodes" },
#define X_NUMDIRTYBUFFERS 11
{ "_numdirtybuffers" },
{ "" },
};
@ -310,10 +313,11 @@ labelkre()
mvprintw(VMSTATROW + 11, VMSTATCOL + 10, "pdpgs");
mvprintw(VMSTATROW + 12, VMSTATCOL + 10, "intrn");
mvprintw(VMSTATROW + 13, VMSTATCOL + 10, "buf");
mvprintw(VMSTATROW + 14, VMSTATCOL + 10, "dirtybuf");
mvprintw(VMSTATROW + 14, VMSTATCOL + 10, "desiredvnodes");
mvprintw(VMSTATROW + 15, VMSTATCOL + 10, "numvnodes");
mvprintw(VMSTATROW + 16, VMSTATCOL + 10, "freevnodes");
mvprintw(VMSTATROW + 15, VMSTATCOL + 10, "desiredvnodes");
mvprintw(VMSTATROW + 16, VMSTATCOL + 10, "numvnodes");
mvprintw(VMSTATROW + 17, VMSTATCOL + 10, "freevnodes");
mvprintw(GENSTATROW, GENSTATCOL, " Csw Trp Sys Int Sof Flt");
@ -330,6 +334,7 @@ labelkre()
mvprintw(DISKROW + 1, DISKCOL, "KB/t");
mvprintw(DISKROW + 2, DISKCOL, "tps");
mvprintw(DISKROW + 3, DISKCOL, "MB/s");
mvprintw(DISKROW + 4, DISKCOL, "%% busy");
/*
* For now, we don't support a fourth disk statistic. So there's
* no point in providing a label for it. If someone can think of a
@ -504,9 +509,10 @@ showkre()
}
putint(s.bufspace/1024, VMSTATROW + 13, VMSTATCOL, 9);
putint(s.desiredvnodes, VMSTATROW + 14, VMSTATCOL, 9);
putint(s.numvnodes, VMSTATROW + 15, VMSTATCOL, 9);
putint(s.freevnodes, VMSTATROW + 16, VMSTATCOL, 9);
putint(s.numdirtybuffers, VMSTATROW + 14, VMSTATCOL, 9);
putint(s.desiredvnodes, VMSTATROW + 15, VMSTATCOL, 9);
putint(s.numvnodes, VMSTATROW + 16, VMSTATCOL, 9);
putint(s.freevnodes, VMSTATROW + 17, VMSTATCOL, 9);
PUTRATE(Cnt.v_vnodein, PAGEROW + 2, PAGECOL + 5, 5);
PUTRATE(Cnt.v_vnodeout, PAGEROW + 2, PAGECOL + 10, 5);
PUTRATE(Cnt.v_swapin, PAGEROW + 2, PAGECOL + 17, 5);
@ -736,6 +742,7 @@ getinfo(s, st)
NREAD(X_FREEVNODES, &s->freevnodes, LONG);
NREAD(X_NCHSTATS, &s->nchstats, sizeof s->nchstats);
NREAD(X_INTRCNT, s->intrcnt, nintr * LONG);
NREAD(X_NUMDIRTYBUFFERS, &s->numdirtybuffers, sizeof(s->numdirtybuffers));
size = sizeof(s->Total);
mib[0] = CTL_VM;
mib[1] = VM_METER;
@ -800,24 +807,39 @@ dinfo(dn, c, now, then)
{
long double transfers_per_second;
long double kb_per_transfer, mb_per_second;
long double busy_seconds;
long double elapsed_time, device_busy;
int di;
di = dev_select[dn].position;
busy_seconds = compute_etime(now->busy_time, then ?
elapsed_time = compute_etime(now->busy_time, then ?
then->busy_time :
now->dinfo->devices[di].dev_creation_time);
device_busy = compute_etime(now->dinfo->devices[di].busy_time, then ?
then->dinfo->devices[di].busy_time :
now->dinfo->devices[di].dev_creation_time);
if (compute_stats(&now->dinfo->devices[di], then ?
&then->dinfo->devices[di] : NULL, busy_seconds,
&then->dinfo->devices[di] : NULL, elapsed_time,
NULL, NULL, NULL,
&kb_per_transfer, &transfers_per_second,
&mb_per_second, NULL, NULL) != 0)
errx(1, "%s", devstat_errbuf);
if ((device_busy == 0) && (transfers_per_second > 5))
/* the device has been 100% busy, fake it because
* as long as the device is 100% busy the busy_time
* field in the devstat struct is not updated */
device_busy = elapsed_time;
if (device_busy > elapsed_time)
/* this normally happens after one or more periods
* where the device has been 100% busy, correct it */
device_busy = elapsed_time;
c = DISKCOL + c * 6;
putlongdouble(kb_per_transfer, DISKROW + 1, c, 5, 2, 0);
putlongdouble(transfers_per_second, DISKROW + 2, c, 5, 0, 0);
putlongdouble(mb_per_second, DISKROW + 3, c, 5, 2, 0);
putlongdouble(device_busy * 100 / elapsed_time, DISKROW + 4, c, 5, 0, 0);
}