Date: Sat, 23 Feb 2008 03:00:11 GMT From: pluknet <pluknet@gmail.com> To: freebsd-i386@FreeBSD.org Subject: Re: i386/74327: [pmap] [patch] mlock() causes physical memory leakage Message-ID: <200802230300.m1N30BwR040068@freefall.freebsd.org>
next in thread | raw e-mail | index | archive | help
The following reply was made to PR i386/74327; it has been noted by GNATS.
From: pluknet <pluknet@gmail.com>
To: bug-followup@freebsd.org, deniskv@mail.ru
Cc:
Subject: Re: i386/74327: [pmap] [patch] mlock() causes physical memory leakage
Date: Sat, 23 Feb 2008 04:16:50 +0300
I tried to reproduce on RELENG_[67] mlock()'ing 20 megs - without success.
Test procedure (output from RELENG_7, same from _6):
[root@notebook /tmp]# for i in 1 2 3 4 5; do ./a; done
64414
69534
64414
69534
64414
69534
64414
69534
64414
69534
---
#include <sys/mman.h>
#include <sys/types.h>
#include <sys/sysctl.h>
#include <stdio.h>
#include <stdlib.h>
#include <strings.h>
int
main(void)
{
u_int count;
size_t len = sizeof(u_int);
const size_t size = 20 * 1024 * 1024;
void *a;
a = malloc(size);
if (a == NULL)
perror("malloc");
bzero(a, size);
sysctlbyname("vm.stats.vm.v_wire_count", &count, &len,
NULL, 0);
printf("%d\n", count);
if (mlock(a, size) == -1)
perror("mlock");
sysctlbyname("vm.stats.vm.v_wire_count", &count, &len,
NULL, 0);
printf("%d\n", count);
exit(0);
}
---
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200802230300.m1N30BwR040068>