lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:	Wed, 02 Dec 2009 17:32:42 +0900
From:	Hidetoshi Seto <seto.hidetoshi@...fujitsu.com>
To:	Stanislaw Gruszka <sgruszka@...hat.com>
CC:	Ingo Molnar <mingo@...e.hu>, Thomas Gleixner <tglx@...utronix.de>,
	Peter Zijlstra <peterz@...radead.org>,
	Spencer Candland <spencer@...ehost.com>,
	Américo Wang <xiyou.wangcong@...il.com>,
	linux-kernel@...r.kernel.org, Oleg Nesterov <oleg@...hat.com>,
	Balbir Singh <balbir@...ibm.com>
Subject: reproducer: invisible utime

This program can reproduce another problem that originally
task_{u,s}time() intended to solve, I think.

Adjustment by task_{u,s}time() was only applied on getrusage()
and /proc/<pid>/stat.  So even in .32-rc8, we can reproduce this
problem on sys_times() which has no adjustment.

I confirmed that my patches fix it, by thread_group_times().


Thanks,
H.Seto

===
/*
 * Sample program to demonstrate invisible utime
 */

#include <stdio.h>
#include <unistd.h>
#include <sys/times.h>

/* 300 million, arbitrary */
#define LOOP (300 * 1000 * 1000)

unsigned long lpt, l, c;

int do_loop(void)
{
	struct tms t1, t2;
	clock_t j1, j2;
	unsigned long i;

	printf("Loop %d times, sleeping %d times every %d.\n", l, l/c, c);
	printf("  start ...\n");
	j1 = times(&t1);
	for (i = 1; i <= l; i++)
		if (!(i % c))
			usleep(0);
	j2 = times(&t2);
	printf("  ... done.\n");

	/* tms_{u,s}time is clock_t */
	printf("  user   : %5d ms\n", (t2.tms_utime - t1.tms_utime) * 10);
	printf("  system : %5d ms\n", (t2.tms_stime - t1.tms_stime) * 10);
	printf("  elapse : %5d ms\n\n", (j2 - j1) * 10);

	return (t2.tms_utime - t1.tms_utime) * 10;
}

int main(int argc, char **argv)
{
	int u0, u1, u2, u3;
	int ticks;

	l = argc > 1 ? atoi(argv[1]) : LOOP;

	printf("### Prep:\n");
	c = l;
	ticks = do_loop();
	lpt = l / ticks;
	printf("loops/tick: %d\n", lpt);
	l = lpt * 1000;
	printf("change loop to %d to short test.\n\n", l);

	printf("### Test:\n");
	c = l;
	u0 = do_loop();
	c = lpt;
	u1 = do_loop();
	c = lpt / 2;
	u2 = do_loop();
	c = lpt / 8;
	u3 = do_loop();
	printf("result: %s\n\n",
		(u0 <= u1) && (u1 <= u2) && (u2 <= u3) ? "GOOD" : "BAD");
}

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ