linux-kernel - Underflow in system time reported by getrusage()

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>] [day] [month] [year] [list]

Message-ID: <1325174416.13595.34.camel@deadeye>
Date:	Thu, 29 Dec 2011 17:00:16 +0100
From:	Ben Hutchings <ben@...adent.org.uk>
To:	Ingo Molnar <mingo@...e.hu>, Peter Zijlstra <peterz@...radead.org>
Cc:	652732@...s.debian.org, Vincent Lefevre <vincent@...c17.net>,
	LKML <linux-kernel@...r.kernel.org>
Subject: Underflow in system time reported by getrusage()

On Tue, 2011-12-20 at 13:14 +0100, Vincent Lefevre wrote:
> Package: linux-2.6
> Version: 3.1.5-1
> Severity: minor
> 
> I've had the following problem (under zsh-beta):
> 
> ypig:~> time ./a.out 0.999
> ./a.out 0.999  2.87s user 18446744073.71s system -2147483648% cpu 2.902 total
> 
> Of course, I can't reproduce it. I suppose this problem is quite rare
> and occurs under some particular conditions.

The value reported for system time looks close to the maximum possible
64-bit number of microseconds, presumably resulting from an underflow.

Looking at the Debian source package zsh-beta, version
4.3.15-dev-0+20111220-1, it appears to use getrusage() where available,
and doesn't appear to do any calculations that might cause such
underflow.

Ben.

> I've search on Google and could only find:
> 
>   http://ftp.zcu.cz/mirrors/linux/kernel/v2.6/snapshots/old/patch-2.6.25-git4.log
> 
> which mentions a "very huge system time" problem on IA64 due to
> interrupts. This is unrelated, but could give an idea about a
> possible cause...
[...]

-- 
Ben Hutchings
Design a system any fool can use, and only a fool will want to use it.

Download attachment "signature.asc" of type "application/pgp-signature" (829 bytes)