From owner-cvs-usrsbin Fri May 9 22:48:32 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id WAA20655 for cvs-usrsbin-outgoing; Fri, 9 May 1997 22:48:32 -0700 (PDT) Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id WAA20645; Fri, 9 May 1997 22:48:23 -0700 (PDT) Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.5/8.6.9) id PAA06971; Sat, 10 May 1997 15:42:38 +1000 Date: Sat, 10 May 1997 15:42:38 +1000 From: Bruce Evans Message-Id: <199705100542.PAA06971@godzilla.zeta.org.au> To: bde@zeta.org.au, brian@awfulhak.org Subject: Re: cvs commit: src/usr.sbin/ppp timer.c Cc: ache@nagual.pp.ru, brian@FreeBSD.ORG, cvs-all@FreeBSD.ORG, CVS-committers@FreeBSD.ORG, cvs-usrsbin@FreeBSD.ORG Sender: owner-cvs-usrsbin@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk >> Actually it is, read the kernel source :-). The kernel checks the interval >> even when it doesn't use it and fails without doing anything if the interval >> is garbage (tv_sec < 0 || tv_sec > 10^8 || tv_usec < 0 || tv_usec >= 10^6). > >Should I fix kern/kern_time.c then ? I checked some other systems: FreeBSD-1.1.5: same as now. Overflows are avoided for starting times between 1970 and 1935 by limiting itimer values to about 3 years. Linux-2.1.29: timevals are converted to jiffies and not used again; some overflows are corrected, others give silly times. Bruce