*BSD News Article 2085


Return to BSD News archive

Path: sserve!manuel!munnari.oz.au!uunet!usc!noiro.acs.uci.edu!ucivax!bvickers
From: bvickers@net1.ics.uci.edu (Brett J. Vickers)
Subject: 386BSD: Bug in gettimeofday()?
Nntp-Posting-Host: net1.ics.uci.edu
Message-ID: <2A6A642D.25091@ics.uci.edu>
Newsgroups: comp.unix.bsd
Organization: Univ. of Calif., Irvine, Info. & Computer Sci. Dept.
Lines: 21
Date: 20 Jul 92 06:58:53 GMT

First off:  Wow!  Awesome system, Bill, Lynne, et al!  Thanks for all
the fine work!

Now for the bad news:  It appears that when I use the time() call,
the system occasionally gets stuck in an infinite loop within the
gettimeofday() function (which gets called by time()).  When it
happens, the process just sits there running up CPU time like it's
going out of style.  Forcing a core dump by using ^\ shows that the
process was stuck in gettimeofday().

I'm not sure if this is a bug in the library function for gettimeofday,
or if it's related to the kernel in any way.  Nevertheless, it's an
unexplainable phenomenon (well, I can't explain it anyway!)

Anyone have any ideas?

--
 ___            _    _     _  _        _
(  _) ___ ___ _( )__( )_  ( )( ) o  __( ) _  ___  ___  ___
(___)(_) (__=) (_)_ (_)_   (__) (_)(_((_)(_'(__=)(_) _(_)
Brett Vickers (bvickers@ics.uci.edu)