Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sdd.hp.com!elroy.jpl.nasa.gov!jpl-devvax!lwall
From: lwall@jpl-devvax.jpl.nasa.gov (Larry Wall)
Newsgroups: comp.lang.perl
Subject: Re: Bug in 4.003
Message-ID: <1991Apr19.184655.8696@jpl-devvax.jpl.nasa.gov>
Date: 19 Apr 91 18:46:55 GMT
References: <VICTOR.91Apr18112016@irt.watson.ibm.com>
Reply-To: lwall@jpl-devvax.JPL.NASA.GOV (Larry Wall)
Distribution: comp
Organization: Jet Propulsion Laboratory, Pasadena, CA
Lines: 21

In article <VICTOR.91Apr18112016@irt.watson.ibm.com> victor@watson.ibm.com writes:
: I built and tested perl-4.003 (from prep) on three platforms: IBM
: BSD/RT, IBM RS/6000, and SparcStation running SunOs 4.  The script
: below causes perl to get a core dump on all 3.
: 
: sub NOPIPE {
:     print join(',',caller(0));
:     shift(@_); die "Couldn't pipe:" . join(' ',@_) . "\n";
: }
: $SIG{'PIPE'} = 'NOPIPE';
: open(MESS,"| $foobar") || die "Couldn't pipe to $foobar\n";
: print MESS "Hello\n";

caller with an argument currently only works if you're running under
the debugger.  I'll fix that in the next patch.

But I should mention that using stdio in a signal handler dealing with
a fault that happened in the stdio routines is risky business.  I'd
use syswrite().

Larry