Return to BSD News archive
Newsgroups: comp.os.386bsd.apps Path: sserve!newshost.anu.edu.au!munnari.oz.au!uniwa!cujo!cproto From: cproto@cs.curtin.edu.au (Computer Protocol) Subject: Ghostscript 2.5.2 dies with floating point exceptions Message-ID: <cproto.733031902@marsh> Summary: Ghostscript 2.5.2 dies with floating point exceptions Keywords: Ghostscript floating point exception Sender: news@cujo.curtin.edu.au (News Manager) Organization: Curtin University of Technology Date: Thu, 25 Mar 1993 03:58:22 GMT Lines: 23 I ported Ghostscript 2.5.2 to 386bsd-0.1 (with most patches installed) without any problems. If I run the executable it dies with a floating point exception. Unfortunately the signal is delivered much after the event. Because of this gdb is not reporting the error at the right source line. I narrowed the fault down to a single function but when I try to single step through it, then the kernel panics. The reason might be that two signals are caused at the same time, i.e. the SIGFPE and the SIGTRAP. 1. Has anybody found the reason for the FPE in Ghostscript? 2. Has anybody fixed the delayed delivery of SIGFPEs? 3. Any ideas why a SIGFPE occuring at the same time as SIGTRAP causes the kernel to panic. 4. Has anybody tried to run savecore saving a dump from /dev/rwd0b to /var/crash which resides on /dev/wd1a. My system just locks up! Regards - Tibor Sashegyi (cproto@cs.curtin.edu.au)