home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.sgi
- Path: sparky!uunet!munnari.oz.au!mips!mips!odin!sgihub!zola!zuni!anchor!olson
- From: olson@anchor.esd.sgi.com (Dave Olson)
- Subject: Re: Problems with signal handling.
- Message-ID: <p8a1m4k@zuni.esd.sgi.com>
- Keywords: signal
- Sender: news@zuni.esd.sgi.com (Net News)
- Organization: Silicon Graphics, Inc. Mountain View, CA
- References: <1992Aug30.210156.11419@cc.umontreal.ca>
- Date: Mon, 31 Aug 92 02:54:36 GMT
- Lines: 27
-
- In <1992Aug30.210156.11419@cc.umontreal.ca> gaumondp@ERE.UMontreal.CA (Gaumond Pierre) writes:
-
- | I am on a 4D380 SGI machine under IRIX 4.0.1.
- |
- | I am using a function to catch "bus error" and "segmentation fault". I call
- | signal(2) once with SIGBUS and once with SIGSEGV. The same function is used to
- | catch both signals.
- |
- | I receive the good signal when the function is entered but, with SIGSEGV, the
- | code is not one of those expected (if I refer to the manual). The numeric value
- | is "4", corresponding to EINTR (interrupted system call).
- |
-
- To the best of my knowledge, there is no argument to signal handlers
- that corresponds to any values of errno, except in a few carefully
- specified cases where the signal is initiated by the kernel (and only
- for BSD and POSIX signals). Are you sure that these are in fact the
- cases that you are dealing with?
-
- The reference to EACCESS instead of EACCES is a bug; there are
- a surprising number of man pages with this problem. I've filed
- internal bugs to get them fixed in the next release.
-
- --
- Let no one tell me that silence gives consent, | Dave Olson
- because whoever is silent dissents. | Silicon Graphics, Inc.
- Maria Isabel Barreno | olson@sgi.com
-