Date: Fri, 3 Aug 2001 08:11:33 -0700 From: William Richard <wdr@tdl.com> To: j mckitrick <jcm@freebsd-uk.eu.org> Cc: freebsd-stable@FreeBSD.ORG Subject: Re: moving to XFree86-4 Message-ID: <01080308113300.00854@saffron.my.domain> In-Reply-To: <20010803154352.A25257@dogma.freebsd-uk.eu.org> References: <20010802112630.A9855@dogma.freebsd-uk.eu.org> <01080212254800.01448@saffron.my.domain> <20010803154352.A25257@dogma.freebsd-uk.eu.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Friday 03 August 2001 07:43, j mckitrick wrote: > | Before you deinstall all of your X ports, make sure the new X > | server works. And make sure you can go back to 3.3.6 relatively > | expeditiously (mv /usr/X11R6 /usr/X11backup). The XFree86-4 X > | server dumped core on my allegedly-supported Matrox Mystique > | shortly after starting, so I never got past the configuration. If > | I hadn't had the old X11R6 directory backed up, I would have been > | mightily screwed. > > I ran XFree86 -configure, then ran it again using that file. I got a > signal 10 error, and a core dump. Does a signal 10 under XFree86 > mean the same as a signal 10 in make world, namely bad memory? From <sys/signal.h>: #define SIGBUS 10 /* bus error */ This would seem to indicate not "bad memory", but "bad bus", if it indicates anything about the hardware at all. But it doesn't necessarily--I occassionally get sig 11s from a few "known suspects" (some KDE-related programs), but they don't adversely affect how I use my machine, and I can make buildworld without a care. If I started getting random sig 11s from programs that I know to work on other people's machines (ppp, for instance, or lpd) that I would begin to view my memory modules with a suspicious eye. It is possible to write a program which generates a sig 11 (or a sig 10) without having bad equipment. Dereferencing an uninitialised pointer is a good way to get a sig 11. So copping a sig 10/11 is not necessarily indicative of bad memory or bad bus, rather, it could indicate bad programmer. It's when you get these errors in programs that are known to work consistently on everyone else's machines (make buildworld on -STABLE, for instance, with no obvious complaints on the freebsd-stable list) that you should consider bad hardware. (Finally, after reading every one else's posts about how XFree86-4 works flawlessly on their machines, let me indulge in a degree of schadenfraude... [ ... ] ... okay, I'm done.) -- Cheers, William Richard wdr@tdl.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?01080308113300.00854>