Date: Thu, 21 Mar 1996 18:41:27 -0700 (MST) From: Terry Lambert <terry@lambert.org> To: julian@ref.tfs.com (JULIAN Elischer) Cc: terry@lambert.org, jfieber@indiana.edu, rich@lamprey.utmb.edu, hackers@FreeBSD.org Subject: Re: Mathematica under FreeBSD Message-ID: <199603220141.SAA01623@phaeton.artisoft.com> In-Reply-To: <199603220126.RAA20452@ref.tfs.com> from "JULIAN Elischer" at Mar 21, 96 05:26:10 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> > I would put in:
> >
> > {
> > int i;
> > printf( "here 1\n");
> > for( i=0; i < 0x7fffffff; i++) /* adjust down for time*/
> > continue;
> > }
> >
> that's silly, he doesn't have the source to that (I believe)
Now *that's* silly. I was talking about the kernel, not the program.
The bug is in the kernel if the program runs on Linux but not FreeBSD.
He needs to find out which Linux system call is blowing chunks, then
he needs to find which line in that call is blowing chunks, then he
needs to breakpoint there with the kernel debugger and find out how
many repetitions are necessary before it blows chunks, then he needs
to set the breakpoint again and step through that many times minus
one, and then he needs to watch it so he can see what's actually
blowing chunks.
And if you think that was a run-on sentence, it's because tracking
a kernel bug without a dump is a run-on procedure.
Further, since most of us don't have Mathematica to test with, it's
probably him or no one.
And running kdb is a question that's better answered by Bruce or
someone else who uses the thing on a regular basis.
Terry Lambert
terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199603220141.SAA01623>