Date: Sat, 12 Dec 1998 23:33:15 -0600 (CST) From: Henry Miller <hank@black-hole.com> To: Dag-Erling Smorgrav <des@flood.ping.uio.no> Cc: Matt Dillon <dillon@FreeBSD.ORG>, cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG Subject: Re: cvs commit: src/sys/kern vfs_syscalls.c Message-ID: <Pine.BSF.3.96.981212232007.5817A-100000@daphne.bogus> In-Reply-To: <xzppv9pgemn.fsf@flood.ping.uio.no>
next in thread | previous in thread | raw e-mail | index | archive | help
On 12 Dec 1998, Dag-Erling Smorgrav wrote:
> Matt Dillon <dillon@FreeBSD.ORG> writes:
> Speaking of kernel dumps, what happens to physical memory pages when
> they're not in use? Are they zeroed? If not, is it possible to add a
> kernel option that zeroes out unused pages, or fills them with a fixed
> pattern (e.g. f001f001 og deadbeef)? That would make kernel dumps more
> compressible...
Woah! When there is a kernel dump, something is wrong. I have seen (not
on freebsd, but on other systems) where memory marked free contained the
critical clue to why the dump occured. When a kernel dumps we cannot
assume that the flags marking memory as free are correct. A system can
write to "free" memory intending a different area, not corrupt anything
until it goes a little farther and overwrites a different subsystem that
executes sometime latter.
I don't know how x86 memory protection works, so maybe the above can't
happen, but I wouldn't count on it.
--
http://blugill.home.ml.org/
hank@black-hole.com
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.96.981212232007.5817A-100000>