Date: Sun, 9 Dec 2001 00:21:37 -0800 (PST) From: Matthew Dillon <dillon@apollo.backplane.com> To: Terry Lambert <tlambert2@mindspring.com> Cc: Varshavchick Alexander <alex@metrocom.ru>, freebsd-hackers@FreeBSD.ORG Subject: Re: 4G phisical memory kernel trap Message-ID: <200112090821.fB98LbQ36051@apollo.backplane.com> References: <Pine.GSO.4.21.0112061018260.9266-100000@apache.metrocom.ru> <3C0F2E1B.7F933520@mindspring.com>
next in thread | previous in thread | raw e-mail | index | archive | help
:Varshavchick Alexander wrote:
:> > For this to work, you really MUST address the machdep.c problems.
:>
:> So how can it be done, are there any patches for machdep.c, or is it
:> solved in 4.4-stable kernel?
:
:Matt Dillon took a sideways stab at addressing a bit of these
:issues. They didn't do everything that I thought was needed,
:but he posted the diffs to the -current list, and may have
:committed them to -current (look for them there).
It's fixed in -stable and -current. Basically I comitted code
that puts a cap on the size of certain KVM blocks that were
previously scaled according to the amount of physical memory.
Since KVM is limited to the upper quarter of VM (1G) by default, the
scaling of these structures blew out our KVM space and paniced
machines with large amounts of memory. I also cut the size of the
swap map reservation in half.
It isn't perfect, but I've tested it on 4G machines and it does work.
What we really need to do is rearrange the kernel such that we can
calculate all the KVM requirements and scale them properly before
actually trying to allocate any of them.
:sometimes useful). I will cover where and how to do this for
:non-4M pages, in my article.
Go for it! I look forward to reading your article Terry!
:-- Terry
-Matt
Matthew Dillon
<dillon@backplane.com>
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200112090821.fB98LbQ36051>