Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 22 Feb 2002 14:10:15 -0800
From:      Terry Lambert <tlambert2@mindspring.com>
To:        Matthew Dillon <dillon@apollo.backplane.com>
Cc:        Andrew Mobbs <andrewm@chiark.greenend.org.uk>, hackers@FreeBSD.ORG
Subject:   Re: Re2: msync performance
Message-ID:  <3C76C1C7.248128A4@mindspring.com>
References:  <15478.31998.459219.178549@chiark.greenend.org.uk> <200202222042.g1MKg4u22700@apollo.backplane.com>

next in thread | previous in thread | raw e-mail | index | archive | help
Matthew Dillon wrote:
>     So, this falls back to your suggested solution.... sort
>     object->memq (it's the actual page queue that is the problem,
>     not the object queue).  Looking at it some more I believe
>     this may be a viable solution.  I am going to work something
>     up.

You will need to put it on two lists, I think.

Though it is not sorted in adjacency order, it is sorted
into LRU order, I think, and simply resorting by adjacency
would destroy the LRU property.

Maybe you could sort it on demand, and not keep the sort
list?

Alternately, does msync() use count as "use" for the
purposes of LRU?  If so, I'm all wet, and sorting it will
work, if it's only done at msync() time (e.g. you won't
be able to use an insertion sort).

-- Terry

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?3C76C1C7.248128A4>