Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 22 Feb 2002 14:50:27 -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:  <3C76CB33.988E246F@mindspring.com>
References:  <15478.31998.459219.178549@chiark.greenend.org.uk> <200202222042.g1MKg4u22700@apollo.backplane.com> <3C76C1C7.248128A4@mindspring.com> <200202222240.g1MMeQP31567@apollo.backplane.com>

next in thread | previous in thread | raw e-mail | index | archive | help
Matthew Dillon wrote:
> :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.
> 
>     Nope, object->memq is not sorted in any order whatsover.

I saw your patch.  Read other message.  8-).

> :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).
> 
>     What 'count' are you refering to?

That's "use count", not "usecount".  It's "count" as in
"has meaning as a significator".  8-).

I mean does msync() preterb the LRU order, by counting as
a recent use?  It's probably wrong to preterb the order,
since what you are doing is forcing pages *out*.  If you
want them *out*, then they probably have lower locality
than the pages you decided not to force out.

In other words, msync() should probably not change the
LRU order.

-- 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?3C76CB33.988E246F>