Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 4 Apr 2000 14:32:43 -0700
From:      Alfred Perlstein <bright@wintelcom.net>
To:        Andrew Gallatin <gallatin@cs.duke.edu>
Cc:        freebsd-hackers@FreeBSD.ORG
Subject:   Re: reducing the number of NFSv3 commit ops
Message-ID:  <20000404143243.S20770@fw.wintelcom.net>
In-Reply-To: <14570.22237.61025.935384@grasshopper.cs.duke.edu>; from gallatin@cs.duke.edu on Tue, Apr 04, 2000 at 04:59:00PM -0400
References:  <14570.10864.359054.10598@grasshopper.cs.duke.edu> <20000404141641.P20770@fw.wintelcom.net> <14570.22237.61025.935384@grasshopper.cs.duke.edu>

next in thread | previous in thread | raw e-mail | index | archive | help
* Andrew Gallatin <gallatin@cs.duke.edu> [000404 14:25] wrote:
> 
> Alfred Perlstein writes:
>  > > 
>  > > Can anybody tell me if doing something like this is fundamentally
>  > > broken?  Is it worth pursuing?
>  > 
>  > http://www.freebsd.org/~alfred/nfs_supercommit_broken.diff
>  > 
>  > only grab as many adjacent blocks as possible, you don't want to
>  > scan the entire file's buffer list for each commit, you also don't
>  > want to interfere with other client's caching forcing sever commits
>  > on thier behalf.
>  > 
> 
> I'll look at that tonight.  But before I do -- why is it broken?
> (the name sorta implies that it us ;)

I'm not sure, i did it a while back and ran out of time to get it
working, it functions in the strategy layer and tries to grab adjacent
commit blocks to the already clustered IO.

I think I may have some math errors or something, I haven't had time
to give it a retry in a while.

-- 
-Alfred Perlstein - [bright@wintelcom.net|alfred@freebsd.org]
"I have the heart of a child; I keep it in a jar on my desk."


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?20000404143243.S20770>