Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 28 Mar 2007 18:59:34 +0100 (BST)
From:      Robert Watson <rwatson@FreeBSD.org>
To:        Ivan Voras <ivoras@fer.hr>
Cc:        freebsd-current@freebsd.org
Subject:   Re: NFS write() calls lead to read() calls?
Message-ID:  <20070328185815.I1185@fledge.watson.org>
In-Reply-To: <eue3ku$mq2$1@sea.gmane.org>
References:  <7ad7ddd90703280238r5dd3f30ftc1641926ecdf44a8@mail.gmail.com> <eudfat$g2m$1@sea.gmane.org> <7ad7ddd90703280611p5c0ca4e1y600315551391a813@mail.gmail.com> <eue3ku$mq2$1@sea.gmane.org>

next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, 28 Mar 2007, Ivan Voras wrote:

> Ulrich Spoerlein wrote:
>
>> What kind of C program or script did you have in mind? My C-foo is very 
>> weak ...
>
> If you have python installed, this is a simple way:

Also good is dd between the file system and /dev/null.  Something worth 
remembering is that some tools (cp(1) in particular) use memory-mapped I/O, 
which may behave differently than raw I/O operations.

Robert N M Watson
Computer Laboratory
University of Cambridge

>
> ----
> print "writing"
> f = file("a_file", "w")
> for x in xrange(1024): # write 1024 MB
>    f.write(' '*1024*1024)
> f.close()
>
> raw_input("press enter to rewrite")
>
> print "rewriting"
> f = file("a_file", "r+")
> for x in xrange(1024): # write 1024 MB again
>    f.write(' '*1024*1024)
> f.close()
> ----
>
> save it to a file, run the file with python interpreter. If you don't observe 
> the weird read-before-write effect, then it's not a problem in FreeBSD / NFS.
>
>



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20070328185815.I1185>