Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 23 Jun 2009 15:14:31 -0500
From:      David Kelly <dkelly@hiwaay.net>
To:        FreeBSD-Questions@FreeBSD.org
Subject:   Re: you're not going to believe this.
Message-ID:  <20090623201431.GA43645@Grumpy.DynDNS.org>
In-Reply-To: <alpine.BSF.2.00.0906232145290.59407@wojtek.tensor.gdynia.pl>
References:  <20090622230729.GA20167@thought.org> <alpine.BSF.2.00.0906230929470.55064@wojtek.tensor.gdynia.pl> <20090623170739.GA33220@slackbox.xs4all.nl> <alpine.BSF.2.00.0906231922250.58723@wojtek.tensor.gdynia.pl> <20090623182225.GC33220@slackbox.xs4all.nl> <alpine.BSF.2.00.0906232145290.59407@wojtek.tensor.gdynia.pl>

next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, Jun 23, 2009 at 09:46:01PM +0200, Wojciech Puchar wrote:
> >>and lifetime.
> >
> >Even a flash filesystem will have to do wear levelling.
> 
> yes - but it don't have to copy blocks that are free. with disk
> emulation - it doesn't know anything about filesystem and don't know
> what blocks are free.

If it is swapping from heavily used blocks to lightly used blocks then
"so what" if there is an "unnecessary" read/write? Perhaps its harder to
determine if unused than to simply move the data. I seem to recall
something like this in comments in the FreeBSD virtual memory manager in
6.0-RELEASE.

Don't want to leave the old data laying around for security reasons so
even if the blocks are unused the formerly heavily used blocks need to
be scrubbed.

As I originally said to Gary Kline, "Don't let someone scare you away
from the 99.8% solution waiting on the 99.9% solution."

-- 
David Kelly N4HHE, dkelly@HiWAAY.net
========================================================================
Whom computers would destroy, they must first drive mad.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20090623201431.GA43645>