Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 24 Apr 2001 12:48:57 +0300
From:      Peter Pentchev <roam@orbitel.bg>
To:        Bob Bishop <rb@gid.co.uk>
Cc:        niek@bergboer.net, freebsd-hackers@FreeBSD.ORG, Mike Silbersack <silby@silby.com>
Subject:   Re: UFS block size vs. write speed
Message-ID:  <20010424124857.C22159@ringworld.oblivion.bg>
In-Reply-To: <4.3.2.7.2.20010424103925.00a81ae0@gid.co.uk>; from rb@gid.co.uk on Tue, Apr 24, 2001 at 10:43:29AM %2B0100
References:  <Pine.BSF.4.31.0104231207360.1931-100000@achilles.silby.com> <20010423134554.A57241@wit379119.student.utwente.nl> <Pine.BSF.4.31.0104231207360.1931-100000@achilles.silby.com> <20010424113536.A61988@wit379119.student.utwente.nl> <4.3.2.7.2.20010424103925.00a81ae0@gid.co.uk>

next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, Apr 24, 2001 at 10:43:29AM +0100, Bob Bishop wrote:
> Hi,
> 
> At 11:35 24/04/01 +0200, Niek Bergboer wrote:
> >[...]
> >In fact, I couldn't care less if the allocated blocks contain random
> >data (rather than zeros), since I'll be overwriting them immediately.
> 
> You *should* care: the blocks are zeroed for security reasons.

But he's only doing a speed test :)  And besides, if the 'random data'
that's in the blocks is only zeroed for security reasons, there's
absolutely no point in zeroing it in the mmap()..  If someone removed
a file and wanted to make sure no one read his data, the blocks should
have been *already* zeroed in the process of removal. (Which, as discussed
on several mailing lists lately, cannot really be guaranteed with
today's caching/remapping/virtual/whatnot FS's :)

G'luck,
Peter

-- 
This sentence is false.

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?20010424124857.C22159>