linux-kernel - Re: AZFS file system proposal

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20080617160628.79efd2d9@mercedes-benz.boeblingen.de.ibm.com>
Date:	Tue, 17 Jun 2008 16:06:28 +0200
From:	Maxim Shchetynin <maxim@...ux.vnet.ibm.com>
To:	linux-fsdevel@...r.kernel.org
Cc:	Jörn Engel <joern@...fs.org>,
	Jan Engelhardt <jengelh@...ozas.de>,
	Matthew Wilcox <matthew@....cx>, linux-kernel@...r.kernel.org
Subject: Re: AZFS file system proposal

On Tue, 17 Jun 2008 12:53:00 +0200
Jörn Engel <joern@...fs.org> wrote:

> On Tue, 17 June 2008 11:35:10 +0200, Jan Engelhardt wrote:
> > On Tuesday 2008-06-17 11:06, Maxim Shchetynin wrote:
> > 
> > >I would say you can either mmap it or place a file system on it.
> > 
> > The help text implies that I *need* azfs to mmap it - and that sounds
> > like a Rube Goldberg machine.
> 
> Some people actually prefer filesystems over raw devices for a variety
> of reasons:
> - each file brings its own address space, which offers memory protection
>   from other processes,
> - files can have owners and permission bits,
> - files hide the fragmentation of the underlying device from users,
> - a file system provides a common and well-understood api for devices
>   with less common or well-understood apis,
> - etc.
> 
> Those reasons are as valid for azfs as for any other filesystem.  I have
> no doubt that azfs is useful.  It probably wouldn't hurt to express the
> merits of the filesystem and the problems it is supposed to solve a
> little better.  So far most criticism was based on the fact that noone
> understood what the hell it was all about.
> 
> My personal question when looking at this is: Why not use ext2?  It

Same reason we don't use ext2 for file system in RAM - we don't need any expensive caching and readahead mechanisms for devices which are as fast as main memory (or almost as fast).

> appears to me that an ext2 mounted with '-o xip' would solve the same
> problems.

-- 
Mit freundlichen Grüßen / met vriendelijke groeten / avec regards

    Maxim V. Shchetynin
    Linux Kernel Entwicklung
    IBM Deutschland Entwicklung GmbH
    Linux für Cell, Abteilung 3250
    Schönaicher Straße 220
    71032 Böblingen

Vorsitzender des Aufsichtsrats: Johann Weihen
Geschäftsführung: Herbert Kircher
Sitz der Gesellschaft: Böblingen
Registriergericht: Amtsgericht Stuttgart, HRB 243294

Fahr nur so schnell wie dein Schutzengel fliegen kann!
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/