linux-kernel - Re: vfs: Add MS_FLUSHONFSYNC mount flag

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] [day] [month] [year] [list]

Date:	Sun, 22 Feb 2009 20:23:36 -0500
From:	Theodore Tso <tytso@....edu>
To:	Jeff Garzik <jeff@...zik.org>
Cc:	Pavel Machek <pavel@....cz>, Eric Sandeen <sandeen@...hat.com>,
	Jan Kara <jack@...e.cz>,
	Fernando Luis V?zquez Cao <fernando@....ntt.co.jp>,
	Alan Cox <alan@...rguk.ukuu.org.uk>,
	kernel list <linux-kernel@...r.kernel.org>,
	Jens Axboe <jens.axboe@...cle.com>, fernando@....ac.jp,
	Ric Wheeler <rwheeler@...hat.com>,
	Andrew Morton <akpm@...ux-foundation.org>
Subject: Re: vfs: Add MS_FLUSHONFSYNC mount flag

On Sun, Feb 22, 2009 at 06:46:51PM -0500, Jeff Garzik wrote:
>
> To be more clear / precise, this means actually performing the  
> guarantees we claim to the user.  For example,
>
> fsync(2) on ext2 should trigger a storage device writeback cache flush  
> [or equivalent guarantee via FUA].
>
> fsync(2) or journal commit on ext3 should trigger a flush [or equivalent  
> guarantee via FUA].
>
> Though, certainly, the user should be able to disable this strict  
> behavior and trade correctness for performance.

No argument here.  The *default* should be that we respect fsync().
However, the user should be able to configure with a reasonable amount
of granularity tradeoffs between safety and performance and/or battery
usage.

						- Ted
--
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/