| To: | Michael Lampe <Michael.Lampe@xxxxxxxxxxxxxxxxxxxxx> |
|---|---|
| Subject: | Re: 2.4.25 & xfs & ide write barriers |
| From: | Danny Cox <danscox@xxxxxxxxxxxxxx> |
| Date: | Wed, 03 Mar 2004 17:44:35 -0500 |
| Cc: | Jeremy Jackson <jerj@xxxxxxxxxxxx>, XFS Mailing List <linux-xfs@xxxxxxxxxxx> |
| In-reply-to: | <40436E6A.5020800@iwr.uni-heidelberg.de> |
| Organization: | No Organization at ALL |
| References: | <403E0A7A.3040505@iwr.uni-heidelberg.de> <20040229221924.GA731@frodo> <4043355E.9080208@iwr.uni-heidelberg.de> <40435121.4050906@coplanar.net> <40436E6A.5020800@iwr.uni-heidelberg.de> |
| Sender: | linux-xfs-bounce@xxxxxxxxxxx |
Jeremy & Michael,
On Mon, 2004-03-01 at 12:10, Michael Lampe wrote:
> Jeremy Jackson wrote:
>
> > SCSI disks
> > use TCQ to achieve write performance without compromising data
> > integrity.
>
> Ordered queue tags can be used to implement write barriers.
>
> > (XFS guys - Linux does this, right?)
>
> No. (I'm not an XFS guy, though.)
Actually, the answer is: "not yet". There is extant code, but the
controllers and disks that can actually make use of it is rather thin
right now (in the ATA camp).
--
kernel, n.: A part of an operating system that preserves the
medieval traditions of sorcery and black art.
Danny
|
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: Desktop Filesystem Benchmarks in 2.6.3, Kristian Köhntopp |
|---|---|
| Next by Date: | Re: [Jfs-discussion] Re: Desktop Filesystem Benchmarks in 2.6.3, Pascal Gienger |
| Previous by Thread: | Re: 2.4.25 & xfs & ide write barriers, Michael Lampe |
| Next by Thread: | mkfs.xfs options with Linux RAID, AndyLiebman |
| Indexes: | [Date] [Thread] [Top] [All Lists] |