Subject: Re: wd, disk write cache, sync cache, and softdep.
To: Daniel Carosone <firstname.lastname@example.org>
From: Jason Thorpe <email@example.com>
Date: 12/16/2004 14:04:04
Content-Type: text/plain; charset=US-ASCII; format=flowed
On Dec 16, 2004, at 12:18 PM, Daniel Carosone wrote:
> The thing I liked about the completion queue idea is that it stays
> etirely within the disk driver layer, and merely restores the disk
> semantics that are assumed by all the upper layers: biodone buffers
> are safely on stable storage.
> The ideas are not incompatible, of course: arrival of one a write with
> one of these force flags would be another condition to trigger a sync
I think you are going down the road of "really bad performance" with
What if the drive has a non-volatile cache?
Why issue two commands for this when one will do (in the case of SCSI
and the FUA bit).
> I wasn't aware that there was a problem for SCSI disks, as Bill
> suggests, by the way. I don't know if the filesystems are smart
> enough, or have the interface to indicate, that a given write should
> be done as a tagged or ordered command, or whether they just wait to
> issue later writes until earlier ones are returned.
You don't have to care about this. The application (in this case, FFS
or whatever) enforces its own barriers of this type, by simply not
starting an I/O before its dependant I/O is known to have completed.
There is no need for us to fiddle with SIMPLE and ORDERED tags... just
let the driver issue SIMPLE all it wants.
-- Jason R. Thorpe <firstname.lastname@example.org>
content-type: application/pgp-signature; x-mac-type=70674453;
content-description: This is a digitally signed message part
content-disposition: inline; filename=PGP.sig
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (Darwin)
-----END PGP SIGNATURE-----