--AVsU0XxfC6ed1aQ6
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Thu, Dec 16, 2004 at 06:13:42PM +1100, Daniel Carosone wrote:
>=20
> Conditions for issuing a synccache in this fashion might include:
>  - all currently pending write requests in the completion queue
>  - some threshold magic number of requests issued since the last one
>  - some timeout since the last one, to try and reduce potential latency
>=20
> but actually I think only the first condition is strictly necessary,

Perhaps the second condition is also a good idea, to prevent a
constant stream of write requests letting the completion queue grow
inordinately (one busy filesystem or dd starving another in another
partition?).  This could be done as a limit on the length of the
completion queue, or perhaps even count all transactions (including
reads) to try and get better balance.

Also, another condition perhaps needs to be added, to match
assumptions about ordered vs unordered/tagged commands on scsi disks:
 - something requested a sync write

--
Dan.

--AVsU0XxfC6ed1aQ6
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.6 (NetBSD)

iD8DBQFBwTpCEAVxvV4N66cRAqN1AJ9Hv9uiCP2O6lAwz6pPw2fowjcGSACg9Qo0
iK2hYQhRf2BQZss8OXUO10s=
=mgqu
-----END PGP SIGNATURE-----

--AVsU0XxfC6ed1aQ6--