Subject: Re: results of the IRC debug patch
To: Neil Ludban <nludban@columbus.rr.com>
From: Bill Studenmund <wrstuden@netbsd.org>
List: port-macppc
Date: 12/10/2004 17:47:52
--GID0FwUMdk1T2AWN
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Sat, Dec 04, 2004 at 12:34:51PM -0500, Neil Ludban wrote:
> The levels blocked by intr_calcmasks() is the union of the IPL levels of
> all devices sharing the IRQ.  Sharing an IRQ between an audio and a USB
> device would be particularly bad: USB is supposed to be at a low priority
> (IPL_USB =3D=3D IPL_BIO), but this device gets elevated to IPL_AUDIO.  Ev=
en
> worse, any code running at IPL_BIO would block the shared IRQ, so the aud=
io
> interrupt will experience high latency.

When ever I thought of priorities, I'd have put audio below IPL_BIO. Maybe=
=20
that's my mac68k damage...

Take care,

Bill

--GID0FwUMdk1T2AWN
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (NetBSD)

iD8DBQFBulHIWz+3JHUci9cRAq6nAJ4oaG5wBkdXY3O6vNoAP6VfjZDupQCcDt5v
rylBKIvtSqZxlnmTAlefwNs=
=tCdX
-----END PGP SIGNATURE-----

--GID0FwUMdk1T2AWN--