Subject: Re: newlock
To: Jason Thorpe <email@example.com>
From: Bill Studenmund <firstname.lastname@example.org>
Date: 09/05/2006 16:39:17
Content-Type: text/plain; charset=us-ascii
On Tue, Sep 05, 2006 at 04:17:48PM -0700, Jason Thorpe wrote:
> On Sep 5, 2006, at 3:40 PM, Bill Studenmund wrote:
> >Another would be an iSCSI initiator where splbio() and splnet() get
> >mingled. And heven help someone trying to do iSCSI over ppp. :-)
> Sorry, this is really splbio() and splsoftnet().
> What this really points to is "we do too much stuff in interrupt =20
> context", and really ought to be deferring more to threads / =20
> continuations / just-in-time processing.
I was thinking something like that would be the best fix, but I don't=20
quite know how to get us there w/o having to have everything at once.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.3 (NetBSD)
-----END PGP SIGNATURE-----