Subject: Re: CVS commit: src/sys/arch/i386/conf
To: YAMAMOTO Takashi <firstname.lastname@example.org>
From: Manuel Bouyer <email@example.com>
Date: 10/04/2005 11:48:57
On Tue, Oct 04, 2005 at 12:54:16PM +0900, YAMAMOTO Takashi wrote:
> > > see sbspace().
> > >
> > > > What would
> > > > be the consequence of this ?
> > >
> > > suboptimal flowcontrol.
> > Shouldn't this be fixed ? Setups with 4k clusters can be desireable on systems
> > using non-ethernet network adapters, or even with gigabit adapters.
> copying more eagerly in sbcompress?
> making cluster size variable?
Well, I don't know. I don't know anything about the socket code.
But making cluster size variable doesn't look a good idea at first glance,
as it would defeat what I'm trying to do in Xen :)
> anyway it doesn't belong source-changes. :)
Sure. But as I don't know anything about this part of the kernel I'll stop
Manuel Bouyer <firstname.lastname@example.org>
NetBSD: 26 ans d'experience feront toujours la difference