Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 10 Jul 2003 17:15:15 -0500 (CDT)
From:      Mike Silbersack <silby@silby.com>
To:        "Alan L. Cox" <alc@imimic.com>
Cc:        cvs-all@FreeBSD.org
Subject:    Re: cvs commit: src/sys/kern subr_param.c sys_pipe.c src/sys/sys   pipe.h
Message-ID:  <20030710171232.T1451@odysseus.silby.com>
In-Reply-To: <3F0DCEFE.2916E707@imimic.com>
References:  <200307080457.h684vRM7009343@gw.catspoiler.org>    <20030708004340.T6733@odysseus.silby.com> <20030710182436.GA6484@HAL9000.homeunix.com> <20030710153734.GA1101@technokratis.com> <3F0DCEFE.2916E707@imimic.com>

next in thread | previous in thread | raw e-mail | index | archive | help

On Thu, 10 Jul 2003, Alan L. Cox wrote:

> The other factor here is PAE.  Some of our autosizing functions take
> into account the amount of physical memory but not the amount of
> available KVA.  The old vnode autosizing was a case in point.
>
> Regards,
> Alan

FWIW, I'm going to add a init_param3 function in subr_param which will be
used to auto-scale values which should scale to the size of the kernel
map, not the size of physical ram.  Pipes will be sized in there, and I
may move over some of the other autosizing values as well.  That should
improve the autosizing situation quite a bit.

Mike "Silby" Silbersack



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030710171232.T1451>