From owner-freebsd-hackers@FreeBSD.ORG Wed Feb 16 12:46:37 2005 Return-Path: Delivered-To: freebsd-hackers@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id B240516A4CE for ; Wed, 16 Feb 2005 12:46:37 +0000 (GMT) Received: from cyrus.watson.org (cyrus.watson.org [204.156.12.53]) by mx1.FreeBSD.org (Postfix) with ESMTP id 3753F43D2D for ; Wed, 16 Feb 2005 12:46:37 +0000 (GMT) (envelope-from robert@fledge.watson.org) Received: from fledge.watson.org (fledge.watson.org [204.156.12.50]) by cyrus.watson.org (Postfix) with SMTP id C40A946B86; Wed, 16 Feb 2005 07:46:36 -0500 (EST) Date: Wed, 16 Feb 2005 12:45:14 +0000 (GMT) From: Robert Watson X-Sender: robert@fledge.watson.org To: Borja Marcos In-Reply-To: <02266ed294653cda2452e7ac0a713c2c@sarenet.es> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: freebsd-hackers@freebsd.org Subject: Re: MBUF statistics X-BeenThere: freebsd-hackers@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Technical Discussions relating to FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 16 Feb 2005 12:46:37 -0000 On Tue, 15 Feb 2005, Borja Marcos wrote: > Looking at the mbuf statistics available in FreeBSD 4 and > FreeBSD 5 I can see that the statistics available in FreeBSD 5 are, > surprisingly, much less comprehensive. Is there any other place where I > can find out how many mbuf requests have been done, how many of them > have waited, how many have failed, etc? > > I think that it's an interesting resource to graph in a Orca > performance monitor for FreeBSD. This statistic appears to have been lost during the conversion to UMA allocating mbufs. I probably won't get to it this week, but I'll try to get that statistic back in the mix in the next week or two. Unfortunately, it may break the user<->kernel ABI for monitoring applications, since there appear to be no spare fields in the mbstat structure. What I'd like to do, actually, is add tracking of allocation failures to UMA, and then use UMA statistics to generate the mbuf output of netstat rather than a separate API. This would also eliminate SMP races present in the maintenance of mbuf statistics (they can "drift" on SMP boxes right now). Robert N M Watson