From owner-freebsd-hackers Fri Sep 24 19:19:41 1999 Delivered-To: freebsd-hackers@freebsd.org Received: from mail-out1.apple.com (mail-out1.apple.com [17.254.0.52]) by hub.freebsd.org (Postfix) with ESMTP id 88BD014C9D for ; Fri, 24 Sep 1999 19:19:39 -0700 (PDT) (envelope-from justin@walker3.apple.com) Received: from mailgate1.apple.com (A17-128-100-225.apple.com [17.128.100.225]) by mail-out1.apple.com (8.9.3/8.9.3) with ESMTP id TAA12455 for ; Fri, 24 Sep 1999 19:19:38 -0700 (PDT) Received: from scv2.apple.com (scv2.apple.com) by mailgate1.apple.com (mailgate1.apple.com- SMTPRS 2.0.15) with ESMTP id for ; Fri, 24 Sep 1999 19:19:33 -0700 Received: from walker3.apple.com (walkeridsl1.apple.com [17.219.158.66]) by scv2.apple.com (8.9.3/8.9.3) with ESMTP id TAA11600 for ; Fri, 24 Sep 1999 19:19:32 -0700 (PDT) Received: (from justin@localhost) by walker3.apple.com (8.9.1/8.9.1) id TAA00818 for hackers@FreeBSD.ORG; Fri, 24 Sep 1999 19:19:50 -0700 (PDT) Message-Id: <199909250219.TAA00818@walker3.apple.com> To: hackers@freebsd.org Subject: Re: mbufs question/problem Date: Fri, 24 Sep 1999 19:19:48 -0700 From: "Justin C. Walker" Reply-To: justin@apple.com X-Mailer: by Apple MailViewer (2.105.dev) Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > From: Dennis > Date: 1999-06-30 15:24:46 +0200 > To: hackers@FreeBSD.ORG > Subject: mbufs question/problem > Delivered-to: freebsd-hackers@freebsd.org > X-Sender: dennis@etinc.com > X-Mailer: QUALCOMM Windows Eudora Pro Version 4.0 > X-Loop: FreeBSD.ORG > Lines: 21 > > > I have a customer who has been experiencing "slow downs" with a freebsd > router....they have substantially increased performance by reducing > MINCLSIZE. I havent tracked the source, but im trying to hypothesize what > it might be. On the surface I cant see any relationship since very few > routines seem dependent on that value (m_devget() in particular, but I dont > believe they are using any driver that use it). Is it possible that they > are running out of small mbufs (they have NMBCLUSTERS set to a very high > value)? The problem, I believe, is the cost of using multiple mbufs per packet when a packet exceeds MHLEN but is less than MINCLSIZE. Probably the best performance, at the expense of memory, is had by setting MINCLSIZE=MHLEN. Regards, Justin -- Justin C. Walker, Curmudgeon-At-Large * Institute for General Semantics | Manager, CoreOS Networking | Men are from Earth. Apple Computer, Inc. | Women are from Earth. 2 Infinite Loop | Deal with it. Cupertino, CA 95014 | *-------------------------------------*-------------------------------* -- Justin C. Walker, Curmudgeon-At-Large * Institute for General Semantics | Manager, CoreOS Networking | Men are from Earth. Apple Computer, Inc. | Women are from Earth. 2 Infinite Loop | Deal with it. Cupertino, CA 95014 | *-------------------------------------*-------------------------------* To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message