From owner-freebsd-bugs Mon May 13 08:01:46 1996 Return-Path: owner-bugs Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id IAA26468 for bugs-outgoing; Mon, 13 May 1996 08:01:46 -0700 (PDT) Received: from halloran-eldar.lcs.mit.edu (halloran-eldar.lcs.mit.edu [18.26.0.159]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id IAB26460 for ; Mon, 13 May 1996 08:01:42 -0700 (PDT) Received: by halloran-eldar.lcs.mit.edu; (5.65/1.1.8.2/19Aug95-0530PM) id AA04357; Mon, 13 May 1996 11:01:33 -0400 Date: Mon, 13 May 1996 11:01:33 -0400 From: Garrett Wollman Message-Id: <9605131501.AA04357@halloran-eldar.lcs.mit.edu> To: Tony Ardolino Cc: bugs@freebsd.org Subject: Freebsd, mbuf cluster size MINCLSIZE In-Reply-To: <31972068.4DCF@netcon.com> References: <31972068.4DCF@netcon.com> Sender: owner-bugs@freebsd.org X-Loop: FreeBSD.org Precedence: bulk < said: > What we don't understand is why is MINCLSIZE being set to 208 bytes when it > should be 100 bytes. Is there something wrong here or is that we just > don't understatnd what going on. You just don't understand what is going on. MINCLSIZE is the point at which the kernel switches from allocating a chain of regular mbufs to allocating clusters. There are a number of factors which go into the choice of this size, including speed, but the most important is memory usage. For small data, the benefits of using a cluster are more than outweighed by both the cost of the memory allocation (although less so than it used to be), and the amount of memory which would be wasted. MINCLSIZE is currently set to require at least two regular mbufs' worth of data before it is considered useful to allocate a cluster. -GAWollman -- Garrett A. Wollman | Shashish is simple, it's discreet, it's brief. ... wollman@lcs.mit.edu | Shashish is the bonding of hearts in spite of distance. Opinions not those of| It is a bond more powerful than absence. We like people MIT, LCS, ANA, or NSA| who like Shashish. - Claude McKenzie + Florent Vollant