From owner-freebsd-current Wed Aug 28 20:28:40 1996 Return-Path: owner-current Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id UAA08450 for current-outgoing; Wed, 28 Aug 1996 20:28:40 -0700 (PDT) Received: (from jmb@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id UAA08441; Wed, 28 Aug 1996 20:28:38 -0700 (PDT) From: "Jonathan M. Bresler" Message-Id: <199608290328.UAA08441@freefall.freebsd.org> Subject: Re: [Q] mbuf 128 vs 1k bytes ?? To: dg@root.com Date: Wed, 28 Aug 1996 20:28:37 -0700 (PDT) Cc: current@freefall.freebsd.org In-Reply-To: <199608290306.UAA20774@root.com> from "David Greenman" at Aug 28, 96 08:06:38 pm X-Mailer: ELM [version 2.4 PL24] Content-Type: text Sender: owner-current@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk David Greenman wrote: > > Because mbufs are no longer allocated out of the malloc pool. The 1K entry > that you are seeing is just some malloc() in the kernel that (probably bogusly) > specified the M_MBUF type. I only have 1 of them on my machine: > > Type InUse MemUse HighUse Limit Requests Limit Limit Size(s) > mbuf 1 1K 1K 19661K 1 0 0 1K > > ...anyway, this is all much to do about nothing. To find out how many > network buffers are in-use, use netstat -m. ah...that's might be all the hint that i needed: how about this code in machdep.c 1.199 1996/08/19 line 365 /* * Finally, allocate mbuf pool. Since mclrefcnt is an off-size * we use the more space efficient malloc in place of kmem_alloc. */ mclrefcnt = (char *)malloc(nmbclusters+PAGE_SIZE/MCLBYTES, M_MBUF, M_NOWAIT); bzero(mclrefcnt, nmbclusters+PAGE_SIZE/MCLBYTES); looking at /sys/sys/malloc.h, i dont see what else this might be classified as ;( jmb -- Jonathan M. Bresler FreeBSD Postmaster jmb@FreeBSD.ORG FreeBSD--4.4BSD Unix for PC clones, source included. http://www.freebsd.org/ PGP 2.6.2 Fingerprint: 31 57 41 56 06 C1 40 13 C5 1C E3 E5 DC 62 0E FB