From owner-freebsd-stable Sat Jan 19 5:43:44 2002 Delivered-To: freebsd-stable@freebsd.org Received: from rwcrmhc53.attbi.com (rwcrmhc53.attbi.com [204.127.198.39]) by hub.freebsd.org (Postfix) with ESMTP id 7A09237B405 for ; Sat, 19 Jan 2002 05:43:38 -0800 (PST) Received: from bmah.dyndns.org ([12.233.149.189]) by rwcrmhc53.attbi.com (InterMail vM.4.01.03.27 201-229-121-127-20010626) with ESMTP id <20020119134338.YPGJ10199.rwcrmhc53.attbi.com@bmah.dyndns.org>; Sat, 19 Jan 2002 13:43:38 +0000 Received: (from bmah@localhost) by bmah.dyndns.org (8.11.6/8.11.6) id g0JDhbN72777; Sat, 19 Jan 2002 05:43:37 -0800 (PST) (envelope-from bmah) Message-Id: <200201191343.g0JDhbN72777@bmah.dyndns.org> X-Mailer: exmh version 2.5 07/13/2001 with nmh-1.0.4 To: Thomas Seck Cc: stable@FreeBSD.ORG, dillon@apollo.backplane.com Subject: Re: what's vnlru? In-reply-to: <20020119132902.GB676@laurel.seck.home> References: <200201151618468.SM01176@141.com> <200201160101.g0G117r64693@apollo.backplane.com> <20020116104904.A2800@shikima.mine.nu> <20020116135318.GA427@laurel.seck.home> <200201190102.g0J12MF37253@apollo.backplane.com> <20020119132902.GB676@laurel.seck.home> Comments: In-reply-to Thomas Seck message dated "Sat, 19 Jan 2002 14:29:02 +0100." From: "Bruce A. Mah" Reply-To: bmah@FreeBSD.ORG X-Face: g~c`.{#4q0"(V*b#g[i~rXgm*w;:nMfz%_RZLma)UgGN&=j`5vXoU^@n5v4:OO)c["!w)nD/!!~e4Sj7LiT'6*wZ83454H""lb{CC%T37O!!'S$S&D}sem7I[A 2V%N&+ X-Image-Url: http://www.employees.org/~bmah/Images/bmah-cisco-small.gif X-Url: http://www.employees.org/~bmah/ Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 19 Jan 2002 05:43:37 -0800 Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG If memory serves me right, Thomas Seck wrote: [snip] > > + The kernel now enforces the kern.maxvnodes limit with > > + a new kernel process called vnlru. The > > + code typically only needs to do work on large-memory systems > > + which access lots of tiny files and prevents us from overflowing > > + the kernel malloc hoppers for vnodes or inodes. [snip] > Hmm, this sounds like one can turn it on or off (which is not the > case, right?) and people then tend to ask how to do it. How about > rewording the second sentence into something like "This code prevents > the system from overflowing the kernel malloc hoppers for vnodes or > inodes when a large number of small files are accessed on large memory > systems"? (Please apply grammar fixes as needed, apologies for my german > high-school english). So...here's what I already committed a day or two ago to the release notes for both CURRENT and 4-STABLE... The kern.maxvnodes limit now properly limits the number of vnodes in use. Previously only vnodes with no cached pages could be freed; this could allow the number of vnodes to grow without limit on large-memory machines accessing many small files. A vnlru kernel thread helps to flush and reuse vnodes. Matt...if you'd like me to replace this with what you wrote or make changes, just give me a yell. Bruce. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message