From owner-freebsd-hackers Sun Apr 18 13: 0:40 1999 Delivered-To: freebsd-hackers@freebsd.org Received: from apollo.backplane.com (apollo.backplane.com [209.157.86.2]) by hub.freebsd.org (Postfix) with ESMTP id 79C8814D3B for ; Sun, 18 Apr 1999 13:00:38 -0700 (PDT) (envelope-from dillon@apollo.backplane.com) Received: (from dillon@localhost) by apollo.backplane.com (8.9.3/8.9.1) id MAA81828; Sun, 18 Apr 1999 12:58:01 -0700 (PDT) (envelope-from dillon) Date: Sun, 18 Apr 1999 12:58:01 -0700 (PDT) From: Matthew Dillon Message-Id: <199904181958.MAA81828@apollo.backplane.com> To: Peter Wemm Cc: "John S. Dyson" , dyson@iquest.net, dg@root.com, hackers@freebsd.org Subject: Re: Directories not VMIO cached at all! References: <19990418164232.4DC7C1F2A@spinner.netplex.com.au> Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG :> the amount of memory used for caching directories. The disadvantage :> is the potentially gross amount of internal fragmentation of memory. :> :> Perhaps before getting rid of B_MALLOC, take a look at the standard :> mix of directory sizes (don't just look at news servers.) If there is an :> extreme bias towards 512 or 2048, then you might consider keeping B_MALLOC. : :Would small block devices/filesystems likely be affected? (ie: msdos, :ext2fs etc) : :Cheers, :-Peter Remember all those VFS/BIO fixes Luoqi made a few weeks ago? They were mainly to fix bugs in VFS/BIO relating to small block filesystems. small block filesystems use VMIO, but since the block size is less then a page the base offset in the struct buf for any given small-block buffer is *NOT* the same as the base offset the page being mapped to that struct buf. We could conceivably do the same thing with directories, but it probably would not be worth the hassle. -Matt Matthew Dillon To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message