Date: Fri, 27 Dec 1996 22:19:01 -0500 (EST) From: "John S. Dyson" <toor@dyson.iquest.net> To: terry@lambert.org (Terry Lambert) Cc: julian@whistle.com, luigi@labinfo.iet.unipi.it, proff@iq.org, danny@panda.hilink.com.au, hackers@freebsd.org Subject: Re: ipretard.c selective tcp/ip queues and throughput limiters Message-ID: <199612280319.WAA07203@dyson.iquest.net> In-Reply-To: <199612272306.QAA25429@phaeton.artisoft.com> from "Terry Lambert" at Dec 27, 96 04:06:08 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> > For instance, for FS's, the VFS is well defined, but the VM interface > for doing actual disk I/O is not well abstracted at all. > The VM code uses direct disk input when the filesystem allows it. Otherwise, the filesystems can do VOP_READ operations as in normal UFS read write operations. The vnode pageout uses VOP_WRITE operations. Refer to vnode_pager (where the VM input output operations happen.) I don't think that there are layering problems anymore (there were initially though.) Note that the filesystem can take over all paging operations now -- used to be they couldn't. John
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199612280319.WAA07203>