Date: Mon, 11 Oct 1999 10:20:02 +0930 From: Greg Lehey <grog@lemis.com> To: Matthew Dillon <dillon@apollo.backplane.com> Cc: Poul-Henning Kamp <phk@critter.freebsd.dk>, Peter Wemm <peter@netplex.com.au>, "Daniel C. Sobral" <dcs@newsguy.com>, cvs-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/vm vm_swap.c Message-ID: <19991011102002.H78191@freebie.lemis.com> In-Reply-To: <199910101624.JAA16685@apollo.backplane.com> References: <15397.939571805@critter.freebsd.dk> <199910101624.JAA16685@apollo.backplane.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sunday, 10 October 1999 at 9:24:55 -0700, Matthew Dillon wrote: >>> useable as anything else as this strategy routine doesn't actually work for >>> normal device access. libkvm doesn't use it, at least not for a very long >>> time. IMHO, the strategy routine should be moved elsewhere for the VM swap >>> code to use as a subroutine and the whole "device" should be deep-six'ed. >> >> But isn't this really just a "striping" function which could be done >> equally well by ccd or vinum ? > > It's similar, but neither ccd nor vinum can guarentee non-blocking > operation in low-memory situations. Vinum will use private memory if it can't get a malloc. > And, anyhow, the swapper is my baby. That in itself isn't a reason. But I think that it's possible for the swapper to know more about the usage of the data, and thus achieve a better load balance across the disks than a virtual device layer with a fixed striping mechanism. Greg -- See complete headers for address, home page and phone numbers finger grog@lemis.com for PGP public key To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19991011102002.H78191>