Date: Mon, 19 May 1997 00:24:55 -0500 (EST) From: "John S. Dyson" <toor@dyson.iquest.net> To: davem@caip.rutgers.edu (David S. Miller) Cc: dyson@FreeBSD.ORG, terry@lambert.org, deischen@iworks.InterWorks.org, freebsd-hackers@FreeBSD.ORG, jb@freebsd1.cimlogic.com.au Subject: Re: GNAT-pthreads integration bugs/questions Message-ID: <199705190524.AAA10655@dyson.iquest.net> In-Reply-To: <199705190428.AAA16419@darkwing.rutgers.edu> from "David S. Miller" at "May 19, 97 00:28:20 am"
next in thread | previous in thread | raw e-mail | index | archive | help
> From: "John S. Dyson" <toor@dyson.iquest.net> > Date: Sun, 18 May 1997 23:24:40 -0500 (EST) > > That's the reason that we don't do it. > > Don't you have to setup some sort of mapping to get at vfs pages > during (for example) a read/write? (so this essentially is > pre-faulting the page, avoiding the trap, and I imagine you then tear > down the mapping after the operation is complete?) > Just create a buffer cache buffer. Right now, we pass the page pointers and mappings along in the buffer, but the drivers right now still use the kernel VM mappings. The buffer cache code currently provides both. If mappings are not provided, they can be created at any time or never. John
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199705190524.AAA10655>