Date: Thu, 17 May 2001 03:28:06 -0400 From: Alfred Perlstein <bright@rush.net> To: Matt Dillon <dillon@earth.backplane.com> Cc: Tor.Egge@fast.no, arch@FreeBSD.ORG Subject: Re: Preliminary O_DIRECT patch (for review only, not yet tested!) Message-ID: <20010517032806.N12365@superconductor.rush.net> In-Reply-To: <200105170647.f4H6lkk88458@earth.backplane.com>; from dillon@earth.backplane.com on Wed, May 16, 2001 at 11:47:46PM -0700 References: <200105162222.f4GMMpC81247@earth.backplane.com> <200105162331.BAA04708@midten.fast.no> <200105170647.f4H6lkk88458@earth.backplane.com>
next in thread | previous in thread | raw e-mail | index | archive | help
* Matt Dillon <dillon@earth.backplane.com> [010517 02:49] wrote: > > I've looked at the rawread/rawwrite issue and I believe it may be > possible to use the already-existing B_VMIO flag coupled with > some VM magic to achieve the equivalent in the buffer cache itself > rather then having to write a rawread/rawwrite function for each > filesystem. Filesystems already support B_VMIO. If it is possible, > then we'll have a general raw I/O solution. I'm not sure what the rawread/rawwrite functions do (I'll review them if/when I get a chance) however if what they do is avoid remapping things into kernel memory it'd be really nice to see them go into the system. Avoiding vm tricks if possible would be nice. -Alfred To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010517032806.N12365>