From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mark Fasheh Date: Mon Aug 30 11:10:59 2004 Subject: [Ocfs2-devel] 2.6 I/O path totally messed up In-Reply-To: <20040830135022.GA17620@lst.de> References: <20040830135022.GA17620@lst.de> Message-ID: <20040830161052.GC15160@ca-server1.us.oracle.com> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: ocfs2-devel@oss.oracle.com On Mon, Aug 30, 2004 at 03:50:22PM +0200, Christoph Hellwig wrote: > The second argumentto generic_file_write_nolock is an iovec, not an > actual userbuffer. The current code doesn't even have the slightest > chance to work. Yeah, I'm embarrassed to say we missed that in the commit for that code. I'll be fixing it up asap. > Maybe it's time to decouple the 2.6 read/write code > from the 2.4 code and implement proper vectored operations everywhere? Probably - we've got limited resources, so patches are welcome :) > (and kill tge O_DIRECT vs O_APPEN hack, and the broken fallback code, > and..) No comment :) --Mark -- Mark Fasheh Software Developer, Oracle Corp mark.fasheh@oracle.com