From mboxrd@z Thu Jan 1 00:00:00 1970 From: Zach Brown Subject: Re: [RFC] loop: issue aio with pages Date: Mon, 26 Oct 2009 15:13:04 -0700 Message-ID: <4AE61EF0.8090309@oracle.com> References: <1256243157-16667-1-git-send-email-zach.brown@oracle.com> <20091025073655.GA6965@infradead.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: linux-fsdevel@vger.kernel.org To: Christoph Hellwig Return-path: Received: from rcsinet12.oracle.com ([148.87.113.124]:42903 "EHLO rgminet12.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754435AbZJZWOe (ORCPT ); Mon, 26 Oct 2009 18:14:34 -0400 In-Reply-To: <20091025073655.GA6965@infradead.org> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: Christoph Hellwig wrote: > Thanks, from the highlevel POV this looks much better than any of the > previous attempts on page-based kernel I/O. Great, I'll keep at it. > Just wondering, shouldn't > the first two patches be a separate series that can be merged even > before the rest is ready? Yeah, I think so. The second patch should be expanded a bit to really pull out all the retrying code. I can work on that series independently. Those first patches were just in this series to save the patch that introduced the aio api for kernel callers from having to worry about calling aio providers that would try to queue iocbs for retrying. - z