From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from bedivere.hansenpartnership.com ([66.63.167.143]:45650 "EHLO bedivere.hansenpartnership.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757958AbcB1Q0s (ORCPT ); Sun, 28 Feb 2016 11:26:48 -0500 Message-ID: <1456676806.2324.4.camel@HansenPartnership.com> Subject: Re: [Lsf-pc] [LSF/MM ATTEND] block: multipage bvecs From: James Bottomley To: Christoph Hellwig , Ming Lei Cc: lsf-pc@lists.linuxfoundation.org, linux-block@vger.kernel.org, Linux FS Devel Date: Sun, 28 Feb 2016 08:26:46 -0800 In-Reply-To: <20160228160716.GA12881@infradead.org> References: <20160228160716.GA12881@infradead.org> Content-Type: text/plain; charset="UTF-8" Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Sender: linux-fsdevel-owner@vger.kernel.org List-ID: On Sun, 2016-02-28 at 08:07 -0800, Christoph Hellwig wrote: > I've been playing around with Kent's old patches a bit as well, > and I think building up larger bvecs from the start would be > very useful, and it might make sense to get some broader exposure > of the issues around it. You mean in bio_add_page() the code which currently aggregates chunks within a page could build a bio vec entry up to the max segment size? I think that is reasonable, especially now the bio splitting code can actually split inside a bio vec entry. James