From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755401Ab2JVPgN (ORCPT ); Mon, 22 Oct 2012 11:36:13 -0400 Received: from mail-pa0-f46.google.com ([209.85.220.46]:57332 "EHLO mail-pa0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753267Ab2JVPgK (ORCPT ); Mon, 22 Oct 2012 11:36:10 -0400 Date: Mon, 22 Oct 2012 08:36:06 -0700 From: Greg Kroah-Hartman To: Tejun Heo Cc: Ben Hutchings , linux-kernel@vger.kernel.org, stable@vger.kernel.org, alan@lxorguk.ukuu.org.uk, Vivek Goyal , Jens Axboe Subject: Re: [ 12/37] block: fix request_queue->flags initialization Message-ID: <20121022153606.GA13399@kroah.com> References: <20121019031302.789593147@linuxfoundation.org> <20121019031304.836071629@linuxfoundation.org> <20121019231604.GF13292@decadent.org.uk> <20121021162529.GA25856@kroah.com> <20121021185618.GA5951@atj.dyndns.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20121021185618.GA5951@atj.dyndns.org> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Sun, Oct 21, 2012 at 11:56:18AM -0700, Tejun Heo wrote: > Hello, Greg. > > On Sun, Oct 21, 2012 at 09:25:29AM -0700, Greg Kroah-Hartman wrote: > > > > commit 60ea8226cbd5c8301f9a39edc574ddabcb8150e0 upstream. > > > > > > > > A queue newly allocated with blk_alloc_queue_node() has only > > > > QUEUE_FLAG_BYPASS set. For request-based drivers, > > > > blk_init_allocated_queue() is called and q->queue_flags is overwritten > > > > with QUEUE_FLAG_DEFAULT which doesn't include BYPASS even though the > > > > initial bypass is still in effect. > > > > > > > > In blk_init_allocated_queue(), or QUEUE_FLAG_DEFAULT to q->queue_flags > > > > instead of overwriting. > > > [...] > > > > > > This is not needed, as there is no QUEUE_FLAG_BYPASS in 3.0.y. > > > > Odd, it doesn't break the build. And for some reason I just missed > > this, does it hurt to have it applied? > > I don't think it will break anything as it simply changes assignment > to |= to avoid overwriting existing flags. That said, any patch can > break anything, so if possible it would be better to drop for 3.0.y. Ok, I'll revert this and push out a 3.0.48 to be safe. thanks, greg k-h