From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-it0-f50.google.com ([209.85.214.50]:32781 "EHLO mail-it0-f50.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932395AbdHWSEs (ORCPT ); Wed, 23 Aug 2017 14:04:48 -0400 Received: by mail-it0-f50.google.com with SMTP id y62so969775itc.0 for ; Wed, 23 Aug 2017 11:04:48 -0700 (PDT) Subject: Re: don't require a struct block_device to submit a bio To: Christoph Hellwig Cc: linux-block@vger.kernel.org, linux-raid@vger.kernel.org, linux-btrfs@vger.kernel.org References: <20170823171032.30529-1-hch@lst.de> From: Jens Axboe Message-ID: Date: Wed, 23 Aug 2017 12:04:45 -0600 MIME-Version: 1.0 In-Reply-To: <20170823171032.30529-1-hch@lst.de> Content-Type: text/plain; charset=utf-8 Sender: linux-btrfs-owner@vger.kernel.org List-ID: On 08/23/2017 11:10 AM, Christoph Hellwig wrote: > Hi Jens, > > this series removes the need to have a struct block_device ready to > submit a bio. We basically don't use it anywhere in the block stack > anyway - we always go for the gendisk or request_queue. The only > exception is partition remapping for which we'll now need an additional > partition index. This helps with cases where we submit I/O from a > character device (nvme or lightnvm passthrough) or a different block > device (upcoming nvme multipath support). Looks fine to me. I can take it through the block tree, when the md/btrfs parts are reviewed by the right parties. -- Jens Axboe