From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from silver.sucs.swan.ac.uk ([137.44.10.1]:57465 "EHLO silver.sucs.swan.ac.uk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751921AbcFOV0T (ORCPT ); Wed, 15 Jun 2016 17:26:19 -0400 Date: Wed, 15 Jun 2016 22:26:17 +0100 From: Sitsofe Wheeler To: Shaohua Li Cc: linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, axboe@fb.com, snitzer@redhat.com, martin.petersen@oracle.com, hch@infradead.org, Kernel-team@fb.com, Dan Carpenter Subject: Re: [PATCH V2] block: correctly fallback for zeroout Message-ID: <20160615212617.GB5410@sucs.org> References: <20160606223357.GA52883@shli-mbp.local> <20160607045049.GA10921@sucs.org> <20160607145824.GA84027@shli-mbp.local> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <20160607145824.GA84027@shli-mbp.local> Sender: linux-block-owner@vger.kernel.org List-Id: linux-block@vger.kernel.org On Tue, Jun 07, 2016 at 07:58:25AM -0700, Shaohua Li wrote: > > I didn't follow. io_err is only and always set when ret == 0. io_err is > meanless if ret != 0, because that means the disk doesn't support discard and > we don't dispatch discard IO. why should we initialized io_err to 0? My mistake - I confused what !ret would mean. Unfortunately the V2 patch no longer cleanly applies to the latest kernel (db06d759d6cf903aeda8c107fd3abd366dd80200 ) so I can't easily test it there. -- Sitsofe | http://sucs.org/~sits/