From mboxrd@z Thu Jan 1 00:00:00 1970 From: Vladislav Bolkhovitin Subject: Re: Is there any plan to support 64bit lun in mainline? Date: Tue, 15 Oct 2013 21:11:11 -0700 Message-ID: <525E11DF.3040004@vlnb.net> References: <525BC836.2030306@suse.de> <1381775449.3752.9.camel@dabdike.lan> <525CDA2A.2060308@suse.de> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: Received: from moutng.kundenserver.de ([212.227.17.8]:62411 "EHLO moutng.kundenserver.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751481Ab3JPELv (ORCPT ); Wed, 16 Oct 2013 00:11:51 -0400 In-Reply-To: <525CDA2A.2060308@suse.de> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: Hannes Reinecke Cc: James Bottomley , "linux-scsi@vger.kernel.org" , Christoph Hellwig , James Smart , Andrew Vasquez Hannes Reinecke, on 10/14/2013 11:01 PM wrote: > And HBAs like lpfc or qla2xxx even have a fast command abort built > into the firmware, where the firmware will not even wait for a > command abort to hit the wire but rather just disable the exchange > internally and return. Doing so is asking for data corruption. Aborts intended to cleanup commands on the target, so they could not anyhow badly interact with future commands. Otherwise it is possible that an old WRITE command stuck in some deep corner inside the target, be bypassed by such pseudo-abort and then got released AFTER its LBAs written by newer data, hence overwrite new data by old data. Vlad