From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754589Ab1JJTTT (ORCPT ); Mon, 10 Oct 2011 15:19:19 -0400 Received: from smtp.ctxuk.citrix.com ([62.200.22.115]:36910 "EHLO SMTP.EU.CITRIX.COM" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754566Ab1JJTTS (ORCPT ); Mon, 10 Oct 2011 15:19:18 -0400 X-IronPort-AV: E=Sophos;i="4.68,518,1312156800"; d="scan'208";a="8311790" Subject: Re: [Xen-devel] [PATCH 3/3] xen/blk[front|back]: Enhance discard support with secure erasing support. From: Ian Campbell To: Konrad Rzeszutek Wilk CC: "xen-devel@lists.xensource.com" , "linux-kernel@vger.kernel.org" , "hch@infradead.org" , Jan Beulich , Li Dongyang In-Reply-To: <20111010175323.GA31210@phenom.oracle.com> References: <1318260494-27985-1-git-send-email-konrad.wilk@oracle.com> <1318260494-27985-4-git-send-email-konrad.wilk@oracle.com> <1318263187.21903.464.camel@zakaz.uk.xensource.com> <20111010164250.GG28646@phenom.oracle.com> <20111010175323.GA31210@phenom.oracle.com> Content-Type: text/plain; charset="ISO-8859-1" Organization: Citrix Systems, Inc. Date: Mon, 10 Oct 2011 20:19:15 +0100 Message-ID: <1318274355.27397.12.camel@dagon.hellion.org.uk> MIME-Version: 1.0 X-Mailer: Evolution 2.32.3 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Where is your tree at the moment? On Mon, 2011-10-10 at 18:53 +0100, Konrad Rzeszutek Wilk wrote: > > > I think an explicit flag variable is likely to be less trouble WRT > > > maintaining compatibility in the future than a bit-field. Also I think > > > you may as well align the struct size to something larger than a byte, > > > either 4 or 8 bytes would make sense. > > > > Ok. Will change it and make it an uint64_t secure_flag > > variable. Later on if there are any "other" flags we can chop it down. > > New patch (it also looks like the patch I sent to xen-devel to update > the blkif.h was never merged) - so let me send right now. > > BTW, it seems that the #pragma pack(push, 4) is used in the > drivers/block/xen-blkback/common.h to compact the structures already so > I don't think we need the aligment. Only around the x86_32 ABI definition, working around the fact that the original 32 bit interface was not 64 bit clean :-(. I think to ensure that the new DISCARD structure is sensibly aligned you probably do want it to be a multiple of 64 bits in size (32 of flags and 32 of pad would do it). > diff --git a/include/xen/interface/io/blkif.h b/include/xen/interface/io/blkif.h > index 9324488..13d040e 100644 > --- a/include/xen/interface/io/blkif.h > +++ b/include/xen/interface/io/blkif.h > @@ -84,6 +84,10 @@ typedef uint64_t blkif_sector_t; > * e07154r6-Data_Set_Management_Proposal_for_ATA-ACS2.doc > * http://www.seagate.com/staticfiles/support/disc/manuals/ > * Interface%20manuals/100293068c.pdf > + * We also provide three extra XenBus options to the discard operation: Who is "We" here? The frontend, backend or both? Do they both need to agree on anything? > + * 'discard-granularity' - Max amount of sectors that can be discarded. ... in a single request? > + * 'discard-alignment' - 4K, 128K, etc aligment on sectors to erased. alignment What size are the sectors which "discard-granularity" is measured in? Is it "discard-alignment"-byte sectors or in base 512-byte sectors? > + * 'discard-secure' - whether the discard can also securely erase data. > */ > #define BLKIF_OP_DISCARD 5 > > @@ -107,6 +111,8 @@ struct blkif_request_rw { > struct blkif_request_discard { > blkif_sector_t sector_number; > uint64_t nr_sectors; > +#define BLKIF_OP_DISCARD_FLAG_SECURE (1<<1) /* ignored if discard-secure=0 */ "1<<0" is unused. > + uint32_t flag; > }; > > struct blkif_request {