From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Martin K. Petersen" Subject: Re: [PATCH 09/14] target/configfs: Expose protection device attributes Date: Mon, 13 Jan 2014 15:19:47 -0500 Message-ID: References: <1389212157-14540-1-git-send-email-nab@daterainc.com> <1389212157-14540-10-git-send-email-nab@daterainc.com> <1389637859.5567.431.camel@haakon3.risingtidesystems.com> <1389639177.12062.21.camel@dabdike.int.hansenpartnership.com> <1389641243.5567.445.camel@haakon3.risingtidesystems.com> Mime-Version: 1.0 Content-Type: text/plain Return-path: In-Reply-To: <1389641243.5567.445.camel@haakon3.risingtidesystems.com> (Nicholas A. Bellinger's message of "Mon, 13 Jan 2014 11:27:23 -0800") Sender: target-devel-owner@vger.kernel.org To: "Nicholas A. Bellinger" Cc: James Bottomley , "Martin K. Petersen" , "Nicholas A. Bellinger" , target-devel , linux-scsi , linux-kernel , Christoph Hellwig , Hannes Reinecke , Sagi Grimberg , Or Gerlitz List-Id: linux-scsi@vger.kernel.org >>>>> "nab" == Nicholas A Bellinger writes: >> What proposed 16 byte scheme? The only DIF proposals I know for >> SBC-4 are 13-185R0 and 12-369R0 and that's a couple of new algorithms >> and types because we cannot change the 8 byte PI. nab> Then I'm probably getting the SBC version wrong.. It's the one nab> that includes using CRC32C for the block guard, and larger space nab> for reference tag as mentioned by MKP. This is the Type 4 we have been shopping among various vendors. It predates and is simpler than HP's proposal (which met resistance in T10 and was subsequently dropped). So we revived our original Type 4 proposal which is 16 bytes of protection information per interval (CRC32C, 48-bit LBA and 6 bytes of app tag). The proposal has been sitting around for a while waiting for SBC-4 to open. -- Martin K. Petersen Oracle Linux Engineering