From mboxrd@z Thu Jan 1 00:00:00 1970 From: Douglas Gilbert Subject: Re: [PATCH, RFC] xfs: batched discard support Date: Thu, 20 Aug 2009 10:58:45 -0400 Message-ID: <4A8D64A5.9060003@interlog.com> References: <20090816004705.GA7347@infradead.org> <20090819203916.GA25296@elte.hu> <4A8CA956.2060406@rtr.ca> <4A8D5442.1000302@redhat.com> Reply-To: dgilbert@interlog.com Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Cc: Mark Lord , Ingo Molnar , Christoph Hellwig , Peter Zijlstra , Paul Mackerras , Linus Torvalds , xfs@oss.sgi.com, linux-fsdevel@vger.kernel.org, linux-scsi@vger.kernel.org, linux-kernel@vger.kernel.org, jens.axboe@oracle.com, IDE/ATA development list , Neil Brown To: Ric Wheeler Return-path: In-Reply-To: <4A8D5442.1000302@redhat.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-fsdevel.vger.kernel.org Ric Wheeler wrote: > On 08/19/2009 09:39 PM, Mark Lord wrote: >> [resending, after fixing the Cc: list; somebody trimmed it earlier] >> >> Jamie Lokier wrote: >> .. >>> I don't remember - does TRIM guarantee the blocks read zeros afterwards? >> .. >> >> No, it doesn't. >> >> A drive can optionally support "deterministic TRIM", whereby it will >> return >> consistent data for any given trimmed sector afterwards, but that >> doesn't mean zeros. >> >> -ml > > Note that returning consistent data is critical for devices that are > used in a RAID group since you will need each RAID block that is used to > compute the parity to continue to return the same data until you > overwrite it with new data :-) > > If we have a device that does not support this (or is misconfigured not > to do this), we should not use those devices in an MD group & do discard > against it... A closer reading of d2015r2-ATAATAPI_Command_Set_-_2_ACS-2.pdf section 7.10.3.2 (latest ACS-2 draft from www.t13.org) shows that there are 3 possible variants for data read from a logical block that has been trimmed (or "unmapped"): a) indeterminate b) determinate c) determinate, return all zeroes In the case of b) the same data is returned for each subsequent read. And that data must not be something that has previously be written to some other LBA! In the case of SCSI (sbc3r19.pdf) case b) is not supported (very sensibly IMO). Another difference I noticed between SCSI and ATA drafts is with the SECURITY ERASE UNIT command which is somewhat similar to the SCSI FORMAT UNIT command (which includes a security erase option). The ATA draft says that all blocks are determinate ("mapped" in the SCSI state model) after a SECURITY ERASE UNIT. The SCSI draft says that all logical blocks may be unmapped after FORMAT UNIT. Doug Gilbert