[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <yq1a77nag7n.fsf@oracle.com>
Date: Thu, 19 Dec 2019 21:22:36 -0500
From: "Martin K. Petersen" <martin.petersen@...cle.com>
To: "Darrick J. Wong" <darrick.wong@...cle.com>
Cc: "Martin K. Petersen" <martin.petersen@...cle.com>,
Kirill Tkhai <ktkhai@...tuozzo.com>, axboe@...nel.dk,
linux-block@...r.kernel.org, linux-kernel@...r.kernel.org,
linux-ext4@...r.kernel.org, tytso@....edu,
adilger.kernel@...ger.ca, ming.lei@...hat.com, osandov@...com,
jthumshirn@...e.de, minwoo.im.dev@...il.com, damien.lemoal@....com,
andrea.parri@...rulasolutions.com, hare@...e.com, tj@...nel.org,
ajay.joshi@....com, sagi@...mberg.me, dsterba@...e.com,
chaitanya.kulkarni@....com, bvanassche@....org,
dhowells@...hat.com, asml.silence@...il.com
Subject: Re: [PATCH RFC 1/3] block: Add support for REQ_OP_ASSIGN_RANGE operation
Darrick,
[Anchor]
> What happens if you anchor and then try to read the results? IO
> error?
Depends on the device. Typically you get the same results as UNMAP.
Anchoring is essentially a way to preallocate blocks on a thinly
provisioned device. If you then run out of actual storage capacity you
won't get a write failure writing to those LBAs. It is intended to
protect block ranges that do not tolerate write failures (journals,
metadata).
--
Martin K. Petersen Oracle Linux Engineering
Powered by blists - more mailing lists