[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <2c38b7cd0aad46ec9f8bf03715109f10@AcuMS.aculab.com>
Date: Fri, 3 Jul 2020 14:13:18 +0000
From: David Laight <David.Laight@...LAB.COM>
To: 'Pavel Machek' <pavel@....cz>, Ming Lei <tom.leiming@...il.com>
CC: Henrique de Moraes Holschuh <hmh@....eng.br>,
Damien Le Moal <Damien.LeMoal@....com>,
Simon Arlott <simon@...iron.net>,
"James E.J. Bottomley" <jejb@...ux.ibm.com>,
"Martin K. Petersen" <martin.petersen@...cle.com>,
Jonathan Corbet <corbet@....net>,
"Linux Kernel Mailing List" <linux-kernel@...r.kernel.org>,
"linux-scsi@...r.kernel.org" <linux-scsi@...r.kernel.org>,
"linux-doc@...r.kernel.org" <linux-doc@...r.kernel.org>
Subject: RE: [PATCH] scsi: sd: stop SSD (non-rotational) disks before reboot
From: Pavel Machek
> Sent: 02 July 2020 22:17
> > > during a FLASH write or erase can cause from weakened cells, to much
> > > larger damage. It is possible to harden the chip or the design against
> > > this, but it is *expensive*. And even if warded off by hardening and no
> > > FLASH damage happens, an erase/program cycle must be done on the whole
> > > erase block to clean up the incomplete program cycle.
> >
> > It should have been SSD's(including FW) responsibility to avoid data loss when
> > the SSD is doing its own BG writing, because power cut can happen any time
> > from SSD's viewpoint.
>
> It should be their responsibility. But we know how well that works
> (not well), so we try hard (and should try hard) to power SSDs down
> cleanly.
I hope modern SSD disks are better than very old CF drives.
I had one where the entire contents got scrambled after an unexpected
power removal.
I suspect it was in the middle of a 'wear levelling' activity.
Even though it was only a FAT filesystem I was glad I didn't
actually need to recover any of the data.
David
-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)
Powered by blists - more mailing lists