[<prev] [next>] [day] [month] [year] [list]
Message-ID: <20141224163823.GA17035@amd>
Date: Wed, 24 Dec 2014 17:38:23 +0100
From: Pavel Machek <pavel@....cz>
To: kernel list <linux-kernel@...r.kernel.org>
Subject: DRAM unreliable under specific access patern
Hi!
It seems that it is easy to induce DRAM bit errors by doing repeated
reads from adjacent memory cells on common hw. Details are at
https://www.ece.cmu.edu/~safari/pubs/kim-isca14.pdf
. Older memory modules seem to work better, and ECC should detect
this. Paper has inner loop that should trigger this.
Workarounds seem to be at hardware level, and tricky, too.
Does anyone have implementation of detector? Any ideas how to work
around it in software?
Pavel
--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
Powered by blists - more mailing lists