From mboxrd@z Thu Jan 1 00:00:00 1970 From: NeilBrown Subject: Re: [PATCH 0/2] patches addresses problem with memory corruption Date: Tue, 1 Feb 2011 10:38:02 +1100 Message-ID: <20110201103802.6513b66a@notabene.brown> References: <20110131151657.13213.16328.stgit@gklab-128-013.igk.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20110131151657.13213.16328.stgit@gklab-128-013.igk.intel.com> Sender: linux-raid-owner@vger.kernel.org To: Adam Kwolek Cc: linux-raid@vger.kernel.org, dan.j.williams@intel.com, ed.ciechanowski@intel.com, wojciech.neubauer@intel.com List-Id: linux-raid.ids On Mon, 31 Jan 2011 16:25:08 +0100 Adam Kwolek wrote: > When for raid0 metadata is updated outside mdmon it can happen that > memory corruption occurs due to too amount of memory is allocated > for imsm device structure. > > 1. imsm: FIX: mdmon crash during 2 raid0 arrays expansion > this patch is almost the same to sent earlier except 2 changes: > a) compilation problem > b) size should affect anchor size also > (added calculation changes extends space_needed variable also > this variable is used for anchor allocation) > so it replaces previous one. > 2. imsm: FIX: sizeof_imsm_dev() can return too small value > size returned by sizeof_imsm_dev() describes minimum size of device > This is correct when both device maps has the same length > When device expands or shrinks we should return size that allows > for storing larger device information to avoid memory corruption. > Both applied (replacing the previous one).. Thanks. NeilBrown > BR > Adam > > --- > > Adam Kwolek (2): > imsm: FIX: sizeof_imsm_dev() can return too small value > imsm: FIX: mdmon crash during 2 raid0 arrays expansion > > > super-intel.c | 20 +++++++++++++++----- > 1 files changed, 15 insertions(+), 5 deletions(-) >