Message-ID: <5436DDEB.5090004@oracle.com>
Date: Thu, 09 Oct 2014 15:11:39 -0400
From: Sasha Levin <sasha.levin@oracle.com>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.0
MIME-Version: 1.0
To: Hugh Dickins <hughd@google.com>
CC: Andrew Morton <akpm@linux-foundation.org>, linux-kernel@vger.kernel.org,
        linux-mm@kvack.org, mgorman@suse.de
Subject: Re: [PATCH 0/5] mm: poison critical mm/ structs
References: <1412041639-23617-1-git-send-email-sasha.levin@oracle.com> <20141001140725.fd7f1d0cf933fbc2aa9fc1b1@linux-foundation.org> <542C749B.1040103@oracle.com> <alpine.LSU.2.11.1410020154500.6444@eggly.anvils>
In-Reply-To: <alpine.LSU.2.11.1410020154500.6444@eggly.anvils>
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: 8bit
Sender: linux-kernel-owner@vger.kernel.org

On 10/02/2014 05:23 AM, Hugh Dickins wrote:
> I'm glad to hear they've confirmed some vm_area_struct corruption:
> any ideas on where that's coming from?

Hugh,

I think that what we're seeing isn't a corruption of vm_area_struct
per-se, but something weirder.

I've poisoned every spot where vm_area_struct is allocated, and yet
there seems to be nothing that's hitting that field before we end
up using a "zeroed out" vm_area_struct.

The results are the same both with and without kasan, there seems
to be no corruption happening anywhere, but we somehow end up with
an empty vm_area_struct.

It also somewhat makes sense considering that we're seeing no slub
corruption either. Either something is zeroing out *exactly*
vm_area_struct, or it's not really corruption...


Thanks,
Sasha
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/