linux-kernel - Re: [PATCH v2 1/2] mm: uninitialized struct page poisoning sanity checking

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20180406124535.k3qyxjfrlo55d5if@xakep.localdomain>
Date:   Fri, 6 Apr 2018 08:45:35 -0400
From:   Pavel Tatashin <pasha.tatashin@...cle.com>
To:     Sasha Levin <Alexander.Levin@...rosoft.com>
Cc:     "steven.sistare@...cle.com" <steven.sistare@...cle.com>,
        "daniel.m.jordan@...cle.com" <daniel.m.jordan@...cle.com>,
        "akpm@...ux-foundation.org" <akpm@...ux-foundation.org>,
        "mgorman@...hsingularity.net" <mgorman@...hsingularity.net>,
        "mhocko@...e.com" <mhocko@...e.com>,
        "linux-mm@...ck.org" <linux-mm@...ck.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "gregkh@...uxfoundation.org" <gregkh@...uxfoundation.org>,
        "vbabka@...e.cz" <vbabka@...e.cz>,
        "bharata@...ux.vnet.ibm.com" <bharata@...ux.vnet.ibm.com>
Subject: Re: [PATCH v2 1/2] mm: uninitialized struct page poisoning sanity
 checking

On 18-04-05 19:22:58, Sasha Levin wrote:
> On Thu, Apr 05, 2018 at 09:49:40AM -0400, Pavel Tatashin wrote:
> >> Hi Sasha,
> >>
> >> I have registered on Azure's portal, and created a VM with 4 CPUs and 16G
> >> of RAM. However, I still was not able to reproduce the boot bug you found.
> >
> >I have also tried to reproduce this issue on Windows 10 + Hyper-V, still
> >unsuccessful.
> 
> I'm not sure why you can't reproduce it. I built a 4.16 kernel + your 6
> patches on top, and booting on a D64s_v3 instance gives me this:

Hi Sasha,

Thank you for running it again, the new trace is cleaner, as we do not get
nested panics within dump_page.

Perhaps a NUMA is required to reproduce this issue. I have tried,
unsuccessfully, on D4S_V3. This is the largest VM allowed with free trial,
as 4 CPU is the limit. D64S_V3 is with 64 CPUs and over $2K a month! :)

Let me study your trace, perhaps I will able to figure out the issue
without reproducing it.

Thank you,
Pasha