linux-kernel - Re: [PATCH] mm, numa: Fix bad pmd by atomically check for pmd_trans_huge when marking page tables prot

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 for Android: free password hash cracker in your pocket

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

Message-ID: <20170410173849.ecy2ysm7twzgcm53@techsingularity.net>
Date:   Mon, 10 Apr 2017 18:38:49 +0100
From:   Mel Gorman <mgorman@...hsingularity.net>
To:     Michal Hocko <mhocko@...nel.org>
Cc:     Andrew Morton <akpm@...ux-foundation.org>,
        Andrea Arcangeli <aarcange@...hat.com>,
        Rik van Riel <riel@...hat.com>,
        Vlastimil Babka <vbabka@...e.cz>, linux-mm@...ck.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] mm, numa: Fix bad pmd by atomically check for
 pmd_trans_huge when marking page tables prot_numa

On Mon, Apr 10, 2017 at 03:53:42PM +0200, Michal Hocko wrote:
> > While this could be fixed with heavy locking, it's only necessary to
> > make a copy of the PMD on the stack during change_pmd_range and avoid
> > races. A new helper is created for this as the check if quite subtle and the
> > existing similar helpful is not suitable. This passed 154 hours of testing
> 
> s@...pful@...per@ I suspect
> 

Yes. I'll wait to see if there is more feedback and if not, resend unless
Andrew decides to pick it up and correct the mistake directly.

> > (usually triggers between 20 minutes and 24 hours) without detecting bad
> > PMDs or corruption. A basic test of an autonuma-intensive workload showed
> > no significant change in behaviour.
> > 
> > Signed-off-by: Mel Gorman <mgorman@...hsingularity.net>
> > Cc: stable@...r.kernel.org
> 
> Acked-by: Michal Hocko <mhocko@...e.com>
> 

Thanks.

> you will probably win the_longest_function_name_contest but I do not
> have much better suggestion.
> 

I know, it's not a type of function that yields a snappy name.

-- 
Mel Gorman
SUSE Labs