linux-kernel - Re: [PATCH] mm: clean up hwpoison page cache page in fault path

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] [day] [month] [year] [list]

Message-ID: <8aafa00865f564d58dfa39a1e2816a8ec0eab097.camel@surriel.com>
Date:   Sun, 13 Feb 2022 14:21:26 -0500
From:   Rik van Riel <riel@...riel.com>
To:     John Hubbard <jhubbard@...dia.com>
Cc:     linux-kernel@...r.kernel.org, kernel-team@...com,
        linux-mm@...ck.org, Andrew Morton <akpm@...ux-foundation.org>,
        Mel Gorman <mgorman@...e.de>,
        Johannes Weiner <hannes@...xchg.org>,
        Matthew Wilcox <willy@...radead.org>
Subject: Re: [PATCH] mm: clean up hwpoison page cache page in fault path

On Sun, 2022-02-13 at 00:56 -0800, John Hubbard wrote:
> On Fri, 11 Feb 2022, Rik van Riel wrote:
> 
> >     
> > This is particularly embarrassing when the page was offlined due to
> > having too many corrected memory errors. Now we are killing tasks
> > due to them trying to access memory that probably isn't even
> > corrupted.
> 
> I'd recommend deleting that paragraph entirely. It's a separate
> question, and it is not necessarily an accurate assessment of that
> question either: the engineers who set the thresholds for "too many
> corrected errors" may not--in fact, probably *will not*--agree with
> your
> feeling that the memory is still working and reliable!

Fair enough. We try to offline pages before we get to
a point where the error correction might no longer be
able to correct the error correctly, but I am pretty
sure I have seen a few odd kernel crashes following a
stream of corrected errors that strongly suggested
corruption had in fact happened.

I'll take that paragraph out if anybody else asks
for further changes for v3 of the patch.

-- 
All Rights Reversed.

Download attachment "signature.asc" of type "application/pgp-signature" (489 bytes)