linux-kernel - Re: [GIT PULL] pstore updates for v6.6-rc1

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: <202308300957.BCD878259@keescook>
Date:   Wed, 30 Aug 2023 10:00:43 -0700
From:   Kees Cook <keescook@...omium.org>
To:     Ard Biesheuvel <ardb@...nel.org>
Cc:     Eric Biggers <ebiggers@...nel.org>,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        Kees Cook <kees@...nel.org>, linux-kernel@...r.kernel.org,
        Enlin Mu <enlin.mu@...soc.com>,
        "Guilherme G. Piccoli" <gpiccoli@...lia.com>,
        "Matthew Wilcox (Oracle)" <willy@...radead.org>,
        Yunlong Xing <yunlong.xing@...soc.com>,
        Yuxiao Zhang <yuxiaozhang@...gle.com>
Subject: Re: [GIT PULL] pstore updates for v6.6-rc1

On Wed, Aug 30, 2023 at 09:48:48AM +0200, Ard Biesheuvel wrote:
> In any case, I'll rate limit the error so it doesn't clutter up the logs.

Great; thanks for looking at it!

A related issue I'm going to tackle is dealing with the risk of
ever-growing record counts for backends that don't treat their storage
as a circular buffer. (e.g. ramoops will overwrite the latest record
when it runs out of empty areas, but EFI will just keep on writing new
records.) It's clear we can't depend on userspace to do this clean-up.
I think pstore tossing the oldest records above a (configurable) limit
(say, 32) per dump type makes sense...

-- 
Kees Cook