linux-hardening - Re: [PATCH] pstore: migrate to crypto acomp interface (take 2)

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: <CAMj1kXGxNP+hnN5yo9Pb_kQC+tcDn+9d30h5QNAUm3JYgoGiwQ@mail.gmail.com>
Date:   Mon, 17 Oct 2022 21:11:02 +0200
From:   Ard Biesheuvel <ardb@...nel.org>
To:     "Guilherme G. Piccoli" <gpiccoli@...lia.com>
Cc:     Kees Cook <keescook@...omium.org>,
        Anton Vorontsov <anton@...msg.org>,
        Colin Cross <ccross@...roid.com>,
        Tony Luck <tony.luck@...el.com>, linux-kernel@...r.kernel.org,
        linux-hardening@...r.kernel.org
Subject: Re: [PATCH] pstore: migrate to crypto acomp interface (take 2)

On Mon, 17 Oct 2022 at 20:23, Guilherme G. Piccoli <gpiccoli@...lia.com> wrote:
>
> On 17/10/2022 15:14, Ard Biesheuvel wrote:
> > [...]
> >
> > So at that point, I wondered what the point is of all this complexity.
> > Do we really need 6 different algorithms to compress a couple of K of
> > ASCII text on a code path that is ice cold by definition? Wouldn't it
> > be better to drop the crypto API altogether here, and just use GZIP
> > via the library interface?
>
> Skipping all the interesting and more complex parts, I'd just want to
> consider zstd maybe?

I just made the point that it doesn't matter. So on the one hand, I
don't have any objections to ZSTD per se. But I do wonder if it is the
best choice when it comes to code size etc. Perhaps one of the
compression algorithms is guaranteed to be compiled in anyway?

> Quite fast and efficient - it's what we're using by
> default on Steam Deck.
>
> I'm not sure what is the gzip library interface - you mean skipping the
> scomp/legacy comp interface, and make use directly of gzip?
>

zlib_deflate() and friends.