linux-kernel - Re: [PATCH] tty: max310x: work around regmap->regcache data corruption

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: <20231204150224.add8b07a59bf737edb0b5c1c@hugovil.com>
Date:   Mon, 4 Dec 2023 15:02:24 -0500
From:   Hugo Villeneuve <hugo@...ovil.com>
To:     Mark Brown <broonie@...nel.org>
Cc:     Jan Kundrát <jan.kundrat@...net.cz>,
        Cosmin Tanislav <cosmin.tanislav@...log.com>,
        linux-serial@...r.kernel.org,
        Andy Shevchenko <andy.shevchenko@...il.com>,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] tty: max310x: work around regmap->regcache data
 corruption

On Mon, 4 Dec 2023 19:48:05 +0000
Mark Brown <broonie@...nel.org> wrote:

> On Mon, Dec 04, 2023 at 02:41:36PM -0500, Hugo Villeneuve wrote:
> 
> > But that is not was my question was about. Here a pseudo code
> > example to select "page" 1:
> 
> > 1. save original value of LCR register.
> > 2. write 0xBF to LCR register
> > 3. access desired register in page 1
> > 4. restore original LCR value saved in step 1
> 
> > How do you do that with regmap range?
> 
> Are you saying that the selector has other, non-selector functions?

Yes! There is no bit or bit range in that register that is used to
select a praticular set of registers or "page". It is only when
writing the special magic value of $BF that the IC switches to "page"
1. And if the value is NOT $BF, then it switches back to "page" 0.

When I told you if you could point to other IC/drivers that had the
same configuration, I tough you were aware of this. That explains some
of the confusion.

> This is truly innovative hardware,...

Well, I would not say innovative, but more "crappy" hardware design :)

> ...generally the selector is just a 
> bitfield that you write paging values to.

This is also what I am accustomed to normally.

>  You'd need to extend the core
> so that it knows about this quirk, right now that's not possible and
> we'll just leave the window pointing at whatever was last accessed.

Ok. I am not sure that adding support for it would make sense, since I
do not know of other ICs that could reuse this very specific and
particular method for switching "paged" registers.

Thank you,
Hugo Villeneuve