linux-kernel - Re: [PATCH v3 0/6] RTL8231 GPIO expander support

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]

Date:   Tue, 1 Jun 2021 12:51:43 +0200
From:   Linus Walleij <linus.walleij@...aro.org>
To:     Michael Walle <michael@...le.cc>
Cc:     Andy Shevchenko <andy.shevchenko@...il.com>,
        Hans de Goede <hdegoede@...hat.com>,
        Sander Vanheule <sander@...nheule.net>,
        Andrew Lunn <andrew@...n.ch>, Pavel Machek <pavel@....cz>,
        Rob Herring <robh+dt@...nel.org>,
        Lee Jones <lee.jones@...aro.org>,
        Mark Brown <broonie@...nel.org>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        "Rafael J . Wysocki" <rafael@...nel.org>,
        Bartosz Golaszewski <bgolaszewski@...libre.com>,
        Linux LED Subsystem <linux-leds@...r.kernel.org>,
        devicetree <devicetree@...r.kernel.org>,
        "open list:GPIO SUBSYSTEM" <linux-gpio@...r.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v3 0/6] RTL8231 GPIO expander support

On Tue, Jun 1, 2021 at 12:18 PM Michael Walle <michael@...le.cc> wrote:
> Am 2021-06-01 11:59, schrieb Linus Walleij:

> > Just regarding all registers/memory cells in a register page
> > as default volatile (which is what we do a lot of the time)
> > has its upsides: bugs like this doesn't happen.
>
> I don't think this is the bug here. If it is really a write-only
> register
> the problem is the read in RMW. Because reading the register will return
> the input value instead of the (previously written) output value.

True that. Write and read semantics differ on the register.

Volatile is used for this and some other things,
like for example interrupts being cleared when a register
is read so it is strictly read-once.

So the regmap config is really important to get right.

IIUC one of the ambitions around Rust is to encode this
in how memory is specified in the language. (I am still
thinking about whether that is really a good idea or not.)

Yours,
Linus Walleij