linux-kernel - Re: [PATCH v2] i2c: Allow ACPI_I2C_OPREGION if I2C is built as a module

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:   Wed, 24 Jan 2018 16:23:50 +0100
From:   Hans de Goede <hdegoede@...hat.com>
To:     Sinan Kaya <okaya@...eaurora.org>,
        Andy Shevchenko <andy.shevchenko@...il.com>
Cc:     Mika Westerberg <mika.westerberg@...ux.intel.com>,
        Wolfram Sang <wsa@...-dreams.de>,
        Eric Auger <eric.auger@...hat.com>, eric.auger.pro@...il.com,
        linux-i2c <linux-i2c@...r.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        "linux-acpi@...r.kernel.org" <linux-acpi@...r.kernel.org>
Subject: Re: [PATCH v2] i2c: Allow ACPI_I2C_OPREGION if I2C is built as a
 module

Hi,

On 24-01-18 16:12, Sinan Kaya wrote:
> On 1/24/2018 9:59 AM, Hans de Goede wrote:
>> TL;DR: I have to NAK this, I'm sorry but with the current state of ACPI we
>> must simply have some stuff builtin to help with probe-ordering issues. Now
>> if the ACPI code where ever to honor the _DEP method everywhere instead of
>> only for battery devices this might change, but even then things will still
>> be tricky.
> 
> Well, the alternative is even worse.

No it is not, if I2C gets builtin things work fine, you proposal does not
fix anything, it merely gives the illusion of being builtin

> Redhat and most other distros configure I2C as a module.

Fedora has stopped building I2C as a module a long time ago already and where
Fedora goes RHEL typically follows.

> With this setup,
> I2C OpRegion support does not get compiled. It doesn't even work let alone to have
> race conditions.

So at least it is consistent then, which makes for a lot easier debugging.

> I2C OpRegion feature is practically dead for most general users unless you recompile
> your own kernel.

Then talk to various distro kernel maintainers and ask them to
set CONFIG_I2C=y.

> There must be a middle ground somewhere.

One thing which comes to mind is to simply not allow building i2c as a module
when ACPI is selected, something like this should work I think:

  config I2C
          tristate "I2C support"
          select RT_MUTEXES
          select IRQ_DOMAIN
+        # force building I2C in on ACPI systems, for opregion availability
+        depends on y || !ACPI

> I had some conversation with Rafael about _DEP support. He is not a big fan :)

Ok.

Regards,

Hans