linux-kernel - Re: [PATCH 5/6] platform/chrome: cros_ec_lpc: Correct ACPI name for Framework Laptop

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: <87jzjk1ibr.fsf@jubnut.com>
Date: Thu, 23 May 2024 19:42:00 +0100
From: Ben Walsh <ben@...nut.com>
To: Tzung-Bi Shih <tzungbi@...nel.org>
Cc: Benson Leung <bleung@...omium.org>, Guenter Roeck <groeck@...omium.org>,
 "Dustin L. Howett" <dustin@...ett.net>, Kieran Levin <ktl@...me.work>,
 Thomas Weißschuh <linux@...ssschuh.net>, Mario Limonciello
 <mario.limonciello@....com>, chrome-platform@...ts.linux.dev,
 linux-kernel@...r.kernel.org
Subject: Re: [PATCH 5/6] platform/chrome: cros_ec_lpc: Correct ACPI name for
 Framework Laptop

Tzung-Bi Shih <tzungbi@...nel.org> writes:

> On Wed, May 15, 2024 at 06:56:30AM +0100, Ben Walsh wrote:
>> Framework Laptops' ACPI exposes the EC as name "PNP0C09". Use this to
>> find the device. This makes it easy to find the AML mutex via the
>> ACPI_COMPANION device.
>>
>> The name "PNP0C09" is part of the ACPI standard, not Chrome-specific,
>> so only recognise the device if the DMI data is recognised too.
>
> I don't quite understand the statement.  Why it needs DMI data?

There are lots of computers with EC chips with ACPI name "PNP0C09"
because it's part of the ACPI standard (for example I have an Intel NUC
with one of these). Most of them don't support the cros_ec protocol, so
the cros_ec driver should ignore these chips. The Framework EC is
unusual in that it's called "PNP0C09" and supports the cros_ec protocol.

Before these patches, the cros_ec code just ignored PNP0C09 because it
wasn't in the match table. The cros_ec_lpc_init logic looked like:

  * dmi_match => ok
  * acpi_name == "GOOG0004" => ok
  * otherwise fail.

After the patch, cros_ec_lpc_init still has this behaviour. We have
"PNP0C09" in the match table so the driver gets hooked up correctly
with the right "ACPI_COMPANION" device, but we don't allow the match
to proceed unless we have the DMI data indicating it's a Framework EC.

>> -	} else if (!cros_ec_lpc_acpi_device_found) {
>> +	} else if (cros_ec_lpc_acpi_device_found <= 0) {
>> +		/* Standard EC "PNP0C09" not supported without DMI data */
>
> Also the way is a bit confusing as "PNP0C09" must be at index 0 in the
> acpi_device_id.

I need some way of saying "will we match PNP0C09?" The table index seems
a simple way of doing it. I could use a strcmp on the table match
instead?

Regarding your other emails, I agree with all your suggestions. Thanks
for reviewing!