linux-kernel - Re: [PATCH] ACPI: resource: Skip IRQ override on Asus Vivobook Go E1404GAB

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] [day] [month] [year] [list]

Message-ID: <7ce7f7cc-870f-4f7f-98c6-95eb784008ff@leemhuis.info>
Date: Thu, 26 Sep 2024 12:18:51 +0200
From: Thorsten Leemhuis <linux@...mhuis.info>
To: Hans de Goede <hdegoede@...hat.com>, Tamim Khan <tamim@...etak.com>,
 linux-acpi@...r.kernel.org
Cc: rafael@...nel.org, lenb@...nel.org, LKML <linux-kernel@...r.kernel.org>,
 Linux kernel regressions list <regressions@...ts.linux.dev>
Subject: Re: [PATCH] ACPI: resource: Skip IRQ override on Asus Vivobook Go
 E1404GAB

On 25.09.24 16:31, Hans de Goede wrote:
> 
> On 25-Sep-24 1:56 PM, Thorsten Leemhuis wrote:
>> [CCing Hans]
> Can you next time maybe also add a bit of text explaining why ?

Sorry, yes, you are right, will keep that in mind.

>> On 05.09.24 12:45, Thorsten Leemhuis wrote:
>>> On 05.09.24 11:51, Thorsten Leemhuis wrote:
>>>> On 03.09.24 03:43, Tamim Khan wrote:
>>>>> Like other Asus Vivobooks, the Asus Vivobook Go E1404GAB has a DSDT
>>>>> that describes IRQ 1 as ActiveLow, while the kernel overrides to Edge_High.
>>>>> This override prevents the internal keyboard from working. This patch fixes
>>>>> this problem by adding this laptop to the table that prevents the kernel from
>>>>> overriding the IRQ.
>>>>>
>>>>> Link: https://bugzilla.kernel.org/show_bug.cgi?id=219212
>>>> Thx for that. FWIW, I by chance noticed another such report about the
>>>> E1404GA: https://bugzilla.kernel.org/show_bug.cgi?id=219224
>>>
>>> TWIMC, shortly after sending this mail I noticed there is another request
>>> for a quirk that was send to the list, bug afaics fall through the
>>> cracks. See here:
>>> https://lore.kernel.org/all/1226760b-4699-4529-bf57-6423938157a3@wanadoo.fr/
>>>
>>> It afaics add a X1704VAP:
> [...]
> 
> Ok, I wonder did you Cc me so that I can write / submit patches upstream
> for these ones?

Not really. Of course it would be nice if you or someone else took care
of that one and...

> It seems that there are 3 missing models:
> - E1404GA: https://bugzilla.kernel.org/show_bug.cgi?id=219224
> - X1704VAP https://lore.kernel.org/all/1226760b-4699-4529-bf57-6423938157a3@wanadoo.fr/
> - B2502CV: https://bugzilla.kernel.org/show_bug.cgi?id=217760#c12
> 
> Which someone needs to submit upstream, right ?

...these as well -- and ideally would even be willing to act as go-to
person from now on in case more of these quirk entries are needed, which
I guess will be the case. But given the backstory (see below) I don't
think you or anyone else is obliged to do this, even if the current
situation is parlty caused by regressions and recent fixes for them.

>> """
>>> adding this section to file drivers/acpi/resource.c helped; if someone could do an official patch it would be great :-|
>>>
>>>         {
>>>                 /* Asus ExpertBook B2502CVA */
>>>                 .matches = {
>>>                         DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
>>>                         DMI_MATCH(DMI_BOARD_NAME, "B2502CVA"),
>>>                 },
>> """
>>
>> This is not a regression, but given how frequently I notice these issues
>> please as a bystander allow me to ask: is there maybe some way to
>> improve the situation so that we do not have to add all these and likely
>> many other Asus Laptops to that file?
> 
> This has been discussed before and unfortunately there is no other way
> to deal with this. There has been some discussion about reading back
> the polarity and edge/level from the hardware for the keyboard IRQ
> to get the values which the BIOS has setup. But it is uncertain if this
> will work and no one is working on this.

A refresher like this was more what I was fishing for. Thx for this.

Now that you wrote it, it makes me think: given the amount of quirk
entries we apparently need I wonder if it might be wise to revisit this
at some point in the not to distant future, as I suspect otherwise
sooner or later Linus might yell at all of us with something along the
lines of "the kernel is clearly broken, why did we work around this with
lots of quirk entries instead of fixing this for real". Something like
that happened in a somewhat (but not exactly) similar situation a year
ago, and it iirc was not the first time.

Writing this lead to another thought: does anyone have contacts to Asus
and could just ask if there is some generic way to detect which of their
Laptops need a quirk?

Ciao, Thorsten