linux-kernel - Re: [PATCH v7 3/3] virt: vmgenid: add support for devicetree bindings

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: <e2de24bf-d154-4b11-b322-8bad7de8a0a6@linaro.org>
Date: Fri, 19 Apr 2024 16:14:35 +0200
From: Krzysztof Kozlowski <krzysztof.kozlowski@...aro.org>
To: "Jason A. Donenfeld" <Jason@...c4.com>, robh@...nel.org
Cc: tytso@....edu, krzk@...nel.org, "Landge, Sudan" <sudanl@...zon.co.uk>,
 conor+dt@...nel.org, devicetree@...r.kernel.org,
 linux-kernel@...r.kernel.org, sudanl@...zon.com, graf@...zon.com,
 dwmw@...zon.co.uk, xmarcalx@...zon.co.uk
Subject: Re: [PATCH v7 3/3] virt: vmgenid: add support for devicetree bindings

On 19/04/2024 16:02, Jason A. Donenfeld wrote:
> Hi Rob,
> 
> On Fri, Apr 19, 2024 at 3:30 PM Landge, Sudan <sudanl@...zon.co.uk> wrote:
>>>   config VMGENID
>>>          tristate "Virtual Machine Generation ID driver"
>>>          default y
>>> -       depends on ACPI
>>> +       depends on ACPI || OF
>>
>> Version 2 of the patches had these flag but were removed in version3
>> following the below review comment from Rob:
>> "One of those is pretty much always enabled, so it can probably be dropped."
>> reference :
>> https://lore.kernel.org/lkml/CAL_JsqJoB5CYajWuntMdQrJZir+ZA-69Q0cwvxcVZAqs-mXC+Q@mail.gmail.com/
> 
> Are you sure about that? I see other drivers that depend on `ACPI ||
> OF`, and this driver is totally useless without either disabled
> (right?), so the dependency makes sense. I think? Not an expert in
> this area, so I'm happy to defer to what you know is best, but I did
> want to lightly pushback a bit.
> 

There are just 18 cases, so I would argue that it's not a real pattern
but could be just growing organically, without really thinking.

The driver could still be probed on some !ACPI and !OF system via driver
name matching.

Keeping it as is seems harmless, but OTOH, if this stays, you miss ||
COMPILE_TEST

Best regards,
Krzysztof