linux-kernel - Re: [PATCH] virt: vmgenid: drop redundant .owner

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]

Date: Tue, 26 Mar 2024 17:10:58 +0100
From: Krzysztof Kozlowski <krzysztof.kozlowski@...aro.org>
To: "Jason A. Donenfeld" <Jason@...c4.com>
Cc: Theodore Ts'o <tytso@....edu>, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] virt: vmgenid: drop redundant .owner

On 26/03/2024 15:17, Jason A. Donenfeld wrote:
> On Tue, Mar 26, 2024 at 09:07:52AM +0100, Krzysztof Kozlowski wrote:
>> On 26/03/2024 08:53, Krzysztof Kozlowski wrote:
>>> Core already sets .owner in acpi_bus_register_driver().
>>>
>>> Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@...aro.org>
>>> ---
>>>  drivers/virt/vmgenid.c | 1 -
>>>  1 file changed, 1 deletion(-)
>>>
>>> diff --git a/drivers/virt/vmgenid.c b/drivers/virt/vmgenid.c
>>> index b67a28da4702..8f6880c3a87f 100644
>>> --- a/drivers/virt/vmgenid.c
>>> +++ b/drivers/virt/vmgenid.c
>>> @@ -88,7 +88,6 @@ static const struct acpi_device_id vmgenid_ids[] = {
>>>  static struct acpi_driver vmgenid_driver = {
>>>  	.name = "vmgenid",
>>>  	.ids = vmgenid_ids,
>>> -	.owner = THIS_MODULE,
>>
>> This does not make sense and is not correct.  I need to fix
>> acpi_bus_register_driver first(). Please ignore.
> 
> Oh, okay. Does that mean this will resurface as part of a series that is
> core fix + individual driver cleanups subsequently?

Yes.

> 
> You made some comment on the other thread indicating that this
> meant the driver is "clearly abandoned" because of "all this owner
> crap". Wondering if this driver here hits some weird edge case that
> isn't covered by changes made "many years ago" or if something else is
> up.

No, it was just my mistake. Driver is fine, sorry for too fast judging.
It turns out only some xxx_driver set .owner field. Which I think leads
to actual problems, but that's a bit bigger patchset I need to prepare.

Best regards,
Krzysztof