Re: [PATCH 2/6] irqchip/gic-v3-its: Fix memleak in its_probe_one()

[Kemeng Shi]

在 2026/6/17 20:07:54, Marc Zyngier 写道:
> On Tue, 16 Jun 2026 02:39:10 +0100,
> Kemeng Shi <shikemeng@xxxxxxxxxxxxxxx> wrote:
>>
>> 在 2026/6/15 16:59:14, Marc Zyngier 写道:
>>> On Mon, 15 Jun 2026 04:29:06 +0100,
>>> Kemeng Shi <shikemeng@xxxxxxxxxxxxxxx> wrote:
>>>>
>>>> Fix collection leak when its_init_domain() failed in its_probe_one().
>>>>
>>>> Signed-off-by: Kemeng Shi <shikemeng@xxxxxxxxxxxxxxx>
>>>> ---
>>>>  drivers/irqchip/irq-gic-v3-its.c | 10 +++++++++-
>>>>  1 file changed, 9 insertions(+), 1 deletion(-)
>>>>
>>>> diff --git a/drivers/irqchip/irq-gic-v3-its.c b/drivers/irqchip/irq-gic-v3-its.c
>>>> index 2b7b546c43c8..df26ddc97ae2 100644
>>>> --- a/drivers/irqchip/irq-gic-v3-its.c
>>>> +++ b/drivers/irqchip/irq-gic-v3-its.c
>>>> @@ -3032,6 +3032,12 @@ static int its_alloc_collections(struct its_node *its)
>>>>  	return 0;
>>>>  }
>>>>  
>>>> +static void its_free_collections(struct its_node *its)
>>>> +{
>>>> +	kfree(its->collections);
>>>> +	its->collections = NULL;
>>>> +}
>>>
>>> Why do we need an extra helper for something that has a single calling
>>> spot? Why is it important to set collections to NULL, given that we're
>>> about to free the structure without even looking further?
>>>
>> The extra helper is added for symmetry with its_alloc_collections(), keeping
>> allocation/deallocation logic paired.
> 
> But there is zero logic here. None. it is a call to kfree(), just
> that, and we don't need any additional abstraction.
> 
>> In this way, we need to only modified
>> allocation/deallocation function if more member is added for collections.
> 
> Well, when we get there, we can always add an abstraction.
> 
>> Setting collections to NULL is a personal habit to quickly catch use-after-free
>> of collections. Could drop this which is not that import.
> 
> I don't see the point of doing that.
> 
Thanks for feedback. I will drop its_free_collections() in release code in next
version.

Thanks,
Kemeng> 	M.
>