Hi,
On 1/30/26 1:38 PM, Sascha Hauer wrote:
> On Tue, Jan 27, 2026 at 09:39:19AM +0100, Ahmad Fatoum wrote:
>> resize_region allows enlarging or reducing a region end while checking
>> for conflicts. This is less effort than removing a region and
>> reallocating it.
>>
>> Signed-off-by: Ahmad Fatoum <[email protected]>
>> ---
>> common/resource.c | 45 ++++++++++++++++++++++++++++++++++++++++++
>> include/linux/ioport.h | 2 ++
>> 2 files changed, 47 insertions(+)
>>
>> diff --git a/common/resource.c b/common/resource.c
>> index e391d268e0bb..39867da63b69 100644
>> --- a/common/resource.c
>> +++ b/common/resource.c
>> @@ -92,6 +92,51 @@ struct resource *__request_region(struct resource *parent,
>> return new;
>> }
>>
>> +int resize_region(struct resource *res, resource_size_t size)
>> +{
>> + struct resource *parent;
>> + struct resource *next;
>> + resource_size_t newend;
>> +
>> + if (!res)
>> + return 0;
>> + if (!size)
>> + return release_region(res);
>
> This frees the resource itself. I am not sure if this is useful. It's an
> invitation for the caller to reference freed memory. I think the caller
> must distinguish between resizing and releasing the resource anyway, so
> he could equally call resize_region() or release_region() directly.
You're right. I did actually have a use after free for 0 size :D
>> +
>> + if (size == resource_size(res))
>> + return 0;
>> +
>> + if (size < resource_size(res)) {
>> + res->end = res->start + size - 1;
>> + return 0;
>> + }
>
> A resource can have child resources. For completeness we should check if
> the child resources are still inside the now smaller parent resource.
Ok will do.
Cheers,
Ahmad
>
> Sascha
>
--
Pengutronix e.K. | |
Steuerwalder Str. 21 | http://www.pengutronix.de/ |
31137 Hildesheim, Germany | Phone: +49-5121-206917-0 |
Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
