Re: [syzbot] [net?] possible deadlock in inet6_getname

[Fernando Fernandez Mancera <fmancera@suse.de>]

On 2/17/26 6:13 PM, Gerd Rausch wrote:
> Hi,
> 
> On 2026-02-17 09:03, Fernando Fernandez Mancera wrote:
>>> rds_tcp_conn_slots_available() can also be called from 
>>> rds_conn_shutdown(),
>>> where no "rds_tcp.ko" backend specific lock is held.
>>>
>>
>>
>> AFAICS, from rds_conn_shutdown() path rds_tcp_conn_slots_available() 
>> is called with fan-out argument as false. Therefore, no need to get 
>> the peer source port.
>>
>> I think that should be fine.
>>
> 
> True, but IMHO subtle and error-prone.
> 
> If someone were to change the code to pass in "fan_out == true"
> from a context not already holding a socket lock,
> would they remember to change rds_tcp_conn_slots_available()
> to acquire that lock?
>

Usually kernel requires the developer to understand when do they need to 
acquire a lock or not. Anyway, what would you suggest? To check whether 
we have acquired the lock or not and do it conditionally?

Thanks,
Fernando.

> Thanks,
> 
>    Gerd
> 
>