[PATCH 2/3] wifi: ath11k: fix leaking peer->tfm_mmic in reset scenario

Baochen Qiang quic_bqiang at quicinc.com
Wed Jan 15 00:09:40 PST 2025



On 1/11/2025 3:15 AM, Jeff Johnson wrote:
> On 8/27/2024 5:33 PM, Karthikeyan Periyasamy wrote:
>> On 8/26/2024 7:19 AM, Baochen Qiang wrote:
> ...
>>> @@ -883,7 +883,7 @@ void ath11k_mac_peer_cleanup_all(struct ath11k *ar)
>>>   	mutex_lock(&ab->tbl_mtx_lock);
>>>   	spin_lock_bh(&ab->base_lock);
>>>   	list_for_each_entry_safe(peer, tmp, &ab->peers, list) {
>>> -		ath11k_peer_rx_tid_cleanup(ar, peer);
>>> +		ath11k_dp_peer_cleanup(ar, peer->vdev_id, peer->sta->addr);
>>
>> peer->tfm_mmic is allocated in ath11k_peer_rx_frag_setup() but its not 
>> cleanup in ath11k_dp_rx_frags_cleanup(), which is not symmetric now. 
>> Instead its freed in ath11k_dp_peer_cleanup(). can you refactor 
>> allocation/deallocation symmetric funcs ?
> 
> There was no action on this review comment.
> 
> Do you plan on submitting a v2?

sorry for not manage to respond timely.

Regarding this comment, my impression is that peer->tfm_mmic is not the only filed being
non-symmetric, and it is not easy to refactor it to achieve symmetry. So my plan is to
leave the patch as is, and a separate patch could be cooked in the future if we want to do
refactor.

> 
> /jeff
> 




More information about the ath11k mailing list