On Tue, Jul 30, 2019 at 03:13:35AM +0000, Tony Chuang wrote: > > Those coex response skb buffers are allocated in rtw_pci_rx_isr(), > > but I do not see where they are freed (seems we do not process > > them in c2h_work which does dev_kfree_skb()). > > You're right, that SKB leaked. Should free them after responded. > I will send v2 to fix it :) FWIW maybe would be better to process coex commands entairly in c2h_work ? Not sure if that would work or really would be better, just an idea you can consider :-) Stanislaw