On 11/4/22 17:01, Johannes Berg wrote:
On Fri, 2022-11-04 at 16:58 +0100, Marek Vasut wrote:
Therefore, to fix this problem, inspect the ETH_P_802_3 frames in
the rsi_91x driver, check the ethertype of the encapsulated frame,
and in case it is ETH_P_PAE, transmit the frame via high-priority
queue just like other ETH_P_PAE frames.
This part seems wrong now.
OK
+bool rsi_is_tx_eapol(struct sk_buff *skb)
+{
+ return !!(IEEE80211_SKB_CB(skb)->control.flags &
+ IEEE80211_TX_CTRL_PORT_CTRL_PROTO);
+}
For how trivial this is now, maybe make it an inline? Feels fairly
pointless to have it as an out-of-line function to call in another file
when it's a simple bit check.
In V2 it was suggested I deduplicate this into a separate function,
since the test is done in multiple places. I would like to keep it
deduplicated.
You can also drop the !! since the return value is bool.
OK
