From: Ben Greear <greearb@xxxxxxxxxxxxxxx> This tries to encapsulate email conversation with Johannes on this topic for posterity's sake. Signed-off-by: Ben Greear <greearb@xxxxxxxxxxxxxxx> --- net/mac80211/tx.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 1fa422782905..05982538c3cf 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -2277,6 +2277,12 @@ netdev_tx_t ieee80211_monitor_start_xmit(struct sk_buff *skb, * isn't always enough to find the interface to use; for proper * VLAN/WDS support we will need a different mechanism (which * likely isn't going to be monitor interfaces). + * + * I had a question about why we need to do this, and the answer + * is that old hostap used this API and expects it to work like this, + * and also monitor vdevs are not directly mapped into the driver + * (and have no chantx in my case, at least), so you cannot directly + * transmit on a monitor port anyway. */ sdata = IEEE80211_DEV_TO_SUB_IF(dev); -- 2.20.1
