> let me suggest an alternative, which makes use of current ieee80211_is_eapol. > in regular Rx data frames, code will look like this: > > + ieee80211_drop_802_1x_pae(rx, size of current 802.11 hdr); > + ieee80211_drop_unencrypted(rx, size of current 802.11 hdr); > ieee80211_data_to_8023(rx) > - ieee80211_drop_802_1x_pae(rx, size of 802.3 hdr); > - ieee80211_drop_unencrypted(rx, size of 802.3 hdr); > > while in A-MSDU frames code will look like this: > > ieee80211_data_to_8023(rx) > for each internal frame (that is any how in 802.2 form): > + ieee80211_drop_802_1x_pae(rx, 0); > + ieee80211_drop_unencrypted(rx, 0); > move from 802.2 to 802.3 > - ieee80211_drop_802_1x_pae(rx, size of 802.3 hdr); > - ieee80211_drop_unencrypted(rx, size of 802.3 hdr); > > I thought it makes good use of the infrastructure we made so far, what > do you think? Yes, that would work too. I'm not entirely sure about the effect on eapol handling but I'll sort that out with that patch then. At worst I get to restructure the code instead of you :) johannes
Attachment:
signature.asc
Description: This is a digitally signed message part
