On Thu, 2019-08-22 at 09:00 +0200, Johannes Berg wrote:
>
> Perhaps it expects the 4-way-HS to already be in 4-addr frame format, or
> something else special in the 4-way-HS if you have WDS?
I think this is actually the right guess.
The working capture you sent me has the EAPOL 2/4 in a 4-addr frame:
ToDS=1, FromDS=1
A1/RA = AP
A2/TA = STA
A3/DA = AP
A4/SA = STA
The non-working capture has the EAPOL 2/4 in 3-addr format, as you'd
expect in the Linux 4-addr AP/STA case:
ToDS=1, FromDS=0
A1/RA = AP
A2/TA,SA = STA
A3/DA = AP
Since it's basically ignoring the message 2 (it just says "handshake
timed out" later) it's almost certainly expecting *only* the 4-addr
format.
As a hack, you could do
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -2623,8 +2623,7 @@ static struct sk_buff *ieee80211_build_hdr(struct ieee80211_sub_if_data *sdata,
memcpy(hdr.addr2, skb->data + ETH_ALEN, ETH_ALEN);
memcpy(hdr.addr3, sdata->u.mgd.bssid, ETH_ALEN);
hdrlen = 24;
- } else if (sdata->u.mgd.use_4addr &&
- cpu_to_be16(ethertype) != sdata->control_port_protocol) {
+ } else if (sdata->u.mgd.use_4addr) {
fc |= cpu_to_le16(IEEE80211_FCTL_FROMDS |
IEEE80211_FCTL_TODS);
/* RA TA DA SA */
in mac80211, then it should send 4-addr frames even for EAPOL.
johannes
