I believe I found a problem that may be larger than my little sandbox.
The problem I see is that sometimes (and quite often when I am using lots
of vdevs and thus the NIC is busy), the keys are set before the EAPOL 4/4
hits the air. When the key is set, the NIC will no longer transmit the
frame because of key-length issues in the tx-descriptor (ath10k wave-2
in this case).
If I add a sleep before setting the key, then it works much more
reliably.
I suspect that there is a fundamental race between the EAPOL packet-tx
logic and the key-set logic, but supplicant appears to act as though
they are natually synchronized.
So, any suggestions on how to do this right?
Thanks,
Ben
greearb@v-f24-64 hostap]$ git diff
diff --git a/src/rsn_supp/wpa.c b/src/rsn_supp/wpa.c
index 8a1d164..50a3006 100644
--- a/src/rsn_supp/wpa.c
+++ b/src/rsn_supp/wpa.c
@@ -1423,6 +1423,11 @@ static void wpa_supplicant_process_3_of_4(struct wpa_sm *sm,
sm->renew_snonce = 1;
if (key_info & WPA_KEY_INFO_INSTALL) {
+ /* Well now...what if the NIC hasn't actually put the 4/4 on the air
+ * yet? If we set the key too soon, it is liable to corrupt the pkt being
+ * sent.. I don't know a good fix, but here is a hack.
+ */
+ os_sleep(0, 10000); /* sleep 10ms */
if (wpa_supplicant_install_ptk(sm, key))
goto failed;
}
--
Ben Greear <greearb@xxxxxxxxxxxxxxx>
Candela Technologies Inc http://www.candelatech.com
