station: check for roam timeout before rearming

A user reported a crash which was due to the roam trigger timeout
being overwritten, followed by a disconnect. Post-disconnect the
timer would fire and result in a crash. Its not clear exactly where
the overwrite was happening but upon code inspection it could
happen in the following scenario:

1. Beacon loss event, start roam timeout
2. Signal low event, no check if timeout is running and the timeout
   gets overwritten.

The reported crash actually didn't appear to be from the above
scenario but something else, so this logic is being hardened and
improved

Now if a roam timeout already exists and trying to be rearmed IWD
will check the time remaining on the current timer and either keep
the active timer or reschedule it to the lesser of the two values
(current or new rearm time). This will avoid cases such as a long
roam timer being active (e.g. 60 seconds) followed by a beacon or
packet loss event which should trigger a more agressive roam
schedule.
This commit is contained in:
James Prestwood 2024-08-29 04:27:58 -07:00 committed by Denis Kenzior
parent 574b0d80dc
commit 61cba6bd28
1 changed files with 29 additions and 19 deletions

View File

@ -102,7 +102,6 @@ struct station {
struct l_queue *owe_hidden_scan_ids; struct l_queue *owe_hidden_scan_ids;
/* Roaming related members */ /* Roaming related members */
struct timespec roam_min_time;
struct l_timeout *roam_trigger_timeout; struct l_timeout *roam_trigger_timeout;
uint32_t roam_scan_id; uint32_t roam_scan_id;
uint8_t preauth_bssid[6]; uint8_t preauth_bssid[6];
@ -1820,7 +1819,6 @@ static void station_roam_state_clear(struct station *station)
station->preparing_roam = false; station->preparing_roam = false;
station->roam_scan_full = false; station->roam_scan_full = false;
station->signal_low = false; station->signal_low = false;
station->roam_min_time.tv_sec = 0;
station->netconfig_after_roam = false; station->netconfig_after_roam = false;
station->last_roam_scan = 0; station->last_roam_scan = 0;
@ -3052,20 +3050,33 @@ static void station_roam_trigger_cb(struct l_timeout *timeout, void *user_data)
static void station_roam_timeout_rearm(struct station *station, int seconds) static void station_roam_timeout_rearm(struct station *station, int seconds)
{ {
struct timespec now, min_timeout; uint64_t remaining;
clock_gettime(CLOCK_MONOTONIC, &now); if (!station->roam_trigger_timeout)
goto new_timeout;
min_timeout = now; /* If we cant get the remaining time just create a new timer */
min_timeout.tv_sec += seconds; if (L_WARN_ON(!l_timeout_remaining(station->roam_trigger_timeout,
&remaining))) {
l_timeout_remove(station->roam_trigger_timeout);
goto new_timeout;
}
if (station->roam_min_time.tv_sec < min_timeout.tv_sec || /* Our current timeout is less than the rearm, keep current */
(station->roam_min_time.tv_sec == min_timeout.tv_sec && if (l_time_before(remaining, seconds * L_USEC_PER_SEC)) {
station->roam_min_time.tv_nsec < min_timeout.tv_nsec)) l_debug("Keeping current roam timeout of %lu seconds",
station->roam_min_time = min_timeout; l_time_to_secs(remaining));
return;
}
seconds = station->roam_min_time.tv_sec - now.tv_sec + l_debug("Rescheduling roam timeout from %lu to %u seconds",
(station->roam_min_time.tv_nsec > now.tv_nsec ? 1 : 0); l_time_to_secs(remaining), seconds);
l_timeout_modify(station->roam_trigger_timeout, seconds);
return;
new_timeout:
l_debug("Arming new roam timer for %u seconds", seconds);
station->roam_trigger_timeout = station->roam_trigger_timeout =
l_timeout_create(seconds, station_roam_trigger_cb, l_timeout_create(seconds, station_roam_trigger_cb,
@ -3223,7 +3234,6 @@ static void station_ok_rssi(struct station *station)
station->roam_trigger_timeout = NULL; station->roam_trigger_timeout = NULL;
station->signal_low = false; station->signal_low = false;
station->roam_min_time.tv_sec = 0;
} }
static void station_event_roamed(struct station *station, struct scan_bss *new) static void station_event_roamed(struct station *station, struct scan_bss *new)
@ -3569,14 +3579,17 @@ static void station_packets_lost(struct station *station, uint32_t num_pkts)
l_debug("Too many roam attempts in %u second timeframe, " l_debug("Too many roam attempts in %u second timeframe, "
"delaying roam", LOSS_ROAM_RATE_LIMIT); "delaying roam", LOSS_ROAM_RATE_LIMIT);
if (station->roam_trigger_timeout)
return;
station_roam_timeout_rearm(station, LOSS_ROAM_RATE_LIMIT); station_roam_timeout_rearm(station, LOSS_ROAM_RATE_LIMIT);
return; return;
} }
if (station->roam_trigger_timeout) {
l_debug("canceling roam timer to roam immediately");
l_timeout_remove(station->roam_trigger_timeout);
station->roam_trigger_timeout = NULL;
}
station_start_roam(station); station_start_roam(station);
} }
@ -3589,9 +3602,6 @@ static void station_beacon_lost(struct station *station)
station_debug_event(station, "beacon-loss-roam"); station_debug_event(station, "beacon-loss-roam");
if (station->roam_trigger_timeout)
return;
station_roam_timeout_rearm(station, LOSS_ROAM_RATE_LIMIT); station_roam_timeout_rearm(station, LOSS_ROAM_RATE_LIMIT);
} }