Fix: Kernel panic due to unregistered timer callback invoked
When traffic monitor support is enabled we fail to deregister
the timer which is started, during the unloading of the driver.
Due to which timer gets triggered after driver unload resulting
in kernel crash.
The fix is to correct the conditional check where we fail to
deregister traffic monitor timer.
Change-Id: I0328d2d8fcf9ae0ff295914aedc0006e0e28b7ab
CRs-Fixed: 572585
diff --git a/CORE/HDD/src/wlan_hdd_softap_tx_rx.c b/CORE/HDD/src/wlan_hdd_softap_tx_rx.c
index c9a248c..0699d49 100644
--- a/CORE/HDD/src/wlan_hdd_softap_tx_rx.c
+++ b/CORE/HDD/src/wlan_hdd_softap_tx_rx.c
@@ -187,7 +187,7 @@
status = wlan_hdd_validate_context(pHddCtx);
- if (0 != status)
+ if (-ENODEV == status)
{
VOS_TRACE( VOS_MODULE_ID_HDD, VOS_TRACE_LEVEL_ERROR,
"%s: HDD context is not valid", __func__);
@@ -209,7 +209,7 @@
vos_lock_destroy(&pHddCtx->traffic_monitor.trafficLock);
pHddCtx->traffic_monitor.isInitialized = 0;
}
- return status;
+ return VOS_STATUS_SUCCESS;
}
/**============================================================================