ipcpd: Fix bad lock in Ethernet IPCP

An unlock was called twice instead of a lock/unlock sequence, causing
a data race.

Signed-off-by: Sander Vrijders <[email protected]>
Signed-off-by: Dimitri Staessens <[email protected]>

1 files changed, 1 insertions, 1 deletions

diff --git a/src/ipcpd/eth/eth.c b/src/ipcpd/eth/eth.c
index af34f68e..9456f4cb 100644
--- a/src/ipcpd/eth/eth.c
+++ b/src/ipcpd/eth/eth.c
@@ -905,8 +905,8 @@ static void * eth_ipcp_sdu_reader(void * o)
 
                         memcpy(frame->buf, &e_frame->payload, length);
                         memcpy(frame->r_addr, e_frame->src_hwaddr, MAC_SIZE);
-                        pthread_mutex_unlock(&eth_data.mgmt_lock);
 
+                        pthread_mutex_lock(&eth_data.mgmt_lock);
                         list_add(&frame->next, &eth_data.mgmt_frames);
                         pthread_cond_signal(&eth_data.mgmt_cond);
                         pthread_mutex_unlock(&eth_data.mgmt_lock);