From: Phil Young <py@napatech.com>
Date: Wed, 11 Nov 2020 19:21:39 +0000 (-0500)
Subject: napatech: Fix potential double release of packet
X-Git-Tag: suricata-6.0.1~70
X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=789dcbd49d077b54726fb09738fb0f42112d00d0;p=thirdparty%2Fsuricata.git

napatech: Fix potential double release of packet

This addresses readmine issue #4018.  There was the potential for a packet
buffer to be released twice in response to an error condition.  This
addresses this by only calling NT_NetRxRelease() when the p->ReleasePacket
is called.
---

diff --git a/src/source-napatech.c b/src/source-napatech.c
index c30a4fb213..a18616e512 100644
--- a/src/source-napatech.c
+++ b/src/source-napatech.c
@@ -1001,12 +1001,10 @@ TmEcode NapatechPacketLoop(ThreadVars *tv, void *data, void *slot)
 
         if (unlikely(PacketSetData(p, (uint8_t *)NT_NET_GET_PKT_L2_PTR(packet_buffer), NT_NET_GET_PKT_WIRE_LENGTH(packet_buffer)))) {
             TmqhOutputPacketpool(ntv->tv, p);
-            NT_NetRxRelease(ntv->rx_stream, packet_buffer);
             SCReturnInt(TM_ECODE_FAILED);
         }
 
         if (unlikely(TmThreadsSlotProcessPkt(ntv->tv, ntv->slot, p) != TM_ECODE_OK)) {
-            NT_NetRxRelease(ntv->rx_stream, packet_buffer);
             SCReturnInt(TM_ECODE_FAILED);
         }