From: Vladimír Čunát <vladimir.cunat@nic.cz>
Date: Mon, 29 May 2023 11:59:00 +0000 (+0200)
Subject: lib/cache pkt_renew(): fix an edge-case bug
X-Git-Tag: v5.7.0~7^2
X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=93169196498c2675936b4e7298c9abed16111cf3;p=thirdparty%2Fknot-resolver.git

lib/cache pkt_renew(): fix an edge-case bug

It could happen that this condition didn't get triggered,
but the structures weren't completely clear.  In particular,
the current section could be past KNOT_ANSWER already.
Let's be more conservative here; pkt_recycle() shouldn't be expensive.

I'm not sure why I only ran into this on the new-policy branch,
but it really seems like bug here on master already.
---

diff --git a/lib/cache/knot_pkt.c b/lib/cache/knot_pkt.c
index 31fa7e9b0..864cb57ca 100644
--- a/lib/cache/knot_pkt.c
+++ b/lib/cache/knot_pkt.c
@@ -11,8 +11,8 @@
 
 int pkt_renew(knot_pkt_t *pkt, const knot_dname_t *name, uint16_t type)
 {
-	/* Update packet question if needed. */
-	if (!knot_dname_is_equal(knot_pkt_qname(pkt), name)
+	/* Clear the packet if needed. */
+	if (pkt->rrset_count != 0 || !knot_dname_is_equal(knot_pkt_qname(pkt), name)
 	    || knot_pkt_qtype(pkt) != type || knot_pkt_qclass(pkt) != KNOT_CLASS_IN) {
 		int ret = kr_pkt_recycle(pkt);
 		if (ret) return kr_error(ret);