dnsmasq: Fix CVE-2026-5172

Pick patch from [1] dnsmasq 2.90 debian bookworm pacthes.

[1] https://sources.debian.org/src/dnsmasq/2.90-4~deb12u2/debian/patches/CVE-2026-5172.patch

Signed-off-by: Hugo SIMELIERE (Schneider Electric) <hsimeliere.opensource@witekio.com>
Reviewed-by: Bruno VERNAY <bruno.vernay@se.com>
Signed-off-by: Anuj Mittal <anuj.mittal@oss.qualcomm.com>
This commit is contained in:
Hugo SIMELIERE (Schneider Electric)
2026-05-20 14:29:08 +02:00
committed by Anuj Mittal
parent 7dda8e9bd7
commit f1d78e9527
2 changed files with 40 additions and 0 deletions
@@ -19,6 +19,7 @@ SRC_URI = "http://www.thekelleys.org.uk/dnsmasq/${@['archive/', ''][float(d.getV
file://CVE-2026-4891.patch \
file://CVE-2026-4892.patch \
file://CVE-2026-4893.patch \
file://CVE-2026-5172.patch \
"
SRC_URI[sha256sum] = "8f6666b542403b5ee7ccce66ea73a4a51cf19dd49392aaccd37231a2c51b303b"
@@ -0,0 +1,39 @@
From f158664062e049ec4604f6e772551a00575011f4 Mon Sep 17 00:00:00 2001
From: Simon Kelley <simon@thekelleys.org.uk>
Date: Mon, 30 Mar 2026 16:24:33 +0100
Subject: [PATCH] Fix buffer overflow vulnerability in extract_addresses()
CVE-2026-5172
Thanks to Hugo Martinez Ray for spotting this.
The value of rdlen for an RR can be a lie, allowing the
call to extract_name() at rfc1025.c:952 to advance the value of p1
past the calculated end of the record. The makes the calculation
of bytes remaining in the RR underflow to a huge number and results
in a massive heap OOB read and certain crash.
CVE: CVE-2026-5172
Upstream-Status: Backport [https://thekelleys.org.uk/gitweb/?p=dnsmasq.git;a=commit;h=073082ddc0aba7b8efa15a688d6183463b65effa]
Signed-off-by: Hugo SIMELIERE (Schneider Electric) <hsimeliere.opensource@witekio.com>
---
src/rfc1035.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/src/rfc1035.c b/src/rfc1035.c
index 387d894a..32dc5711 100644
--- a/src/rfc1035.c
+++ b/src/rfc1035.c
@@ -932,7 +932,8 @@ int extract_addresses(struct dns_header *header, size_t qlen, char *name, time_t
/* Name, extract it then re-encode. */
int len;
- if (!extract_name(header, qlen, &p1, name, 1, 0))
+ /* rdlen may lie, and extract_name() advances p1 past where it says the record ends. */
+ if (!extract_name(header, qlen, &p1, name, 1, 0) || (p1 > endrr))
{
blockdata_free(addr.rrblock.rrdata);
return 2;
--
2.43.0