ipv4: Unmask upper DSCP bits in fib_compute_spec_dst()
authorIdo Schimmel <idosch@nvidia.com>
Wed, 21 Aug 2024 12:52:46 +0000 (15:52 +0300)
committerJakub Kicinski <kuba@kernel.org>
Thu, 22 Aug 2024 23:59:57 +0000 (16:59 -0700)
As explained in commit 35ebf65e851c ("ipv4: Create and use
fib_compute_spec_dst() helper."), the function is used - for example -
to determine the source address for an ICMP reply. If we are responding
to a multicast or broadcast packet, the source address is set to the
source address that we would use if we were to send a packet to the
unicast source of the original packet. This address is determined by
performing a FIB lookup and using the preferred source address of the
resulting route.

Unmask the upper DSCP bits of the DS field of the packet that triggered
the reply so that in the future the FIB lookup could be performed
according to the full DSCP value.

No functional changes intended since the upper DSCP bits are masked when
comparing against the TOS selectors in FIB rules and routes.

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Guillaume Nault <gnault@redhat.com>
Acked-by: Florian Westphal <fw@strlen.de>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://patch.msgid.link/20240821125251.1571445-8-idosch@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
net/ipv4/fib_frontend.c

index 8b740f575da1a5a25ac0d09f1f2beaae769646f8..793e6781399a4fa575a51e5508d0cac0663e36e1 100644 (file)
@@ -293,7 +293,7 @@ __be32 fib_compute_spec_dst(struct sk_buff *skb)
                        .flowi4_iif = LOOPBACK_IFINDEX,
                        .flowi4_l3mdev = l3mdev_master_ifindex_rcu(dev),
                        .daddr = ip_hdr(skb)->saddr,
-                       .flowi4_tos = ip_hdr(skb)->tos & IPTOS_RT_MASK,
+                       .flowi4_tos = ip_hdr(skb)->tos & INET_DSCP_MASK,
                        .flowi4_scope = scope,
                        .flowi4_mark = vmark ? skb->mark : 0,
                };