From: Lorenz Bauer lmb@isovalent.com Date: Mon, 26 Jun 2023 16:08:58 +0100
Contrary to TCP, UDP reuseport groups can contain TCP_ESTABLISHED sockets. To support these properly we remember whether a group has a connected socket and skip the fast reuseport early-return. In effect we continue scoring all reuseport sockets and then choose the one with the highest score.
The current code fails to re-calculate the score for the result of lookup_reuseport. According to Kuniyuki Iwashima:
1) SO_INCOMING_CPU is set -> selected sk might have +1 score 2) BPF prog returns ESTABLISHED and/or SO_INCOMING_CPU sk -> selected sk will have more than 8Using the old score could trigger more lookups depending on the order that sockets are created.
sk -> sk (SO_INCOMING_CPU) -> sk (ESTABLISHED) | | `-> select the next SO_INCOMING_CPU sk | `-> select itself (We should save this lookup)Fixes: efc6b6f6c311 ("udp: Improve load balancing for SO_REUSEPORT.") Signed-off-by: Lorenz Bauer lmb@isovalent.com
Reviewed-by: Kuniyuki Iwashima kuniyu@amazon.com
1 minor nit below.
net/ipv4/udp.c | 20 +++++++++++++++----- net/ipv6/udp.c | 19 ++++++++++++++----- 2 files changed, 29 insertions(+), 10 deletions(-)
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index fd3dae081f3a..5ef478d2c408 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -450,14 +450,24 @@ static struct sock *udp4_lib_lookup2(struct net *net, score = compute_score(sk, net, saddr, sport, daddr, hnum, dif, sdif); if (score > badness) {
result = lookup_reuseport(net, sk, skb,saddr, sport, daddr, hnum);
badness = score;result = lookup_reuseport(net, sk, skb, saddr, sport, daddr, hnum);if (!result) {result = sk;continue;}/* Fall back to scoring if group has connections */
if (result && !reuseport_has_conns(sk))
if (!reuseport_has_conns(sk)) return result;
result = result ? : sk;badness = score;
/* Reuseport logic returned an error, keep original score. */if (IS_ERR(result))continue;badness = compute_score(result, net, saddr, sport,daddr, hnum, dif, sdif);
Unnecessary blank line here.
}} return result; diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index e5a337e6b970..8b3cb1d7da7c 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -193,14 +193,23 @@ static struct sock *udp6_lib_lookup2(struct net *net, score = compute_score(sk, net, saddr, sport, daddr, hnum, dif, sdif); if (score > badness) {
result = lookup_reuseport(net, sk, skb,saddr, sport, daddr, hnum);
badness = score;result = lookup_reuseport(net, sk, skb, saddr, sport, daddr, hnum);if (!result) {result = sk;continue;}/* Fall back to scoring if group has connections */
if (result && !reuseport_has_conns(sk))
if (!reuseport_has_conns(sk)) return result;
result = result ? : sk;badness = score;
/* Reuseport logic returned an error, keep original score. */if (IS_ERR(result))continue;badness = compute_score(sk, net, saddr, sport,daddr, hnum, dif, sdif);-- 2.40.1