tcp: ensure PMTU updates are processed during fastopen
Subsystems (3)
Jakub Kicinski
maintainerPATCH 1/1 tcp: ensure PMTU updates are processed during fastopen
tp->rx_opt.mss_clamp is not populated, yet, during TFO send so we rise it to the local MSS. tp->mss_cache is not updated, however:
tcp_v6_connect():
tp->rx_opt.mss_clamp = IPV6_MIN_MTU - headers;
tcp_connect():
tcp_connect_init():
tp->mss_cache = min(mtu, tp->rx_opt.mss_clamp)
tcp_send_syn_data():
tp->rx_opt.mss_clamp = tp->advmssAfter recent fixes to ICMPv6 PTB handling we started dropping PMTU updates higher than tp->mss_cache. Because of the stale tp->mss_cache value PMTU updates during TFO are always dropped.
Thanks to Wei for helping zero in on the problem and the fix!
--- net/ipv4/tcp_output.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
net/ipv4/tcp_output.c +4 -1
@@ -3719,6 +3719,7 @@ static void tcp_connect_queue_skb(struct sock *sk, struct sk_buff *skb)*/ static int tcp_send_syn_data (struct sock *sk,struct sk_buff *syn) { +struct inet_connection_sock *icsk =inet_csk (sk);struct tcp_sock *tp =tcp_sk (sk);struct tcp_fastopen_request *fo = tp->fastopen_req;int space, err =0 ;
@@ -3733,8 +3734,10 @@ static int tcp_send_syn_data(struct sock *sk, struct sk_buff *syn)* private TCP options. The cost is reduced data space in SYN :( */ tp->rx_opt.mss_clamp =tcp_mss_clamp (tp, tp->rx_opt.mss_clamp); +/* Sync mss_cache after updating the mss_clamp */ +tcp_sync_mss (sk, icsk->icsk_pmtu_cookie); - space =__tcp_mtu_to_mss (sk,inet_csk (sk)->icsk_pmtu_cookie) - + space =__tcp_mtu_to_mss (sk, icsk->icsk_pmtu_cookie) -MAX_TCP_OPTION_SPACE ; space =min_t (size_t, space, fo->size);
Signature · 1 line
2.34.1
Quoted 18 lines · Jakub Kicinski
tp->rx_opt.mss_clampis not populated, yet, during TFO send so we rise it to the local MSS.tp->mss_cacheis not updated, however:
tcp_v6_connect():
tp->rx_opt.mss_clamp=IPV6_MIN_MTU- headers;
tcp_connect():tcp_connect_init(): tp->mss_cache = min(mtu, tp->rx_opt.mss_clamp) tcp_send_syn_data(): tp->rx_opt.mss_clamp = tp->advmssAfter recent fixes to ICMPv6 PTB handling we started dropping PMTU updates higher than
tp->mss_cache. Because of the staletp->mss_cachevalue PMTU updates during TFO are always dropped.Thanks to Wei for helping zero in on the problem and the fix!
Quoted 8 lines
Fixes c7bb4b89033b ("ipv6: tcp: drop silly ICMPv6 packet too big messages")Reported-by Andre Nash <[email protected]>Reported-by Neil Spring <[email protected]>Reviewed-by Wei Wang <[email protected]>Acked-by Yuchung Cheng <[email protected]>Acked-by Martin KaFai Lau <[email protected]>Signed-off-by Jakub Kicinski <[email protected]>---
Do you have a packetdrill test by any chance ?
Thanks!
Jakub Kicinski
maintainerQuoted 1 line · On Mon, 21 Mar 2022 12:32:34 -0700 Eric Dumazet wrote:
Do you have a packetdrill test by any chance ?
I don't :( I was using veths to repro.
This patch was applied to netdev/net.git (master) by Jakub Kicinski <[email protected]>:
Hello:
This patch was applied to netdev/net.git (master)
by Jakub Kicinski <[email protected]>:
Quoted 6 lines · On Mon, 21 Mar 2022 09:59:57 -0700 you wrote:
tp->rx_opt.mss_clampis not populated, yet, during TFO send so we rise it to the local MSS.tp->mss_cacheis not updated, however:
tcp_v6_connect():
tp->rx_opt.mss_clamp=IPV6_MIN_MTU- headers;
tcp_connect():
Quoted 6 lines
tcp_connect_init(): tp->mss_cache = min(mtu, tp->rx_opt.mss_clamp) tcp_send_syn_data(): tp->rx_opt.mss_clamp = tp->advmss[...]
Here is the summary with links:
- [net] tcp: ensure PMTU updates are processed during fastopen
https://git.kernel.org/netdev/net/c/ed0c99dc0f49
You are awesome, thank you!
Signature · 2 lines
Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/patchwork/pwbot.html