Skip to contentSkip to conversation

net: fix the RTO timer retransmitting skb every 1ms if linear option is enabled

PATCH Applied Jason Xing 4 participants 6 messages Aug 11, 2023 – Aug 15, 2023

Subsystems (3)
lore
Share & tools
lore

Jason Xing

author
#0 lore raw

PATCH v2 1/1 net: fix the RTO timer retransmitting skb every 1ms if linear option is enabled

From: Jason Xing <[email protected]>

In the real workload, I encountered an issue which could cause the RTO timer to retransmit the skb per 1ms with linear option enabled. The amount of lost-retransmitted skbs can go up to 1000+ instantly.

The root cause is that if the icsk_rto happens to be zero in the 6th round (which is the TCP_THIN_LINEAR_RETRIES value), then it will always be zero due to the changed calculation method in tcp_retransmit_timer() as follows:

icsk->icsk_rto = min(icsk->icsk_rto << 1, TCP_RTO_MAX);

Above line could be converted to
icsk->icsk_rto = min(0 << 1, TCP_RTO_MAX) = 0

Therefore, the timer expires so quickly without any doubt.

I read through the RFC 6298 and found that the RTO value can be rounded up to a certain value, in Linux, say TCP_RTO_MIN as default, which is regarded as the lower bound in this patch as suggested by Eric.

Fixes 36e31b0af587 ("net: TCP thin linear timeouts")

---
v2:
1) nit: alway->always and the indentation style suggested by Simon.

---
 net/ipv4/tcp_timer.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
net/ipv4/tcp_timer.c +3 -1
index d45c96c7f5a4..69795b273419 100644
@@ -599,7 +599,9 @@ void tcp_retransmit_timer(struct sock *sk)
 	    tcp_stream_is_thin(tp) &&
 	    icsk->icsk_retransmits <= TCP_THIN_LINEAR_RETRIES) {
 		icsk->icsk_backoff = 0;
-		icsk->icsk_rto = min(__tcp_set_rto(tp), TCP_RTO_MAX);
+		icsk->icsk_rto = clamp(__tcp_set_rto(tp),
+				       tcp_rto_min(sk),
+				       TCP_RTO_MAX);
 	} else if (sk->sk_state != TCP_SYN_SENT ||
 		   icsk->icsk_backoff >
 		   READ_ONCE(net->ipv4.sysctl_tcp_syn_linear_timeouts)) {
Signature · 1 line
2.37.3
author
#1 lore raw
Quoted 18 lines · Jason Xing

From: Jason Xing <[email protected]>

In the real workload, I encountered an issue which could cause the RTO timer to retransmit the skb per 1ms with linear option enabled. The amount of lost-retransmitted skbs can go up to 1000+ instantly.

The root cause is that if the icsk_rto happens to be zero in the 6th round (which is the TCP_THIN_LINEAR_RETRIES value), then it will always be zero due to the changed calculation method in tcp_retransmit_timer() as follows:

icsk->icsk_rto = min(icsk->icsk_rto << 1, TCP_RTO_MAX);

Above line could be converted to
icsk->icsk_rto = min(0 << 1, TCP_RTO_MAX) = 0

Therefore, the timer expires so quickly without any doubt.

Quoted 7 lines

I read through the RFC 6298 and found that the RTO value can be rounded up to a certain value, in Linux, say TCP_RTO_MIN as default, which is regarded as the lower bound in this patch as suggested by Eric.

Fixes 36e31b0af587 ("net: TCP thin linear timeouts")

Hello maintainers,

I wonder why someone in the patchwork[1] changed this v2 patch into Superseded status without comments? Should I convert it to a new status or something else?

[1]: https://patchwork.kernel.org/project/netdevbpf/patch/[email protected]/

Thanks,
Jason

Quoted 25 lines

---
v2:
1) nit: alway->always and the indentation style suggested by Simon.

---
 net/ipv4/tcp_timer.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
net/ipv4/tcp_timer.c +3 -2
index d45c96c7f5a4..69795b273419 100644
@@ -599,7 +599,9 @@ void tcp_retransmit_timer(struct sock *sk)
            tcp_stream_is_thin(tp) &&
            icsk->icsk_retransmits <= TCP_THIN_LINEAR_RETRIES) {
                icsk->icsk_backoff = 0;
-               icsk->icsk_rto = min(__tcp_set_rto(tp), TCP_RTO_MAX);
+               icsk->icsk_rto = clamp(__tcp_set_rto(tp),
+                                      tcp_rto_min(sk),
+                                      TCP_RTO_MAX);
        } else if (sk->sk_state != TCP_SYN_SENT ||
                   icsk->icsk_backoff >
                   READ_ONCE(net->ipv4.sysctl_tcp_syn_linear_timeouts)) {
--

2.37.3

maintainer
#2 lore raw
Quoted 3 lines · On Tue, 15 Aug 2023 10:08:13 +0800 Jason Xing wrote:

I wonder why someone in the patchwork[1] changed this v2 patch into Superseded status without comments? Should I convert it to a new status or something else?

Ack, looks like a mistake, let me bring it back.

Signature · 1 line
pw-bot: under-review
author
#3 lore raw
Quoted 7 lines · Jakub Kicinski
Quoted 3 lines · On Tue, 15 Aug 2023 10:08:13 +0800 Jason Xing wrote:

I wonder why someone in the patchwork[1] changed this v2 patch into Superseded status without comments? Should I convert it to a new status or something else?

Ack, looks like a mistake, let me bring it back.

Thanks.

Jason

Quoted 2 lines

--
pw-bot: under-review

Reviewed-by on v2 1/1
#4 lore raw
In reply to Jason Xing
Quoted 18 lines · Jason Xing

From: Jason Xing <[email protected]>

In the real workload, I encountered an issue which could cause the RTO timer to retransmit the skb per 1ms with linear option enabled. The amount of lost-retransmitted skbs can go up to 1000+ instantly.

The root cause is that if the icsk_rto happens to be zero in the 6th round (which is the TCP_THIN_LINEAR_RETRIES value), then it will always be zero due to the changed calculation method in tcp_retransmit_timer() as follows:

icsk->icsk_rto = min(icsk->icsk_rto << 1, TCP_RTO_MAX);

Above line could be converted to
icsk->icsk_rto = min(0 << 1, TCP_RTO_MAX) = 0

Therefore, the timer expires so quickly without any doubt.

Quoted 7 lines

I read through the RFC 6298 and found that the RTO value can be rounded up to a certain value, in Linux, say TCP_RTO_MIN as default, which is regarded as the lower bound in this patch as suggested by Eric.

Fixes 36e31b0af587 ("net: TCP thin linear timeouts")
#5 lore raw
In reply to Jason Xing
This patch was applied to netdev/net.git (main) by David S. Miller <[email protected]>:

Hello:

This patch was applied to netdev/net.git (main)
by David S. Miller <[email protected]>:

Quoted 11 lines · On Fri, 11 Aug 2023 10:37:47 +0800 you wrote:

From: Jason Xing <[email protected]>

In the real workload, I encountered an issue which could cause the RTO timer to retransmit the skb per 1ms with linear option enabled. The amount of lost-retransmitted skbs can go up to 1000+ instantly.

The root cause is that if the icsk_rto happens to be zero in the 6th round (which is the TCP_THIN_LINEAR_RETRIES value), then it will always be zero due to the changed calculation method in tcp_retransmit_timer() as follows:

[...]

Here is the summary with links:
- [v2,net] net: fix the RTO timer retransmitting skb every 1ms if linear option is enabled

    https://git.kernel.org/netdev/net/c/e4dd0d3a2f64

You are awesome, thank you!

Signature · 2 lines
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html