[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1495110825.6465.22.camel@edumazet-glaptop3.roam.corp.google.com>
Date: Thu, 18 May 2017 05:33:45 -0700
From: Eric Dumazet <eric.dumazet@...il.com>
To: Eric Dumazet <edumazet@...gle.com>
Cc: "David S . Miller" <davem@...emloft.net>,
Neal Cardwell <ncardwell@...gle.com>,
Yuchung Cheng <ycheng@...gle.com>,
Soheil Hassas Yeganeh <soheil@...gle.com>,
Wei Wang <weiwan@...gle.com>, netdev <netdev@...r.kernel.org>
Subject: Re: [PATCH net-next 15/15] tcp: switch TCP TS option (RFC 7323) to
1ms clock
On Tue, 2017-05-16 at 14:00 -0700, Eric Dumazet wrote:
> TCP Timestamps option is defined in RFC 7323
>
> Traditionally on linux, it has been tied to the internal
> 'jiffies' variable, because it had been a cheap and good enough
> generator.
...
> @@ -3003,7 +3004,7 @@ void tcp_rearm_rto(struct sock *sk)
> struct sk_buff *skb = tcp_write_queue_head(sk);
> const u32 rto_time_stamp =
> tcp_skb_timestamp(skb) + rto;
> - s32 delta = (s32)(rto_time_stamp - tcp_time_stamp);
> + s32 delta = (s32)(rto_time_stamp - tcp_jiffies32);
> /* delta may not be positive if the socket is locked
> * when the retrans timer fires and is rescheduled.
> */
RTO is broken, I will send a fix after tests.
Powered by blists - more mailing lists