Re: [RFC PATCHv2 net-next 1/2] tcp: RTO Restart (RTOR)

[Per Hurtig <per.hurtig@kau.se>]

[-- Attachment #1: Type: text/plain, Size: 2677 bytes --]


> On 08 Dec 2015, at 11:50, Ilpo Järvinen <ilpo.jarvinen@helsinki.fi> wrote:
> 
> On Tue, 8 Dec 2015, Per Hurtig wrote:
> 
>> This patch implements the RTO restart modification (RTOR). When data is
>> ACKed, and the RTO timer is restarted, the time elapsed since the last
>> outstanding segment was transmitted is subtracted from the calculated RTO
>> value. This way, the RTO timer will expire after exactly RTO seconds, and
>> not RTO + RTT [+ delACK] seconds.
>> 
>> This patch also implements a new sysctl (tcp_timer_restart) that is used
>> to control the timer restart behavior.
>> 
>> Signed-off-by: Per Hurtig <per.hurtig@kau.se>
>> ---
>> Documentation/networking/ip-sysctl.txt | 12 ++++++++++++
>> include/net/tcp.h                      |  6 ++++++
>> net/ipv4/sysctl_net_ipv4.c             | 10 ++++++++++
>> net/ipv4/tcp_input.c                   | 29 +++++++++++++++++++++++++++++
>> 4 files changed, 57 insertions(+)
>> 
>> diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
>> index 2ea4c45..4094128 100644
>> --- a/Documentation/networking/ip-sysctl.txt
>> +++ b/Documentation/networking/ip-sysctl.txt
>> @@ -591,6 +591,18 @@ tcp_syn_retries - INTEGER
>> 	with the current initial RTO of 1second. With this the final timeout
>> 	for an active TCP connection attempt will happen after 127seconds.
>> 
>> +tcp_timer_restart - INTEGER
>> +	Controls how the RTO and PTO timers are restarted (RTOR and TLPR).
>> +	If set (per timer or combined) the timers are restarted with
>> +	respect to the earliest outstanding segment, to not extend tail loss
>> +	latency unnecessarily.
>> +	Possible values:
>> +		0 disables RTOR and TLPR.
>> +		1 enables RTOR.
>> +		2 enables TLPR.
>> +		3 enables RTOR and TLPR.
>> +	Default: 3
>> +
>> tcp_timestamps - BOOLEAN
>> 	Enable timestamps as defined in RFC1323.
>> 
>> diff --git a/include/net/tcp.h b/include/net/tcp.h
>> index f80e74c..833efb7 100644
>> --- a/include/net/tcp.h
>> +++ b/include/net/tcp.h
>> @@ -76,6 +76,11 @@ void tcp_time_wait(struct sock *sk, int state, int timeo);
>> /* After receiving this amount of duplicate ACKs fast retransmit starts. */
>> #define TCP_FASTRETRANS_THRESH 3
>> 
>> +/* Disable RTO Restart if the number of outstanding segments is at least. */
>> +#define TCP_TIMER_RTORESTART	1
>> +#define TCP_TIMER_TLPRESTART	2
>> +#define TCP_RTORESTART_THRESH	4
> 
> Unfortunately the comment got now separated from the actual define.
> 
> 
> --
> i.
Darn, I knew I missed something. Well, I’ll fix that in the next round. Suppose there are more things that could be improved.

Per


[-- Attachment #2: Message signed with OpenPGP using GPGMail --]
[-- Type: application/pgp-signature, Size: 455 bytes --]