tcp: improve setsockopt() TCP_USER_TIMEOUT accuracy

Based on:

https://patchwork.kernel.org/patch/10516195/

Every time the TCP retransmission timer fires. It checks to see if there is a timeout before scheduling the next retransmit timer. The retransmit interval between each retransmission increases exponentially. The issue is that in order for the timeout to occur the retransmit timer needs to fire again. If the user timeout check happens after the 9th retransmit for example. It needs to wait for the 10th retransmit timer to fire in order to evaluate whether a timeout has occurred or not. If the interval is large enough then the timeout will be inaccurate.

For example with a TCP_USER_TIMEOUT of 10 seconds without patch:

1st retransmit:

22:25:18.973488 IP host1.49310 > host2.search-agent: Flags [.]

Last retransmit:

22:25:26.205499 IP host1.49310 > host2.search-agent: Flags [.]

Timeout:

send: Connection timed out
Sun Jul 1 22:25:34 EDT 2018

We can see that last retransmit took ~7 seconds. Which pushed the total timeout to ~15 seconds instead of the expected 10 seconds. This gets more inaccurate the larger the TCP_USER_TIMEOUT value. As the interval increases.

Add tcp_clamp_rto_to_user_timeout() to determine if the user rto has expired. Or whether the rto interval needs to be recalculated. Use the original interval if user rto is not set.

Test results with the patch is the expected 10 second timeout:

1st retransmit:

01:37:59.022555 IP host1.49310 > host2.search-agent: Flags [.]

Last retransmit:

01:38:06.486558 IP host1.49310 > host2.search-agent: Flags [.]

Timeout:

send: Connection timed out
Mon Jul 2 01:38:09 EDT 2018

Jon Maxwell (3):
tcp: convert icsk_user_timeout from jiffies to msecs
tcp: Add tcp_retransmit_time() helper routine
tcp: Add tcp_clamp_rto_to_user_timeout() helper to improve accuracy

 net/ipv4/tcp.c       |  4 ++--
 net/ipv4/tcp_timer.c | 51 ++++++++++++++++++++++++++++++++++++++-------------
 2 files changed, 40 insertions(+), 15 deletions(-)

2.13.6

I must confess I am lost with your submissions.

Patchwork is also lost ( https://patchwork.ozlabs.org/project/netdev/list/ )

Are you really using git format-patch ?

Normally all the patches should have a common [PATCH Vx net-next] string

git format-patch -o ../output --cover-letter --subject-prefix "PATCH V3 net-next" HEAD~3

After committing the patches in my net-next git branch I used git send-mail:

git send-email --identity=XXX --cover-letter --annotate origin --compose --signoff

and manually updated it based on an example of yours:

https://lwn.net/Articles/706491/

I can see the 3 patches that I just submitted on:

https://patchwork.ozlabs.org/project/netdev/list/

Flagged as under review by DaveM.

I dunno, each patch belongs to a separate patch series, this is not expected.

Compare to what happens on https://patchwork.ozlabs.org/project/netdev/list/ for a proper submission.

If you click on https://patchwork.ozlabs.org/project/netdev/list/?series=56100 for example, you can see 5 patch in a series.

Standard workflow :

git-format-patch ...

<edit cover letter>

git send-email --to "David S. Miller <[email protected]>" \
	--cc "netdev <[email protected]>" \
	--cc "Eric Dumazet <[email protected]>" \
	00*

Okay I see what you mean looking some examples on:

https://patchwork.ozlabs.org/project/netdev/list/

I'll resubmit with the correct patch series numbers but with the same detailed description.