netdev - Re: [PATCH net-next 2/2] tcp_cubic: refine Hystart delay threshold

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:	Tue, 9 Dec 2014 13:22:23 -0800
From:	Dave Taht <dave.taht@...il.com>
To:	David Miller <davem@...emloft.net>
Cc:	Eric Dumazet <eric.dumazet@...il.com>,
	"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
	Nandita Dukkipati <nanditad@...gle.com>,
	Neal Cardwell <ncardwell@...gle.com>,
	Yuchung Cheng <ycheng@...gle.com>, sangtae.ha@...il.com
Subject: Re: [PATCH net-next 2/2] tcp_cubic: refine Hystart delay threshold

On Tue, Dec 9, 2014 at 11:58 AM, David Miller <davem@...emloft.net> wrote:
> From: Eric Dumazet <eric.dumazet@...il.com>
> Date: Thu, 04 Dec 2014 16:13:49 -0800
>
>> From: Eric Dumazet <edumazet@...gle.com>
>>
>> In commit 2b4636a5f8ca ("tcp_cubic: make the delay threshold of HyStart
>> less sensitive"), HYSTART_DELAY_MIN was changed to 4 ms.
>>
>> The remaining problem is that using delay_min + (delay_min/16) as the
>> threshold is too sensitive.
>>
>> 6.25 % of variation is too small for rtt above 60 ms, which are not
>> uncommon.
>  ...
>> Signed-off-by: Eric Dumazet <edumazet@...gle.com>
>
> Applied.

Thank you (in a weird way!) for invalidating several weeks worth of
long rtt tests of some aqm and fq variants.

http://snapon.lab.bufferbloat.net/~d/comprehensive.puck/long_rtt_baseline.png

/me goes off to pull and recompile, and try again.

-- 
Dave Täht

http://www.bufferbloat.net/projects/bloat/wiki/Upcoming_Talks
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html