From: David Miller <davem@davemloft.net>
To: ilpo.jarvinen@helsinki.fi
Cc: netdev@vger.kernel.org
Subject: Re: [PATCH net-next 02/17] tcp: don't backtrack to sacked skbs
Date: Mon, 02 Mar 2009 03:01:03 -0800 (PST) [thread overview]
Message-ID: <20090302.030103.89533524.davem@davemloft.net> (raw)
In-Reply-To: <1235832282317-git-send-email-ilpo.jarvinen@helsinki.fi>
From: "Ilpo Järvinen" <ilpo.jarvinen@helsinki.fi>
Date: Sat, 28 Feb 2009 16:44:27 +0200
> From: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
>
> Backtracking to sacked skbs is a horrible performance killer
> since the hint cannot be advanced successfully past them...
> ...And it's totally unnecessary too.
>
> In theory this is 2.6.27..28 regression but I doubt anybody
> can make .28 to have worse performance because of other TCP
> improvements.
>
> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Applied.
next prev parent reply other threads:[~2009-03-02 11:01 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-02-28 14:44 [PATCH 0/17]: tcp & more tcp Ilpo Järvinen
2009-02-28 14:44 ` [PATCH net-2.6 01/17] tcp: fix retrans_out leaks Ilpo Järvinen
2009-02-28 14:44 ` [PATCH net-next 02/17] tcp: don't backtrack to sacked skbs Ilpo Järvinen
2009-02-28 14:44 ` [PATCH net-next 03/17] tcp: fix lost_cnt_hint miscounts Ilpo Järvinen
2009-02-28 14:44 ` [PATCH net-next 04/17] tcp: deferring in middle of queue makes very little sense Ilpo Järvinen
2009-02-28 14:44 ` [PATCH net-next 05/17] tcp: Don't clear hints when tcp_fragmenting Ilpo Järvinen
2009-02-28 14:44 ` [PATCH net-next 06/17] tcp: fix corner case issue in segmentation during rexmitting Ilpo Järvinen
2009-02-28 14:44 ` [PATCH net-next 07/17] tcp: remove redundant code from tcp_mark_lost_retrans Ilpo Järvinen
2009-02-28 14:44 ` [PATCH net-next 08/17] tcp: separate timeout marking loop to it's own function Ilpo Järvinen
2009-02-28 14:44 ` [PATCH net-next 09/17] tcp: cleanup ca_state mess in tcp_timer Ilpo Järvinen
2009-02-28 14:44 ` [PATCH net-next 10/17] tcp: drop unnecessary local var in collapse Ilpo Järvinen
2009-02-28 14:44 ` [PATCH net-next 11/17] htcp: merge icsk_ca_state compare Ilpo Järvinen
2009-02-28 14:44 ` [PATCH net-next 12/17] tcp: add helper for AI algorithm Ilpo Järvinen
2009-02-28 14:44 ` [PATCH net-next 13/17] tcp: kill eff_sacks "cache", the sole user can calculate itself Ilpo Järvinen
2009-02-28 14:44 ` [PATCH net-next 14/17] tcp: remove pointless .dsack code Ilpo Järvinen
2009-02-28 14:44 ` [PATCH net-next 15/17] tcp: kill pointless if () in sack_remove Ilpo Järvinen
2009-02-28 14:44 ` [PATCH net-next 16/17] tcp: in sendmsg/pages open code the real goto target Ilpo Järvinen
2009-02-28 14:44 ` [PATCH net-next 17/17] tcp: get rid of two unnecessary u16s in TCP skb flags copying Ilpo Järvinen
2009-03-01 2:39 ` Andi Kleen
2009-03-01 3:08 ` David Miller
2009-03-01 21:44 ` Andi Kleen
2009-03-02 11:03 ` David Miller
2009-03-02 11:03 ` [PATCH net-next 16/17] tcp: in sendmsg/pages open code the real goto target David Miller
2009-03-02 11:03 ` [PATCH net-next 15/17] tcp: kill pointless if () in sack_remove David Miller
2009-03-02 11:03 ` [PATCH net-next 14/17] tcp: remove pointless .dsack code David Miller
2009-03-02 11:54 ` Ilpo Järvinen
2009-03-02 11:57 ` David Miller
2009-03-02 11:02 ` [PATCH net-next 13/17] tcp: kill eff_sacks "cache", the sole user can calculate itself David Miller
2009-03-02 11:02 ` [PATCH net-next 12/17] tcp: add helper for AI algorithm David Miller
2009-03-02 11:02 ` [PATCH net-next 11/17] htcp: merge icsk_ca_state compare David Miller
2009-03-02 11:02 ` [PATCH net-next 10/17] tcp: drop unnecessary local var in collapse David Miller
2009-03-02 11:02 ` [PATCH net-next 09/17] tcp: cleanup ca_state mess in tcp_timer David Miller
2009-03-02 11:02 ` [PATCH net-next 08/17] tcp: separate timeout marking loop to it's own function David Miller
2009-03-02 11:01 ` [PATCH net-next 06/17] tcp: fix corner case issue in segmentation during rexmitting David Miller
2009-03-02 11:01 ` [PATCH net-next 07/17] tcp: remove redundant code from tcp_mark_lost_retrans David Miller
2009-03-02 11:01 ` [PATCH net-next 05/17] tcp: Don't clear hints when tcp_fragmenting David Miller
2009-03-02 11:01 ` [PATCH net-next 04/17] tcp: deferring in middle of queue makes very little sense David Miller
2009-03-02 11:01 ` [PATCH net-next 03/17] tcp: fix lost_cnt_hint miscounts David Miller
2009-03-02 11:01 ` David Miller [this message]
2009-03-01 8:22 ` [PATCH net-2.6 01/17] tcp: fix retrans_out leaks David Miller
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20090302.030103.89533524.davem@davemloft.net \
--to=davem@davemloft.net \
--cc=ilpo.jarvinen@helsinki.fi \
--cc=netdev@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.