public inbox for netdev@vger.kernel.org 
 help / color / mirror / Atom feed
From: David Miller <davem@davemloft•net>
To: hadi@cyberus•ca
Cc: jgarzik@pobox•com, auke-jan.h.kok@intel•com,
	greearb@candelatech•com, john.ronciak@intel•com,
	Robert.Olsson@data•slu.se, jesse.brandeburg@intel•com,
	netdev@vger•kernel.org
Subject: Re: [e1000]: flow control on by default - good idea really?
Date: Wed, 05 Jul 2006 11:21:09 -0700 (PDT)	[thread overview]
Message-ID: <20060705.112109.74731123.davem@davemloft.net> (raw)
In-Reply-To: <1152033116.5276.22.camel@jzny2>

From: jamal <hadi@cyberus•ca>
Date: Tue, 04 Jul 2006 13:11:56 -0400

> Clearly, this is a bad thing. Yes, the device in the first instance was
> at fault. But i have argued in the past that NAPI does just fine without
> flow control being turned on, so even chewing 5% of bandwidth on flow
> control is a bad thing..

I think it should be on by default.

If the machine is really busy, proper flow control keeps the
sender from overflowing the RX ring of the receiver.

Yes, with NAPI this happens less, but it does still happen.  And
I know that this is probably one of the least tested code paths
in every single driver, RX ring emptying out, and in fact the Tigon2
and the earliest Tigon3 chips would lock up on you if the RX ring
was emptied completely requiring a full chip reset when the watchdog
timer fired.

  parent reply	other threads:[~2006-07-05 18:20 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-07-04 17:11 [e1000]: flow control on by default - good idea really? jamal
2006-07-04 19:20 ` jamal
2006-07-05 16:23   ` Auke Kok
2006-07-05 20:37     ` Krzysztof Oledzki
2006-07-05 18:22   ` David Miller
2006-07-05 18:32     ` Auke Kok
2006-07-05 20:45       ` Krzysztof Oledzki
2006-07-05 21:13         ` Auke Kok
2006-07-06 13:03         ` jamal
2006-07-06 18:25           ` Auke Kok
2006-07-07  3:09             ` jamal
2006-07-07  6:59           ` David Miller
2006-07-07 12:28             ` jamal
2006-07-20 20:15               ` Bug in e1000 + semantics of flow control WAS(Re: " jamal
2006-08-03 12:29                 ` jamal
2006-10-16 18:55               ` Auke Kok
2006-10-17 13:05                 ` jamal
2006-10-17 17:18                   ` Auke Kok
2006-10-17 18:25                   ` Stephen Hemminger
2006-10-17 21:02                   ` Auke Kok
2006-10-18 13:35                     ` jamal
2006-10-18 14:57                       ` Auke Kok
2006-10-17 21:46                   ` David Miller
2006-07-05 16:57 ` Robert Olsson
2006-07-05 18:21 ` David Miller [this message]
  -- strict thread matches above, loose matches on Subject: below --
2006-07-07  4:43 Michael Chan

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=20060705.112109.74731123.davem@davemloft.net \
    --to=davem@davemloft$(echo .)net \
    --cc=Robert.Olsson@data$(echo .)slu.se \
    --cc=auke-jan.h.kok@intel$(echo .)com \
    --cc=greearb@candelatech$(echo .)com \
    --cc=hadi@cyberus$(echo .)ca \
    --cc=jesse.brandeburg@intel$(echo .)com \
    --cc=jgarzik@pobox$(echo .)com \
    --cc=john.ronciak@intel$(echo .)com \
    --cc=netdev@vger$(echo .)kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox