From: Timur Tabi <timur@codeaurora.org>
To: David Laight <David.Laight@ACULAB.COM>,
"David S. Miller" <davem@davemloft.net>,
"netdev@vger.kernel.org" <netdev@vger.kernel.org>
Subject: Re: [PATCH 1/2] [for 4.13] net: qcom/emac: disable flow control autonegotiation by default
Date: Wed, 2 Aug 2017 10:08:35 -0500 [thread overview]
Message-ID: <115c4698-cfcb-83dc-e70b-89207ef326a8@codeaurora.org> (raw)
In-Reply-To: <063D6719AE5E284EB5DD2968C1650D6DD00491F6@AcuExch.aculab.com>
On 08/02/2017 09:51 AM, David Laight wrote:
> Sending pause frames just tells the adjacent switch not to send you packets
> (because you'll discard them).
> Since the idea is to avoid the discards, the switch will buffer the
> packets it would have sent.
> The buffers in the switch then fill up with packets it isn't sending you.
I was under the impression that the switch forwards the pause frames to
other devices, so that the transmitting NIC can stop sending the data,
but your explanation makes a lot more sense. If the EMAC never stops
sending pause frames, then the switch's buffers will fill up, disabling
all other devices. If the switch does not have per-port buffers, then
it makes sense when the buffer is full, it blocks all ports.
> The switch then runs out of buffers, it has 2 choices:
> 1) Throw the packets away.
> 2) Send 'pause' frames to the sources.
> If it sends 'pause' frames the entire network will very quickly lock up.
> If it discards the packets they might as well have been discarded by the
> receiving MAC.
>
> Doesn't this mean that pause frames are 99.999% useless??
Pause frames are intended for situations where the receiving CPU is
temporarily overwhelmed and just needs a second or two to resume
processing incoming packets. That makes sense on a dinky single-core
32-bit CPU.
--
Qualcomm Datacenter Technologies, Inc. as an affiliate of Qualcomm
Technologies, Inc. Qualcomm Technologies, Inc. is a member of the
Code Aurora Forum, a Linux Foundation Collaborative Project.
next prev parent reply other threads:[~2017-08-02 15:08 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-08-01 21:37 [PATCH 0/2] net: qcom/emac: fixes for pause frame floods Timur Tabi
2017-08-01 21:37 ` [PATCH 1/2] [for 4.13] net: qcom/emac: disable flow control autonegotiation by default Timur Tabi
2017-08-01 21:55 ` Florian Fainelli
2017-08-01 22:02 ` Timur Tabi
2017-08-01 22:08 ` Florian Fainelli
2017-08-01 23:15 ` Andrew Lunn
2017-08-02 0:56 ` Timur Tabi
2017-08-02 2:58 ` Andrew Lunn
2017-08-02 3:22 ` Timur Tabi
2017-08-02 13:48 ` David Laight
2017-08-02 14:21 ` Timur Tabi
2017-08-02 14:51 ` David Laight
2017-08-02 15:08 ` Timur Tabi [this message]
2017-08-02 15:38 ` David Laight
2017-08-02 17:54 ` David Miller
2017-08-02 18:23 ` Timur Tabi
2017-08-02 18:35 ` David Miller
2017-08-02 18:39 ` Timur Tabi
2017-08-02 23:15 ` David Miller
2017-08-03 1:00 ` Timur Tabi
2017-08-02 18:36 ` David Miller
2017-08-01 21:37 ` [PATCH 2/2] net: qcom/emac: add software control for pause frame mode Timur Tabi
2017-08-01 21:51 ` Florian Fainelli
2017-08-01 22:00 ` Timur Tabi
2017-08-01 22:06 ` Florian Fainelli
2017-09-12 22:07 ` Timur Tabi
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=115c4698-cfcb-83dc-e70b-89207ef326a8@codeaurora.org \
--to=timur@codeaurora.org \
--cc=David.Laight@ACULAB.COM \
--cc=davem@davemloft.net \
--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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).