From: Andrew Lunn <andrew@lunn.ch>
To: Sagi Grimberg <sagi@grimberg.me>
Cc: Tariq Toukan <tariqt@mellanox.com>,
Jes Sorensen <jsorensen@fb.com>,
Saeed Mahameed <saeedm@dev.mellanox.co.il>,
Networking <netdev@vger.kernel.org>,
Leon Romanovsky <leonro@mellanox.com>,
Saeed Mahameed <saeedm@mellanox.com>,
Kernel Team <kernel-team@fb.com>, Christoph Hellwig <hch@lst.de>,
Thomas Gleixner <tglx@linutronix.de>
Subject: Re: mlx5 broken affinity
Date: Thu, 2 Nov 2017 13:13:30 +0100 [thread overview]
Message-ID: <20171102121330.GD4772@lunn.ch> (raw)
In-Reply-To: <83d3944f-8a31-eb31-93db-294906630b0e@grimberg.me>
> >This means that if your NIC is on NUMA #1, and you reduce the number of
> >channels, you might end up working only with the cores on the far NUMA.
> >Not good!
> We deliberated on this before, and concluded that application affinity
> and device affinity are equally important. If you have a real use case
> that shows otherwise, its perfectly doable to start from the device home
> node.
Hi Sagi
You might find this talk at NetDev interesting:
https://www.netdevconf.org/2.2/session.html?shochat-devicemgmt-talk
Andrew
next prev parent reply other threads:[~2017-11-02 12:13 UTC|newest]
Thread overview: 47+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-11-01 16:19 mlx5 broken affinity Jes Sorensen
2017-11-01 17:21 ` Sagi Grimberg
2017-11-01 18:20 ` Jes Sorensen
2017-11-01 22:41 ` Saeed Mahameed
2017-11-01 23:02 ` Jes Sorensen
2017-11-02 8:28 ` Tariq Toukan
2017-11-02 10:08 ` Sagi Grimberg
2017-11-02 12:13 ` Andrew Lunn [this message]
2017-11-02 14:48 ` Jes Sorensen
2017-11-02 16:14 ` Sagi Grimberg
2017-11-02 17:13 ` Jes Sorensen
2017-11-02 18:10 ` Thomas Gleixner
2017-11-05 8:36 ` Sagi Grimberg
2017-11-07 15:07 ` Thomas Gleixner
2017-11-08 7:27 ` Sagi Grimberg
2017-11-08 12:21 ` David Laight
2017-11-08 16:13 ` Jens Axboe
2017-11-09 10:09 ` Christoph Hellwig
2017-11-09 15:18 ` Jens Axboe
2017-11-09 15:08 ` Saeed Mahameed
2017-11-09 15:40 ` Sagi Grimberg
2017-11-08 16:19 ` Jes Sorensen
2017-11-08 17:33 ` Thomas Gleixner
2017-11-09 10:50 ` Sagi Grimberg
2017-11-09 14:19 ` Thomas Gleixner
2017-11-09 15:21 ` Jens Axboe
2017-11-09 17:03 ` Thomas Gleixner
2017-11-09 20:11 ` Jens Axboe
2017-11-09 21:23 ` Thomas Gleixner
2017-11-09 21:30 ` Jens Axboe
2017-11-09 21:42 ` [RFD] Managed interrupt affinities [ Was: mlx5 broken affinity ] Thomas Gleixner
2017-11-10 5:56 ` Saeed Mahameed
2017-11-10 13:03 ` Thomas Gleixner
2017-11-13 19:20 ` Sagi Grimberg
2017-11-13 20:51 ` Thomas Gleixner
2017-11-13 21:13 ` Sagi Grimberg
2017-11-13 21:33 ` Thomas Gleixner
2017-11-13 21:49 ` Sagi Grimberg
2017-11-14 10:15 ` Thomas Gleixner
2017-11-09 16:01 ` mlx5 broken affinity Sagi Grimberg
2017-11-09 16:09 ` Jens Axboe
2017-11-09 17:07 ` Thomas Gleixner
2017-11-09 20:12 ` Jens Axboe
2017-11-09 21:25 ` Thomas Gleixner
2017-11-09 15:19 ` Jens Axboe
2017-11-09 22:03 ` Jes Sorensen
2017-11-02 7:57 ` Sagi Grimberg
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=20171102121330.GD4772@lunn.ch \
--to=andrew@lunn.ch \
--cc=hch@lst.de \
--cc=jsorensen@fb.com \
--cc=kernel-team@fb.com \
--cc=leonro@mellanox.com \
--cc=netdev@vger.kernel.org \
--cc=saeedm@dev.mellanox.co.il \
--cc=saeedm@mellanox.com \
--cc=sagi@grimberg.me \
--cc=tariqt@mellanox.com \
--cc=tglx@linutronix.de \
/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.