All of lore.kernel.org
 help / color / mirror / Atom feed
From: Mel Gorman <mgorman@techsingularity.net>
To: Tariq Toukan <tariqt@mellanox.com>
Cc: Jesper Dangaard Brouer <brouer@redhat.com>,
	"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
	akpm@linux-foundation.org, linux-mm <linux-mm@kvack.org>,
	Saeed Mahameed <saeedm@mellanox.com>
Subject: Re: Page allocator order-0 optimizations merged
Date: Wed, 22 Mar 2017 23:40:04 +0000	[thread overview]
Message-ID: <20170322234004.kffsce4owewgpqnm@techsingularity.net> (raw)
In-Reply-To: <83a0e3ef-acfa-a2af-2770-b9a92bda41bb@mellanox.com>

On Wed, Mar 22, 2017 at 07:39:17PM +0200, Tariq Toukan wrote:
> > > > This modification may slow allocations from IRQ context slightly
> > > > but the
> > > > main gain from the per-cpu allocator is that it scales better for
> > > > allocations from multiple contexts.  There is an implicit
> > > > assumption that
> > > > intensive allocations from IRQ contexts on multiple CPUs from a single
> > > > NUMA node are rare
> Hi Mel, Jesper, and all.
> 
> This assumption contradicts regular multi-stream traffic that is naturally
> handled
> over close numa cores.  I compared iperf TCP multistream (8 streams)
> over CX4 (mlx5 driver) with kernels v4.10 (before this series) vs
> kernel v4.11-rc1 (with this series).
> I disabled the page-cache (recycle) mechanism to stress the page allocator,
> and see a drastic degradation in BW, from 47.5 G in v4.10 to 31.4 G in
> v4.11-rc1 (34% drop).
> I noticed queued_spin_lock_slowpath occupies 62.87% of CPU time.

Can you get the stack trace for the spin lock slowpath to confirm it's
from IRQ context?

-- 
Mel Gorman
SUSE Labs

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  reply	other threads:[~2017-03-22 23:40 UTC|newest]

Thread overview: 63+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-02-27 20:25 [merged] mm-page_alloc-only-use-per-cpu-allocator-for-irq-safe-requests.patch removed from -mm tree akpm
2017-03-01 13:48 ` Page allocator order-0 optimizations merged Jesper Dangaard Brouer
2017-03-01 17:36   ` Tariq Toukan
2017-03-01 17:36     ` Tariq Toukan
2017-03-22 17:39     ` Tariq Toukan
2017-03-22 17:39       ` Tariq Toukan
2017-03-22 23:40       ` Mel Gorman [this message]
2017-03-23 13:43         ` Jesper Dangaard Brouer
2017-03-23 14:51           ` Mel Gorman
2017-03-26  8:21             ` Tariq Toukan
2017-03-26 10:17               ` Tariq Toukan
2017-03-27  7:32                 ` Pankaj Gupta
2017-03-27  8:55                   ` Jesper Dangaard Brouer
2017-03-27 12:28                     ` Mel Gorman
2017-03-27 12:39                     ` Jesper Dangaard Brouer
2017-03-27 13:32                       ` Mel Gorman
2017-03-28  7:32                         ` Tariq Toukan
2017-03-28  8:29                           ` Jesper Dangaard Brouer
2017-03-28 16:05                           ` Tariq Toukan
2017-03-28 18:24                             ` Jesper Dangaard Brouer
2017-03-29  7:13                               ` Tariq Toukan
2017-03-28  8:28                         ` Pankaj Gupta
2017-03-27 14:15                       ` Matthew Wilcox
2017-03-27 14:15                         ` Matthew Wilcox
2017-03-27 15:15                         ` Jesper Dangaard Brouer
2017-03-27 16:58                           ` in_irq_or_nmi() Matthew Wilcox
2017-03-27 16:58                             ` in_irq_or_nmi() Matthew Wilcox
2017-03-27 16:58                             ` in_irq_or_nmi() Matthew Wilcox
2017-03-29  8:12                             ` in_irq_or_nmi() Peter Zijlstra
2017-03-29  8:12                               ` in_irq_or_nmi() Peter Zijlstra
2017-03-29  8:12                               ` in_irq_or_nmi() Peter Zijlstra
2017-03-29  8:59                               ` in_irq_or_nmi() Jesper Dangaard Brouer
2017-03-29  8:59                                 ` in_irq_or_nmi() Jesper Dangaard Brouer
2017-03-29  9:19                                 ` in_irq_or_nmi() Peter Zijlstra
2017-03-29  9:19                                   ` in_irq_or_nmi() Peter Zijlstra
2017-03-29  9:19                                   ` in_irq_or_nmi() Peter Zijlstra
2017-03-29 18:12                                   ` in_irq_or_nmi() Matthew Wilcox
2017-03-29 18:12                                     ` in_irq_or_nmi() Matthew Wilcox
2017-03-29 19:11                                     ` in_irq_or_nmi() Jesper Dangaard Brouer
2017-03-29 19:11                                       ` in_irq_or_nmi() Jesper Dangaard Brouer
2017-03-29 19:44                                       ` in_irq_or_nmi() and RFC patch Jesper Dangaard Brouer
2017-03-29 19:44                                         ` Jesper Dangaard Brouer
2017-03-30  6:49                                         ` Peter Zijlstra
2017-03-30  6:49                                           ` Peter Zijlstra
2017-03-30  7:12                                           ` Jesper Dangaard Brouer
2017-03-30  7:12                                             ` Jesper Dangaard Brouer
2017-03-30  7:35                                             ` Peter Zijlstra
2017-03-30  7:35                                               ` Peter Zijlstra
2017-03-30  9:46                                               ` Jesper Dangaard Brouer
2017-03-30  9:46                                                 ` Jesper Dangaard Brouer
2017-03-30 13:04                                         ` Mel Gorman
2017-03-30 13:04                                           ` Mel Gorman
2017-03-30 15:07                                           ` Jesper Dangaard Brouer
2017-03-30 15:07                                             ` Jesper Dangaard Brouer
2017-04-03 12:05                                             ` Mel Gorman
2017-04-03 12:05                                               ` Mel Gorman
2017-04-05  8:53                                               ` Mel Gorman
2017-04-05  8:53                                                 ` Mel Gorman
2017-04-10 14:31   ` Page allocator order-0 optimizations merged zhong jiang
2017-04-10 14:31     ` zhong jiang
2017-04-10 15:10     ` Mel Gorman
2017-04-11  1:54       ` zhong jiang
2017-04-11  1:54         ` zhong jiang

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=20170322234004.kffsce4owewgpqnm@techsingularity.net \
    --to=mgorman@techsingularity.net \
    --cc=akpm@linux-foundation.org \
    --cc=brouer@redhat.com \
    --cc=linux-mm@kvack.org \
    --cc=netdev@vger.kernel.org \
    --cc=saeedm@mellanox.com \
    --cc=tariqt@mellanox.com \
    /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.