bpf.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jesper Dangaard Brouer <hawk@kernel.org>
To: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: "Toke Høiland-Jørgensen" <toke@redhat.com>,
	bpf@vger.kernel.org, netdev@vger.kernel.org
Subject: Re: [PATCH RFC net-next 1/2] net: Reference bpf_redirect_info via task_struct on PREEMPT_RT.
Date: Tue, 20 Feb 2024 11:42:57 +0100	[thread overview]
Message-ID: <0b1c8247-ccfb-4228-bd64-53583329aaa7@kernel.org> (raw)
In-Reply-To: <20240220101741.PZwhANsA@linutronix.de>



On 20/02/2024 11.17, Sebastian Andrzej Siewior wrote:
> On 2024-02-20 10:17:53 [+0100], Jesper Dangaard Brouer wrote:
>>
>>
>> On 19/02/2024 20.01, Toke Høiland-Jørgensen wrote:
>>> may be simpler to use pktgen, and at 10G rates that shouldn't become a
>>> bottleneck either. The pktgen_sample03_burst_single_flow.sh script in
>>> samples/pktgen in the kernel source tree is fine for this usage.
>>
>> Example of running script:
>>   ./pktgen_sample03_burst_single_flow.sh -vi mlx5p1 -d 198.18.1.1 -m
>> ec:0d:9a:db:11:c4 -t 12
>>
>> Notice the last parameter, which is number threads to start.
>> If you have a ixgbe NIC driver, then I recommend -t 2 even if you have more
>> CPUs.
> 
> I get
> | Summary                 8,435,690 rx/s                  0 err/s

This seems low...
Have you remembered to disable Ethernet flow-control?

  # ethtool -A ixgbe1 rx off tx off
  # ethtool -A i40e2 rx off tx off


> | Summary                 8,436,294 rx/s                  0 err/s

You want to see the "extended" info via cmdline (or Ctrl+\)

  # xdp-bench drop -e eth1


> 
> with "-t 8 -b 64". I started with 2 and then increased until rx/s was
> falling again. I have ixgbe on the sending side and i40e on the

With ixgbe on the sending side, my testlab shows I need -t 2.

With -t 2 :
Summary                14,678,170 rx/s                  0 err/s
   receive total        14,678,170 pkt/s        14,678,170 drop/s 
         0 error/s
     cpu:1              14,678,170 pkt/s        14,678,170 drop/s 
         0 error/s
   xdp_exception                 0 hit/s

with -t 4:

Summary                10,255,385 rx/s                  0 err/s
   receive total        10,255,385 pkt/s        10,255,385 drop/s 
         0 error/s
     cpu:1              10,255,385 pkt/s        10,255,385 drop/s 
         0 error/s
   xdp_exception                 0 hit/s

> receiving side. I tried to receive on ixgbe but this ended with -ENOMEM
> | # xdp-bench drop eth1
> | Failed to attach XDP program: Cannot allocate memory
> 
> This is v6.8-rc5 on both sides. Let me see where this is coming from…
> 

Another pitfall with ixgbe is that it does a full link reset when
adding/removing XDP prog on device.  This can be annoying if connected
back-to-back, because "remote" pktgen will stop on link reset.

--Jesper

  reply	other threads:[~2024-02-20 10:43 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-02-13 14:58 [PATCH RFC net-next 0/2] Use per-task storage for XDP-redirects on PREEMPT_RT Sebastian Andrzej Siewior
2024-02-13 14:58 ` [PATCH RFC net-next 1/2] net: Reference bpf_redirect_info via task_struct " Sebastian Andrzej Siewior
2024-02-13 20:50   ` Jesper Dangaard Brouer
2024-02-14 12:19     ` Sebastian Andrzej Siewior
2024-02-14 13:23       ` Toke Høiland-Jørgensen
2024-02-14 14:28         ` Sebastian Andrzej Siewior
2024-02-14 16:08           ` Toke Høiland-Jørgensen
2024-02-14 16:36             ` Sebastian Andrzej Siewior
2024-02-15 20:23               ` Toke Høiland-Jørgensen
2024-02-16 16:57                 ` Sebastian Andrzej Siewior
2024-02-19 19:01                   ` Toke Høiland-Jørgensen
2024-02-20  9:17                     ` Jesper Dangaard Brouer
2024-02-20 10:17                       ` Sebastian Andrzej Siewior
2024-02-20 10:42                         ` Jesper Dangaard Brouer [this message]
2024-02-20 12:08                           ` Sebastian Andrzej Siewior
2024-02-20 12:57                             ` Jesper Dangaard Brouer
2024-02-20 15:32                               ` Sebastian Andrzej Siewior
2024-02-22  9:22                                 ` Sebastian Andrzej Siewior
2024-02-22 10:10                                   ` Jesper Dangaard Brouer
2024-02-22 10:58                                     ` Sebastian Andrzej Siewior
2024-02-20 12:10                           ` Dave Taht
2024-02-14 16:13   ` Toke Høiland-Jørgensen
2024-02-15  9:04     ` Sebastian Andrzej Siewior
2024-02-15 12:11       ` Toke Høiland-Jørgensen
2024-02-13 14:58 ` [PATCH RFC net-next 2/2] net: Move per-CPU flush-lists to bpf_xdp_storage " Sebastian Andrzej Siewior

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=0b1c8247-ccfb-4228-bd64-53583329aaa7@kernel.org \
    --to=hawk@kernel.org \
    --cc=bigeasy@linutronix.de \
    --cc=bpf@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=toke@redhat.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 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).