From: Jakub Kicinski <kuba@kernel.org>
To: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Cc: Maxim Mikityanskiy <maximmi@nvidia.com>,
bpf@vger.kernel.org, ast@kernel.org, daniel@iogearbox.net,
magnus.karlsson@intel.com, bjorn@kernel.org,
netdev@vger.kernel.org, brouer@redhat.com,
alexandr.lobakin@intel.com, Tariq Toukan <tariqt@nvidia.com>,
Saeed Mahameed <saeedm@nvidia.com>
Subject: Re: [PATCH bpf-next 00/10] xsk: stop softirq processing on full XSK Rx queue
Date: Mon, 11 Apr 2022 10:07:46 -0700 [thread overview]
Message-ID: <20220411100746.1231a5a6@kernel.org> (raw)
In-Reply-To: <YlRNPuHdN5RTZjDn@boxer>
On Mon, 11 Apr 2022 17:46:06 +0200 Maciej Fijalkowski wrote:
> > +1
> > cover letter refers to busy poll, but did that test enable prefer busy
> > poll w/ the timeout configured right? It seems like similar goal can
> > be achieved with just that.
>
> AF_XDP busy poll where app and driver runs on same core, without
> configuring gro_flush_timeout and napi_defer_hard_irqs does not bring much
> value, so all of the busy poll tests were done with:
>
> echo 2 | sudo tee /sys/class/net/ens4f1/napi_defer_hard_irqs
> echo 200000 | sudo tee /sys/class/net/ens4f1/gro_flush_timeout
>
> That said, performance can still suffer and packets would not make it up
> to user space even with timeout being configured in the case I'm trying to
> improve.
Does the system not break out of busy poll then? See if the NAPI
hrtimer fires or not.
It's a 2k queue IIRC, with timeout of 200us, so 10Mpps at least.
What rate is l2fwd sustaining without these patches?
You may have to either increase the timeout or increase the frequency
of repoll from user space (with a smaller budget).
next prev parent reply other threads:[~2022-04-11 17:07 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-04-05 11:06 [PATCH bpf-next 00/10] xsk: stop softirq processing on full XSK Rx queue Maciej Fijalkowski
2022-04-05 11:06 ` [PATCH bpf-next 01/10] xsk: improve xdp_do_redirect() error codes Maciej Fijalkowski
2022-04-05 12:18 ` Jesper Dangaard Brouer
2022-04-05 11:06 ` [PATCH bpf-next 02/10] xsk: diversify return codes in xsk_rcv_check() Maciej Fijalkowski
2022-04-05 13:00 ` Jesper Dangaard Brouer
2022-04-05 13:35 ` Maciej Fijalkowski
2022-04-05 11:06 ` [PATCH bpf-next 03/10] ice: xsk: terminate NAPI when XSK Rx queue gets full Maciej Fijalkowski
2022-04-05 11:34 ` Alexander Lobakin
2022-04-05 12:02 ` Maciej Fijalkowski
2022-04-05 11:06 ` [PATCH bpf-next 04/10] i40e: " Maciej Fijalkowski
2022-04-05 13:04 ` Jesper Dangaard Brouer
2022-04-06 16:04 ` Maciej Fijalkowski
2022-04-05 11:06 ` [PATCH bpf-next 05/10] ixgbe: " Maciej Fijalkowski
2022-04-05 12:36 ` Jesper Dangaard Brouer
2022-04-05 13:52 ` Maciej Fijalkowski
2022-04-05 11:06 ` [PATCH bpf-next 06/10] ice: xsk: diversify return values from xsk_wakeup call paths Maciej Fijalkowski
2022-04-05 11:06 ` [PATCH bpf-next 07/10] i40e: " Maciej Fijalkowski
2022-04-05 11:06 ` [PATCH bpf-next 08/10] ixgbe: " Maciej Fijalkowski
2022-04-05 11:06 ` [PATCH bpf-next 09/10] ice: xsk: avoid refilling single Rx descriptors Maciej Fijalkowski
2022-04-05 11:06 ` [PATCH bpf-next 10/10] xsk: drop ternary operator from xskq_cons_has_entries Maciej Fijalkowski
2022-04-07 10:49 ` [PATCH bpf-next 00/10] xsk: stop softirq processing on full XSK Rx queue Maxim Mikityanskiy
2022-04-08 9:08 ` Maciej Fijalkowski
2022-04-08 12:48 ` Maxim Mikityanskiy
2022-04-08 18:17 ` Jakub Kicinski
2022-04-11 15:46 ` Maciej Fijalkowski
2022-04-11 17:07 ` Jakub Kicinski [this message]
2022-04-11 15:35 ` Maciej Fijalkowski
2022-04-13 10:40 ` Maxim Mikityanskiy
2022-04-13 15:12 ` Magnus Karlsson
2022-04-13 15:26 ` Maciej Fijalkowski
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=20220411100746.1231a5a6@kernel.org \
--to=kuba@kernel.org \
--cc=alexandr.lobakin@intel.com \
--cc=ast@kernel.org \
--cc=bjorn@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=brouer@redhat.com \
--cc=daniel@iogearbox.net \
--cc=maciej.fijalkowski@intel.com \
--cc=magnus.karlsson@intel.com \
--cc=maximmi@nvidia.com \
--cc=netdev@vger.kernel.org \
--cc=saeedm@nvidia.com \
--cc=tariqt@nvidia.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).