From: Stephen Hemminger <stephen@networkplumber.org>
To: "Morten Brørup" <mb@smartsharesystems.com>
Cc: "Lukáš Šišmiš" <sismis@cesnet.cz>,
anatoly.burakov@intel.com, ian.stokes@intel.com, dev@dpdk.org,
bruce.richardson@intel.com
Subject: Re: [PATCH] net: increase the maximum of RX/TX descriptors
Date: Tue, 5 Nov 2024 07:55:14 -0800 [thread overview]
Message-ID: <20241105075514.45ebea5c@hermes.local> (raw)
In-Reply-To: <98CBD80474FA8B44BF855DF32C47DC35E9F876@smartserver.smartshare.dk>
On Tue, 5 Nov 2024 09:49:39 +0100
Morten Brørup <mb@smartsharesystems.com> wrote:
> >
> > I suspect AF_PACKET provides an intermediate step which can buffer more
> > or spread out the work.
>
> Agree. It's a Linux scheduling issue.
>
> With DPDK polling, there is no interrupt in the kernel scheduler.
> If the CPU core running the DPDK polling thread is running some other thread when the packets arrive on the hardware, the DPDK polling thread is NOT scheduled immediately, but has to wait for the kernel scheduler to switch to this thread instead of the other thread.
> Quite a lot of time can pass before this happens - the kernel scheduler does not know that the DPDK polling thread has urgent work pending.
> And the number of RX descriptors needs to be big enough to absorb all packets arriving during the scheduling delay.
> It is not well described how to *guarantee* that nothing but the DPDK polling thread runs on a dedicated CPU core.
That why any non-trivial DPDK application needs to run on isolated cpu's.
next prev parent reply other threads:[~2024-11-05 15:55 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-10-29 12:48 Lukas Sismis
2024-10-29 14:37 ` Morten Brørup
2024-10-30 13:58 ` Lukáš Šišmiš
2024-10-30 15:20 ` Stephen Hemminger
2024-10-30 15:40 ` Lukáš Šišmiš
2024-10-30 15:58 ` Bruce Richardson
2024-10-30 16:06 ` Stephen Hemminger
2024-11-05 8:49 ` Morten Brørup
2024-11-05 15:55 ` Stephen Hemminger [this message]
2024-11-05 16:50 ` Morten Brørup
2024-11-05 21:20 ` Lukáš Šišmiš
2024-10-30 15:06 ` [PATCH v2 1/2] net/ixgbe: " Lukas Sismis
2024-10-30 15:06 ` [PATCH v2 2/2] net/ice: " Lukas Sismis
2024-10-30 15:42 ` [PATCH v3 1/1] net/bonding: make bonding functions stable Lukas Sismis
2024-10-30 15:42 ` [PATCH v3 1/2] net/ixgbe: increase the maximum of RX/TX descriptors Lukas Sismis
2024-10-30 16:26 ` Morten Brørup
2024-11-01 11:16 ` Bruce Richardson
2024-10-30 15:42 ` [PATCH v3 2/2] net/ice: " Lukas Sismis
2024-10-30 16:26 ` Morten Brørup
2024-10-31 2:24 ` [PATCH v3 1/1] net/bonding: make bonding functions stable lihuisong (C)
2024-11-06 2:14 ` Ferruh Yigit
-- strict thread matches above, loose matches on Subject: below --
2024-10-29 12:46 [PATCH] net: increase the maximum of RX/TX descriptors Lukas Sismis
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=20241105075514.45ebea5c@hermes.local \
--to=stephen@networkplumber.org \
--cc=anatoly.burakov@intel.com \
--cc=bruce.richardson@intel.com \
--cc=dev@dpdk.org \
--cc=ian.stokes@intel.com \
--cc=mb@smartsharesystems.com \
--cc=sismis@cesnet.cz \
/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).