From: Stephen Hemminger <stephen@networkplumber.org>
To: dev@dpdk.org
Cc: Long Li <longli@microsoft.com>, Aman Singh <aman.deep.singh@intel.com>
Subject: Re: [PATCH v2] app/testpmd: add configurable flow count for txonly multi-flow
Date: Mon, 2 Mar 2026 12:00:43 -0800 [thread overview]
Message-ID: <20260302120043.7817639a@phoenix.local> (raw)
In-Reply-To: <20260227224405.61074-1-stephen@networkplumber.org>
On Fri, 27 Feb 2026 14:43:19 -0800
Stephen Hemminger <stephen@networkplumber.org> wrote:
> The txonly multi-flow mode generates 64 unique UDP source ports per
> lcore by cycling the high byte from 0xC0 to 0xFF. On SmartNICs with
> limited hardware flow table caching, this fixed count can exhaust the
> flow cache and degrade receive-side performance.
>
> Add --txonly-flows=N command line parameter and 'set txonly-flows'
> runtime command to limit the number of unique source ports per lcore
> to between 1 and 64. The default remains 64 to preserve existing
> behavior.
>
> The source port encoding is unchanged: the low byte carries the lcore
> ID (avoiding atomics) and the high byte cycles through N values
> starting at 0xC0. Total unique flows = txonly_nb_flows * active_lcores.
>
> Reported-by: Long Li <longli@microsoft.com>
> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
> Tested-by: Long Li <longli@microsoft.com>
Applied to next-net (after fixing spelling errors).
prev parent reply other threads:[~2026-03-02 20:00 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-02-26 19:16 [PATCH] app/testpmd: reduce txonly multi-flow src_port range Long Li
2026-02-26 21:35 ` Stephen Hemminger
2026-02-26 23:21 ` [RFC] app/testpmd: add configurable flow count for txonly multi-flow Stephen Hemminger
2026-02-26 23:51 ` [EXTERNAL] " Long Li
2026-02-27 22:43 ` [PATCH v2] " Stephen Hemminger
2026-03-02 20:00 ` Stephen Hemminger [this message]
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=20260302120043.7817639a@phoenix.local \
--to=stephen@networkplumber.org \
--cc=aman.deep.singh@intel.com \
--cc=dev@dpdk.org \
--cc=longli@microsoft.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