From: "Morten Brørup" <mb@smartsharesystems.com>
To: "David Marchand" <david.marchand@redhat.com>, <dev@dpdk.org>
Cc: <thomas@monjalon.net>, <ferruh.yigit@amd.com>, <stable@dpdk.org>,
"Olivier Matz" <olivier.matz@6wind.com>,
"Jijiang Liu" <jijiang.liu@intel.com>,
"Andrew Rybchenko" <andrew.rybchenko@oktetlabs.ru>,
<konstantin.ananyev@huawei.com>,
"Ferruh Yigit" <ferruh.yigit@amd.com>,
"Kaiwen Deng" <kaiwenx.deng@intel.com>, <qiming.yang@intel.com>,
<yidingx.zhou@intel.com>,
"Aman Singh" <aman.deep.singh@intel.com>,
"Yuying Zhang" <yuying.zhang@intel.com>,
"Thomas Monjalon" <thomas@monjalon.net>,
"Jerin Jacob" <jerinj@marvell.com>
Subject: RE: [PATCH v2 3/8] mbuf: fix Tx checksum offload examples
Date: Fri, 5 Apr 2024 18:20:08 +0200 [thread overview]
Message-ID: <98CBD80474FA8B44BF855DF32C47DC35E9F36C@smartserver.smartshare.dk> (raw)
In-Reply-To: <20240405144604.906695-4-david.marchand@redhat.com>
> From: David Marchand [mailto:david.marchand@redhat.com]
> Sent: Friday, 5 April 2024 16.46
>
> Mandate use of rte_eth_tx_prepare() in the mbuf Tx checksum offload
> examples.
I strongly disagree with this change!
It will cause a huge performance degradation for shaping applications:
A packet will be processed and finalized at an output or forwarding pipeline stage, where some other fields might also be written, so zeroing e.g. the out_ip checksum at this stage has low cost (no new cache misses).
Then, the packet might be queued for QoS or similar.
If rte_eth_tx_prepare() must be called at the egress pipeline stage, it has to write to the packet and cause a cache miss per packet, instead of simply passing on the packet to the NIC hardware.
It must be possible to finalize the packet at the output/forwarding pipeline stage!
Also, how is rte_eth_tx_prepare() supposed to work for cloned packets egressing on different NIC hardware?
In theory, it might get even worse if we make this opaque instead of transparent and standardized:
One PMD might reset out_ip checksum to 0x0000, and another PMD might reset it to 0xFFFF.
I can only see one solution:
We need to standardize on common minimum requirements for how to prepare packets for each TX offload.
next prev parent reply other threads:[~2024-04-05 16:20 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20240405125039.897933-1-david.marchand@redhat.com>
2024-04-05 12:49 ` [PATCH " David Marchand
2024-04-05 12:49 ` [PATCH 4/8] app/testpmd: fix outer IP checksum offload David Marchand
2024-04-05 12:49 ` [PATCH 5/8] net: fix outer UDP checksum in Intel prepare helper David Marchand
2024-04-05 12:49 ` [PATCH 6/8] net/i40e: fix outer UDP checksum offload for X710 David Marchand
2024-04-05 12:49 ` [PATCH 7/8] net/iavf: remove outer UDP checksum offload for X710 VF David Marchand
[not found] ` <20240405144604.906695-1-david.marchand@redhat.com>
2024-04-05 14:45 ` [PATCH v2 3/8] mbuf: fix Tx checksum offload examples David Marchand
2024-04-05 16:20 ` Morten Brørup [this message]
2024-04-08 10:12 ` David Marchand
2024-04-09 13:38 ` Konstantin Ananyev
2024-04-09 14:44 ` Morten Brørup
2024-04-10 10:35 ` Konstantin Ananyev
2024-04-10 12:20 ` Morten Brørup
2024-04-12 12:46 ` Konstantin Ananyev
2024-04-12 14:44 ` Morten Brørup
2024-04-12 15:17 ` Konstantin Ananyev
2024-04-12 15:54 ` Morten Brørup
2024-04-16 9:16 ` Konstantin Ananyev
2024-04-16 11:36 ` Konstantin Ananyev
2024-04-15 15:07 ` Ferruh Yigit
2024-04-16 7:14 ` Morten Brørup
2024-04-16 9:26 ` Konstantin Ananyev
2024-04-05 14:45 ` [PATCH v2 4/8] app/testpmd: fix outer IP checksum offload David Marchand
2024-04-05 14:45 ` [PATCH v2 5/8] net: fix outer UDP checksum in Intel prepare helper David Marchand
2024-04-05 14:46 ` [PATCH v2 6/8] net/i40e: fix outer UDP checksum offload for X710 David Marchand
2024-04-05 14:46 ` [PATCH v2 7/8] net/iavf: remove outer UDP checksum offload for X710 VF David Marchand
[not found] ` <20240418082023.1767998-1-david.marchand@redhat.com>
2024-04-18 8:20 ` [PATCH v3 3/7] app/testpmd: fix outer IP checksum offload David Marchand
2024-04-18 8:20 ` [PATCH v3 4/7] net: fix outer UDP checksum in Intel prepare helper David Marchand
2024-04-18 8:20 ` [PATCH v3 5/7] net/i40e: fix outer UDP checksum offload for X710 David Marchand
2024-04-18 8:20 ` [PATCH v3 6/7] net/iavf: remove outer UDP checksum offload for X710 VF David Marchand
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=98CBD80474FA8B44BF855DF32C47DC35E9F36C@smartserver.smartshare.dk \
--to=mb@smartsharesystems.com \
--cc=aman.deep.singh@intel.com \
--cc=andrew.rybchenko@oktetlabs.ru \
--cc=david.marchand@redhat.com \
--cc=dev@dpdk.org \
--cc=ferruh.yigit@amd.com \
--cc=jerinj@marvell.com \
--cc=jijiang.liu@intel.com \
--cc=kaiwenx.deng@intel.com \
--cc=konstantin.ananyev@huawei.com \
--cc=olivier.matz@6wind.com \
--cc=qiming.yang@intel.com \
--cc=stable@dpdk.org \
--cc=thomas@monjalon.net \
--cc=yidingx.zhou@intel.com \
--cc=yuying.zhang@intel.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).