From: Aurelien Aptel <aaptel@nvidia.com>
To: Eric Dumazet <edumazet@google.com>
Cc: linux-nvme@lists.infradead.org, netdev@vger.kernel.org,
sagi@grimberg.me, hch@lst.de, kbusch@kernel.org, axboe@fb.com,
chaitanyak@nvidia.com, davem@davemloft.net, kuba@kernel.org,
Boris Pismenny <borisp@nvidia.com>,
aurelien.aptel@gmail.com, smalin@nvidia.com, malin1024@gmail.com,
ogerlitz@nvidia.com, yorayz@nvidia.com, galshalom@nvidia.com,
mgurtovoy@nvidia.com, tariqt@nvidia.com, gus@collabora.com,
pabeni@redhat.com, dsahern@kernel.org, ast@kernel.org,
jacob.e.keller@intel.com
Subject: Re: [PATCH v28 01/20] net: Introduce direct data placement tcp offload
Date: Thu, 05 Jun 2025 14:54:54 +0300 [thread overview]
Message-ID: <253v7paxv1t.fsf@nvidia.com> (raw)
In-Reply-To: <CANn89iJ6HROtg3m3z8Ac61e0Ex5HvgOTNavfG_W0j97B0XMZkw@mail.gmail.com>
Eric Dumazet <edumazet@google.com> writes:
> Adding one bit in all skbs for such a narrow case is not very convincing to me.
>
> I would prefer a disable/enable bit in the receiving socket, or a
> global static key.
We can move the bit to the socket, within the sock_read_rx cacheline group:
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -420,7 +420,7 @@ struct sock {
#endif
u8 sk_userlocks;
int sk_rcvbuf;
-
+ bool sk_no_condense;
struct sk_filter __rcu *sk_filter;
union {
struct socket_wq __rcu *sk_wq;
And then check for it in skb_condense().
We are still evaluating it but it looks fine for us.
Would that work for you?
Thanks
next prev parent reply other threads:[~2025-06-05 11:55 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-04-30 8:57 [PATCH v28 00/20] nvme-tcp receive offloads Aurelien Aptel
2025-04-30 8:57 ` [PATCH v28 01/20] net: Introduce direct data placement tcp offload Aurelien Aptel
2025-05-14 7:12 ` Eric Dumazet
2025-05-16 14:47 ` Aurelien Aptel
2025-05-16 22:31 ` Jakub Kicinski
2025-05-17 7:38 ` Eric Dumazet
2025-05-22 15:01 ` Aurelien Aptel
2025-06-04 12:33 ` Aurelien Aptel
2025-06-04 12:55 ` Eric Dumazet
2025-06-05 11:54 ` Aurelien Aptel [this message]
2025-06-05 12:44 ` Eric Dumazet
2025-04-30 8:57 ` [PATCH v28 02/20] netlink: add new family to manage ULP_DDP enablement and stats Aurelien Aptel
2025-04-30 8:57 ` [PATCH v28 03/20] iov_iter: skip copy if src == dst for direct data placement Aurelien Aptel
2025-04-30 8:57 ` [PATCH v28 04/20] net/tls,core: export get_netdev_for_sock Aurelien Aptel
2025-04-30 8:57 ` [PATCH v28 05/20] nvme-tcp: Add DDP offload control path Aurelien Aptel
2025-04-30 8:57 ` [PATCH v28 06/20] nvme-tcp: Add DDP data-path Aurelien Aptel
2025-04-30 8:57 ` [PATCH v28 07/20] nvme-tcp: RX DDGST offload Aurelien Aptel
2025-04-30 8:57 ` [PATCH v28 08/20] nvme-tcp: Deal with netdevice DOWN events Aurelien Aptel
2025-04-30 8:57 ` [PATCH v28 09/20] Documentation: add ULP DDP offload documentation Aurelien Aptel
2025-04-30 8:57 ` [PATCH v28 10/20] net/mlx5e: Rename from tls to transport static params Aurelien Aptel
2025-04-30 8:57 ` [PATCH v28 11/20] net/mlx5e: Refactor ico sq polling to get budget Aurelien Aptel
2025-04-30 8:57 ` [PATCH v28 12/20] net/mlx5: Add NVMEoTCP caps, HW bits, 128B CQE and enumerations Aurelien Aptel
2025-04-30 8:57 ` [PATCH v28 13/20] net/mlx5e: NVMEoTCP, offload initialization Aurelien Aptel
2025-04-30 8:57 ` [PATCH v28 14/20] net/mlx5e: TCP flow steering for nvme-tcp acceleration Aurelien Aptel
2025-04-30 8:57 ` [PATCH v28 15/20] net/mlx5e: NVMEoTCP, use KLM UMRs for buffer registration Aurelien Aptel
2025-04-30 8:57 ` [PATCH v28 16/20] net/mlx5e: NVMEoTCP, queue init/teardown Aurelien Aptel
2025-04-30 8:57 ` [PATCH v28 17/20] net/mlx5e: NVMEoTCP, ddp setup and resync Aurelien Aptel
2025-04-30 8:57 ` [PATCH v28 18/20] net/mlx5e: NVMEoTCP, async ddp invalidation Aurelien Aptel
2025-04-30 8:57 ` [PATCH v28 19/20] net/mlx5e: NVMEoTCP, data-path for DDP+DDGST offload Aurelien Aptel
2025-04-30 8:57 ` [PATCH v28 20/20] net/mlx5e: NVMEoTCP, statistics Aurelien Aptel
2025-04-30 12:52 ` [PATCH v28 00/20] nvme-tcp receive offloads Gustavo Padovan
2025-05-05 20:43 ` Jakub Kicinski
2025-05-05 21:51 ` Keith Busch
2025-05-05 22:51 ` Jakub Kicinski
2025-05-13 12:56 ` Aurelien Aptel
2025-05-13 14:36 ` Jakub Kicinski
2025-05-06 13:34 ` Sagi Grimberg
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=253v7paxv1t.fsf@nvidia.com \
--to=aaptel@nvidia.com \
--cc=ast@kernel.org \
--cc=aurelien.aptel@gmail.com \
--cc=axboe@fb.com \
--cc=borisp@nvidia.com \
--cc=chaitanyak@nvidia.com \
--cc=davem@davemloft.net \
--cc=dsahern@kernel.org \
--cc=edumazet@google.com \
--cc=galshalom@nvidia.com \
--cc=gus@collabora.com \
--cc=hch@lst.de \
--cc=jacob.e.keller@intel.com \
--cc=kbusch@kernel.org \
--cc=kuba@kernel.org \
--cc=linux-nvme@lists.infradead.org \
--cc=malin1024@gmail.com \
--cc=mgurtovoy@nvidia.com \
--cc=netdev@vger.kernel.org \
--cc=ogerlitz@nvidia.com \
--cc=pabeni@redhat.com \
--cc=sagi@grimberg.me \
--cc=smalin@nvidia.com \
--cc=tariqt@nvidia.com \
--cc=yorayz@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).