From: Hannes Reinecke <hare@suse.de>
To: Sagi Grimberg <sagi@grimberg.me>, Jakub Kicinski <kuba@kernel.org>
Cc: Christoph Hellwig <hch@lst.de>, Keith Busch <kbusch@kernel.org>,
linux-nvme@lists.infradead.org,
Chuck Lever <chuck.lever@oracle.com>,
kernel-tls-handshake@lists.linux.dev
Subject: Re: [PATCH 08/18] nvme-tcp: do not set MSG_SENDPAGE_NOTLAST
Date: Mon, 17 Apr 2023 17:35:46 +0200 [thread overview]
Message-ID: <cce7cbdb-0866-58ec-8bf1-d0eea7ce9fe1@suse.de> (raw)
In-Reply-To: <5df08f86-2599-6e47-2fdc-bda8d70edaf2@grimberg.me>
On 4/17/23 17:28, Sagi Grimberg wrote:
>
>>> nak from me.
>>>
>>>> MSG_SENDPAGE_NOTLAST was introduced with commit 35f9c09fe9c7
>>>> ("tcp: tcp_sendpages() should call tcp_push() once") to fix issues
>>>> with the ->sendfile() call, and the implication seems to be that
>>>> the flag should be strictly internal to sendfile.
>>>
>>> no, the use case was to allow sendpage consumers to build larger
>>> tso packets for large payloads (splice being the primary one), nvme-tcp
>>> also benefits from it (as do others), just like sendfile.
>>>
>>>> Hence we shouldn't be setting it on nvme-tcp, and anyway doing
>>>> so will cause TLS sending invalid data.
>>>
>>> That is something else. I don't recall the full details on the
>>> discussion with Jakub, but it was not my intention to strike
>>> it altogether, just remove it if !sendpage_ok.
>>>
>> But then we're running afoul with tls_sw_do_sendpage(), which uses
>> MSG_SENDPAGE_NOTLAST as a record indicator. Forwarding
>> MSG_SENDPAGE_NOTLAST to tls results in incorrect TLS records on the
>> wire and a transmission stall.
>>
>> Took me days to figure that out.
>>
>> If you have a better idea, I'm all ears.
>
> So sendfile is unsupported over kTLS? Or is there some special
> case for it in the upper layer?
>
Dunno. Seems to work, at least according to commit d452d48b9f8b
("tls: prevent oversized sendfile() hangs by ignoring MSG_MORE")
Which is also where I got the notion about sendfile() from.
If MSG_SENDPAGE_NOTLAST is valid for non-sendfile users then
tls_sw_do_sendpage() has a bug.
Cheers,
Hannes
--
Dr. Hannes Reinecke Kernel Storage Architect
hare@suse.de +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew
Myers, Andrew McDonald, Martje Boudien Moerman
next prev parent reply other threads:[~2023-04-17 15:35 UTC|newest]
Thread overview: 49+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-04-17 13:02 [PATCHv3 00/18] nvme: In-kernel TLS support for TCP Hannes Reinecke
2023-04-17 13:02 ` [PATCH 01/18] nvme-keyring: register '.nvme' keyring Hannes Reinecke
2023-04-17 13:02 ` [PATCH 02/18] nvme-keyring: define a 'psk' keytype Hannes Reinecke
2023-04-17 13:02 ` [PATCH 03/18] nvme: add TCP TSAS definitions Hannes Reinecke
2023-04-17 13:02 ` [PATCH 04/18] nvme-tcp: add definitions for TLS cipher suites Hannes Reinecke
2023-04-17 13:02 ` [PATCH 05/18] nvme-keyring: implement nvme_tls_psk_default() Hannes Reinecke
2023-04-17 13:02 ` [PATCH 06/18] net/tls: implement ->read_sock() Hannes Reinecke
2023-04-17 15:14 ` Sagi Grimberg
2023-04-17 13:02 ` [PATCH 07/18] net/tls: sanitize MSG_EOR handling Hannes Reinecke
2023-04-17 15:19 ` Sagi Grimberg
2023-04-17 20:10 ` Jakub Kicinski
2023-04-18 10:07 ` Sagi Grimberg
2023-04-18 10:24 ` Hannes Reinecke
2023-04-18 10:39 ` Hannes Reinecke
2023-04-18 10:43 ` Sagi Grimberg
2023-04-18 11:02 ` Hannes Reinecke
2023-04-18 18:28 ` Jakub Kicinski
2023-04-17 13:02 ` [PATCH 08/18] nvme-tcp: do not set MSG_SENDPAGE_NOTLAST Hannes Reinecke
2023-04-17 15:10 ` Sagi Grimberg
2023-04-17 15:25 ` Hannes Reinecke
2023-04-17 15:28 ` Sagi Grimberg
2023-04-17 15:35 ` Hannes Reinecke [this message]
2023-04-17 20:16 ` Jakub Kicinski
2023-04-18 10:33 ` Hannes Reinecke
2023-04-18 18:21 ` Jakub Kicinski
2023-04-18 18:25 ` Hannes Reinecke
2023-04-17 13:02 ` [PATCH 09/18] security/keys: export key_lookup() Hannes Reinecke
2023-04-17 13:02 ` [PATCH 10/18] nvme/tcp: allocate socket file Hannes Reinecke
2023-04-17 13:02 ` [PATCH 11/18] nvme-tcp: enable TLS handshake upcall Hannes Reinecke
2023-04-17 15:26 ` Sagi Grimberg
2023-04-17 15:28 ` Hannes Reinecke
2023-04-17 15:31 ` Sagi Grimberg
2023-04-17 15:36 ` Hannes Reinecke
2023-04-18 5:52 ` Daniel Wagner
2023-04-18 9:38 ` Hannes Reinecke
2023-04-18 10:12 ` Sagi Grimberg
2023-04-18 10:28 ` Hannes Reinecke
2023-04-18 10:32 ` Sagi Grimberg
2023-04-18 10:33 ` Hannes Reinecke
2023-04-17 13:02 ` [PATCH 12/18] nvme-tcp: control message handling for recvmsg() Hannes Reinecke
2023-04-17 15:24 ` Sagi Grimberg
2023-04-17 15:26 ` Hannes Reinecke
2023-04-17 13:02 ` [PATCH 13/18] nvme-fabrics: parse options 'keyring' and 'tls_key' Hannes Reinecke
2023-04-17 13:02 ` [PATCH 14/18] nvmet: make TCP sectype settable via configfs Hannes Reinecke
2023-04-17 13:02 ` [PATCH 15/18] nvmet-tcp: allocate socket file Hannes Reinecke
2023-04-17 15:29 ` Sagi Grimberg
2023-04-17 13:03 ` [PATCH 16/18] nvmet-tcp: enable TLS handshake upcall Hannes Reinecke
2023-04-17 13:03 ` [PATCH 17/18] nvmet-tcp: control messages for recvmsg() Hannes Reinecke
2023-04-17 13:03 ` [PATCH 18/18] nvmet-tcp: add configfs attribute 'param_keyring' Hannes Reinecke
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=cce7cbdb-0866-58ec-8bf1-d0eea7ce9fe1@suse.de \
--to=hare@suse.de \
--cc=chuck.lever@oracle.com \
--cc=hch@lst.de \
--cc=kbusch@kernel.org \
--cc=kernel-tls-handshake@lists.linux.dev \
--cc=kuba@kernel.org \
--cc=linux-nvme@lists.infradead.org \
--cc=sagi@grimberg.me \
/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).