From: David Howells <dhowells@redhat.com>
To: Steve French <sfrench@samba.org>
Cc: David Howells <dhowells@redhat.com>,
Paulo Alcantara <pc@manguebit.org>,
Shyam Prasad N <sprasad@microsoft.com>,
Tom Talpey <tom@talpey.com>, Stefan Metzmacher <metze@samba.org>,
Mina Almasry <almasrymina@google.com>,
linux-cifs@vger.kernel.org, linux-kernel@vger.kernel.org,
netfs@lists.linux.dev, linux-fsdevel@vger.kernel.org
Subject: [RFC PATCH 17/36] cifs: Don't use corking
Date: Tue, 19 May 2026 11:21:35 +0100 [thread overview]
Message-ID: <20260519102158.592165-18-dhowells@redhat.com> (raw)
In-Reply-To: <20260519102158.592165-1-dhowells@redhat.com>
We don't need to cork the TCP socket if we send each message in a single
sendmsg() and set MSG_EOR to mark the last packet of the message with a
don't-append flag.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Steve French <sfrench@samba.org>
cc: Paulo Alcantara <pc@manguebit.org>
cc: Shyam Prasad N <sprasad@microsoft.com>
cc: Tom Talpey <tom@talpey.com>
cc: linux-cifs@vger.kernel.org
cc: netfs@lists.linux.dev
cc: linux-fsdevel@vger.kernel.org
---
fs/smb/client/transport.c | 13 ++++---------
1 file changed, 4 insertions(+), 9 deletions(-)
diff --git a/fs/smb/client/transport.c b/fs/smb/client/transport.c
index 3ead26f76112..e43fce1099a9 100644
--- a/fs/smb/client/transport.c
+++ b/fs/smb/client/transport.c
@@ -271,9 +271,11 @@ static int smb_sendmsg(struct TCP_Server_Info *server, const struct iov_iter *it
struct msghdr msg = {
/*
* MSG_SPLICE_PAGES causes tcp_sendmsg() to splice in the pages
- * in the iterator rather than copying from them.
+ * in the iterator rather than copying from them; MSG_EOR
+ * indicates that the last TCP packet we create should be
+ * marked no-append with regards to the next sendmsg.
*/
- .msg_flags = MSG_NOSIGNAL | MSG_SPLICE_PAGES,
+ .msg_flags = MSG_NOSIGNAL | MSG_SPLICE_PAGES | MSG_EOR,
.msg_iter = *iter,
};
int retries = 0;
@@ -362,10 +364,6 @@ __smb_send_rqst(struct TCP_Server_Info *server, struct iov_iter *iter)
goto out;
}
- rc = 0;
- /* cork the socket */
- tcp_sock_set_cork(ssocket->sk, true);
-
/*
* We should not allow signals to interrupt the network send because
* any partial send will cause session reconnects thus increasing
@@ -381,9 +379,6 @@ __smb_send_rqst(struct TCP_Server_Info *server, struct iov_iter *iter)
sigprocmask(SIG_SETMASK, &oldmask, NULL);
- /* uncork it */
- tcp_sock_set_cork(ssocket->sk, false);
-
if (sent > 0) {
/*
* If signal is pending but we have already sent the whole
next prev parent reply other threads:[~2026-05-19 10:23 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20260519102158.592165-1-dhowells@redhat.com>
2026-05-19 10:21 ` [RFC PATCH 01/36] net: Perform special handling for a splice from a bvecq David Howells
2026-05-19 10:21 ` [RFC PATCH 02/36] netfs: Add a facility to splice TCP receive buffers into " David Howells
2026-05-19 10:21 ` [RFC PATCH 03/36] netfs: Add some TCP receive queue helpers David Howells
2026-05-19 10:21 ` [RFC PATCH 04/36] cifs, nls: Provide unicode size determination func David Howells
2026-05-19 10:21 ` [RFC PATCH 05/36] cifs: Introduce an ALIGN8() macro David Howells
2026-05-19 10:21 ` [RFC PATCH 06/36] cifs: Rename mid_q_entry to smb_message David Howells
2026-05-19 10:21 ` [RFC PATCH 07/36] cifs: Add "Has dynamic part" flag form SMB2/3 StructureSize LSB David Howells
2026-05-19 10:21 ` [RFC PATCH 09/36] cifs: Institute message managing struct David Howells
2026-05-19 10:21 ` [RFC PATCH 10/36] cifs: Split crypt_message() into encrypt and decrypt variants David Howells
2026-05-19 10:21 ` [RFC PATCH 11/36] cifs: Add new AEAD alloc and setup routines that draw from an iterator David Howells
2026-05-19 10:21 ` [RFC PATCH 12/36] cifs: [WIP] Rewrite base Rx to put data off the socket into a bvecq David Howells
2026-05-19 10:21 ` [RFC PATCH 13/36] cifs: Remove validate_t2() David Howells
2026-05-19 10:21 ` [RFC PATCH 14/36] cifs: Remove cifs_io_subrequest::got_bytes David Howells
2026-05-19 10:21 ` [RFC PATCH 15/36] cifs: Pass smb_message to cifs_verify_signature() David Howells
2026-05-19 10:21 ` [RFC PATCH 16/36] cifs: Rewrite base TCP transmission David Howells
2026-05-19 10:36 ` Stefan Metzmacher
2026-05-19 10:21 ` David Howells [this message]
2026-05-19 10:21 ` [RFC PATCH 20/36] cifs: Pass smb_message structs down into the transport layer David Howells
2026-05-19 10:21 ` [RFC PATCH 21/36] cifs: Add a tracepoint to trace the smb_message refcount David Howells
2026-05-19 10:21 ` [RFC PATCH 22/36] cifs: Trace smb1/2_copy_to_prepped_buffers() David Howells
2026-05-19 10:21 ` [RFC PATCH 23/36] cifs: Clean up mid->callback_data and kill off mid->creator David Howells
2026-05-19 10:21 ` [RFC PATCH 24/36] cifs: Add netmem allocation functions David Howells
2026-05-19 10:21 ` [RFC PATCH 25/36] cifs: Add more pieces to smb_message David Howells
2026-05-19 10:21 ` [RFC PATCH 26/36] cifs: Convert SMB2 Negotiate Protocol request David Howells
2026-05-19 10:21 ` [RFC PATCH 27/36] cifs: Convert SMB2 Session Setup request David Howells
2026-05-19 10:21 ` [RFC PATCH 28/36] cifs: Convert SMB2 Logoff request David Howells
2026-05-19 10:21 ` [RFC PATCH 29/36] cifs: Convert SMB2 Tree Connect request David Howells
2026-05-19 10:21 ` [RFC PATCH 30/36] cifs: Convert SMB2 Tree Disconnect request David Howells
2026-05-19 10:21 ` [RFC PATCH 31/36] cifs: Convert SMB2 Read request David Howells
2026-05-19 10:21 ` [RFC PATCH 32/36] cifs: Convert SMB2 Write request David Howells
2026-05-19 10:21 ` [RFC PATCH 33/36] cifs: [WIP] Don't copy new-style smb_messages to a set of pages David Howells
2026-05-19 10:21 ` [RFC PATCH 34/36] cifs: [WIP] Rearrange Create request subfuncs David Howells
2026-05-19 10:21 ` [RFC PATCH 35/36] cifs: [WIP] Convert SMB2 Posix Mkdir request David Howells
2026-05-19 10:21 ` [RFC PATCH 36/36] cifs: [WIP] Convert SMB2 Open request David Howells
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=20260519102158.592165-18-dhowells@redhat.com \
--to=dhowells@redhat.com \
--cc=almasrymina@google.com \
--cc=linux-cifs@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=metze@samba.org \
--cc=netfs@lists.linux.dev \
--cc=pc@manguebit.org \
--cc=sfrench@samba.org \
--cc=sprasad@microsoft.com \
--cc=tom@talpey.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