diff mbox series

[net-next,1/9] mptcp: add a new sysctl for make after break timeout

Message ID 20231023-send-net-next-20231023-2-v1-1-9dc60939d371@kernel.org (mailing list archive)
State Accepted
Commit d866ae9aaa4325f1097e8b7a50f202348ca89b87
Delegated to: Netdev Maintainers
Headers show
Series mptcp: Features and fixes for v6.7 | expand

Checks

Context Check Description
netdev/series_format success Posting correctly formatted
netdev/tree_selection success Clearly marked for net-next, async
netdev/fixes_present success Fixes tag not required for -next series
netdev/header_inline success No static functions without inline keyword in header files
netdev/build_32bit success Errors and warnings before: 1368 this patch: 1368
netdev/cc_maintainers warning 2 maintainers not CCed: linux-doc@vger.kernel.org corbet@lwn.net
netdev/build_clang success Errors and warnings before: 1389 this patch: 1389
netdev/verify_signedoff success Signed-off-by tag matches author and committer
netdev/deprecated_api success None detected
netdev/check_selftest success No net selftest shell script
netdev/verify_fixes success No Fixes tag
netdev/build_allmodconfig_warn success Errors and warnings before: 1393 this patch: 1393
netdev/checkpatch warning WARNING: line length of 85 exceeds 80 columns WARNING: line length of 88 exceeds 80 columns
netdev/kdoc success Errors and warnings before: 0 this patch: 0
netdev/source_inline success Was 0 now: 0

Commit Message

Mat Martineau Oct. 23, 2023, 8:44 p.m. UTC 
From: Paolo Abeni <pabeni@redhat.com>

The MPTCP protocol allows sockets with no alive subflows to stay
in ESTABLISHED status for and user-defined timeout, to allow for
later subflows creation.

Currently such timeout is constant - TCP_TIMEWAIT_LEN. Let the
user-space configure them via a newly added sysctl, to better cope
with busy servers and simplify (make them faster) the relevant
pktdrill tests.

Note that the new know does not apply to orphaned MPTCP socket
waiting for the data_fin handshake completion: they always wait
TCP_TIMEWAIT_LEN.

Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <martineau@kernel.org>
---
 Documentation/networking/mptcp-sysctl.rst | 11 +++++++++++
 net/mptcp/ctrl.c                          | 16 ++++++++++++++++
 net/mptcp/protocol.c                      |  6 +++---
 net/mptcp/protocol.h                      |  1 +
 4 files changed, 31 insertions(+), 3 deletions(-)

Comments

Jakub Kicinski Oct. 25, 2023, 12:54 a.m. UTC | #1 
On Mon, 23 Oct 2023 13:44:34 -0700 Mat Martineau wrote:
> +		.procname = "close_timeout",
> +		.maxlen = sizeof(unsigned int),
> +		.mode = 0644,
> +		.proc_handler = proc_dointvec_jiffies,

Silly question - proc_dointvec_jiffies() works fine for unsigned types?
Paolo Abeni Oct. 25, 2023, 1:48 p.m. UTC | #2 
On Tue, 2023-10-24 at 17:54 -0700, Jakub Kicinski wrote:
> On Mon, 23 Oct 2023 13:44:34 -0700 Mat Martineau wrote:
> > +		.procname = "close_timeout",
> > +		.maxlen = sizeof(unsigned int),
> > +		.mode = 0644,
> > +		.proc_handler = proc_dointvec_jiffies,
> 
> Silly question - proc_dointvec_jiffies() works fine for unsigned types?

AFAICS, yes: proc_dointvec_jiffies() internally uses
__do_proc_dointvec(), which parses the buffer to a 'long' local
variable and then do_proc_dointvec_jiffies_conv() converts it to
integer taking care of possible overflows.

Cheers,

Paolo
diff mbox series

Patch

diff --git a/Documentation/networking/mptcp-sysctl.rst b/Documentation/networking/mptcp-sysctl.rst
index 15f1919d640c..69975ce25a02 100644
--- a/Documentation/networking/mptcp-sysctl.rst
+++ b/Documentation/networking/mptcp-sysctl.rst
@@ -25,6 +25,17 @@  add_addr_timeout - INTEGER (seconds)
 
 	Default: 120
 
+close_timeout - INTEGER (seconds)
+	Set the make-after-break timeout: in absence of any close or
+	shutdown syscall, MPTCP sockets will maintain the status
+	unchanged for such time, after the last subflow removal, before
+	moving to TCP_CLOSE.
+
+	The default value matches TCP_TIMEWAIT_LEN. This is a per-namespace
+	sysctl.
+
+	Default: 60
+
 checksum_enabled - BOOLEAN
 	Control whether DSS checksum can be enabled.
 
diff --git a/net/mptcp/ctrl.c b/net/mptcp/ctrl.c
index e72b518c5d02..13fe0748dde8 100644
--- a/net/mptcp/ctrl.c
+++ b/net/mptcp/ctrl.c
@@ -27,6 +27,7 @@  struct mptcp_pernet {
 #endif
 
 	unsigned int add_addr_timeout;
+	unsigned int close_timeout;
 	unsigned int stale_loss_cnt;
 	u8 mptcp_enabled;
 	u8 checksum_enabled;
@@ -65,6 +66,13 @@  unsigned int mptcp_stale_loss_cnt(const struct net *net)
 	return mptcp_get_pernet(net)->stale_loss_cnt;
 }
 
+unsigned int mptcp_close_timeout(const struct sock *sk)
+{
+	if (sock_flag(sk, SOCK_DEAD))
+		return TCP_TIMEWAIT_LEN;
+	return mptcp_get_pernet(sock_net(sk))->close_timeout;
+}
+
 int mptcp_get_pm_type(const struct net *net)
 {
 	return mptcp_get_pernet(net)->pm_type;
@@ -79,6 +87,7 @@  static void mptcp_pernet_set_defaults(struct mptcp_pernet *pernet)
 {
 	pernet->mptcp_enabled = 1;
 	pernet->add_addr_timeout = TCP_RTO_MAX;
+	pernet->close_timeout = TCP_TIMEWAIT_LEN;
 	pernet->checksum_enabled = 0;
 	pernet->allow_join_initial_addr_port = 1;
 	pernet->stale_loss_cnt = 4;
@@ -141,6 +150,12 @@  static struct ctl_table mptcp_sysctl_table[] = {
 		.mode = 0644,
 		.proc_handler = proc_dostring,
 	},
+	{
+		.procname = "close_timeout",
+		.maxlen = sizeof(unsigned int),
+		.mode = 0644,
+		.proc_handler = proc_dointvec_jiffies,
+	},
 	{}
 };
 
@@ -163,6 +178,7 @@  static int mptcp_pernet_new_table(struct net *net, struct mptcp_pernet *pernet)
 	table[4].data = &pernet->stale_loss_cnt;
 	table[5].data = &pernet->pm_type;
 	table[6].data = &pernet->scheduler;
+	table[7].data = &pernet->close_timeout;
 
 	hdr = register_net_sysctl_sz(net, MPTCP_SYSCTL_PATH, table,
 				     ARRAY_SIZE(mptcp_sysctl_table));
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 886ab689a8ae..a21f8ed26343 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -2391,8 +2391,8 @@  static void __mptcp_close_ssk(struct sock *sk, struct sock *ssk,
 	if (msk->in_accept_queue && msk->first == ssk &&
 	    (sock_flag(sk, SOCK_DEAD) || sock_flag(ssk, SOCK_DEAD))) {
 		/* ensure later check in mptcp_worker() will dispose the msk */
-		mptcp_set_close_tout(sk, tcp_jiffies32 - (TCP_TIMEWAIT_LEN + 1));
 		sock_set_flag(sk, SOCK_DEAD);
+		mptcp_set_close_tout(sk, tcp_jiffies32 - (mptcp_close_timeout(sk) + 1));
 		lock_sock_nested(ssk, SINGLE_DEPTH_NESTING);
 		mptcp_subflow_drop_ctx(ssk);
 		goto out_release;
@@ -2516,7 +2516,7 @@  static bool mptcp_close_tout_expired(const struct sock *sk)
 		return false;
 
 	return time_after32(tcp_jiffies32,
-		  inet_csk(sk)->icsk_mtup.probe_timestamp + TCP_TIMEWAIT_LEN);
+		  inet_csk(sk)->icsk_mtup.probe_timestamp + mptcp_close_timeout(sk));
 }
 
 static void mptcp_check_fastclose(struct mptcp_sock *msk)
@@ -2659,7 +2659,7 @@  void mptcp_reset_tout_timer(struct mptcp_sock *msk, unsigned long fail_tout)
 		return;
 
 	close_timeout = inet_csk(sk)->icsk_mtup.probe_timestamp - tcp_jiffies32 + jiffies +
-			TCP_TIMEWAIT_LEN;
+			mptcp_close_timeout(sk);
 
 	/* the close timeout takes precedence on the fail one, and here at least one of
 	 * them is active
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 3612545fa62e..02556921bc6c 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -613,6 +613,7 @@  unsigned int mptcp_get_add_addr_timeout(const struct net *net);
 int mptcp_is_checksum_enabled(const struct net *net);
 int mptcp_allow_join_id0(const struct net *net);
 unsigned int mptcp_stale_loss_cnt(const struct net *net);
+unsigned int mptcp_close_timeout(const struct sock *sk);
 int mptcp_get_pm_type(const struct net *net);
 const char *mptcp_get_scheduler(const struct net *net);
 void mptcp_subflow_fully_established(struct mptcp_subflow_context *subflow,