From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: stable@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
patches@lists.linux.dev, Paolo Abeni <pabeni@redhat.com>,
Mat Martineau <martineau@kernel.org>,
Matthieu Baerts <matthieu.baerts@tessares.net>,
Jakub Kicinski <kuba@kernel.org>
Subject: [PATCH 5.15 02/15] mptcp: consolidate fallback and non fallback state machine
Date: Mon, 3 Jul 2023 20:54:47 +0200 [thread overview]
Message-ID: <20230703184518.962554478@linuxfoundation.org> (raw)
In-Reply-To: <20230703184518.896751186@linuxfoundation.org>
From: Paolo Abeni <pabeni@redhat.com>
commit 81c1d029016001f994ce1c46849c5e9900d8eab8 upstream.
An orphaned msk releases the used resources via the worker,
when the latter first see the msk in CLOSED status.
If the msk status transitions to TCP_CLOSE in the release callback
invoked by the worker's final release_sock(), such instance of the
workqueue will not take any action.
Additionally the MPTCP code prevents scheduling the worker once the
socket reaches the CLOSE status: such msk resources will be leaked.
The only code path that can trigger the above scenario is the
__mptcp_check_send_data_fin() in fallback mode.
Address the issue removing the special handling of fallback socket
in __mptcp_check_send_data_fin(), consolidating the state machine
for fallback and non fallback socket.
Since non-fallback sockets do not send and do not receive data_fin,
the mptcp code can update the msk internal status to match the next
step in the SM every time data fin (ack) should be generated or
received.
As a consequence we can remove a bunch of checks for fallback from
the fastpath.
Fixes: 6e628cd3a8f7 ("mptcp: use mptcp release_cb for delayed tasks")
Cc: stable@vger.kernel.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
net/mptcp/protocol.c | 39 +++++++++++++++------------------------
net/mptcp/subflow.c | 17 ++++++++++-------
2 files changed, 25 insertions(+), 31 deletions(-)
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -51,7 +51,7 @@ enum {
static struct percpu_counter mptcp_sockets_allocated;
static void __mptcp_destroy_sock(struct sock *sk);
-static void __mptcp_check_send_data_fin(struct sock *sk);
+static void mptcp_check_send_data_fin(struct sock *sk);
DEFINE_PER_CPU(struct mptcp_delegated_action, mptcp_delegated_actions);
static struct net_device mptcp_napi_dev;
@@ -355,8 +355,7 @@ static bool mptcp_pending_data_fin_ack(s
{
struct mptcp_sock *msk = mptcp_sk(sk);
- return !__mptcp_check_fallback(msk) &&
- ((1 << sk->sk_state) &
+ return ((1 << sk->sk_state) &
(TCPF_FIN_WAIT1 | TCPF_CLOSING | TCPF_LAST_ACK)) &&
msk->write_seq == READ_ONCE(msk->snd_una);
}
@@ -509,9 +508,6 @@ static bool mptcp_check_data_fin(struct
u64 rcv_data_fin_seq;
bool ret = false;
- if (__mptcp_check_fallback(msk))
- return ret;
-
/* Need to ack a DATA_FIN received from a peer while this side
* of the connection is in ESTABLISHED, FIN_WAIT1, or FIN_WAIT2.
* msk->rcv_data_fin was set when parsing the incoming options
@@ -549,7 +545,8 @@ static bool mptcp_check_data_fin(struct
}
ret = true;
- mptcp_send_ack(msk);
+ if (!__mptcp_check_fallback(msk))
+ mptcp_send_ack(msk);
mptcp_close_wake_up(sk);
}
return ret;
@@ -1612,7 +1609,7 @@ out:
if (!mptcp_timer_pending(sk))
mptcp_reset_timer(sk);
if (copied)
- __mptcp_check_send_data_fin(sk);
+ mptcp_check_send_data_fin(sk);
}
static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk)
@@ -2451,7 +2448,6 @@ static void mptcp_worker(struct work_str
if (unlikely((1 << state) & (TCPF_CLOSE | TCPF_LISTEN)))
goto unlock;
- mptcp_check_data_fin_ack(sk);
mptcp_flush_join_list(msk);
mptcp_check_fastclose(msk);
@@ -2462,7 +2458,8 @@ static void mptcp_worker(struct work_str
if (test_and_clear_bit(MPTCP_WORK_EOF, &msk->flags))
mptcp_check_for_eof(msk);
- __mptcp_check_send_data_fin(sk);
+ mptcp_check_send_data_fin(sk);
+ mptcp_check_data_fin_ack(sk);
mptcp_check_data_fin(sk);
/* There is no point in keeping around an orphaned sk timedout or
@@ -2591,6 +2588,12 @@ void mptcp_subflow_shutdown(struct sock
pr_debug("Fallback");
ssk->sk_shutdown |= how;
tcp_shutdown(ssk, how);
+
+ /* simulate the data_fin ack reception to let the state
+ * machine move forward
+ */
+ WRITE_ONCE(mptcp_sk(sk)->snd_una, mptcp_sk(sk)->snd_nxt);
+ mptcp_schedule_work(sk);
} else {
pr_debug("Sending DATA_FIN on subflow %p", ssk);
tcp_send_ack(ssk);
@@ -2630,7 +2633,7 @@ static int mptcp_close_state(struct sock
return next & TCP_ACTION_FIN;
}
-static void __mptcp_check_send_data_fin(struct sock *sk)
+static void mptcp_check_send_data_fin(struct sock *sk)
{
struct mptcp_subflow_context *subflow;
struct mptcp_sock *msk = mptcp_sk(sk);
@@ -2648,18 +2651,6 @@ static void __mptcp_check_send_data_fin(
WRITE_ONCE(msk->snd_nxt, msk->write_seq);
- /* fallback socket will not get data_fin/ack, can move to the next
- * state now
- */
- if (__mptcp_check_fallback(msk)) {
- if ((1 << sk->sk_state) & (TCPF_CLOSING | TCPF_LAST_ACK)) {
- inet_sk_state_store(sk, TCP_CLOSE);
- mptcp_close_wake_up(sk);
- } else if (sk->sk_state == TCP_FIN_WAIT1) {
- inet_sk_state_store(sk, TCP_FIN_WAIT2);
- }
- }
-
mptcp_flush_join_list(msk);
mptcp_for_each_subflow(msk, subflow) {
struct sock *tcp_sk = mptcp_subflow_tcp_sock(subflow);
@@ -2680,7 +2671,7 @@ static void __mptcp_wr_shutdown(struct s
WRITE_ONCE(msk->write_seq, msk->write_seq + 1);
WRITE_ONCE(msk->snd_data_fin_enable, 1);
- __mptcp_check_send_data_fin(sk);
+ mptcp_check_send_data_fin(sk);
}
static void __mptcp_destroy_sock(struct sock *sk)
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -1653,14 +1653,16 @@ static void subflow_state_change(struct
{
struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
struct sock *parent = subflow->conn;
+ struct mptcp_sock *msk;
__subflow_state_change(sk);
+ msk = mptcp_sk(parent);
if (subflow_simultaneous_connect(sk)) {
mptcp_propagate_sndbuf(parent, sk);
mptcp_do_fallback(sk);
- mptcp_rcv_space_init(mptcp_sk(parent), sk);
- pr_fallback(mptcp_sk(parent));
+ mptcp_rcv_space_init(msk, sk);
+ pr_fallback(msk);
subflow->conn_finished = 1;
mptcp_set_connected(parent);
}
@@ -1676,11 +1678,12 @@ static void subflow_state_change(struct
subflow_sched_work_if_closed(mptcp_sk(parent), sk);
- if (__mptcp_check_fallback(mptcp_sk(parent)) &&
- !subflow->rx_eof && subflow_is_done(sk)) {
- subflow->rx_eof = 1;
- mptcp_subflow_eof(parent);
- }
+ /* when the fallback subflow closes the rx side, trigger a 'dummy'
+ * ingress data fin, so that the msk state will follow along
+ */
+ if (__mptcp_check_fallback(msk) && subflow_is_done(sk) && msk->first == sk &&
+ mptcp_update_rcv_data_fin(msk, READ_ONCE(msk->ack_seq), true))
+ mptcp_schedule_work(parent);
}
static int subflow_ulp_init(struct sock *sk)
next prev parent reply other threads:[~2023-07-03 18:57 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-07-03 18:54 [PATCH 5.15 00/15] 5.15.120-rc1 review Greg Kroah-Hartman
2023-07-03 18:54 ` [PATCH 5.15 01/15] mptcp: fix possible divide by zero in recvmsg() Greg Kroah-Hartman
2023-07-03 18:54 ` Greg Kroah-Hartman [this message]
2023-07-03 18:54 ` [PATCH 5.15 03/15] mm, hwpoison: try to recover from copy-on write faults Greg Kroah-Hartman
2023-07-03 18:54 ` [PATCH 5.15 04/15] mm, hwpoison: when copy-on-write hits poison, take page offline Greg Kroah-Hartman
2023-07-03 18:54 ` [PATCH 5.15 05/15] drm/amdgpu: Set vmbo destroy after pt bo is created Greg Kroah-Hartman
2023-07-03 18:54 ` [PATCH 5.15 06/15] x86/microcode/AMD: Load late on both threads too Greg Kroah-Hartman
2023-07-03 18:54 ` [PATCH 5.15 07/15] x86/smp: Use dedicated cache-line for mwait_play_dead() Greg Kroah-Hartman
2023-07-03 18:54 ` [PATCH 5.15 08/15] can: isotp: isotp_sendmsg(): fix return error fix on TX path Greg Kroah-Hartman
2023-07-03 18:54 ` [PATCH 5.15 09/15] bpf: ensure main program has an extable Greg Kroah-Hartman
2023-07-03 18:54 ` [PATCH 5.15 10/15] HID: wacom: Use ktime_t rather than int when dealing with timestamps Greg Kroah-Hartman
2023-07-03 18:54 ` [PATCH 5.15 11/15] HID: logitech-hidpp: add HIDPP_QUIRK_DELAYED_INIT for the T651 Greg Kroah-Hartman
2023-07-03 18:54 ` [PATCH 5.15 12/15] Revert "thermal/drivers/mediatek: Use devm_of_iomap to avoid resource leak in mtk_thermal_probe" Greg Kroah-Hartman
2023-07-03 18:54 ` [PATCH 5.15 13/15] perf symbols: Symbol lookup with kcore can fail if multiple segments match stext Greg Kroah-Hartman
2023-07-03 18:54 ` [PATCH 5.15 14/15] scripts/tags.sh: Resolve gtags empty index generation Greg Kroah-Hartman
2023-07-03 18:55 ` [PATCH 5.15 15/15] drm/amdgpu: Validate VM ioctl flags Greg Kroah-Hartman
2023-07-04 7:00 ` [PATCH 5.15 00/15] 5.15.120-rc1 review Naresh Kamboju
2023-07-04 7:16 ` Helge Deller
2023-07-04 7:31 ` Greg Kroah-Hartman
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=20230703184518.962554478@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=kuba@kernel.org \
--cc=martineau@kernel.org \
--cc=matthieu.baerts@tessares.net \
--cc=pabeni@redhat.com \
--cc=patches@lists.linux.dev \
--cc=stable@vger.kernel.org \
/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