mptcp: sched: split get_subflow interface into two

get_retrans() interface of the burst packet scheduler invokes a sleeping
function mptcp_pm_subflow_chk_stale(), which calls __lock_sock_fast().
So get_retrans() interface should be set with BPF_F_SLEEPABLE flag in
BPF. But get_send() interface of this scheduler can't be set with
BPF_F_SLEEPABLE flag since it's invoked in ack_update_msk() under mptcp
data lock.

So this patch has to split get_subflow() interface of packet scheduer into
two interfaces: get_send() and get_retrans(). Then we can set get_retrans()
interface alone with BPF_F_SLEEPABLE flag.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250221-net-next-mptcp-pm-misc-cleanup-3-v1-8-2b70ab1cee79@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

2 files changed, 27 insertions, 13 deletions

diff --git a/include/net/mptcp.h b/include/net/mptcp.h
index 814b5f2e3ed5..2c85ca92bb1c 100644
--- a/include/net/mptcp.h
+++ b/include/net/mptcp.h
@@ -103,13 +103,14 @@ struct mptcp_out_options {
 #define MPTCP_SUBFLOWS_MAX	8
 
 struct mptcp_sched_data {
-	bool	reinject;
 	u8	subflows;
 	struct mptcp_subflow_context *contexts[MPTCP_SUBFLOWS_MAX];
 };
 
 struct mptcp_sched_ops {
-	int (*get_subflow)(struct mptcp_sock *msk,
+	int (*get_send)(struct mptcp_sock *msk,
+			struct mptcp_sched_data *data);
+	int (*get_retrans)(struct mptcp_sock *msk,
 			   struct mptcp_sched_data *data);
 
 	char			name[MPTCP_SCHED_NAME_MAX];
diff --git a/net/mptcp/sched.c b/net/mptcp/sched.c
index df7dbcfa3b71..94dc4b3ad82f 100644
--- a/net/mptcp/sched.c
+++ b/net/mptcp/sched.c
@@ -16,13 +16,25 @@
 static DEFINE_SPINLOCK(mptcp_sched_list_lock);
 static LIST_HEAD(mptcp_sched_list);
 
-static int mptcp_sched_default_get_subflow(struct mptcp_sock *msk,
+static int mptcp_sched_default_get_send(struct mptcp_sock *msk,
+					struct mptcp_sched_data *data)
+{
+	struct sock *ssk;
+
+	ssk = mptcp_subflow_get_send(msk);
+	if (!ssk)
+		return -EINVAL;
+
+	mptcp_subflow_set_scheduled(mptcp_subflow_ctx(ssk), true);
+	return 0;
+}
+
+static int mptcp_sched_default_get_retrans(struct mptcp_sock *msk,
 					   struct mptcp_sched_data *data)
 {
 	struct sock *ssk;
 
-	ssk = data->reinject ? mptcp_subflow_get_retrans(msk) :
-			       mptcp_subflow_get_send(msk);
+	ssk = mptcp_subflow_get_retrans(msk);
 	if (!ssk)
 		return -EINVAL;
 
@@ -31,7 +43,8 @@ static int mptcp_sched_default_get_subflow(struct mptcp_sock *msk,
 }
 
 static struct mptcp_sched_ops mptcp_sched_default = {
-	.get_subflow	= mptcp_sched_default_get_subflow,
+	.get_send	= mptcp_sched_default_get_send,
+	.get_retrans	= mptcp_sched_default_get_retrans,
 	.name		= "default",
 	.owner		= THIS_MODULE,
 };
@@ -73,7 +86,7 @@ void mptcp_get_available_schedulers(char *buf, size_t maxlen)
 
 int mptcp_register_scheduler(struct mptcp_sched_ops *sched)
 {
-	if (!sched->get_subflow)
+	if (!sched->get_send)
 		return -EINVAL;
 
 	spin_lock(&mptcp_sched_list_lock);
@@ -164,10 +177,9 @@ int mptcp_sched_get_send(struct mptcp_sock *msk)
 			return 0;
 	}
 
-	data.reinject = false;
 	if (msk->sched == &mptcp_sched_default || !msk->sched)
-		return mptcp_sched_default_get_subflow(msk, &data);
-	return msk->sched->get_subflow(msk, &data);
+		return mptcp_sched_default_get_send(msk, &data);
+	return msk->sched->get_send(msk, &data);
 }
 
 int mptcp_sched_get_retrans(struct mptcp_sock *msk)
@@ -186,8 +198,9 @@ int mptcp_sched_get_retrans(struct mptcp_sock *msk)
 			return 0;
 	}
 
-	data.reinject = true;
 	if (msk->sched == &mptcp_sched_default || !msk->sched)
-		return mptcp_sched_default_get_subflow(msk, &data);
-	return msk->sched->get_subflow(msk, &data);
+		return mptcp_sched_default_get_retrans(msk, &data);
+	if (msk->sched->get_retrans)
+		return msk->sched->get_retrans(msk, &data);
+	return msk->sched->get_send(msk, &data);
 }