[PATCH mptcp-next v2 00/16] register default scheduler

[PATCH mptcp-next v2 00/16] register default scheduler

[Geliang Tang]

v2:
- split "mptcp: add mptcp_sched_data argument" into 5 squash-to patches.
- split "mptcp: register default scheduler" into 3 patches.
- depends on "refactor push pending" v7

Register the burst scheduler as the default one.

Geliang Tang (16):
  Squash to "mptcp: add struct mptcp_sched_ops"
  Squash to "mptcp: add get_subflow wrappers" II
  Squash to "mptcp: simplify push_pending"
  Squash to "mptcp: multi subflows push_pending"
  Squash to "mptcp: simplify subflow_push_pending"
  Squash to "mptcp: multi subflows subflow_push_pending"
  Squash to "mptcp: multi subflows retrans support"
  mptcp: use snd_burst of data instead of msk
  mptcp: define default scheduler
  mptcp: register default scheduler
  mptcp: add mptcp_sched_data_set_contexts helper
  Squash to "bpf: Add bpf_mptcp_sched_ops"
  Squash to "bpf: Add bpf_mptcp_sched_kfunc_set"
  Squash to "selftests/bpf: Add bpf_first scheduler"
  Squash to "selftests/bpf: Add bpf_bkup scheduler"
  Squash to "selftests/bpf: Add bpf_rr scheduler"

 include/net/mptcp.h                           |  7 +-
 net/mptcp/bpf.c                               |  1 +
 net/mptcp/protocol.c                          | 33 ++++---
 net/mptcp/protocol.h                          | 13 ++-
 net/mptcp/sched.c                             | 92 ++++++++++---------
 tools/testing/selftests/bpf/bpf_tcp_helpers.h |  9 +-
 .../selftests/bpf/progs/mptcp_bpf_bkup.c      | 10 +-
 .../selftests/bpf/progs/mptcp_bpf_first.c     | 10 +-
 .../selftests/bpf/progs/mptcp_bpf_rr.c        | 10 +-
 9 files changed, 116 insertions(+), 69 deletions(-)

-- 
2.35.3

[PATCH mptcp-next v2 01/16] Squash to "mptcp: add struct mptcp_sched_ops"

[Geliang Tang]

New api:
 - add new data_init
 - add snd_burst in mptcp_sched_data
 - add an int return value for get_subflow

Signed-off-by: Geliang Tang <geliang.tang@suse.com>
---
 include/net/mptcp.h | 7 +++++--
 net/mptcp/sched.c   | 2 +-
 2 files changed, 6 insertions(+), 3 deletions(-)

diff --git a/include/net/mptcp.h b/include/net/mptcp.h
index c25939b2af68..cac5b0d512fd 100644
--- a/include/net/mptcp.h
+++ b/include/net/mptcp.h
@@ -101,12 +101,15 @@ struct mptcp_out_options {
 
 struct mptcp_sched_data {
 	bool	reinject;
+	int	snd_burst;
 	struct mptcp_subflow_context *contexts[MPTCP_SUBFLOWS_MAX];
 };
 
 struct mptcp_sched_ops {
-	void (*get_subflow)(const struct mptcp_sock *msk,
-			    struct mptcp_sched_data *data);
+	void (*data_init)(const struct mptcp_sock *msk,
+			  struct mptcp_sched_data *data);
+	int (*get_subflow)(const struct mptcp_sock *msk,
+			   struct mptcp_sched_data *data);
 
 	char			name[MPTCP_SCHED_NAME_MAX];
 	struct module		*owner;
diff --git a/net/mptcp/sched.c b/net/mptcp/sched.c
index 9b128714055a..6324a8a76382 100644
--- a/net/mptcp/sched.c
+++ b/net/mptcp/sched.c
@@ -33,7 +33,7 @@ struct mptcp_sched_ops *mptcp_sched_find(const char *name)
 
 int mptcp_register_scheduler(struct mptcp_sched_ops *sched)
 {
-	if (!sched->get_subflow)
+	if (!sched->data_init || !sched->get_subflow)
 		return -EINVAL;
 
 	spin_lock(&mptcp_sched_list_lock);
-- 
2.35.3

[PATCH mptcp-next v2 02/16] Squash to "mptcp: add get_subflow wrappers" II

[Geliang Tang]

Use new api, use the return value of get_subflow. Drop
mptcp_sched_data_init, use sched->data_init() instead.

And some cleanups.

Signed-off-by: Geliang Tang <geliang.tang@suse.com>
---
 net/mptcp/protocol.h |  6 ++++--
 net/mptcp/sched.c    | 47 ++++++++++----------------------------------
 2 files changed, 14 insertions(+), 39 deletions(-)

diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 68a2b22a47eb..ee9e7e55dfe0 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -640,8 +640,10 @@ void mptcp_subflow_set_scheduled(struct mptcp_subflow_context *subflow,
 				 bool scheduled);
 struct sock *mptcp_subflow_get_send(struct mptcp_sock *msk);
 struct sock *mptcp_subflow_get_retrans(struct mptcp_sock *msk);
-int mptcp_sched_get_send(struct mptcp_sock *msk);
-int mptcp_sched_get_retrans(struct mptcp_sock *msk);
+int mptcp_sched_get_send(struct mptcp_sock *msk,
+			 struct mptcp_sched_data *data);
+int mptcp_sched_get_retrans(struct mptcp_sock *msk,
+			    struct mptcp_sched_data *data);
 
 static inline bool __tcp_can_send(const struct sock *ssk)
 {
diff --git a/net/mptcp/sched.c b/net/mptcp/sched.c
index 6324a8a76382..c459b01f9a9d 100644
--- a/net/mptcp/sched.c
+++ b/net/mptcp/sched.c
@@ -94,34 +94,9 @@ void mptcp_subflow_set_scheduled(struct mptcp_subflow_context *subflow,
 	WRITE_ONCE(subflow->scheduled, scheduled);
 }
 
-static int mptcp_sched_data_init(struct mptcp_sock *msk, bool reinject,
-				 struct mptcp_sched_data *data)
+int mptcp_sched_get_send(struct mptcp_sock *msk,
+			 struct mptcp_sched_data *data)
 {
-	struct mptcp_subflow_context *subflow;
-	int i = 0;
-
-	data->reinject = reinject;
-
-	mptcp_for_each_subflow(msk, subflow) {
-		if (i == MPTCP_SUBFLOWS_MAX) {
-			pr_warn_once("too many subflows");
-			break;
-		}
-		mptcp_subflow_set_scheduled(subflow, false);
-		data->contexts[i++] = subflow;
-	}
-
-	for (; i < MPTCP_SUBFLOWS_MAX; i++)
-		data->contexts[i] = NULL;
-
-	msk->snd_burst = 0;
-
-	return 0;
-}
-
-int mptcp_sched_get_send(struct mptcp_sock *msk)
-{
-	struct mptcp_sched_data data;
 	struct sock *ssk = NULL;
 
 	sock_owned_by_me((const struct sock *)msk);
@@ -145,15 +120,14 @@ int mptcp_sched_get_send(struct mptcp_sock *msk)
 		return 0;
 	}
 
-	mptcp_sched_data_init(msk, false, &data);
-	msk->sched->get_subflow(msk, &data);
-
-	return 0;
+	data->reinject = false;
+	msk->sched->data_init(msk, data);
+	return msk->sched->get_subflow(msk, data);
 }
 
-int mptcp_sched_get_retrans(struct mptcp_sock *msk)
+int mptcp_sched_get_retrans(struct mptcp_sock *msk,
+			    struct mptcp_sched_data *data)
 {
-	struct mptcp_sched_data data;
 	struct sock *ssk = NULL;
 
 	sock_owned_by_me((const struct sock *)msk);
@@ -170,8 +144,7 @@ int mptcp_sched_get_retrans(struct mptcp_sock *msk)
 		return 0;
 	}
 
-	mptcp_sched_data_init(msk, true, &data);
-	msk->sched->get_subflow(msk, &data);
-
-	return 0;
+	data->reinject = true;
+	msk->sched->data_init(msk, data);
+	return msk->sched->get_subflow(msk, data);
 }
-- 
2.35.3

[PATCH mptcp-next v2 03/16] Squash to "mptcp: simplify push_pending"

[Geliang Tang]

Add a parameter data for __subflow_push_pending.

Signed-off-by: Geliang Tang <geliang.tang@suse.com>
---
 net/mptcp/protocol.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 5ddceeb1d334..2cd8e18d917c 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -1512,7 +1512,8 @@ void mptcp_check_and_set_pending(struct sock *sk)
 }
 
 static int __subflow_push_pending(struct sock *sk, struct sock *ssk,
-				  struct mptcp_sendmsg_info *info)
+				  struct mptcp_sendmsg_info *info,
+				  struct mptcp_sched_data *data)
 {
 	struct mptcp_sock *msk = mptcp_sk(sk);
 	struct mptcp_data_frag *dfrag;
@@ -1561,6 +1562,7 @@ static int __subflow_push_pending(struct sock *sk, struct sock *ssk,
 void __mptcp_push_pending(struct sock *sk, unsigned int flags)
 {
 	struct mptcp_sock *msk = mptcp_sk(sk);
+	struct mptcp_sched_data data;
 	int ret = 0;
 
 again:
@@ -1575,7 +1577,7 @@ void __mptcp_push_pending(struct sock *sk, unsigned int flags)
 				struct sock *ssk = mptcp_subflow_tcp_sock(subflow);
 
 				lock_sock(ssk);
-				ret = __subflow_push_pending(sk, ssk, &info);
+				ret = __subflow_push_pending(sk, ssk, &info, &data);
 				release_sock(ssk);
 
 				if (ret <= 0) {
-- 
2.35.3

[PATCH mptcp-next v2 04/16] Squash to "mptcp: multi subflows push_pending"

[Geliang Tang]

Pass data to mptcp_sched_get_send.

Signed-off-by: Geliang Tang <geliang.tang@suse.com>
---
 net/mptcp/protocol.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 2cd8e18d917c..694c07855be9 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -1566,7 +1566,7 @@ void __mptcp_push_pending(struct sock *sk, unsigned int flags)
 	int ret = 0;
 
 again:
-	while (mptcp_send_head(sk) && !mptcp_sched_get_send(msk)) {
+	while (mptcp_send_head(sk) && !mptcp_sched_get_send(msk, &data)) {
 		struct mptcp_subflow_context *subflow;
 		struct mptcp_sendmsg_info info = {
 			.flags = flags,
-- 
2.35.3

[PATCH mptcp-next v2 05/16] Squash to "mptcp: simplify subflow_push_pending"

[Geliang Tang]

Pass data to __subflow_push_pending.

Signed-off-by: Geliang Tang <geliang.tang@suse.com>
---
 net/mptcp/protocol.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 694c07855be9..0f062bf63447 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -1606,6 +1606,7 @@ static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk,
 	struct mptcp_sendmsg_info info = {
 		.data_lock_held = true,
 	};
+	struct mptcp_sched_data data;
 	int ret = 0;
 
 	info.flags = 0;
@@ -1615,7 +1616,7 @@ static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk,
 		 * spooling the first chunk of data
 		 */
 		if (first) {
-			ret = __subflow_push_pending(sk, ssk, &info);
+			ret = __subflow_push_pending(sk, ssk, &info, &data);
 			if (ret <= 0) {
 				if (ret == -EAGAIN)
 					goto again;
-- 
2.35.3

[PATCH mptcp-next v2 06/16] Squash to "mptcp: multi subflows subflow_push_pending"

[Geliang Tang]

Pass data to mptcp_sched_get_send and __subflow_push_pending.

Signed-off-by: Geliang Tang <geliang.tang@suse.com>
---
 net/mptcp/protocol.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 0f062bf63447..18e439e180dd 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -1626,7 +1626,7 @@ static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk,
 			continue;
 		}
 
-		if (mptcp_sched_get_send(msk))
+		if (mptcp_sched_get_send(msk, &data))
 			goto out;
 
 		mptcp_for_each_subflow(msk, subflow) {
@@ -1647,7 +1647,7 @@ static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk,
 					goto out;
 				}
 
-				ret = __subflow_push_pending(sk, ssk, &info);
+				ret = __subflow_push_pending(sk, ssk, &info, &data);
 				if (ret <= 0) {
 					if (ret == -EAGAIN)
 						goto again;
-- 
2.35.3

[PATCH mptcp-next v2 07/16] Squash to "mptcp: multi subflows retrans support"

[Geliang Tang]

Pass data to mptcp_sched_get_retrans.

Signed-off-by: Geliang Tang <geliang.tang@suse.com>
---
 net/mptcp/protocol.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 18e439e180dd..0d9fae5718cc 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -2489,6 +2489,7 @@ static void __mptcp_retrans(struct sock *sk)
 	struct mptcp_subflow_context *subflow;
 	struct mptcp_sendmsg_info info = {};
 	struct mptcp_data_frag *dfrag;
+	struct mptcp_sched_data data;
 	struct sock *ssk;
 	int ret, err;
 	u16 len = 0;
@@ -2496,7 +2497,7 @@ static void __mptcp_retrans(struct sock *sk)
 	mptcp_clean_una_wakeup(sk);
 
 	/* first check ssk: need to kick "stale" logic */
-	err = mptcp_sched_get_retrans(msk);
+	err = mptcp_sched_get_retrans(msk, &data);
 	dfrag = mptcp_rtx_head(sk);
 	if (!dfrag) {
 		if (mptcp_data_fin_enabled(msk)) {
-- 
2.35.3

[PATCH mptcp-next v2 08/16] mptcp: use snd_burst of data instead of msk

[Geliang Tang]

Use data->snd_burst instead of msk->snd_burst.

Signed-off-by: Geliang Tang <geliang.tang@suse.com>
---
 net/mptcp/protocol.c | 12 ++++++------
 net/mptcp/protocol.h |  4 ++--
 net/mptcp/sched.c    |  2 +-
 3 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 0d9fae5718cc..956c8ea0edcf 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -1406,7 +1406,8 @@ bool mptcp_subflow_active(struct mptcp_subflow_context *subflow)
  * returns the subflow that will transmit the next DSS
  * additionally updates the rtx timeout
  */
-struct sock *mptcp_subflow_get_send(struct mptcp_sock *msk)
+struct sock *mptcp_subflow_get_send(const struct mptcp_sock *msk,
+				    struct mptcp_sched_data *data)
 {
 	struct subflow_send_info send_info[SSK_MODE_MAX];
 	struct mptcp_subflow_context *subflow;
@@ -1476,7 +1477,7 @@ struct sock *mptcp_subflow_get_send(struct mptcp_sock *msk)
 	subflow->avg_pacing_rate = div_u64((u64)subflow->avg_pacing_rate * wmem +
 					   READ_ONCE(ssk->sk_pacing_rate) * burst,
 					   burst + wmem);
-	msk->snd_burst = burst;
+	data->snd_burst = burst;
 	return ssk;
 }
 
@@ -1488,8 +1489,6 @@ static void mptcp_update_post_push(struct mptcp_sock *msk,
 
 	dfrag->already_sent += sent;
 
-	msk->snd_burst -= sent;
-
 	snd_nxt_new += dfrag->already_sent;
 
 	/* snd_nxt_new can be smaller than snd_nxt in case mptcp
@@ -1535,12 +1534,13 @@ static int __subflow_push_pending(struct sock *sk, struct sock *ssk,
 			info->sent += ret;
 			copied += ret;
 			len -= ret;
+			data->snd_burst -= ret;
 
 			mptcp_update_post_push(msk, dfrag, ret);
 		}
 		WRITE_ONCE(msk->first_pending, mptcp_send_next(sk));
 
-		if (msk->snd_burst <= 0 ||
+		if (data->snd_burst <= 0 ||
 		    !sk_stream_memory_free(ssk) ||
 		    !mptcp_subflow_active(mptcp_subflow_ctx(ssk))) {
 			err = copied ? : -EAGAIN;
@@ -1616,6 +1616,7 @@ static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk,
 		 * spooling the first chunk of data
 		 */
 		if (first) {
+			data.snd_burst = 0;
 			ret = __subflow_push_pending(sk, ssk, &info, &data);
 			if (ret <= 0) {
 				if (ret == -EAGAIN)
@@ -2282,7 +2283,6 @@ bool __mptcp_retransmit_pending_data(struct sock *sk)
 	mptcp_data_unlock(sk);
 
 	msk->first_pending = rtx_head;
-	msk->snd_burst = 0;
 
 	/* be sure to clear the "sent status" on all re-injected fragments */
 	list_for_each_entry(cur, &msk->rtx_queue, list) {
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index ee9e7e55dfe0..3f97bacca67d 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -257,7 +257,6 @@ struct mptcp_sock {
 	atomic64_t	rcv_wnd_sent;
 	u64		rcv_data_fin_seq;
 	int		rmem_fwd_alloc;
-	int		snd_burst;
 	int		old_wspace;
 	u64		recovery_snd_nxt;	/* in recovery mode accept up to this seq;
 						 * recovery related fields are under data_lock
@@ -638,7 +637,8 @@ int mptcp_init_sched(struct mptcp_sock *msk,
 void mptcp_release_sched(struct mptcp_sock *msk);
 void mptcp_subflow_set_scheduled(struct mptcp_subflow_context *subflow,
 				 bool scheduled);
-struct sock *mptcp_subflow_get_send(struct mptcp_sock *msk);
+struct sock *mptcp_subflow_get_send(const struct mptcp_sock *msk,
+				    struct mptcp_sched_data *data);
 struct sock *mptcp_subflow_get_retrans(struct mptcp_sock *msk);
 int mptcp_sched_get_send(struct mptcp_sock *msk,
 			 struct mptcp_sched_data *data);
diff --git a/net/mptcp/sched.c b/net/mptcp/sched.c
index c459b01f9a9d..f9440b935c76 100644
--- a/net/mptcp/sched.c
+++ b/net/mptcp/sched.c
@@ -113,7 +113,7 @@ int mptcp_sched_get_send(struct mptcp_sock *msk,
 	}
 
 	if (!msk->sched) {
-		ssk = mptcp_subflow_get_send(msk);
+		ssk = mptcp_subflow_get_send(msk, data);
 		if (!ssk)
 			return -EINVAL;
 		mptcp_subflow_set_scheduled(mptcp_subflow_ctx(ssk), true);
-- 
2.35.3

[PATCH mptcp-next v2 09/16] mptcp: define default scheduler

[Geliang Tang]

This patch defines the default packet scheduler mptcp_sched_default.
Skip deleting this default scheduler in mptcp_unregister_scheduler().

Use data->snd_burst instead of msk->snd_burst.

Signed-off-by: Geliang Tang <geliang.tang@suse.com>
---
 net/mptcp/protocol.c |  2 +-
 net/mptcp/protocol.h |  2 +-
 net/mptcp/sched.c    | 30 ++++++++++++++++++++++++++++++
 3 files changed, 32 insertions(+), 2 deletions(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 956c8ea0edcf..3f0829ea2ad9 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -2211,7 +2211,7 @@ static void mptcp_timeout_timer(struct timer_list *t)
  *
  * A backup subflow is returned only if that is the only kind available.
  */
-struct sock *mptcp_subflow_get_retrans(struct mptcp_sock *msk)
+struct sock *mptcp_subflow_get_retrans(const struct mptcp_sock *msk)
 {
 	struct sock *backup = NULL, *pick = NULL;
 	struct mptcp_subflow_context *subflow;
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 3f97bacca67d..d3cc84b15929 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -639,7 +639,7 @@ void mptcp_subflow_set_scheduled(struct mptcp_subflow_context *subflow,
 				 bool scheduled);
 struct sock *mptcp_subflow_get_send(const struct mptcp_sock *msk,
 				    struct mptcp_sched_data *data);
-struct sock *mptcp_subflow_get_retrans(struct mptcp_sock *msk);
+struct sock *mptcp_subflow_get_retrans(const struct mptcp_sock *msk);
 int mptcp_sched_get_send(struct mptcp_sock *msk,
 			 struct mptcp_sched_data *data);
 int mptcp_sched_get_retrans(struct mptcp_sock *msk,
diff --git a/net/mptcp/sched.c b/net/mptcp/sched.c
index f9440b935c76..14b3efb347bd 100644
--- a/net/mptcp/sched.c
+++ b/net/mptcp/sched.c
@@ -16,6 +16,33 @@
 static DEFINE_SPINLOCK(mptcp_sched_list_lock);
 static LIST_HEAD(mptcp_sched_list);
 
+static void mptcp_sched_default_data_init(const struct mptcp_sock *msk,
+					  struct mptcp_sched_data *data)
+{
+	data->snd_burst = 0;
+}
+
+static int mptcp_sched_default_get_subflow(const 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, data);
+	if (!ssk)
+		return -EINVAL;
+
+	mptcp_subflow_set_scheduled(mptcp_subflow_ctx(ssk), true);
+	return 0;
+}
+
+static struct mptcp_sched_ops mptcp_sched_default = {
+	.data_init	= mptcp_sched_default_data_init,
+	.get_subflow    = mptcp_sched_default_get_subflow,
+	.name           = "default",
+	.owner          = THIS_MODULE,
+};
+
 /* Must be called with rcu read lock held */
 struct mptcp_sched_ops *mptcp_sched_find(const char *name)
 {
@@ -50,6 +77,9 @@ int mptcp_register_scheduler(struct mptcp_sched_ops *sched)
 
 void mptcp_unregister_scheduler(struct mptcp_sched_ops *sched)
 {
+	if (sched == &mptcp_sched_default)
+		return;
+
 	spin_lock(&mptcp_sched_list_lock);
 	list_del_rcu(&sched->list);
 	spin_unlock(&mptcp_sched_list_lock);
-- 
2.35.3

[PATCH mptcp-next v2 10/16] mptcp: register default scheduler

[Geliang Tang]

This patch registers mptcp_sched_default in mptcp_sched_init(), which is
invoked in mptcp_proto_init().

Set msk->sched to the default scheduler when the input parameter of
mptcp_init_sched() is NULL.

With these changes, msk->sched will no longer be NULL. Drop the code that
checks msk->sched to be NULL in mptcp_sched_get_send() and
mptcp_sched_get_retrans().

Signed-off-by: Geliang Tang <geliang.tang@suse.com>
---
 net/mptcp/protocol.c |  1 +
 net/mptcp/protocol.h |  1 +
 net/mptcp/sched.c    | 28 ++++++----------------------
 3 files changed, 8 insertions(+), 22 deletions(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 3f0829ea2ad9..13d65542a4bd 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -3890,6 +3890,7 @@ void __init mptcp_proto_init(void)
 
 	mptcp_subflow_init();
 	mptcp_pm_init();
+	mptcp_sched_init();
 	mptcp_token_init();
 
 	if (proto_register(&mptcp_prot, MPTCP_USE_SLAB) != 0)
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index d3cc84b15929..669584978213 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -632,6 +632,7 @@ void mptcp_info2sockaddr(const struct mptcp_addr_info *info,
 struct mptcp_sched_ops *mptcp_sched_find(const char *name);
 int mptcp_register_scheduler(struct mptcp_sched_ops *sched);
 void mptcp_unregister_scheduler(struct mptcp_sched_ops *sched);
+void mptcp_sched_init(void);
 int mptcp_init_sched(struct mptcp_sock *msk,
 		     struct mptcp_sched_ops *sched);
 void mptcp_release_sched(struct mptcp_sock *msk);
diff --git a/net/mptcp/sched.c b/net/mptcp/sched.c
index 14b3efb347bd..cc9865a79ffe 100644
--- a/net/mptcp/sched.c
+++ b/net/mptcp/sched.c
@@ -85,11 +85,16 @@ void mptcp_unregister_scheduler(struct mptcp_sched_ops *sched)
 	spin_unlock(&mptcp_sched_list_lock);
 }
 
+void mptcp_sched_init(void)
+{
+	mptcp_register_scheduler(&mptcp_sched_default);
+}
+
 int mptcp_init_sched(struct mptcp_sock *msk,
 		     struct mptcp_sched_ops *sched)
 {
 	if (!sched)
-		goto out;
+		sched = &mptcp_sched_default;
 
 	if (!bpf_try_module_get(sched, sched->owner))
 		return -EBUSY;
@@ -100,7 +105,6 @@ int mptcp_init_sched(struct mptcp_sock *msk,
 
 	pr_debug("sched=%s", msk->sched->name);
 
-out:
 	return 0;
 }
 
@@ -127,8 +131,6 @@ void mptcp_subflow_set_scheduled(struct mptcp_subflow_context *subflow,
 int mptcp_sched_get_send(struct mptcp_sock *msk,
 			 struct mptcp_sched_data *data)
 {
-	struct sock *ssk = NULL;
-
 	sock_owned_by_me((const struct sock *)msk);
 
 	/* the following check is moved out of mptcp_subflow_get_send */
@@ -142,14 +144,6 @@ int mptcp_sched_get_send(struct mptcp_sock *msk,
 		return -EINVAL;
 	}
 
-	if (!msk->sched) {
-		ssk = mptcp_subflow_get_send(msk, data);
-		if (!ssk)
-			return -EINVAL;
-		mptcp_subflow_set_scheduled(mptcp_subflow_ctx(ssk), true);
-		return 0;
-	}
-
 	data->reinject = false;
 	msk->sched->data_init(msk, data);
 	return msk->sched->get_subflow(msk, data);
@@ -158,22 +152,12 @@ int mptcp_sched_get_send(struct mptcp_sock *msk,
 int mptcp_sched_get_retrans(struct mptcp_sock *msk,
 			    struct mptcp_sched_data *data)
 {
-	struct sock *ssk = NULL;
-
 	sock_owned_by_me((const struct sock *)msk);
 
 	/* the following check is moved out of mptcp_subflow_get_retrans */
 	if (__mptcp_check_fallback(msk))
 		return -EINVAL;
 
-	if (!msk->sched) {
-		ssk = mptcp_subflow_get_retrans(msk);
-		if (!ssk)
-			return -EINVAL;
-		mptcp_subflow_set_scheduled(mptcp_subflow_ctx(ssk), true);
-		return 0;
-	}
-
 	data->reinject = true;
 	msk->sched->data_init(msk, data);
 	return msk->sched->get_subflow(msk, data);
-- 
2.35.3

[PATCH mptcp-next v2 11/16] mptcp: add mptcp_sched_data_set_contexts helper

[Geliang Tang]

Add a new helper mptcp_sched_data_set_contexts(), it will be used in
the BPF contexts.

Signed-off-by: Geliang Tang <geliang.tang@suse.com>
---
 net/mptcp/sched.c | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

diff --git a/net/mptcp/sched.c b/net/mptcp/sched.c
index cc9865a79ffe..99d7df8d9b58 100644
--- a/net/mptcp/sched.c
+++ b/net/mptcp/sched.c
@@ -128,6 +128,25 @@ void mptcp_subflow_set_scheduled(struct mptcp_subflow_context *subflow,
 	WRITE_ONCE(subflow->scheduled, scheduled);
 }
 
+void mptcp_sched_data_set_contexts(const struct mptcp_sock *msk,
+				   struct mptcp_sched_data *data)
+{
+	struct mptcp_subflow_context *subflow;
+	int i = 0;
+
+	mptcp_for_each_subflow(msk, subflow) {
+		if (i == MPTCP_SUBFLOWS_MAX) {
+			pr_warn_once("too many subflows");
+			break;
+		}
+		mptcp_subflow_set_scheduled(subflow, false);
+		data->contexts[i++] = subflow;
+	}
+
+	for (; i < MPTCP_SUBFLOWS_MAX; i++)
+		data->contexts[i] = NULL;
+}
+
 int mptcp_sched_get_send(struct mptcp_sock *msk,
 			 struct mptcp_sched_data *data)
 {
-- 
2.35.3

[PATCH mptcp-next v2 12/16] Squash to "bpf: Add bpf_mptcp_sched_ops"

[Geliang Tang]

Use new API.

Signed-off-by: Geliang Tang <geliang.tang@suse.com>
---
 tools/testing/selftests/bpf/bpf_tcp_helpers.h | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/bpf_tcp_helpers.h b/tools/testing/selftests/bpf/bpf_tcp_helpers.h
index c8792e6f125a..ee64b29bf9bb 100644
--- a/tools/testing/selftests/bpf/bpf_tcp_helpers.h
+++ b/tools/testing/selftests/bpf/bpf_tcp_helpers.h
@@ -240,6 +240,7 @@ struct mptcp_subflow_context {
 
 struct mptcp_sched_data {
 	bool	reinject;
+	int	snd_burst;
 	struct mptcp_subflow_context *contexts[MPTCP_SUBFLOWS_MAX];
 } __attribute__((preserve_access_index));
 
@@ -249,8 +250,10 @@ struct mptcp_sched_ops {
 	void (*init)(const struct mptcp_sock *msk);
 	void (*release)(const struct mptcp_sock *msk);
 
-	void (*get_subflow)(const struct mptcp_sock *msk,
-			    struct mptcp_sched_data *data);
+	void (*data_init)(const struct mptcp_sock *msk,
+			  struct mptcp_sched_data *data);
+	int (*get_subflow)(const struct mptcp_sock *msk,
+			   struct mptcp_sched_data *data);
 	void *owner;
 };
 
@@ -264,5 +267,7 @@ struct mptcp_sock {
 
 extern void mptcp_subflow_set_scheduled(struct mptcp_subflow_context *subflow,
 					bool scheduled) __ksym;
+extern void mptcp_sched_data_set_contexts(const struct mptcp_sock *msk,
+					  struct mptcp_sched_data *data) __ksym;
 
 #endif
-- 
2.35.3

[PATCH mptcp-next v2 13/16] Squash to "bpf: Add bpf_mptcp_sched_kfunc_set"

[Geliang Tang]

Add mptcp_sched_data_set_contexts in kfunc_set.

Signed-off-by: Geliang Tang <geliang.tang@suse.com>
---
 net/mptcp/bpf.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/net/mptcp/bpf.c b/net/mptcp/bpf.c
index 0a768898990f..03decb05755f 100644
--- a/net/mptcp/bpf.c
+++ b/net/mptcp/bpf.c
@@ -164,6 +164,7 @@ struct bpf_struct_ops bpf_mptcp_sched_ops = {
 
 BTF_SET8_START(bpf_mptcp_sched_kfunc_ids)
 BTF_ID_FLAGS(func, mptcp_subflow_set_scheduled)
+BTF_ID_FLAGS(func, mptcp_sched_data_set_contexts)
 BTF_SET8_END(bpf_mptcp_sched_kfunc_ids)
 
 static const struct btf_kfunc_id_set bpf_mptcp_sched_kfunc_set = {
-- 
2.35.3

[PATCH mptcp-next v2 14/16] Squash to "selftests/bpf: Add bpf_first scheduler"

[Geliang Tang]

Use new API.

Signed-off-by: Geliang Tang <geliang.tang@suse.com>
---
 tools/testing/selftests/bpf/progs/mptcp_bpf_first.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/progs/mptcp_bpf_first.c b/tools/testing/selftests/bpf/progs/mptcp_bpf_first.c
index fcd733e88b02..e4caa2dd8c6f 100644
--- a/tools/testing/selftests/bpf/progs/mptcp_bpf_first.c
+++ b/tools/testing/selftests/bpf/progs/mptcp_bpf_first.c
@@ -16,16 +16,24 @@ void BPF_PROG(mptcp_sched_first_release, const struct mptcp_sock *msk)
 {
 }
 
-void BPF_STRUCT_OPS(bpf_first_get_subflow, const struct mptcp_sock *msk,
+void BPF_STRUCT_OPS(bpf_first_data_init, const struct mptcp_sock *msk,
 		    struct mptcp_sched_data *data)
+{
+	mptcp_sched_data_set_contexts(msk, data);
+}
+
+int BPF_STRUCT_OPS(bpf_first_get_subflow, const struct mptcp_sock *msk,
+		   struct mptcp_sched_data *data)
 {
 	mptcp_subflow_set_scheduled(data->contexts[0], true);
+	return 0;
 }
 
 SEC(".struct_ops")
 struct mptcp_sched_ops first = {
 	.init		= (void *)mptcp_sched_first_init,
 	.release	= (void *)mptcp_sched_first_release,
+	.data_init	= (void *)bpf_first_data_init,
 	.get_subflow	= (void *)bpf_first_get_subflow,
 	.name		= "bpf_first",
 };
-- 
2.35.3

[PATCH mptcp-next v2 15/16] Squash to "selftests/bpf: Add bpf_bkup scheduler"

[Geliang Tang]

Use new API.

Signed-off-by: Geliang Tang <geliang.tang@suse.com>
---
 tools/testing/selftests/bpf/progs/mptcp_bpf_bkup.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/progs/mptcp_bpf_bkup.c b/tools/testing/selftests/bpf/progs/mptcp_bpf_bkup.c
index 949e053e980c..b2724426676e 100644
--- a/tools/testing/selftests/bpf/progs/mptcp_bpf_bkup.c
+++ b/tools/testing/selftests/bpf/progs/mptcp_bpf_bkup.c
@@ -16,8 +16,14 @@ void BPF_PROG(mptcp_sched_bkup_release, const struct mptcp_sock *msk)
 {
 }
 
-void BPF_STRUCT_OPS(bpf_bkup_get_subflow, const struct mptcp_sock *msk,
+void BPF_STRUCT_OPS(bpf_bkup_data_init, const struct mptcp_sock *msk,
 		    struct mptcp_sched_data *data)
+{
+	mptcp_sched_data_set_contexts(msk, data);
+}
+
+int BPF_STRUCT_OPS(bpf_bkup_get_subflow, const struct mptcp_sock *msk,
+		   struct mptcp_sched_data *data)
 {
 	int nr = 0;
 
@@ -32,12 +38,14 @@ void BPF_STRUCT_OPS(bpf_bkup_get_subflow, const struct mptcp_sock *msk,
 	}
 
 	mptcp_subflow_set_scheduled(data->contexts[nr], true);
+	return 0;
 }
 
 SEC(".struct_ops")
 struct mptcp_sched_ops bkup = {
 	.init		= (void *)mptcp_sched_bkup_init,
 	.release	= (void *)mptcp_sched_bkup_release,
+	.data_init	= (void *)bpf_bkup_data_init,
 	.get_subflow	= (void *)bpf_bkup_get_subflow,
 	.name		= "bpf_bkup",
 };
-- 
2.35.3

[PATCH mptcp-next v2 16/16] Squash to "selftests/bpf: Add bpf_rr scheduler"

[Geliang Tang]

Use new API.

Signed-off-by: Geliang Tang <geliang.tang@suse.com>
---
 tools/testing/selftests/bpf/progs/mptcp_bpf_rr.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/progs/mptcp_bpf_rr.c b/tools/testing/selftests/bpf/progs/mptcp_bpf_rr.c
index b7156f6aae8b..2d27523b5a7c 100644
--- a/tools/testing/selftests/bpf/progs/mptcp_bpf_rr.c
+++ b/tools/testing/selftests/bpf/progs/mptcp_bpf_rr.c
@@ -17,8 +17,14 @@ void BPF_PROG(mptcp_sched_rr_release, const struct mptcp_sock *msk)
 {
 }
 
-void BPF_STRUCT_OPS(bpf_rr_get_subflow, const struct mptcp_sock *msk,
+void BPF_STRUCT_OPS(bpf_rr_data_init, const struct mptcp_sock *msk,
 		    struct mptcp_sched_data *data)
+{
+	mptcp_sched_data_set_contexts(msk, data);
+}
+
+int BPF_STRUCT_OPS(bpf_rr_get_subflow, const struct mptcp_sock *msk,
+		   struct mptcp_sched_data *data)
 {
 	int nr = 0;
 
@@ -37,12 +43,14 @@ void BPF_STRUCT_OPS(bpf_rr_get_subflow, const struct mptcp_sock *msk,
 
 	mptcp_subflow_set_scheduled(data->contexts[nr], true);
 	last_snd = data->contexts[nr]->tcp_sock;
+	return 0;
 }
 
 SEC(".struct_ops")
 struct mptcp_sched_ops rr = {
 	.init		= (void *)mptcp_sched_rr_init,
 	.release	= (void *)mptcp_sched_rr_release,
+	.data_init	= (void *)bpf_rr_data_init,
 	.get_subflow	= (void *)bpf_rr_get_subflow,
 	.name		= "bpf_rr",
 };
-- 
2.35.3

Re: [PATCH mptcp-next v2 00/16] register default scheduler

[Mat Martineau]

On Wed, 12 Oct 2022, Geliang Tang wrote:

> v2:
> - split "mptcp: add mptcp_sched_data argument" into 5 squash-to patches.
> - split "mptcp: register default scheduler" into 3 patches.
> - depends on "refactor push pending" v7
>

Thanks for rebasing this Geliang.

In the review for the "refactor push pending" series, I proposed that the 
series focus on the single-subflow refactoring for now. Does it make sense 
to rearrange these two patch series in to four different series like this:

1. refactor push pending (single subflow)
2. register default scheduler (single subflow)
3. Add multi-subflow scheduling
4. Update BPF scheduler for multi-subflow

I'm trying to think of a way to focus on *one patch set at a time* so we 
can make progress more easily. The large patch sets with lots of squash-to 
patches are getting hard to keep track of.

Thanks,

Mat

> Register the burst scheduler as the default one.
>
> Geliang Tang (16):
>  Squash to "mptcp: add struct mptcp_sched_ops"
>  Squash to "mptcp: add get_subflow wrappers" II
>  Squash to "mptcp: simplify push_pending"
>  Squash to "mptcp: multi subflows push_pending"
>  Squash to "mptcp: simplify subflow_push_pending"
>  Squash to "mptcp: multi subflows subflow_push_pending"
>  Squash to "mptcp: multi subflows retrans support"
>  mptcp: use snd_burst of data instead of msk
>  mptcp: define default scheduler
>  mptcp: register default scheduler
>  mptcp: add mptcp_sched_data_set_contexts helper
>  Squash to "bpf: Add bpf_mptcp_sched_ops"
>  Squash to "bpf: Add bpf_mptcp_sched_kfunc_set"
>  Squash to "selftests/bpf: Add bpf_first scheduler"
>  Squash to "selftests/bpf: Add bpf_bkup scheduler"
>  Squash to "selftests/bpf: Add bpf_rr scheduler"
>
> include/net/mptcp.h                           |  7 +-
> net/mptcp/bpf.c                               |  1 +
> net/mptcp/protocol.c                          | 33 ++++---
> net/mptcp/protocol.h                          | 13 ++-
> net/mptcp/sched.c                             | 92 ++++++++++---------
> tools/testing/selftests/bpf/bpf_tcp_helpers.h |  9 +-
> .../selftests/bpf/progs/mptcp_bpf_bkup.c      | 10 +-
> .../selftests/bpf/progs/mptcp_bpf_first.c     | 10 +-
> .../selftests/bpf/progs/mptcp_bpf_rr.c        | 10 +-
> 9 files changed, 116 insertions(+), 69 deletions(-)
>
> -- 
> 2.35.3
>
>
>

--
Mat Martineau
Intel