sctp: add fair capacity stream scheduler

As it says in rfc8260#section-3.5 about the fair capacity scheduler:

A fair capacity distribution between the streams is used. This
scheduler considers the lengths of the messages of each stream and
schedules them in a specific way to maintain an equal capacity for
all streams. The details are implementation dependent. interleaving
user messages allows for a better realization of the fair capacity
usage.

This patch adds Fair Capacity Scheduler based on the foundations added
by commit 5bbbbe32a431 ("sctp: introduce stream scheduler foundations"):

A fc_list and a fc_length are added into struct sctp_stream_out_ext and
a fc_list is added into struct sctp_stream. In .enqueue, when there are
chunks enqueued into a stream, this stream will be linked into stream->
fc_list by its fc_list ordered by its fc_length. In .dequeue, it always
picks up the 1st skb from stream->fc_list. In .dequeue_done, fc_length
is increased by chunk's len and update its location in stream->fc_list
according to the its new fc_length.

Note that when the new fc_length overflows in .dequeue_done, instead of
resetting all fc_lengths to 0, we only reduced them by U32_MAX / 4 to
avoid a moment of imbalance in the scheduling, as Marcelo suggested.

Signed-off-by: Xin Long <[e-mail]>
Acked-by: Marcelo Ricardo Leitner <[e-mail]>
Signed-off-by: Paolo Abeni <[e-mail]>

include/net/sctp/stream_sched.h

···
 			     struct sctp_sched_ops *sched_ops);
 void sctp_sched_ops_prio_init(void);
 void sctp_sched_ops_rr_init(void);
+void sctp_sched_ops_fc_init(void);
 
 #endif /* __sctp_stream_sched_h__ */

include/net/sctp/structs.h

···
 		struct {
 			struct list_head rr_list;
 		};
+		struct {
+			struct list_head fc_list;
+			__u32 fc_length;
+		};
 	};
 };
 
···
 			struct list_head rr_list;
 			/* The next stream in line */
 			struct sctp_stream_out_ext *rr_next;
+		};
+		struct {
+			struct list_head fc_list;
 		};
 	};
 	struct sctp_stream_interleave *si;

include/uapi/linux/sctp.h

···
 	SCTP_SS_DEFAULT = SCTP_SS_FCFS,
 	SCTP_SS_PRIO,
 	SCTP_SS_RR,
-	SCTP_SS_MAX = SCTP_SS_RR
+	SCTP_SS_FC,
+	SCTP_SS_MAX = SCTP_SS_FC
 };
 
 /* Probe Interval socket option */

net/sctp/Makefile

···
 	  tsnmap.o bind_addr.o socket.o primitive.o \
 	  output.o input.o debug.o stream.o auth.o \
 	  offload.o stream_sched.o stream_sched_prio.o \
-	  stream_sched_rr.o stream_interleave.o
+	  stream_sched_rr.o stream_sched_fc.o \
+	  stream_interleave.o
 
 sctp_diag-y := diag.o
 

net/sctp/stream_sched.c

···
 	sctp_sched_ops_fcfs_init();
 	sctp_sched_ops_prio_init();
 	sctp_sched_ops_rr_init();
+	sctp_sched_ops_fc_init();
 }
 
 static void sctp_sched_free_sched(struct sctp_stream *stream)

net/sctp/stream_sched_fc.c

···
+// SPDX-License-Identifier: GPL-2.0-or-later
+/* SCTP kernel implementation
+ * (C) Copyright Red Hat Inc. 2022
+ *
+ * This file is part of the SCTP kernel implementation
+ *
+ * These functions manipulate sctp stream queue/scheduling.
+ *
+ * Please send any bug reports or fixes you make to the
+ * email addresched(es):
+ *    lksctp developers <[e-mail]>
+ *
+ * Written or modified by:
+ *    Xin Long <[e-mail]>
+ */
+
+#include <linux/list.h>
+#include <net/sctp/sctp.h>
+#include <net/sctp/sm.h>
+#include <net/sctp/stream_sched.h>
+
+/* Fair Capacity handling
+ * RFC 8260 section 3.5
+ */
+static void sctp_sched_fc_unsched_all(struct sctp_stream *stream);
+
+static int sctp_sched_fc_set(struct sctp_stream *stream, __u16 sid,
+			     __u16 weight, gfp_t gfp)
+{
+	return 0;
+}
+
+static int sctp_sched_fc_get(struct sctp_stream *stream, __u16 sid,
+			     __u16 *value)
+{
+	return 0;
+}
+
+static int sctp_sched_fc_init(struct sctp_stream *stream)
+{
+	INIT_LIST_HEAD(&stream->fc_list);
+
+	return 0;
+}
+
+static int sctp_sched_fc_init_sid(struct sctp_stream *stream, __u16 sid,
+				  gfp_t gfp)
+{
+	struct sctp_stream_out_ext *soute = SCTP_SO(stream, sid)->ext;
+
+	INIT_LIST_HEAD(&soute->fc_list);
+	soute->fc_length = 0;
+
+	return 0;
+}
+
+static void sctp_sched_fc_free_sid(struct sctp_stream *stream, __u16 sid)
+{
+}
+
+static void sctp_sched_fc_sched(struct sctp_stream *stream,
+				struct sctp_stream_out_ext *soute)
+{
+	struct sctp_stream_out_ext *pos;
+
+	if (!list_empty(&soute->fc_list))
+		return;
+
+	list_for_each_entry(pos, &stream->fc_list, fc_list)
+		if (pos->fc_length >= soute->fc_length)
+			break;
+	list_add_tail(&soute->fc_list, &pos->fc_list);
+}
+
+static void sctp_sched_fc_enqueue(struct sctp_outq *q,
+				  struct sctp_datamsg *msg)
+{
+	struct sctp_stream *stream;
+	struct sctp_chunk *ch;
+	__u16 sid;
+
+	ch = list_first_entry(&msg->chunks, struct sctp_chunk, frag_list);
+	sid = sctp_chunk_stream_no(ch);
+	stream = &q->asoc->stream;
+	sctp_sched_fc_sched(stream, SCTP_SO(stream, sid)->ext);
+}
+
+static struct sctp_chunk *sctp_sched_fc_dequeue(struct sctp_outq *q)
+{
+	struct sctp_stream *stream = &q->asoc->stream;
+	struct sctp_stream_out_ext *soute;
+	struct sctp_chunk *ch;
+
+	/* Bail out quickly if queue is empty */
+	if (list_empty(&q->out_chunk_list))
+		return NULL;
+
+	/* Find which chunk is next */
+	if (stream->out_curr)
+		soute = stream->out_curr->ext;
+	else
+		soute = list_entry(stream->fc_list.next, struct sctp_stream_out_ext, fc_list);
+	ch = list_entry(soute->outq.next, struct sctp_chunk, stream_list);
+
+	sctp_sched_dequeue_common(q, ch);
+	return ch;
+}
+
+static void sctp_sched_fc_dequeue_done(struct sctp_outq *q,
+				       struct sctp_chunk *ch)
+{
+	struct sctp_stream *stream = &q->asoc->stream;
+	struct sctp_stream_out_ext *soute, *pos;
+	__u16 sid, i;
+
+	sid = sctp_chunk_stream_no(ch);
+	soute = SCTP_SO(stream, sid)->ext;
+	/* reduce all fc_lengths by U32_MAX / 4 if the current fc_length overflows. */
+	if (soute->fc_length > U32_MAX - ch->skb->len) {
+		for (i = 0; i < stream->outcnt; i++) {
+			pos = SCTP_SO(stream, i)->ext;
+			if (!pos)
+				continue;
+			if (pos->fc_length <= (U32_MAX >> 2)) {
+				pos->fc_length = 0;
+				continue;
+			}
+			pos->fc_length -= (U32_MAX >> 2);
+		}
+	}
+	soute->fc_length += ch->skb->len;
+
+	if (list_empty(&soute->outq)) {
+		list_del_init(&soute->fc_list);
+		return;
+	}
+
+	pos = soute;
+	list_for_each_entry_continue(pos, &stream->fc_list, fc_list)
+		if (pos->fc_length >= soute->fc_length)
+			break;
+	list_move_tail(&soute->fc_list, &pos->fc_list);
+}
+
+static void sctp_sched_fc_sched_all(struct sctp_stream *stream)
+{
+	struct sctp_association *asoc;
+	struct sctp_chunk *ch;
+
+	asoc = container_of(stream, struct sctp_association, stream);
+	list_for_each_entry(ch, &asoc->outqueue.out_chunk_list, list) {
+		__u16 sid = sctp_chunk_stream_no(ch);
+
+		if (SCTP_SO(stream, sid)->ext)
+			sctp_sched_fc_sched(stream, SCTP_SO(stream, sid)->ext);
+	}
+}
+
+static void sctp_sched_fc_unsched_all(struct sctp_stream *stream)
+{
+	struct sctp_stream_out_ext *soute, *tmp;
+
+	list_for_each_entry_safe(soute, tmp, &stream->fc_list, fc_list)
+		list_del_init(&soute->fc_list);
+}
+
+static struct sctp_sched_ops sctp_sched_fc = {
+	.set = sctp_sched_fc_set,
+	.get = sctp_sched_fc_get,
+	.init = sctp_sched_fc_init,
+	.init_sid = sctp_sched_fc_init_sid,
+	.free_sid = sctp_sched_fc_free_sid,
+	.enqueue = sctp_sched_fc_enqueue,
+	.dequeue = sctp_sched_fc_dequeue,
+	.dequeue_done = sctp_sched_fc_dequeue_done,
+	.sched_all = sctp_sched_fc_sched_all,
+	.unsched_all = sctp_sched_fc_unsched_all,
+};
+
+void sctp_sched_ops_fc_init(void)
+{
+	sctp_sched_ops_register(SCTP_SS_FC, &sctp_sched_fc);
+}