linux/net/sctp/stream_sched.c

276 lines
6.4 KiB
C
Raw Normal View History

sctp: introduce stream scheduler foundations This patch introduces the hooks necessary to do stream scheduling, as per RFC Draft ndata. It also introduces the first scheduler, which is what we do today but now factored out: first come first served (FCFS). With stream scheduling now we have to track which chunk was enqueued on which stream and be able to select another other than the in front of the main outqueue. So we introduce a list on sctp_stream_out_ext structure for this purpose. We reuse sctp_chunk->transmitted_list space for the list above, as the chunk cannot belong to the two lists at the same time. By using the union in there, we can have distinct names for these moments. sctp_sched_ops are the operations expected to be implemented by each scheduler. The dequeueing is a bit particular to this implementation but it is to match how we dequeue packets today. We first dequeue and then check if it fits the packet and if not, we requeue it at head. Thus why we don't have a peek operation but have dequeue_done instead, which is called once the chunk can be safely considered as transmitted. The check removed from sctp_outq_flush is now performed by sctp_stream_outq_migrate, which is only called during assoc setup. (sctp_sendmsg() also checks for it) The only operation that is foreseen but not yet added here is a way to signalize that a new packet is starting or that the packet is done, for round robin scheduler per packet, but is intentionally left to the patch that actually implements it. Support for I-DATA chunks, also described in this RFC, with user message interleaving is straightforward as it just requires the schedulers to probe for the feature and ignore datamsg boundaries when dequeueing. See-also: https://tools.ietf.org/html/draft-ietf-tsvwg-sctp-ndata-13 Tested-by: Xin Long <lucien.xin@gmail.com> Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-10-03 22:20:13 +00:00
/* SCTP kernel implementation
* (C) Copyright Red Hat Inc. 2017
*
* This file is part of the SCTP kernel implementation
*
* These functions manipulate sctp stream queue/scheduling.
*
* This SCTP implementation is free software;
* you can redistribute it and/or modify it under the terms of
* the GNU General Public License as published by
* the Free Software Foundation; either version 2, or (at your option)
* any later version.
*
* This SCTP implementation is distributed in the hope that it
* will be useful, but WITHOUT ANY WARRANTY; without even the implied
* ************************
* warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
* See the GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with GNU CC; see the file COPYING. If not, see
* <http://www.gnu.org/licenses/>.
*
* Please send any bug reports or fixes you make to the
* email addresched(es):
* lksctp developers <linux-sctp@vger.kernel.org>
*
* Written or modified by:
* Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
*/
#include <linux/list.h>
#include <net/sctp/sctp.h>
#include <net/sctp/sm.h>
#include <net/sctp/stream_sched.h>
/* First Come First Serve (a.k.a. FIFO)
* RFC DRAFT ndata Section 3.1
*/
static int sctp_sched_fcfs_set(struct sctp_stream *stream, __u16 sid,
__u16 value, gfp_t gfp)
{
return 0;
}
static int sctp_sched_fcfs_get(struct sctp_stream *stream, __u16 sid,
__u16 *value)
{
*value = 0;
return 0;
}
static int sctp_sched_fcfs_init(struct sctp_stream *stream)
{
return 0;
}
static int sctp_sched_fcfs_init_sid(struct sctp_stream *stream, __u16 sid,
gfp_t gfp)
{
return 0;
}
static void sctp_sched_fcfs_free(struct sctp_stream *stream)
{
}
static void sctp_sched_fcfs_enqueue(struct sctp_outq *q,
struct sctp_datamsg *msg)
{
}
static struct sctp_chunk *sctp_sched_fcfs_dequeue(struct sctp_outq *q)
{
struct sctp_stream *stream = &q->asoc->stream;
struct sctp_chunk *ch = NULL;
struct list_head *entry;
if (list_empty(&q->out_chunk_list))
goto out;
if (stream->out_curr) {
ch = list_entry(stream->out_curr->ext->outq.next,
struct sctp_chunk, stream_list);
} else {
entry = q->out_chunk_list.next;
ch = list_entry(entry, struct sctp_chunk, list);
}
sctp_sched_dequeue_common(q, ch);
out:
return ch;
}
static void sctp_sched_fcfs_dequeue_done(struct sctp_outq *q,
struct sctp_chunk *chunk)
{
}
static void sctp_sched_fcfs_sched_all(struct sctp_stream *stream)
{
}
static void sctp_sched_fcfs_unsched_all(struct sctp_stream *stream)
{
}
static struct sctp_sched_ops sctp_sched_fcfs = {
.set = sctp_sched_fcfs_set,
.get = sctp_sched_fcfs_get,
.init = sctp_sched_fcfs_init,
.init_sid = sctp_sched_fcfs_init_sid,
.free = sctp_sched_fcfs_free,
.enqueue = sctp_sched_fcfs_enqueue,
.dequeue = sctp_sched_fcfs_dequeue,
.dequeue_done = sctp_sched_fcfs_dequeue_done,
.sched_all = sctp_sched_fcfs_sched_all,
.unsched_all = sctp_sched_fcfs_unsched_all,
};
/* API to other parts of the stack */
sctp: introduce priority based stream scheduler This patch introduces RFC Draft ndata section 3.4 Priority Based Scheduler (SCTP_SS_PRIO). It works by having a struct sctp_stream_priority for each priority configured. This struct is then enlisted on a queue ordered per priority if, and only if, there is a stream with data queued, so that dequeueing is very straightforward: either finish current datamsg or simply dequeue from the highest priority queued, which is the next stream pointed, and that's it. If there are multiple streams assigned with the same priority and with data queued, it will do round robin amongst them while respecting datamsgs boundaries (when not using idata chunks), to be reasonably fair. We intentionally don't maintain a list of priorities nor a list of all streams with the same priority to save memory. The first would mean at least 2 other pointers per priority (which, for 1000 priorities, that can mean 16kB) and the second would also mean 2 other pointers but per stream. As SCTP supports up to 65535 streams on a given asoc, that's 1MB. This impacts when giving a priority to some stream, as we have to find out if the new priority is already being used and if we can free the old one, and also when tearing down. The new fields in struct sctp_stream_out_ext and sctp_stream are added under a union because that memory is to be shared with other schedulers. See-also: https://tools.ietf.org/html/draft-ietf-tsvwg-sctp-ndata-13 Tested-by: Xin Long <lucien.xin@gmail.com> Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-10-03 22:20:16 +00:00
extern struct sctp_sched_ops sctp_sched_prio;
extern struct sctp_sched_ops sctp_sched_rr;
sctp: introduce priority based stream scheduler This patch introduces RFC Draft ndata section 3.4 Priority Based Scheduler (SCTP_SS_PRIO). It works by having a struct sctp_stream_priority for each priority configured. This struct is then enlisted on a queue ordered per priority if, and only if, there is a stream with data queued, so that dequeueing is very straightforward: either finish current datamsg or simply dequeue from the highest priority queued, which is the next stream pointed, and that's it. If there are multiple streams assigned with the same priority and with data queued, it will do round robin amongst them while respecting datamsgs boundaries (when not using idata chunks), to be reasonably fair. We intentionally don't maintain a list of priorities nor a list of all streams with the same priority to save memory. The first would mean at least 2 other pointers per priority (which, for 1000 priorities, that can mean 16kB) and the second would also mean 2 other pointers but per stream. As SCTP supports up to 65535 streams on a given asoc, that's 1MB. This impacts when giving a priority to some stream, as we have to find out if the new priority is already being used and if we can free the old one, and also when tearing down. The new fields in struct sctp_stream_out_ext and sctp_stream are added under a union because that memory is to be shared with other schedulers. See-also: https://tools.ietf.org/html/draft-ietf-tsvwg-sctp-ndata-13 Tested-by: Xin Long <lucien.xin@gmail.com> Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-10-03 22:20:16 +00:00
static struct sctp_sched_ops *sctp_sched_ops[] = {
sctp: introduce stream scheduler foundations This patch introduces the hooks necessary to do stream scheduling, as per RFC Draft ndata. It also introduces the first scheduler, which is what we do today but now factored out: first come first served (FCFS). With stream scheduling now we have to track which chunk was enqueued on which stream and be able to select another other than the in front of the main outqueue. So we introduce a list on sctp_stream_out_ext structure for this purpose. We reuse sctp_chunk->transmitted_list space for the list above, as the chunk cannot belong to the two lists at the same time. By using the union in there, we can have distinct names for these moments. sctp_sched_ops are the operations expected to be implemented by each scheduler. The dequeueing is a bit particular to this implementation but it is to match how we dequeue packets today. We first dequeue and then check if it fits the packet and if not, we requeue it at head. Thus why we don't have a peek operation but have dequeue_done instead, which is called once the chunk can be safely considered as transmitted. The check removed from sctp_outq_flush is now performed by sctp_stream_outq_migrate, which is only called during assoc setup. (sctp_sendmsg() also checks for it) The only operation that is foreseen but not yet added here is a way to signalize that a new packet is starting or that the packet is done, for round robin scheduler per packet, but is intentionally left to the patch that actually implements it. Support for I-DATA chunks, also described in this RFC, with user message interleaving is straightforward as it just requires the schedulers to probe for the feature and ignore datamsg boundaries when dequeueing. See-also: https://tools.ietf.org/html/draft-ietf-tsvwg-sctp-ndata-13 Tested-by: Xin Long <lucien.xin@gmail.com> Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-10-03 22:20:13 +00:00
&sctp_sched_fcfs,
sctp: introduce priority based stream scheduler This patch introduces RFC Draft ndata section 3.4 Priority Based Scheduler (SCTP_SS_PRIO). It works by having a struct sctp_stream_priority for each priority configured. This struct is then enlisted on a queue ordered per priority if, and only if, there is a stream with data queued, so that dequeueing is very straightforward: either finish current datamsg or simply dequeue from the highest priority queued, which is the next stream pointed, and that's it. If there are multiple streams assigned with the same priority and with data queued, it will do round robin amongst them while respecting datamsgs boundaries (when not using idata chunks), to be reasonably fair. We intentionally don't maintain a list of priorities nor a list of all streams with the same priority to save memory. The first would mean at least 2 other pointers per priority (which, for 1000 priorities, that can mean 16kB) and the second would also mean 2 other pointers but per stream. As SCTP supports up to 65535 streams on a given asoc, that's 1MB. This impacts when giving a priority to some stream, as we have to find out if the new priority is already being used and if we can free the old one, and also when tearing down. The new fields in struct sctp_stream_out_ext and sctp_stream are added under a union because that memory is to be shared with other schedulers. See-also: https://tools.ietf.org/html/draft-ietf-tsvwg-sctp-ndata-13 Tested-by: Xin Long <lucien.xin@gmail.com> Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-10-03 22:20:16 +00:00
&sctp_sched_prio,
&sctp_sched_rr,
sctp: introduce stream scheduler foundations This patch introduces the hooks necessary to do stream scheduling, as per RFC Draft ndata. It also introduces the first scheduler, which is what we do today but now factored out: first come first served (FCFS). With stream scheduling now we have to track which chunk was enqueued on which stream and be able to select another other than the in front of the main outqueue. So we introduce a list on sctp_stream_out_ext structure for this purpose. We reuse sctp_chunk->transmitted_list space for the list above, as the chunk cannot belong to the two lists at the same time. By using the union in there, we can have distinct names for these moments. sctp_sched_ops are the operations expected to be implemented by each scheduler. The dequeueing is a bit particular to this implementation but it is to match how we dequeue packets today. We first dequeue and then check if it fits the packet and if not, we requeue it at head. Thus why we don't have a peek operation but have dequeue_done instead, which is called once the chunk can be safely considered as transmitted. The check removed from sctp_outq_flush is now performed by sctp_stream_outq_migrate, which is only called during assoc setup. (sctp_sendmsg() also checks for it) The only operation that is foreseen but not yet added here is a way to signalize that a new packet is starting or that the packet is done, for round robin scheduler per packet, but is intentionally left to the patch that actually implements it. Support for I-DATA chunks, also described in this RFC, with user message interleaving is straightforward as it just requires the schedulers to probe for the feature and ignore datamsg boundaries when dequeueing. See-also: https://tools.ietf.org/html/draft-ietf-tsvwg-sctp-ndata-13 Tested-by: Xin Long <lucien.xin@gmail.com> Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-10-03 22:20:13 +00:00
};
int sctp_sched_set_sched(struct sctp_association *asoc,
enum sctp_sched_type sched)
{
struct sctp_sched_ops *n = sctp_sched_ops[sched];
struct sctp_sched_ops *old = asoc->outqueue.sched;
struct sctp_datamsg *msg = NULL;
struct sctp_chunk *ch;
int i, ret = 0;
if (old == n)
return ret;
if (sched > SCTP_SS_MAX)
return -EINVAL;
if (old) {
old->free(&asoc->stream);
/* Give the next scheduler a clean slate. */
for (i = 0; i < asoc->stream.outcnt; i++) {
void *p = asoc->stream.out[i].ext;
if (!p)
continue;
p += offsetofend(struct sctp_stream_out_ext, outq);
memset(p, 0, sizeof(struct sctp_stream_out_ext) -
offsetofend(struct sctp_stream_out_ext, outq));
}
}
asoc->outqueue.sched = n;
n->init(&asoc->stream);
for (i = 0; i < asoc->stream.outcnt; i++) {
if (!asoc->stream.out[i].ext)
continue;
ret = n->init_sid(&asoc->stream, i, GFP_KERNEL);
if (ret)
goto err;
}
/* We have to requeue all chunks already queued. */
list_for_each_entry(ch, &asoc->outqueue.out_chunk_list, list) {
if (ch->msg == msg)
continue;
msg = ch->msg;
n->enqueue(&asoc->outqueue, msg);
}
return ret;
err:
n->free(&asoc->stream);
asoc->outqueue.sched = &sctp_sched_fcfs; /* Always safe */
return ret;
}
int sctp_sched_get_sched(struct sctp_association *asoc)
{
int i;
for (i = 0; i <= SCTP_SS_MAX; i++)
if (asoc->outqueue.sched == sctp_sched_ops[i])
return i;
return 0;
}
int sctp_sched_set_value(struct sctp_association *asoc, __u16 sid,
__u16 value, gfp_t gfp)
{
if (sid >= asoc->stream.outcnt)
return -EINVAL;
if (!asoc->stream.out[sid].ext) {
int ret;
ret = sctp_stream_init_ext(&asoc->stream, sid);
if (ret)
return ret;
}
return asoc->outqueue.sched->set(&asoc->stream, sid, value, gfp);
}
int sctp_sched_get_value(struct sctp_association *asoc, __u16 sid,
__u16 *value)
{
if (sid >= asoc->stream.outcnt)
return -EINVAL;
if (!asoc->stream.out[sid].ext)
return 0;
return asoc->outqueue.sched->get(&asoc->stream, sid, value);
}
void sctp_sched_dequeue_done(struct sctp_outq *q, struct sctp_chunk *ch)
{
if (!list_is_last(&ch->frag_list, &ch->msg->chunks)) {
struct sctp_stream_out *sout;
__u16 sid;
/* datamsg is not finish, so save it as current one,
* in case application switch scheduler or a higher
* priority stream comes in.
*/
sid = sctp_chunk_stream_no(ch);
sout = &q->asoc->stream.out[sid];
q->asoc->stream.out_curr = sout;
return;
}
q->asoc->stream.out_curr = NULL;
q->sched->dequeue_done(q, ch);
}
/* Auxiliary functions for the schedulers */
void sctp_sched_dequeue_common(struct sctp_outq *q, struct sctp_chunk *ch)
{
list_del_init(&ch->list);
list_del_init(&ch->stream_list);
q->out_qlen -= ch->skb->len;
}
int sctp_sched_init_sid(struct sctp_stream *stream, __u16 sid, gfp_t gfp)
{
struct sctp_sched_ops *sched = sctp_sched_ops_from_stream(stream);
INIT_LIST_HEAD(&stream->out[sid].ext->outq);
return sched->init_sid(stream, sid, gfp);
}
struct sctp_sched_ops *sctp_sched_ops_from_stream(struct sctp_stream *stream)
{
struct sctp_association *asoc;
asoc = container_of(stream, struct sctp_association, stream);
return asoc->outqueue.sched;
}