writeback: split writeback_inodes_wb
[linux-2.6/next.git] / net / sched / sch_multiq.c
blob6ae251279fc2f53c86aa339d39bff13b71c79991
1 /*
2 * Copyright (c) 2008, Intel Corporation.
4 * This program is free software; you can redistribute it and/or modify it
5 * under the terms and conditions of the GNU General Public License,
6 * version 2, as published by the Free Software Foundation.
8 * This program is distributed in the hope it will be useful, but WITHOUT
9 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
10 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
11 * more details.
13 * You should have received a copy of the GNU General Public License along with
14 * this program; if not, write to the Free Software Foundation, Inc., 59 Temple
15 * Place - Suite 330, Boston, MA 02111-1307 USA.
17 * Author: Alexander Duyck <alexander.h.duyck@intel.com>
20 #include <linux/module.h>
21 #include <linux/slab.h>
22 #include <linux/types.h>
23 #include <linux/kernel.h>
24 #include <linux/string.h>
25 #include <linux/errno.h>
26 #include <linux/skbuff.h>
27 #include <net/netlink.h>
28 #include <net/pkt_sched.h>
31 struct multiq_sched_data {
32 u16 bands;
33 u16 max_bands;
34 u16 curband;
35 struct tcf_proto *filter_list;
36 struct Qdisc **queues;
40 static struct Qdisc *
41 multiq_classify(struct sk_buff *skb, struct Qdisc *sch, int *qerr)
43 struct multiq_sched_data *q = qdisc_priv(sch);
44 u32 band;
45 struct tcf_result res;
46 int err;
48 *qerr = NET_XMIT_SUCCESS | __NET_XMIT_BYPASS;
49 err = tc_classify(skb, q->filter_list, &res);
50 #ifdef CONFIG_NET_CLS_ACT
51 switch (err) {
52 case TC_ACT_STOLEN:
53 case TC_ACT_QUEUED:
54 *qerr = NET_XMIT_SUCCESS | __NET_XMIT_STOLEN;
55 case TC_ACT_SHOT:
56 return NULL;
58 #endif
59 band = skb_get_queue_mapping(skb);
61 if (band >= q->bands)
62 return q->queues[0];
64 return q->queues[band];
67 static int
68 multiq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
70 struct Qdisc *qdisc;
71 int ret;
73 qdisc = multiq_classify(skb, sch, &ret);
74 #ifdef CONFIG_NET_CLS_ACT
75 if (qdisc == NULL) {
77 if (ret & __NET_XMIT_BYPASS)
78 sch->qstats.drops++;
79 kfree_skb(skb);
80 return ret;
82 #endif
84 ret = qdisc_enqueue(skb, qdisc);
85 if (ret == NET_XMIT_SUCCESS) {
86 sch->bstats.bytes += qdisc_pkt_len(skb);
87 sch->bstats.packets++;
88 sch->q.qlen++;
89 return NET_XMIT_SUCCESS;
91 if (net_xmit_drop_count(ret))
92 sch->qstats.drops++;
93 return ret;
96 static struct sk_buff *multiq_dequeue(struct Qdisc *sch)
98 struct multiq_sched_data *q = qdisc_priv(sch);
99 struct Qdisc *qdisc;
100 struct sk_buff *skb;
101 int band;
103 for (band = 0; band < q->bands; band++) {
104 /* cycle through bands to ensure fairness */
105 q->curband++;
106 if (q->curband >= q->bands)
107 q->curband = 0;
109 /* Check that target subqueue is available before
110 * pulling an skb to avoid head-of-line blocking.
112 if (!__netif_subqueue_stopped(qdisc_dev(sch), q->curband)) {
113 qdisc = q->queues[q->curband];
114 skb = qdisc->dequeue(qdisc);
115 if (skb) {
116 sch->q.qlen--;
117 return skb;
121 return NULL;
125 static struct sk_buff *multiq_peek(struct Qdisc *sch)
127 struct multiq_sched_data *q = qdisc_priv(sch);
128 unsigned int curband = q->curband;
129 struct Qdisc *qdisc;
130 struct sk_buff *skb;
131 int band;
133 for (band = 0; band < q->bands; band++) {
134 /* cycle through bands to ensure fairness */
135 curband++;
136 if (curband >= q->bands)
137 curband = 0;
139 /* Check that target subqueue is available before
140 * pulling an skb to avoid head-of-line blocking.
142 if (!__netif_subqueue_stopped(qdisc_dev(sch), curband)) {
143 qdisc = q->queues[curband];
144 skb = qdisc->ops->peek(qdisc);
145 if (skb)
146 return skb;
149 return NULL;
153 static unsigned int multiq_drop(struct Qdisc *sch)
155 struct multiq_sched_data *q = qdisc_priv(sch);
156 int band;
157 unsigned int len;
158 struct Qdisc *qdisc;
160 for (band = q->bands-1; band >= 0; band--) {
161 qdisc = q->queues[band];
162 if (qdisc->ops->drop) {
163 len = qdisc->ops->drop(qdisc);
164 if (len != 0) {
165 sch->q.qlen--;
166 return len;
170 return 0;
174 static void
175 multiq_reset(struct Qdisc *sch)
177 u16 band;
178 struct multiq_sched_data *q = qdisc_priv(sch);
180 for (band = 0; band < q->bands; band++)
181 qdisc_reset(q->queues[band]);
182 sch->q.qlen = 0;
183 q->curband = 0;
186 static void
187 multiq_destroy(struct Qdisc *sch)
189 int band;
190 struct multiq_sched_data *q = qdisc_priv(sch);
192 tcf_destroy_chain(&q->filter_list);
193 for (band = 0; band < q->bands; band++)
194 qdisc_destroy(q->queues[band]);
196 kfree(q->queues);
199 static int multiq_tune(struct Qdisc *sch, struct nlattr *opt)
201 struct multiq_sched_data *q = qdisc_priv(sch);
202 struct tc_multiq_qopt *qopt;
203 int i;
205 if (!netif_is_multiqueue(qdisc_dev(sch)))
206 return -EOPNOTSUPP;
207 if (nla_len(opt) < sizeof(*qopt))
208 return -EINVAL;
210 qopt = nla_data(opt);
212 qopt->bands = qdisc_dev(sch)->real_num_tx_queues;
214 sch_tree_lock(sch);
215 q->bands = qopt->bands;
216 for (i = q->bands; i < q->max_bands; i++) {
217 if (q->queues[i] != &noop_qdisc) {
218 struct Qdisc *child = q->queues[i];
219 q->queues[i] = &noop_qdisc;
220 qdisc_tree_decrease_qlen(child, child->q.qlen);
221 qdisc_destroy(child);
225 sch_tree_unlock(sch);
227 for (i = 0; i < q->bands; i++) {
228 if (q->queues[i] == &noop_qdisc) {
229 struct Qdisc *child, *old;
230 child = qdisc_create_dflt(qdisc_dev(sch),
231 sch->dev_queue,
232 &pfifo_qdisc_ops,
233 TC_H_MAKE(sch->handle,
234 i + 1));
235 if (child) {
236 sch_tree_lock(sch);
237 old = q->queues[i];
238 q->queues[i] = child;
240 if (old != &noop_qdisc) {
241 qdisc_tree_decrease_qlen(old,
242 old->q.qlen);
243 qdisc_destroy(old);
245 sch_tree_unlock(sch);
249 return 0;
252 static int multiq_init(struct Qdisc *sch, struct nlattr *opt)
254 struct multiq_sched_data *q = qdisc_priv(sch);
255 int i, err;
257 q->queues = NULL;
259 if (opt == NULL)
260 return -EINVAL;
262 q->max_bands = qdisc_dev(sch)->num_tx_queues;
264 q->queues = kcalloc(q->max_bands, sizeof(struct Qdisc *), GFP_KERNEL);
265 if (!q->queues)
266 return -ENOBUFS;
267 for (i = 0; i < q->max_bands; i++)
268 q->queues[i] = &noop_qdisc;
270 err = multiq_tune(sch,opt);
272 if (err)
273 kfree(q->queues);
275 return err;
278 static int multiq_dump(struct Qdisc *sch, struct sk_buff *skb)
280 struct multiq_sched_data *q = qdisc_priv(sch);
281 unsigned char *b = skb_tail_pointer(skb);
282 struct tc_multiq_qopt opt;
284 opt.bands = q->bands;
285 opt.max_bands = q->max_bands;
287 NLA_PUT(skb, TCA_OPTIONS, sizeof(opt), &opt);
289 return skb->len;
291 nla_put_failure:
292 nlmsg_trim(skb, b);
293 return -1;
296 static int multiq_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
297 struct Qdisc **old)
299 struct multiq_sched_data *q = qdisc_priv(sch);
300 unsigned long band = arg - 1;
302 if (new == NULL)
303 new = &noop_qdisc;
305 sch_tree_lock(sch);
306 *old = q->queues[band];
307 q->queues[band] = new;
308 qdisc_tree_decrease_qlen(*old, (*old)->q.qlen);
309 qdisc_reset(*old);
310 sch_tree_unlock(sch);
312 return 0;
315 static struct Qdisc *
316 multiq_leaf(struct Qdisc *sch, unsigned long arg)
318 struct multiq_sched_data *q = qdisc_priv(sch);
319 unsigned long band = arg - 1;
321 return q->queues[band];
324 static unsigned long multiq_get(struct Qdisc *sch, u32 classid)
326 struct multiq_sched_data *q = qdisc_priv(sch);
327 unsigned long band = TC_H_MIN(classid);
329 if (band - 1 >= q->bands)
330 return 0;
331 return band;
334 static unsigned long multiq_bind(struct Qdisc *sch, unsigned long parent,
335 u32 classid)
337 return multiq_get(sch, classid);
341 static void multiq_put(struct Qdisc *q, unsigned long cl)
345 static int multiq_dump_class(struct Qdisc *sch, unsigned long cl,
346 struct sk_buff *skb, struct tcmsg *tcm)
348 struct multiq_sched_data *q = qdisc_priv(sch);
350 tcm->tcm_handle |= TC_H_MIN(cl);
351 tcm->tcm_info = q->queues[cl-1]->handle;
352 return 0;
355 static int multiq_dump_class_stats(struct Qdisc *sch, unsigned long cl,
356 struct gnet_dump *d)
358 struct multiq_sched_data *q = qdisc_priv(sch);
359 struct Qdisc *cl_q;
361 cl_q = q->queues[cl - 1];
362 cl_q->qstats.qlen = cl_q->q.qlen;
363 if (gnet_stats_copy_basic(d, &cl_q->bstats) < 0 ||
364 gnet_stats_copy_queue(d, &cl_q->qstats) < 0)
365 return -1;
367 return 0;
370 static void multiq_walk(struct Qdisc *sch, struct qdisc_walker *arg)
372 struct multiq_sched_data *q = qdisc_priv(sch);
373 int band;
375 if (arg->stop)
376 return;
378 for (band = 0; band < q->bands; band++) {
379 if (arg->count < arg->skip) {
380 arg->count++;
381 continue;
383 if (arg->fn(sch, band+1, arg) < 0) {
384 arg->stop = 1;
385 break;
387 arg->count++;
391 static struct tcf_proto **multiq_find_tcf(struct Qdisc *sch, unsigned long cl)
393 struct multiq_sched_data *q = qdisc_priv(sch);
395 if (cl)
396 return NULL;
397 return &q->filter_list;
400 static const struct Qdisc_class_ops multiq_class_ops = {
401 .graft = multiq_graft,
402 .leaf = multiq_leaf,
403 .get = multiq_get,
404 .put = multiq_put,
405 .walk = multiq_walk,
406 .tcf_chain = multiq_find_tcf,
407 .bind_tcf = multiq_bind,
408 .unbind_tcf = multiq_put,
409 .dump = multiq_dump_class,
410 .dump_stats = multiq_dump_class_stats,
413 static struct Qdisc_ops multiq_qdisc_ops __read_mostly = {
414 .next = NULL,
415 .cl_ops = &multiq_class_ops,
416 .id = "multiq",
417 .priv_size = sizeof(struct multiq_sched_data),
418 .enqueue = multiq_enqueue,
419 .dequeue = multiq_dequeue,
420 .peek = multiq_peek,
421 .drop = multiq_drop,
422 .init = multiq_init,
423 .reset = multiq_reset,
424 .destroy = multiq_destroy,
425 .change = multiq_tune,
426 .dump = multiq_dump,
427 .owner = THIS_MODULE,
430 static int __init multiq_module_init(void)
432 return register_qdisc(&multiq_qdisc_ops);
435 static void __exit multiq_module_exit(void)
437 unregister_qdisc(&multiq_qdisc_ops);
440 module_init(multiq_module_init)
441 module_exit(multiq_module_exit)
443 MODULE_LICENSE("GPL");