From bb949fbd1878973c3539d9aecff52f284482a937 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 8 Jul 2008 16:55:56 -0700 Subject: netdev: Create netdev_queue abstraction. A netdev_queue is an entity managed by a qdisc. Currently there is one RX and one TX queue, and a netdev_queue merely contains a backpointer to the net_device. The Qdisc struct is augmented with a netdev_queue pointer as well. Eventually the 'dev' Qdisc member will go away and we will have the resulting hierarchy: net_device --> netdev_queue --> Qdisc Also, qdisc_alloc() and qdisc_create_dflt() now take a netdev_queue pointer argument. Signed-off-by: David S. Miller --- net/sched/sch_prio.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net/sched/sch_prio.c') diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c index 5532f10..ca58a03 100644 --- a/net/sched/sch_prio.c +++ b/net/sched/sch_prio.c @@ -281,7 +281,8 @@ static int prio_tune(struct Qdisc *sch, struct nlattr *opt) for (i=0; ibands; i++) { if (q->queues[i] == &noop_qdisc) { struct Qdisc *child; - child = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, + child = qdisc_create_dflt(sch->dev, sch->dev_queue, + &pfifo_qdisc_ops, TC_H_MAKE(sch->handle, i + 1)); if (child) { sch_tree_lock(sch); -- cgit v1.1 From 5ce2d488fe039ddd86a638496cf704df86c74eeb Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 8 Jul 2008 17:06:30 -0700 Subject: pkt_sched: Remove 'dev' member of struct Qdisc. It can be obtained via the netdev_queue. So create a helper routine, qdisc_dev(), to make the transformations nicer looking. Now, qdisc_alloc() now no longer needs a net_device pointer argument. Signed-off-by: David S. Miller --- net/sched/sch_prio.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) (limited to 'net/sched/sch_prio.c') diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c index ca58a03..39157f7 100644 --- a/net/sched/sch_prio.c +++ b/net/sched/sch_prio.c @@ -136,7 +136,8 @@ prio_dequeue(struct Qdisc* sch) * pulling an skb. This way we avoid excessive requeues * for slower queues. */ - if (!__netif_subqueue_stopped(sch->dev, (q->mq ? prio : 0))) { + if (!__netif_subqueue_stopped(qdisc_dev(sch), + (q->mq ? prio : 0))) { qdisc = q->queues[prio]; skb = qdisc->dequeue(qdisc); if (skb) { @@ -165,8 +166,8 @@ static struct sk_buff *rr_dequeue(struct Qdisc* sch) * for slower queues. If the queue is stopped, try the * next queue. */ - if (!__netif_subqueue_stopped(sch->dev, - (q->mq ? q->curband : 0))) { + if (!__netif_subqueue_stopped(qdisc_dev(sch), + (q->mq ? q->curband : 0))) { qdisc = q->queues[q->curband]; skb = qdisc->dequeue(qdisc); if (skb) { @@ -249,10 +250,10 @@ static int prio_tune(struct Qdisc *sch, struct nlattr *opt) if (q->mq) { if (sch->parent != TC_H_ROOT) return -EINVAL; - if (netif_is_multiqueue(sch->dev)) { + if (netif_is_multiqueue(qdisc_dev(sch))) { if (q->bands == 0) - q->bands = sch->dev->egress_subqueue_count; - else if (q->bands != sch->dev->egress_subqueue_count) + q->bands = qdisc_dev(sch)->egress_subqueue_count; + else if (q->bands != qdisc_dev(sch)->egress_subqueue_count) return -EINVAL; } else return -EOPNOTSUPP; @@ -281,7 +282,7 @@ static int prio_tune(struct Qdisc *sch, struct nlattr *opt) for (i=0; ibands; i++) { if (q->queues[i] == &noop_qdisc) { struct Qdisc *child; - child = qdisc_create_dflt(sch->dev, sch->dev_queue, + child = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue, &pfifo_qdisc_ops, TC_H_MAKE(sch->handle, i + 1)); if (child) { -- cgit v1.1 From 1d8ae3fdeb001b8f534a6782c261aba6ec1779f5 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 15 Jul 2008 02:52:19 -0700 Subject: pkt_sched: Remove RR scheduler. This actually fixes a bug added by the RR scheduler changes. The ->bands and ->prio2band parameters were being set outside of the sch_tree_lock() and thus could result in strange behavior and inconsistencies. It might be possible, in the new design (where there will be one qdisc per device TX queue) to allow similar functionality via a TX hash algorithm for RR but I really see no reason to export this aspect of how these multiqueue cards actually implement the scheduling of the the individual DMA TX rings and the single physical MAC/PHY port. Signed-off-by: David S. Miller --- net/sched/sch_prio.c | 136 ++++++--------------------------------------------- 1 file changed, 16 insertions(+), 120 deletions(-) (limited to 'net/sched/sch_prio.c') diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c index 39157f7..536ca47 100644 --- a/net/sched/sch_prio.c +++ b/net/sched/sch_prio.c @@ -24,11 +24,9 @@ struct prio_sched_data { int bands; - int curband; /* for round-robin */ struct tcf_proto *filter_list; u8 prio2band[TC_PRIO_MAX+1]; struct Qdisc *queues[TCQ_PRIO_BANDS]; - int mq; }; @@ -55,17 +53,14 @@ prio_classify(struct sk_buff *skb, struct Qdisc *sch, int *qerr) if (!q->filter_list || err < 0) { if (TC_H_MAJ(band)) band = 0; - band = q->prio2band[band&TC_PRIO_MAX]; - goto out; + return q->queues[q->prio2band[band&TC_PRIO_MAX]]; } band = res.classid; } band = TC_H_MIN(band) - 1; if (band >= q->bands) - band = q->prio2band[0]; -out: - if (q->mq) - skb_set_queue_mapping(skb, band); + return q->queues[q->prio2band[0]]; + return q->queues[band]; } @@ -123,68 +118,23 @@ prio_requeue(struct sk_buff *skb, struct Qdisc* sch) } -static struct sk_buff * -prio_dequeue(struct Qdisc* sch) +static struct sk_buff *prio_dequeue(struct Qdisc* sch) { - struct sk_buff *skb; struct prio_sched_data *q = qdisc_priv(sch); int prio; - struct Qdisc *qdisc; for (prio = 0; prio < q->bands; prio++) { - /* Check if the target subqueue is available before - * pulling an skb. This way we avoid excessive requeues - * for slower queues. - */ - if (!__netif_subqueue_stopped(qdisc_dev(sch), - (q->mq ? prio : 0))) { - qdisc = q->queues[prio]; - skb = qdisc->dequeue(qdisc); - if (skb) { - sch->q.qlen--; - return skb; - } + struct Qdisc *qdisc = q->queues[prio]; + struct sk_buff *skb = qdisc->dequeue(qdisc); + if (skb) { + sch->q.qlen--; + return skb; } } return NULL; } -static struct sk_buff *rr_dequeue(struct Qdisc* sch) -{ - struct sk_buff *skb; - struct prio_sched_data *q = qdisc_priv(sch); - struct Qdisc *qdisc; - int bandcount; - - /* Only take one pass through the queues. If nothing is available, - * return nothing. - */ - for (bandcount = 0; bandcount < q->bands; bandcount++) { - /* Check if the target subqueue is available before - * pulling an skb. This way we avoid excessive requeues - * for slower queues. If the queue is stopped, try the - * next queue. - */ - if (!__netif_subqueue_stopped(qdisc_dev(sch), - (q->mq ? q->curband : 0))) { - qdisc = q->queues[q->curband]; - skb = qdisc->dequeue(qdisc); - if (skb) { - sch->q.qlen--; - q->curband++; - if (q->curband >= q->bands) - q->curband = 0; - return skb; - } - } - q->curband++; - if (q->curband >= q->bands) - q->curband = 0; - } - return NULL; -} - static unsigned int prio_drop(struct Qdisc* sch) { struct prio_sched_data *q = qdisc_priv(sch); @@ -229,45 +179,22 @@ static int prio_tune(struct Qdisc *sch, struct nlattr *opt) { struct prio_sched_data *q = qdisc_priv(sch); struct tc_prio_qopt *qopt; - struct nlattr *tb[TCA_PRIO_MAX + 1]; - int err; int i; - err = nla_parse_nested_compat(tb, TCA_PRIO_MAX, opt, NULL, qopt, - sizeof(*qopt)); - if (err < 0) - return err; - - q->bands = qopt->bands; - /* If we're multiqueue, make sure the number of incoming bands - * matches the number of queues on the device we're associating with. - * If the number of bands requested is zero, then set q->bands to - * dev->egress_subqueue_count. Also, the root qdisc must be the - * only one that is enabled for multiqueue, since it's the only one - * that interacts with the underlying device. - */ - q->mq = nla_get_flag(tb[TCA_PRIO_MQ]); - if (q->mq) { - if (sch->parent != TC_H_ROOT) - return -EINVAL; - if (netif_is_multiqueue(qdisc_dev(sch))) { - if (q->bands == 0) - q->bands = qdisc_dev(sch)->egress_subqueue_count; - else if (q->bands != qdisc_dev(sch)->egress_subqueue_count) - return -EINVAL; - } else - return -EOPNOTSUPP; - } + if (nla_len(opt) < sizeof(*qopt)) + return -EINVAL; + qopt = nla_data(opt); - if (q->bands > TCQ_PRIO_BANDS || q->bands < 2) + if (qopt->bands > TCQ_PRIO_BANDS || qopt->bands < 2) return -EINVAL; for (i=0; i<=TC_PRIO_MAX; i++) { - if (qopt->priomap[i] >= q->bands) + if (qopt->priomap[i] >= qopt->bands) return -EINVAL; } sch_tree_lock(sch); + q->bands = qopt->bands; memcpy(q->prio2band, qopt->priomap, TC_PRIO_MAX+1); for (i=q->bands; imq) { - if (nla_put_flag(skb, TCA_PRIO_MQ) < 0) - goto nla_put_failure; - } nla_nest_compat_end(skb, nest); return skb->len; @@ -509,44 +432,17 @@ static struct Qdisc_ops prio_qdisc_ops __read_mostly = { .owner = THIS_MODULE, }; -static struct Qdisc_ops rr_qdisc_ops __read_mostly = { - .next = NULL, - .cl_ops = &prio_class_ops, - .id = "rr", - .priv_size = sizeof(struct prio_sched_data), - .enqueue = prio_enqueue, - .dequeue = rr_dequeue, - .requeue = prio_requeue, - .drop = prio_drop, - .init = prio_init, - .reset = prio_reset, - .destroy = prio_destroy, - .change = prio_tune, - .dump = prio_dump, - .owner = THIS_MODULE, -}; - static int __init prio_module_init(void) { - int err; - - err = register_qdisc(&prio_qdisc_ops); - if (err < 0) - return err; - err = register_qdisc(&rr_qdisc_ops); - if (err < 0) - unregister_qdisc(&prio_qdisc_ops); - return err; + return register_qdisc(&prio_qdisc_ops); } static void __exit prio_module_exit(void) { unregister_qdisc(&prio_qdisc_ops); - unregister_qdisc(&rr_qdisc_ops); } module_init(prio_module_init) module_exit(prio_module_exit) MODULE_LICENSE("GPL"); -MODULE_ALIAS("sch_rr"); -- cgit v1.1 From 5f86173bdf15981ca49d0434f638b68f70a35644 Mon Sep 17 00:00:00 2001 From: Jussi Kivilinna Date: Sun, 20 Jul 2008 00:08:04 -0700 Subject: net_sched: Add qdisc_enqueue wrapper Signed-off-by: Jussi Kivilinna Signed-off-by: David S. Miller --- net/sched/sch_prio.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net/sched/sch_prio.c') diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c index 536ca47..d29c2f8 100644 --- a/net/sched/sch_prio.c +++ b/net/sched/sch_prio.c @@ -81,7 +81,8 @@ prio_enqueue(struct sk_buff *skb, struct Qdisc *sch) } #endif - if ((ret = qdisc->enqueue(skb, qdisc)) == NET_XMIT_SUCCESS) { + ret = qdisc_enqueue(skb, qdisc); + if (ret == NET_XMIT_SUCCESS) { sch->bstats.bytes += skb->len; sch->bstats.packets++; sch->q.qlen++; -- cgit v1.1 From 0abf77e55a2459aa9905be4b226e4729d5b4f0cb Mon Sep 17 00:00:00 2001 From: Jussi Kivilinna Date: Sun, 20 Jul 2008 00:08:27 -0700 Subject: net_sched: Add accessor function for packet length for qdiscs Signed-off-by: Jussi Kivilinna Signed-off-by: David S. Miller --- net/sched/sch_prio.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/sched/sch_prio.c') diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c index d29c2f8..f849243 100644 --- a/net/sched/sch_prio.c +++ b/net/sched/sch_prio.c @@ -83,7 +83,7 @@ prio_enqueue(struct sk_buff *skb, struct Qdisc *sch) ret = qdisc_enqueue(skb, qdisc); if (ret == NET_XMIT_SUCCESS) { - sch->bstats.bytes += skb->len; + sch->bstats.bytes += qdisc_pkt_len(skb); sch->bstats.packets++; sch->q.qlen++; return NET_XMIT_SUCCESS; -- cgit v1.1