OpenCloudOS-Kernel/net/sched/sch_generic.c

850 lines
20 KiB
C
Raw Normal View History

/*
* net/sched/sch_generic.c Generic packet scheduler routines.
*
* This program 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 of the License, or (at your option) any later version.
*
* Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
* Jamal Hadi Salim, <hadi@cyberus.ca> 990601
* - Ingress support
*/
#include <linux/bitops.h>
#include <linux/module.h>
#include <linux/types.h>
#include <linux/kernel.h>
#include <linux/sched.h>
#include <linux/string.h>
#include <linux/errno.h>
#include <linux/netdevice.h>
#include <linux/skbuff.h>
#include <linux/rtnetlink.h>
#include <linux/init.h>
#include <linux/rcupdate.h>
#include <linux/list.h>
#include <net/pkt_sched.h>
/* Main transmission queue. */
/* Modifications to data participating in scheduling must be protected with
* qdisc_lock(qdisc) spinlock.
*
* The idea is the following:
* - enqueue, dequeue are serialized via qdisc root lock
* - ingress filtering is also serialized via qdisc root lock
* - updates to tree and tree walking are only done under the rtnl mutex.
*/
static inline int dev_requeue_skb(struct sk_buff *skb, struct Qdisc *q)
{
q->gso_skb = skb;
q->qstats.requeues++;
net: Avoid enqueuing skb for default qdiscs dev_queue_xmit enqueue's a skb and calls qdisc_run which dequeue's the skb and xmits it. In most cases, the skb that is enqueue'd is the same one that is dequeue'd (unless the queue gets stopped or multiple cpu's write to the same queue and ends in a race with qdisc_run). For default qdiscs, we can remove the redundant enqueue/dequeue and simply xmit the skb since the default qdisc is work-conserving. The patch uses a new flag - TCQ_F_CAN_BYPASS to identify the default fast queue. The controversial part of the patch is incrementing qlen when a skb is requeued - this is to avoid checks like the second line below: + } else if ((q->flags & TCQ_F_CAN_BYPASS) && !qdisc_qlen(q) && >> !q->gso_skb && + !test_and_set_bit(__QDISC_STATE_RUNNING, &q->state)) { Results of a 2 hour testing for multiple netperf sessions (1, 2, 4, 8, 12 sessions on a 4 cpu system-X). The BW numbers are aggregate Mb/s across iterations tested with this version on System-X boxes with Chelsio 10gbps cards: ---------------------------------- Size | ORG BW NEW BW | ---------------------------------- 128K | 156964 159381 | 256K | 158650 162042 | ---------------------------------- Changes from ver1: 1. Move sch_direct_xmit declaration from sch_generic.h to pkt_sched.h 2. Update qdisc basic statistics for direct xmit path. 3. Set qlen to zero in qdisc_reset. 4. Changed some function names to more meaningful ones. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-08-06 09:44:21 +08:00
q->q.qlen++; /* it's still part of the queue */
__netif_schedule(q);
return 0;
}
static inline struct sk_buff *dequeue_skb(struct Qdisc *q)
{
struct sk_buff *skb = q->gso_skb;
if (unlikely(skb)) {
struct net_device *dev = qdisc_dev(q);
struct netdev_queue *txq;
/* check the reason of requeuing without tx lock first */
txq = netdev_get_tx_queue(dev, skb_get_queue_mapping(skb));
net: Avoid enqueuing skb for default qdiscs dev_queue_xmit enqueue's a skb and calls qdisc_run which dequeue's the skb and xmits it. In most cases, the skb that is enqueue'd is the same one that is dequeue'd (unless the queue gets stopped or multiple cpu's write to the same queue and ends in a race with qdisc_run). For default qdiscs, we can remove the redundant enqueue/dequeue and simply xmit the skb since the default qdisc is work-conserving. The patch uses a new flag - TCQ_F_CAN_BYPASS to identify the default fast queue. The controversial part of the patch is incrementing qlen when a skb is requeued - this is to avoid checks like the second line below: + } else if ((q->flags & TCQ_F_CAN_BYPASS) && !qdisc_qlen(q) && >> !q->gso_skb && + !test_and_set_bit(__QDISC_STATE_RUNNING, &q->state)) { Results of a 2 hour testing for multiple netperf sessions (1, 2, 4, 8, 12 sessions on a 4 cpu system-X). The BW numbers are aggregate Mb/s across iterations tested with this version on System-X boxes with Chelsio 10gbps cards: ---------------------------------- Size | ORG BW NEW BW | ---------------------------------- 128K | 156964 159381 | 256K | 158650 162042 | ---------------------------------- Changes from ver1: 1. Move sch_direct_xmit declaration from sch_generic.h to pkt_sched.h 2. Update qdisc basic statistics for direct xmit path. 3. Set qlen to zero in qdisc_reset. 4. Changed some function names to more meaningful ones. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-08-06 09:44:21 +08:00
if (!netif_tx_queue_stopped(txq) &&
!netif_tx_queue_frozen(txq)) {
q->gso_skb = NULL;
net: Avoid enqueuing skb for default qdiscs dev_queue_xmit enqueue's a skb and calls qdisc_run which dequeue's the skb and xmits it. In most cases, the skb that is enqueue'd is the same one that is dequeue'd (unless the queue gets stopped or multiple cpu's write to the same queue and ends in a race with qdisc_run). For default qdiscs, we can remove the redundant enqueue/dequeue and simply xmit the skb since the default qdisc is work-conserving. The patch uses a new flag - TCQ_F_CAN_BYPASS to identify the default fast queue. The controversial part of the patch is incrementing qlen when a skb is requeued - this is to avoid checks like the second line below: + } else if ((q->flags & TCQ_F_CAN_BYPASS) && !qdisc_qlen(q) && >> !q->gso_skb && + !test_and_set_bit(__QDISC_STATE_RUNNING, &q->state)) { Results of a 2 hour testing for multiple netperf sessions (1, 2, 4, 8, 12 sessions on a 4 cpu system-X). The BW numbers are aggregate Mb/s across iterations tested with this version on System-X boxes with Chelsio 10gbps cards: ---------------------------------- Size | ORG BW NEW BW | ---------------------------------- 128K | 156964 159381 | 256K | 158650 162042 | ---------------------------------- Changes from ver1: 1. Move sch_direct_xmit declaration from sch_generic.h to pkt_sched.h 2. Update qdisc basic statistics for direct xmit path. 3. Set qlen to zero in qdisc_reset. 4. Changed some function names to more meaningful ones. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-08-06 09:44:21 +08:00
q->q.qlen--;
} else
skb = NULL;
} else {
skb = q->dequeue(q);
}
return skb;
}
[NET]: qdisc_restart - readability changes plus one bug fix. New changes : - Incorporated Peter Waskiewicz's comments. - Re-added back one warning message (on driver returning wrong value). Previous changes : - Converted to use switch/case code which looks neater. - "if (ret == NETDEV_TX_LOCKED && lockless)" is buggy, and the lockless check should be removed, since driver will return NETDEV_TX_LOCKED only if lockless is true and driver has to do the locking. In the original code as well as the latest code, this code can result in a bug where if LLTX is not set for a driver (lockless == 0) but the driver is written wrongly to do a trylock (despite LLTX being set), the driver returns LOCKED. But since lockless is zero, the packet is requeue'd instead of calling collision code which will issue warning and free up the skb. Instead this skb will be retried with this driver next time, and the same result will ensue. Removing this check will catch these driver bugs instead of hiding the problem. I am keeping this change to readability section since : a. it is confusing to check two things as it is; and b. it is difficult to keep this check in the changed 'switch' code. - Changed some names, like try_get_tx_pkt to dev_dequeue_skb (as that is the work being done and easier to understand) and do_dev_requeue to dev_requeue_skb, merged handle_dev_cpu_collision and tx_islocked to dev_handle_collision (handle_dev_cpu_collision is a small routine with only one caller, so there is no need to have two separate routines which also results in getting rid of two macros, etc. - Removed an XXX comment as it should never fail (I suspect this was related to batch skb WIP, Jamal ?). Converted some functions to original coding style of having the return values and the function name on same line, eg prio2list. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-06-25 10:56:09 +08:00
static inline int handle_dev_cpu_collision(struct sk_buff *skb,
struct netdev_queue *dev_queue,
[NET]: qdisc_restart - readability changes plus one bug fix. New changes : - Incorporated Peter Waskiewicz's comments. - Re-added back one warning message (on driver returning wrong value). Previous changes : - Converted to use switch/case code which looks neater. - "if (ret == NETDEV_TX_LOCKED && lockless)" is buggy, and the lockless check should be removed, since driver will return NETDEV_TX_LOCKED only if lockless is true and driver has to do the locking. In the original code as well as the latest code, this code can result in a bug where if LLTX is not set for a driver (lockless == 0) but the driver is written wrongly to do a trylock (despite LLTX being set), the driver returns LOCKED. But since lockless is zero, the packet is requeue'd instead of calling collision code which will issue warning and free up the skb. Instead this skb will be retried with this driver next time, and the same result will ensue. Removing this check will catch these driver bugs instead of hiding the problem. I am keeping this change to readability section since : a. it is confusing to check two things as it is; and b. it is difficult to keep this check in the changed 'switch' code. - Changed some names, like try_get_tx_pkt to dev_dequeue_skb (as that is the work being done and easier to understand) and do_dev_requeue to dev_requeue_skb, merged handle_dev_cpu_collision and tx_islocked to dev_handle_collision (handle_dev_cpu_collision is a small routine with only one caller, so there is no need to have two separate routines which also results in getting rid of two macros, etc. - Removed an XXX comment as it should never fail (I suspect this was related to batch skb WIP, Jamal ?). Converted some functions to original coding style of having the return values and the function name on same line, eg prio2list. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-06-25 10:56:09 +08:00
struct Qdisc *q)
{
[NET]: qdisc_restart - readability changes plus one bug fix. New changes : - Incorporated Peter Waskiewicz's comments. - Re-added back one warning message (on driver returning wrong value). Previous changes : - Converted to use switch/case code which looks neater. - "if (ret == NETDEV_TX_LOCKED && lockless)" is buggy, and the lockless check should be removed, since driver will return NETDEV_TX_LOCKED only if lockless is true and driver has to do the locking. In the original code as well as the latest code, this code can result in a bug where if LLTX is not set for a driver (lockless == 0) but the driver is written wrongly to do a trylock (despite LLTX being set), the driver returns LOCKED. But since lockless is zero, the packet is requeue'd instead of calling collision code which will issue warning and free up the skb. Instead this skb will be retried with this driver next time, and the same result will ensue. Removing this check will catch these driver bugs instead of hiding the problem. I am keeping this change to readability section since : a. it is confusing to check two things as it is; and b. it is difficult to keep this check in the changed 'switch' code. - Changed some names, like try_get_tx_pkt to dev_dequeue_skb (as that is the work being done and easier to understand) and do_dev_requeue to dev_requeue_skb, merged handle_dev_cpu_collision and tx_islocked to dev_handle_collision (handle_dev_cpu_collision is a small routine with only one caller, so there is no need to have two separate routines which also results in getting rid of two macros, etc. - Removed an XXX comment as it should never fail (I suspect this was related to batch skb WIP, Jamal ?). Converted some functions to original coding style of having the return values and the function name on same line, eg prio2list. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-06-25 10:56:09 +08:00
int ret;
if (unlikely(dev_queue->xmit_lock_owner == smp_processor_id())) {
[NET]: qdisc_restart - readability changes plus one bug fix. New changes : - Incorporated Peter Waskiewicz's comments. - Re-added back one warning message (on driver returning wrong value). Previous changes : - Converted to use switch/case code which looks neater. - "if (ret == NETDEV_TX_LOCKED && lockless)" is buggy, and the lockless check should be removed, since driver will return NETDEV_TX_LOCKED only if lockless is true and driver has to do the locking. In the original code as well as the latest code, this code can result in a bug where if LLTX is not set for a driver (lockless == 0) but the driver is written wrongly to do a trylock (despite LLTX being set), the driver returns LOCKED. But since lockless is zero, the packet is requeue'd instead of calling collision code which will issue warning and free up the skb. Instead this skb will be retried with this driver next time, and the same result will ensue. Removing this check will catch these driver bugs instead of hiding the problem. I am keeping this change to readability section since : a. it is confusing to check two things as it is; and b. it is difficult to keep this check in the changed 'switch' code. - Changed some names, like try_get_tx_pkt to dev_dequeue_skb (as that is the work being done and easier to understand) and do_dev_requeue to dev_requeue_skb, merged handle_dev_cpu_collision and tx_islocked to dev_handle_collision (handle_dev_cpu_collision is a small routine with only one caller, so there is no need to have two separate routines which also results in getting rid of two macros, etc. - Removed an XXX comment as it should never fail (I suspect this was related to batch skb WIP, Jamal ?). Converted some functions to original coding style of having the return values and the function name on same line, eg prio2list. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-06-25 10:56:09 +08:00
/*
* Same CPU holding the lock. It may be a transient
* configuration error, when hard_start_xmit() recurses. We
* detect it by checking xmit owner and drop the packet when
* deadloop is detected. Return OK to try the next skb.
*/
kfree_skb(skb);
[NET]: qdisc_restart - readability changes plus one bug fix. New changes : - Incorporated Peter Waskiewicz's comments. - Re-added back one warning message (on driver returning wrong value). Previous changes : - Converted to use switch/case code which looks neater. - "if (ret == NETDEV_TX_LOCKED && lockless)" is buggy, and the lockless check should be removed, since driver will return NETDEV_TX_LOCKED only if lockless is true and driver has to do the locking. In the original code as well as the latest code, this code can result in a bug where if LLTX is not set for a driver (lockless == 0) but the driver is written wrongly to do a trylock (despite LLTX being set), the driver returns LOCKED. But since lockless is zero, the packet is requeue'd instead of calling collision code which will issue warning and free up the skb. Instead this skb will be retried with this driver next time, and the same result will ensue. Removing this check will catch these driver bugs instead of hiding the problem. I am keeping this change to readability section since : a. it is confusing to check two things as it is; and b. it is difficult to keep this check in the changed 'switch' code. - Changed some names, like try_get_tx_pkt to dev_dequeue_skb (as that is the work being done and easier to understand) and do_dev_requeue to dev_requeue_skb, merged handle_dev_cpu_collision and tx_islocked to dev_handle_collision (handle_dev_cpu_collision is a small routine with only one caller, so there is no need to have two separate routines which also results in getting rid of two macros, etc. - Removed an XXX comment as it should never fail (I suspect this was related to batch skb WIP, Jamal ?). Converted some functions to original coding style of having the return values and the function name on same line, eg prio2list. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-06-25 10:56:09 +08:00
if (net_ratelimit())
printk(KERN_WARNING "Dead loop on netdevice %s, "
"fix it urgently!\n", dev_queue->dev->name);
[NET]: qdisc_restart - readability changes plus one bug fix. New changes : - Incorporated Peter Waskiewicz's comments. - Re-added back one warning message (on driver returning wrong value). Previous changes : - Converted to use switch/case code which looks neater. - "if (ret == NETDEV_TX_LOCKED && lockless)" is buggy, and the lockless check should be removed, since driver will return NETDEV_TX_LOCKED only if lockless is true and driver has to do the locking. In the original code as well as the latest code, this code can result in a bug where if LLTX is not set for a driver (lockless == 0) but the driver is written wrongly to do a trylock (despite LLTX being set), the driver returns LOCKED. But since lockless is zero, the packet is requeue'd instead of calling collision code which will issue warning and free up the skb. Instead this skb will be retried with this driver next time, and the same result will ensue. Removing this check will catch these driver bugs instead of hiding the problem. I am keeping this change to readability section since : a. it is confusing to check two things as it is; and b. it is difficult to keep this check in the changed 'switch' code. - Changed some names, like try_get_tx_pkt to dev_dequeue_skb (as that is the work being done and easier to understand) and do_dev_requeue to dev_requeue_skb, merged handle_dev_cpu_collision and tx_islocked to dev_handle_collision (handle_dev_cpu_collision is a small routine with only one caller, so there is no need to have two separate routines which also results in getting rid of two macros, etc. - Removed an XXX comment as it should never fail (I suspect this was related to batch skb WIP, Jamal ?). Converted some functions to original coding style of having the return values and the function name on same line, eg prio2list. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-06-25 10:56:09 +08:00
ret = qdisc_qlen(q);
} else {
/*
* Another cpu is holding lock, requeue & delay xmits for
* some time.
*/
__get_cpu_var(netdev_rx_stat).cpu_collision++;
ret = dev_requeue_skb(skb, q);
}
[NET]: qdisc_restart - readability changes plus one bug fix. New changes : - Incorporated Peter Waskiewicz's comments. - Re-added back one warning message (on driver returning wrong value). Previous changes : - Converted to use switch/case code which looks neater. - "if (ret == NETDEV_TX_LOCKED && lockless)" is buggy, and the lockless check should be removed, since driver will return NETDEV_TX_LOCKED only if lockless is true and driver has to do the locking. In the original code as well as the latest code, this code can result in a bug where if LLTX is not set for a driver (lockless == 0) but the driver is written wrongly to do a trylock (despite LLTX being set), the driver returns LOCKED. But since lockless is zero, the packet is requeue'd instead of calling collision code which will issue warning and free up the skb. Instead this skb will be retried with this driver next time, and the same result will ensue. Removing this check will catch these driver bugs instead of hiding the problem. I am keeping this change to readability section since : a. it is confusing to check two things as it is; and b. it is difficult to keep this check in the changed 'switch' code. - Changed some names, like try_get_tx_pkt to dev_dequeue_skb (as that is the work being done and easier to understand) and do_dev_requeue to dev_requeue_skb, merged handle_dev_cpu_collision and tx_islocked to dev_handle_collision (handle_dev_cpu_collision is a small routine with only one caller, so there is no need to have two separate routines which also results in getting rid of two macros, etc. - Removed an XXX comment as it should never fail (I suspect this was related to batch skb WIP, Jamal ?). Converted some functions to original coding style of having the return values and the function name on same line, eg prio2list. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-06-25 10:56:09 +08:00
return ret;
}
/*
net: Avoid enqueuing skb for default qdiscs dev_queue_xmit enqueue's a skb and calls qdisc_run which dequeue's the skb and xmits it. In most cases, the skb that is enqueue'd is the same one that is dequeue'd (unless the queue gets stopped or multiple cpu's write to the same queue and ends in a race with qdisc_run). For default qdiscs, we can remove the redundant enqueue/dequeue and simply xmit the skb since the default qdisc is work-conserving. The patch uses a new flag - TCQ_F_CAN_BYPASS to identify the default fast queue. The controversial part of the patch is incrementing qlen when a skb is requeued - this is to avoid checks like the second line below: + } else if ((q->flags & TCQ_F_CAN_BYPASS) && !qdisc_qlen(q) && >> !q->gso_skb && + !test_and_set_bit(__QDISC_STATE_RUNNING, &q->state)) { Results of a 2 hour testing for multiple netperf sessions (1, 2, 4, 8, 12 sessions on a 4 cpu system-X). The BW numbers are aggregate Mb/s across iterations tested with this version on System-X boxes with Chelsio 10gbps cards: ---------------------------------- Size | ORG BW NEW BW | ---------------------------------- 128K | 156964 159381 | 256K | 158650 162042 | ---------------------------------- Changes from ver1: 1. Move sch_direct_xmit declaration from sch_generic.h to pkt_sched.h 2. Update qdisc basic statistics for direct xmit path. 3. Set qlen to zero in qdisc_reset. 4. Changed some function names to more meaningful ones. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-08-06 09:44:21 +08:00
* Transmit one skb, and handle the return status as required. Holding the
* __QDISC_STATE_RUNNING bit guarantees that only one CPU can execute this
* function.
[NET]: qdisc_restart - readability changes plus one bug fix. New changes : - Incorporated Peter Waskiewicz's comments. - Re-added back one warning message (on driver returning wrong value). Previous changes : - Converted to use switch/case code which looks neater. - "if (ret == NETDEV_TX_LOCKED && lockless)" is buggy, and the lockless check should be removed, since driver will return NETDEV_TX_LOCKED only if lockless is true and driver has to do the locking. In the original code as well as the latest code, this code can result in a bug where if LLTX is not set for a driver (lockless == 0) but the driver is written wrongly to do a trylock (despite LLTX being set), the driver returns LOCKED. But since lockless is zero, the packet is requeue'd instead of calling collision code which will issue warning and free up the skb. Instead this skb will be retried with this driver next time, and the same result will ensue. Removing this check will catch these driver bugs instead of hiding the problem. I am keeping this change to readability section since : a. it is confusing to check two things as it is; and b. it is difficult to keep this check in the changed 'switch' code. - Changed some names, like try_get_tx_pkt to dev_dequeue_skb (as that is the work being done and easier to understand) and do_dev_requeue to dev_requeue_skb, merged handle_dev_cpu_collision and tx_islocked to dev_handle_collision (handle_dev_cpu_collision is a small routine with only one caller, so there is no need to have two separate routines which also results in getting rid of two macros, etc. - Removed an XXX comment as it should never fail (I suspect this was related to batch skb WIP, Jamal ?). Converted some functions to original coding style of having the return values and the function name on same line, eg prio2list. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-06-25 10:56:09 +08:00
*
* Returns to the caller:
* 0 - queue is empty or throttled.
* >0 - queue is not empty.
*/
net: Avoid enqueuing skb for default qdiscs dev_queue_xmit enqueue's a skb and calls qdisc_run which dequeue's the skb and xmits it. In most cases, the skb that is enqueue'd is the same one that is dequeue'd (unless the queue gets stopped or multiple cpu's write to the same queue and ends in a race with qdisc_run). For default qdiscs, we can remove the redundant enqueue/dequeue and simply xmit the skb since the default qdisc is work-conserving. The patch uses a new flag - TCQ_F_CAN_BYPASS to identify the default fast queue. The controversial part of the patch is incrementing qlen when a skb is requeued - this is to avoid checks like the second line below: + } else if ((q->flags & TCQ_F_CAN_BYPASS) && !qdisc_qlen(q) && >> !q->gso_skb && + !test_and_set_bit(__QDISC_STATE_RUNNING, &q->state)) { Results of a 2 hour testing for multiple netperf sessions (1, 2, 4, 8, 12 sessions on a 4 cpu system-X). The BW numbers are aggregate Mb/s across iterations tested with this version on System-X boxes with Chelsio 10gbps cards: ---------------------------------- Size | ORG BW NEW BW | ---------------------------------- 128K | 156964 159381 | 256K | 158650 162042 | ---------------------------------- Changes from ver1: 1. Move sch_direct_xmit declaration from sch_generic.h to pkt_sched.h 2. Update qdisc basic statistics for direct xmit path. 3. Set qlen to zero in qdisc_reset. 4. Changed some function names to more meaningful ones. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-08-06 09:44:21 +08:00
int sch_direct_xmit(struct sk_buff *skb, struct Qdisc *q,
struct net_device *dev, struct netdev_queue *txq,
spinlock_t *root_lock)
{
int ret = NETDEV_TX_BUSY;
/* And release qdisc */
spin_unlock(root_lock);
HARD_TX_LOCK(dev, txq, smp_processor_id());
if (!netif_tx_queue_stopped(txq) &&
!netif_tx_queue_frozen(txq)) {
ret = dev_hard_start_xmit(skb, dev, txq);
/* an error implies that the skb was consumed */
if (ret < 0)
ret = NETDEV_TX_OK;
/* all NET_XMIT codes map to NETDEV_TX_OK */
ret &= ~NET_XMIT_MASK;
}
HARD_TX_UNLOCK(dev, txq);
spin_lock(root_lock);
[NET]: qdisc_restart - readability changes plus one bug fix. New changes : - Incorporated Peter Waskiewicz's comments. - Re-added back one warning message (on driver returning wrong value). Previous changes : - Converted to use switch/case code which looks neater. - "if (ret == NETDEV_TX_LOCKED && lockless)" is buggy, and the lockless check should be removed, since driver will return NETDEV_TX_LOCKED only if lockless is true and driver has to do the locking. In the original code as well as the latest code, this code can result in a bug where if LLTX is not set for a driver (lockless == 0) but the driver is written wrongly to do a trylock (despite LLTX being set), the driver returns LOCKED. But since lockless is zero, the packet is requeue'd instead of calling collision code which will issue warning and free up the skb. Instead this skb will be retried with this driver next time, and the same result will ensue. Removing this check will catch these driver bugs instead of hiding the problem. I am keeping this change to readability section since : a. it is confusing to check two things as it is; and b. it is difficult to keep this check in the changed 'switch' code. - Changed some names, like try_get_tx_pkt to dev_dequeue_skb (as that is the work being done and easier to understand) and do_dev_requeue to dev_requeue_skb, merged handle_dev_cpu_collision and tx_islocked to dev_handle_collision (handle_dev_cpu_collision is a small routine with only one caller, so there is no need to have two separate routines which also results in getting rid of two macros, etc. - Removed an XXX comment as it should never fail (I suspect this was related to batch skb WIP, Jamal ?). Converted some functions to original coding style of having the return values and the function name on same line, eg prio2list. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-06-25 10:56:09 +08:00
switch (ret) {
case NETDEV_TX_OK:
/* Driver sent out skb successfully */
ret = qdisc_qlen(q);
break;
case NETDEV_TX_LOCKED:
/* Driver try lock failed */
ret = handle_dev_cpu_collision(skb, txq, q);
[NET]: qdisc_restart - readability changes plus one bug fix. New changes : - Incorporated Peter Waskiewicz's comments. - Re-added back one warning message (on driver returning wrong value). Previous changes : - Converted to use switch/case code which looks neater. - "if (ret == NETDEV_TX_LOCKED && lockless)" is buggy, and the lockless check should be removed, since driver will return NETDEV_TX_LOCKED only if lockless is true and driver has to do the locking. In the original code as well as the latest code, this code can result in a bug where if LLTX is not set for a driver (lockless == 0) but the driver is written wrongly to do a trylock (despite LLTX being set), the driver returns LOCKED. But since lockless is zero, the packet is requeue'd instead of calling collision code which will issue warning and free up the skb. Instead this skb will be retried with this driver next time, and the same result will ensue. Removing this check will catch these driver bugs instead of hiding the problem. I am keeping this change to readability section since : a. it is confusing to check two things as it is; and b. it is difficult to keep this check in the changed 'switch' code. - Changed some names, like try_get_tx_pkt to dev_dequeue_skb (as that is the work being done and easier to understand) and do_dev_requeue to dev_requeue_skb, merged handle_dev_cpu_collision and tx_islocked to dev_handle_collision (handle_dev_cpu_collision is a small routine with only one caller, so there is no need to have two separate routines which also results in getting rid of two macros, etc. - Removed an XXX comment as it should never fail (I suspect this was related to batch skb WIP, Jamal ?). Converted some functions to original coding style of having the return values and the function name on same line, eg prio2list. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-06-25 10:56:09 +08:00
break;
default:
/* Driver returned NETDEV_TX_BUSY - requeue skb */
if (unlikely (ret != NETDEV_TX_BUSY && net_ratelimit()))
printk(KERN_WARNING "BUG %s code %d qlen %d\n",
dev->name, ret, q->q.qlen);
ret = dev_requeue_skb(skb, q);
[NET]: qdisc_restart - readability changes plus one bug fix. New changes : - Incorporated Peter Waskiewicz's comments. - Re-added back one warning message (on driver returning wrong value). Previous changes : - Converted to use switch/case code which looks neater. - "if (ret == NETDEV_TX_LOCKED && lockless)" is buggy, and the lockless check should be removed, since driver will return NETDEV_TX_LOCKED only if lockless is true and driver has to do the locking. In the original code as well as the latest code, this code can result in a bug where if LLTX is not set for a driver (lockless == 0) but the driver is written wrongly to do a trylock (despite LLTX being set), the driver returns LOCKED. But since lockless is zero, the packet is requeue'd instead of calling collision code which will issue warning and free up the skb. Instead this skb will be retried with this driver next time, and the same result will ensue. Removing this check will catch these driver bugs instead of hiding the problem. I am keeping this change to readability section since : a. it is confusing to check two things as it is; and b. it is difficult to keep this check in the changed 'switch' code. - Changed some names, like try_get_tx_pkt to dev_dequeue_skb (as that is the work being done and easier to understand) and do_dev_requeue to dev_requeue_skb, merged handle_dev_cpu_collision and tx_islocked to dev_handle_collision (handle_dev_cpu_collision is a small routine with only one caller, so there is no need to have two separate routines which also results in getting rid of two macros, etc. - Removed an XXX comment as it should never fail (I suspect this was related to batch skb WIP, Jamal ?). Converted some functions to original coding style of having the return values and the function name on same line, eg prio2list. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-06-25 10:56:09 +08:00
break;
}
if (ret && (netif_tx_queue_stopped(txq) ||
netif_tx_queue_frozen(txq)))
ret = 0;
[NET]: qdisc_restart - readability changes plus one bug fix. New changes : - Incorporated Peter Waskiewicz's comments. - Re-added back one warning message (on driver returning wrong value). Previous changes : - Converted to use switch/case code which looks neater. - "if (ret == NETDEV_TX_LOCKED && lockless)" is buggy, and the lockless check should be removed, since driver will return NETDEV_TX_LOCKED only if lockless is true and driver has to do the locking. In the original code as well as the latest code, this code can result in a bug where if LLTX is not set for a driver (lockless == 0) but the driver is written wrongly to do a trylock (despite LLTX being set), the driver returns LOCKED. But since lockless is zero, the packet is requeue'd instead of calling collision code which will issue warning and free up the skb. Instead this skb will be retried with this driver next time, and the same result will ensue. Removing this check will catch these driver bugs instead of hiding the problem. I am keeping this change to readability section since : a. it is confusing to check two things as it is; and b. it is difficult to keep this check in the changed 'switch' code. - Changed some names, like try_get_tx_pkt to dev_dequeue_skb (as that is the work being done and easier to understand) and do_dev_requeue to dev_requeue_skb, merged handle_dev_cpu_collision and tx_islocked to dev_handle_collision (handle_dev_cpu_collision is a small routine with only one caller, so there is no need to have two separate routines which also results in getting rid of two macros, etc. - Removed an XXX comment as it should never fail (I suspect this was related to batch skb WIP, Jamal ?). Converted some functions to original coding style of having the return values and the function name on same line, eg prio2list. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-06-25 10:56:09 +08:00
return ret;
}
net: Avoid enqueuing skb for default qdiscs dev_queue_xmit enqueue's a skb and calls qdisc_run which dequeue's the skb and xmits it. In most cases, the skb that is enqueue'd is the same one that is dequeue'd (unless the queue gets stopped or multiple cpu's write to the same queue and ends in a race with qdisc_run). For default qdiscs, we can remove the redundant enqueue/dequeue and simply xmit the skb since the default qdisc is work-conserving. The patch uses a new flag - TCQ_F_CAN_BYPASS to identify the default fast queue. The controversial part of the patch is incrementing qlen when a skb is requeued - this is to avoid checks like the second line below: + } else if ((q->flags & TCQ_F_CAN_BYPASS) && !qdisc_qlen(q) && >> !q->gso_skb && + !test_and_set_bit(__QDISC_STATE_RUNNING, &q->state)) { Results of a 2 hour testing for multiple netperf sessions (1, 2, 4, 8, 12 sessions on a 4 cpu system-X). The BW numbers are aggregate Mb/s across iterations tested with this version on System-X boxes with Chelsio 10gbps cards: ---------------------------------- Size | ORG BW NEW BW | ---------------------------------- 128K | 156964 159381 | 256K | 158650 162042 | ---------------------------------- Changes from ver1: 1. Move sch_direct_xmit declaration from sch_generic.h to pkt_sched.h 2. Update qdisc basic statistics for direct xmit path. 3. Set qlen to zero in qdisc_reset. 4. Changed some function names to more meaningful ones. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-08-06 09:44:21 +08:00
/*
* NOTE: Called under qdisc_lock(q) with locally disabled BH.
*
* __QDISC_STATE_RUNNING guarantees only one CPU can process
* this qdisc at a time. qdisc_lock(q) serializes queue accesses for
* this queue.
*
* netif_tx_lock serializes accesses to device driver.
*
* qdisc_lock(q) and netif_tx_lock are mutually exclusive,
* if one is grabbed, another must be free.
*
* Note, that this procedure can be called by a watchdog timer
*
* Returns to the caller:
* 0 - queue is empty or throttled.
* >0 - queue is not empty.
*
*/
static inline int qdisc_restart(struct Qdisc *q)
{
struct netdev_queue *txq;
struct net_device *dev;
spinlock_t *root_lock;
struct sk_buff *skb;
/* Dequeue packet */
skb = dequeue_skb(q);
if (unlikely(!skb))
return 0;
root_lock = qdisc_lock(q);
dev = qdisc_dev(q);
txq = netdev_get_tx_queue(dev, skb_get_queue_mapping(skb));
return sch_direct_xmit(skb, q, dev, txq, root_lock);
}
void __qdisc_run(struct Qdisc *q)
{
unsigned long start_time = jiffies;
while (qdisc_restart(q)) {
/*
* Postpone processing if
* 1. another process needs the CPU;
* 2. we've been doing it for too long.
*/
if (need_resched() || jiffies != start_time) {
__netif_schedule(q);
break;
}
}
clear_bit(__QDISC_STATE_RUNNING, &q->state);
}
unsigned long dev_trans_start(struct net_device *dev)
{
unsigned long val, res = dev->trans_start;
unsigned int i;
for (i = 0; i < dev->num_tx_queues; i++) {
val = netdev_get_tx_queue(dev, i)->trans_start;
if (val && time_after(val, res))
res = val;
}
dev->trans_start = res;
return res;
}
EXPORT_SYMBOL(dev_trans_start);
static void dev_watchdog(unsigned long arg)
{
struct net_device *dev = (struct net_device *)arg;
[NET]: Add netif_tx_lock Various drivers use xmit_lock internally to synchronise with their transmission routines. They do so without setting xmit_lock_owner. This is fine as long as netpoll is not in use. With netpoll it is possible for deadlocks to occur if xmit_lock_owner isn't set. This is because if a printk occurs while xmit_lock is held and xmit_lock_owner is not set can cause netpoll to attempt to take xmit_lock recursively. While it is possible to resolve this by getting netpoll to use trylock, it is suboptimal because netpoll's sole objective is to maximise the chance of getting the printk out on the wire. So delaying or dropping the message is to be avoided as much as possible. So the only alternative is to always set xmit_lock_owner. The following patch does this by introducing the netif_tx_lock family of functions that take care of setting/unsetting xmit_lock_owner. I renamed xmit_lock to _xmit_lock to indicate that it should not be used directly. I didn't provide irq versions of the netif_tx_lock functions since xmit_lock is meant to be a BH-disabling lock. This is pretty much a straight text substitution except for a small bug fix in winbond. It currently uses netif_stop_queue/spin_unlock_wait to stop transmission. This is unsafe as an IRQ can potentially wake up the queue. So it is safer to use netif_tx_disable. The hamradio bits used spin_lock_irq but it is unnecessary as xmit_lock must never be taken in an IRQ handler. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-10 03:20:56 +08:00
netif_tx_lock(dev);
if (!qdisc_tx_is_noop(dev)) {
if (netif_device_present(dev) &&
netif_running(dev) &&
netif_carrier_ok(dev)) {
int some_queue_timedout = 0;
unsigned int i;
unsigned long trans_start;
for (i = 0; i < dev->num_tx_queues; i++) {
struct netdev_queue *txq;
txq = netdev_get_tx_queue(dev, i);
/*
* old device drivers set dev->trans_start
*/
trans_start = txq->trans_start ? : dev->trans_start;
if (netif_tx_queue_stopped(txq) &&
time_after(jiffies, (trans_start +
dev->watchdog_timeo))) {
some_queue_timedout = 1;
break;
}
}
if (some_queue_timedout) {
char drivername[64];
WARN_ONCE(1, KERN_INFO "NETDEV WATCHDOG: %s (%s): transmit queue %u timed out\n",
dev->name, netdev_drivername(dev, drivername, 64), i);
dev->netdev_ops->ndo_tx_timeout(dev);
}
if (!mod_timer(&dev->watchdog_timer,
round_jiffies(jiffies +
dev->watchdog_timeo)))
dev_hold(dev);
}
}
[NET]: Add netif_tx_lock Various drivers use xmit_lock internally to synchronise with their transmission routines. They do so without setting xmit_lock_owner. This is fine as long as netpoll is not in use. With netpoll it is possible for deadlocks to occur if xmit_lock_owner isn't set. This is because if a printk occurs while xmit_lock is held and xmit_lock_owner is not set can cause netpoll to attempt to take xmit_lock recursively. While it is possible to resolve this by getting netpoll to use trylock, it is suboptimal because netpoll's sole objective is to maximise the chance of getting the printk out on the wire. So delaying or dropping the message is to be avoided as much as possible. So the only alternative is to always set xmit_lock_owner. The following patch does this by introducing the netif_tx_lock family of functions that take care of setting/unsetting xmit_lock_owner. I renamed xmit_lock to _xmit_lock to indicate that it should not be used directly. I didn't provide irq versions of the netif_tx_lock functions since xmit_lock is meant to be a BH-disabling lock. This is pretty much a straight text substitution except for a small bug fix in winbond. It currently uses netif_stop_queue/spin_unlock_wait to stop transmission. This is unsafe as an IRQ can potentially wake up the queue. So it is safer to use netif_tx_disable. The hamradio bits used spin_lock_irq but it is unnecessary as xmit_lock must never be taken in an IRQ handler. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-10 03:20:56 +08:00
netif_tx_unlock(dev);
dev_put(dev);
}
void __netdev_watchdog_up(struct net_device *dev)
{
if (dev->netdev_ops->ndo_tx_timeout) {
if (dev->watchdog_timeo <= 0)
dev->watchdog_timeo = 5*HZ;
if (!mod_timer(&dev->watchdog_timer,
round_jiffies(jiffies + dev->watchdog_timeo)))
dev_hold(dev);
}
}
static void dev_watchdog_up(struct net_device *dev)
{
__netdev_watchdog_up(dev);
}
static void dev_watchdog_down(struct net_device *dev)
{
[NET]: Add netif_tx_lock Various drivers use xmit_lock internally to synchronise with their transmission routines. They do so without setting xmit_lock_owner. This is fine as long as netpoll is not in use. With netpoll it is possible for deadlocks to occur if xmit_lock_owner isn't set. This is because if a printk occurs while xmit_lock is held and xmit_lock_owner is not set can cause netpoll to attempt to take xmit_lock recursively. While it is possible to resolve this by getting netpoll to use trylock, it is suboptimal because netpoll's sole objective is to maximise the chance of getting the printk out on the wire. So delaying or dropping the message is to be avoided as much as possible. So the only alternative is to always set xmit_lock_owner. The following patch does this by introducing the netif_tx_lock family of functions that take care of setting/unsetting xmit_lock_owner. I renamed xmit_lock to _xmit_lock to indicate that it should not be used directly. I didn't provide irq versions of the netif_tx_lock functions since xmit_lock is meant to be a BH-disabling lock. This is pretty much a straight text substitution except for a small bug fix in winbond. It currently uses netif_stop_queue/spin_unlock_wait to stop transmission. This is unsafe as an IRQ can potentially wake up the queue. So it is safer to use netif_tx_disable. The hamradio bits used spin_lock_irq but it is unnecessary as xmit_lock must never be taken in an IRQ handler. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-10 03:20:56 +08:00
netif_tx_lock_bh(dev);
if (del_timer(&dev->watchdog_timer))
dev_put(dev);
[NET]: Add netif_tx_lock Various drivers use xmit_lock internally to synchronise with their transmission routines. They do so without setting xmit_lock_owner. This is fine as long as netpoll is not in use. With netpoll it is possible for deadlocks to occur if xmit_lock_owner isn't set. This is because if a printk occurs while xmit_lock is held and xmit_lock_owner is not set can cause netpoll to attempt to take xmit_lock recursively. While it is possible to resolve this by getting netpoll to use trylock, it is suboptimal because netpoll's sole objective is to maximise the chance of getting the printk out on the wire. So delaying or dropping the message is to be avoided as much as possible. So the only alternative is to always set xmit_lock_owner. The following patch does this by introducing the netif_tx_lock family of functions that take care of setting/unsetting xmit_lock_owner. I renamed xmit_lock to _xmit_lock to indicate that it should not be used directly. I didn't provide irq versions of the netif_tx_lock functions since xmit_lock is meant to be a BH-disabling lock. This is pretty much a straight text substitution except for a small bug fix in winbond. It currently uses netif_stop_queue/spin_unlock_wait to stop transmission. This is unsafe as an IRQ can potentially wake up the queue. So it is safer to use netif_tx_disable. The hamradio bits used spin_lock_irq but it is unnecessary as xmit_lock must never be taken in an IRQ handler. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-10 03:20:56 +08:00
netif_tx_unlock_bh(dev);
}
[NET]: Make NAPI polling independent of struct net_device objects. Several devices have multiple independant RX queues per net device, and some have a single interrupt doorbell for several queues. In either case, it's easier to support layouts like that if the structure representing the poll is independant from the net device itself. The signature of the ->poll() call back goes from: int foo_poll(struct net_device *dev, int *budget) to int foo_poll(struct napi_struct *napi, int budget) The caller is returned the number of RX packets processed (or the number of "NAPI credits" consumed if you want to get abstract). The callee no longer messes around bumping dev->quota, *budget, etc. because that is all handled in the caller upon return. The napi_struct is to be embedded in the device driver private data structures. Furthermore, it is the driver's responsibility to disable all NAPI instances in it's ->stop() device close handler. Since the napi_struct is privatized into the driver's private data structures, only the driver knows how to get at all of the napi_struct instances it may have per-device. With lots of help and suggestions from Rusty Russell, Roland Dreier, Michael Chan, Jeff Garzik, and Jamal Hadi Salim. Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra, Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan. [ Ported to current tree and all drivers converted. Integrated Stephen's follow-on kerneldoc additions, and restored poll_list handling to the old style to fix mutual exclusion issues. -DaveM ] Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 07:41:36 +08:00
/**
* netif_carrier_on - set carrier
* @dev: network device
*
* Device has detected that carrier.
*/
void netif_carrier_on(struct net_device *dev)
{
if (test_and_clear_bit(__LINK_STATE_NOCARRIER, &dev->state)) {
if (dev->reg_state == NETREG_UNINITIALIZED)
return;
linkwatch_fire_event(dev);
if (netif_running(dev))
__netdev_watchdog_up(dev);
}
}
EXPORT_SYMBOL(netif_carrier_on);
[NET]: Make NAPI polling independent of struct net_device objects. Several devices have multiple independant RX queues per net device, and some have a single interrupt doorbell for several queues. In either case, it's easier to support layouts like that if the structure representing the poll is independant from the net device itself. The signature of the ->poll() call back goes from: int foo_poll(struct net_device *dev, int *budget) to int foo_poll(struct napi_struct *napi, int budget) The caller is returned the number of RX packets processed (or the number of "NAPI credits" consumed if you want to get abstract). The callee no longer messes around bumping dev->quota, *budget, etc. because that is all handled in the caller upon return. The napi_struct is to be embedded in the device driver private data structures. Furthermore, it is the driver's responsibility to disable all NAPI instances in it's ->stop() device close handler. Since the napi_struct is privatized into the driver's private data structures, only the driver knows how to get at all of the napi_struct instances it may have per-device. With lots of help and suggestions from Rusty Russell, Roland Dreier, Michael Chan, Jeff Garzik, and Jamal Hadi Salim. Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra, Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan. [ Ported to current tree and all drivers converted. Integrated Stephen's follow-on kerneldoc additions, and restored poll_list handling to the old style to fix mutual exclusion issues. -DaveM ] Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 07:41:36 +08:00
/**
* netif_carrier_off - clear carrier
* @dev: network device
*
* Device has detected loss of carrier.
*/
void netif_carrier_off(struct net_device *dev)
{
if (!test_and_set_bit(__LINK_STATE_NOCARRIER, &dev->state)) {
if (dev->reg_state == NETREG_UNINITIALIZED)
return;
linkwatch_fire_event(dev);
}
}
EXPORT_SYMBOL(netif_carrier_off);
/* "NOOP" scheduler: the best scheduler, recommended for all interfaces
under all circumstances. It is difficult to invent anything faster or
cheaper.
*/
static int noop_enqueue(struct sk_buff *skb, struct Qdisc * qdisc)
{
kfree_skb(skb);
return NET_XMIT_CN;
}
static struct sk_buff *noop_dequeue(struct Qdisc * qdisc)
{
return NULL;
}
struct Qdisc_ops noop_qdisc_ops __read_mostly = {
.id = "noop",
.priv_size = 0,
.enqueue = noop_enqueue,
.dequeue = noop_dequeue,
.peek = noop_dequeue,
.owner = THIS_MODULE,
};
static struct netdev_queue noop_netdev_queue = {
.qdisc = &noop_qdisc,
.qdisc_sleeping = &noop_qdisc,
};
struct Qdisc noop_qdisc = {
.enqueue = noop_enqueue,
.dequeue = noop_dequeue,
.flags = TCQ_F_BUILTIN,
.ops = &noop_qdisc_ops,
.list = LIST_HEAD_INIT(noop_qdisc.list),
.q.lock = __SPIN_LOCK_UNLOCKED(noop_qdisc.q.lock),
.dev_queue = &noop_netdev_queue,
};
EXPORT_SYMBOL(noop_qdisc);
static struct Qdisc_ops noqueue_qdisc_ops __read_mostly = {
.id = "noqueue",
.priv_size = 0,
.enqueue = noop_enqueue,
.dequeue = noop_dequeue,
.peek = noop_dequeue,
.owner = THIS_MODULE,
};
static struct Qdisc noqueue_qdisc;
static struct netdev_queue noqueue_netdev_queue = {
.qdisc = &noqueue_qdisc,
.qdisc_sleeping = &noqueue_qdisc,
};
static struct Qdisc noqueue_qdisc = {
.enqueue = NULL,
.dequeue = noop_dequeue,
.flags = TCQ_F_BUILTIN,
.ops = &noqueue_qdisc_ops,
.list = LIST_HEAD_INIT(noqueue_qdisc.list),
.q.lock = __SPIN_LOCK_UNLOCKED(noqueue_qdisc.q.lock),
.dev_queue = &noqueue_netdev_queue,
};
static const u8 prio2band[TC_PRIO_MAX+1] =
{ 1, 2, 2, 2, 1, 2, 0, 0 , 1, 1, 1, 1, 1, 1, 1, 1 };
/* 3-band FIFO queue: old style, but should be a bit faster than
generic prio+fifo combination.
*/
#define PFIFO_FAST_BANDS 3
/*
* Private data for a pfifo_fast scheduler containing:
* - queues for the three band
* - bitmap indicating which of the bands contain skbs
*/
struct pfifo_fast_priv {
u32 bitmap;
struct sk_buff_head q[PFIFO_FAST_BANDS];
};
/*
* Convert a bitmap to the first band number where an skb is queued, where:
* bitmap=0 means there are no skbs on any band.
* bitmap=1 means there is an skb on band 0.
* bitmap=7 means there are skbs on all 3 bands, etc.
*/
static const int bitmap2band[] = {-1, 0, 1, 0, 2, 0, 1, 0};
static inline struct sk_buff_head *band2list(struct pfifo_fast_priv *priv,
int band)
{
return priv->q + band;
}
static int pfifo_fast_enqueue(struct sk_buff *skb, struct Qdisc* qdisc)
{
if (skb_queue_len(&qdisc->q) < qdisc_dev(qdisc)->tx_queue_len) {
int band = prio2band[skb->priority & TC_PRIO_MAX];
struct pfifo_fast_priv *priv = qdisc_priv(qdisc);
struct sk_buff_head *list = band2list(priv, band);
priv->bitmap |= (1 << band);
qdisc->q.qlen++;
return __qdisc_enqueue_tail(skb, qdisc, list);
}
return qdisc_drop(skb, qdisc);
}
static struct sk_buff *pfifo_fast_dequeue(struct Qdisc* qdisc)
{
struct pfifo_fast_priv *priv = qdisc_priv(qdisc);
int band = bitmap2band[priv->bitmap];
if (likely(band >= 0)) {
struct sk_buff_head *list = band2list(priv, band);
struct sk_buff *skb = __qdisc_dequeue_head(qdisc, list);
qdisc->q.qlen--;
if (skb_queue_empty(list))
priv->bitmap &= ~(1 << band);
return skb;
}
return NULL;
}
static struct sk_buff *pfifo_fast_peek(struct Qdisc* qdisc)
{
struct pfifo_fast_priv *priv = qdisc_priv(qdisc);
int band = bitmap2band[priv->bitmap];
if (band >= 0) {
struct sk_buff_head *list = band2list(priv, band);
return skb_peek(list);
}
return NULL;
}
static void pfifo_fast_reset(struct Qdisc* qdisc)
{
int prio;
struct pfifo_fast_priv *priv = qdisc_priv(qdisc);
for (prio = 0; prio < PFIFO_FAST_BANDS; prio++)
__qdisc_reset_queue(qdisc, band2list(priv, prio));
priv->bitmap = 0;
qdisc->qstats.backlog = 0;
qdisc->q.qlen = 0;
}
static int pfifo_fast_dump(struct Qdisc *qdisc, struct sk_buff *skb)
{
struct tc_prio_qopt opt = { .bands = PFIFO_FAST_BANDS };
memcpy(&opt.priomap, prio2band, TC_PRIO_MAX+1);
NLA_PUT(skb, TCA_OPTIONS, sizeof(opt), &opt);
return skb->len;
nla_put_failure:
return -1;
}
static int pfifo_fast_init(struct Qdisc *qdisc, struct nlattr *opt)
{
int prio;
struct pfifo_fast_priv *priv = qdisc_priv(qdisc);
for (prio = 0; prio < PFIFO_FAST_BANDS; prio++)
skb_queue_head_init(band2list(priv, prio));
return 0;
}
struct Qdisc_ops pfifo_fast_ops __read_mostly = {
.id = "pfifo_fast",
.priv_size = sizeof(struct pfifo_fast_priv),
.enqueue = pfifo_fast_enqueue,
.dequeue = pfifo_fast_dequeue,
.peek = pfifo_fast_peek,
.init = pfifo_fast_init,
.reset = pfifo_fast_reset,
.dump = pfifo_fast_dump,
.owner = THIS_MODULE,
};
struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue,
struct Qdisc_ops *ops)
{
void *p;
struct Qdisc *sch;
unsigned int size;
int err = -ENOBUFS;
/* ensure that the Qdisc and the private data are 32-byte aligned */
size = QDISC_ALIGN(sizeof(*sch));
size += ops->priv_size + (QDISC_ALIGNTO - 1);
p = kzalloc(size, GFP_KERNEL);
if (!p)
goto errout;
sch = (struct Qdisc *) QDISC_ALIGN((unsigned long) p);
sch->padded = (char *) sch - (char *) p;
INIT_LIST_HEAD(&sch->list);
skb_queue_head_init(&sch->q);
sch->ops = ops;
sch->enqueue = ops->enqueue;
sch->dequeue = ops->dequeue;
sch->dev_queue = dev_queue;
dev_hold(qdisc_dev(sch));
atomic_set(&sch->refcnt, 1);
return sch;
errout:
return ERR_PTR(err);
}
struct Qdisc * qdisc_create_dflt(struct net_device *dev,
struct netdev_queue *dev_queue,
struct Qdisc_ops *ops,
unsigned int parentid)
{
struct Qdisc *sch;
sch = qdisc_alloc(dev_queue, ops);
if (IS_ERR(sch))
goto errout;
sch->parent = parentid;
if (!ops->init || ops->init(sch, NULL) == 0)
return sch;
qdisc_destroy(sch);
errout:
return NULL;
}
EXPORT_SYMBOL(qdisc_create_dflt);
/* Under qdisc_lock(qdisc) and BH! */
void qdisc_reset(struct Qdisc *qdisc)
{
const struct Qdisc_ops *ops = qdisc->ops;
if (ops->reset)
ops->reset(qdisc);
net: Avoid enqueuing skb for default qdiscs dev_queue_xmit enqueue's a skb and calls qdisc_run which dequeue's the skb and xmits it. In most cases, the skb that is enqueue'd is the same one that is dequeue'd (unless the queue gets stopped or multiple cpu's write to the same queue and ends in a race with qdisc_run). For default qdiscs, we can remove the redundant enqueue/dequeue and simply xmit the skb since the default qdisc is work-conserving. The patch uses a new flag - TCQ_F_CAN_BYPASS to identify the default fast queue. The controversial part of the patch is incrementing qlen when a skb is requeued - this is to avoid checks like the second line below: + } else if ((q->flags & TCQ_F_CAN_BYPASS) && !qdisc_qlen(q) && >> !q->gso_skb && + !test_and_set_bit(__QDISC_STATE_RUNNING, &q->state)) { Results of a 2 hour testing for multiple netperf sessions (1, 2, 4, 8, 12 sessions on a 4 cpu system-X). The BW numbers are aggregate Mb/s across iterations tested with this version on System-X boxes with Chelsio 10gbps cards: ---------------------------------- Size | ORG BW NEW BW | ---------------------------------- 128K | 156964 159381 | 256K | 158650 162042 | ---------------------------------- Changes from ver1: 1. Move sch_direct_xmit declaration from sch_generic.h to pkt_sched.h 2. Update qdisc basic statistics for direct xmit path. 3. Set qlen to zero in qdisc_reset. 4. Changed some function names to more meaningful ones. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-08-06 09:44:21 +08:00
if (qdisc->gso_skb) {
kfree_skb(qdisc->gso_skb);
qdisc->gso_skb = NULL;
qdisc->q.qlen = 0;
}
}
EXPORT_SYMBOL(qdisc_reset);
void qdisc_destroy(struct Qdisc *qdisc)
{
const struct Qdisc_ops *ops = qdisc->ops;
if (qdisc->flags & TCQ_F_BUILTIN ||
!atomic_dec_and_test(&qdisc->refcnt))
return;
#ifdef CONFIG_NET_SCHED
qdisc_list_del(qdisc);
qdisc_put_stab(qdisc->stab);
#endif
gen_kill_estimator(&qdisc->bstats, &qdisc->rate_est);
if (ops->reset)
ops->reset(qdisc);
if (ops->destroy)
ops->destroy(qdisc);
module_put(ops->owner);
dev_put(qdisc_dev(qdisc));
kfree_skb(qdisc->gso_skb);
kfree((char *) qdisc - qdisc->padded);
}
EXPORT_SYMBOL(qdisc_destroy);
/* Attach toplevel qdisc to device queue. */
struct Qdisc *dev_graft_qdisc(struct netdev_queue *dev_queue,
struct Qdisc *qdisc)
{
struct Qdisc *oqdisc = dev_queue->qdisc_sleeping;
spinlock_t *root_lock;
root_lock = qdisc_lock(oqdisc);
spin_lock_bh(root_lock);
/* Prune old scheduler */
if (oqdisc && atomic_read(&oqdisc->refcnt) <= 1)
qdisc_reset(oqdisc);
/* ... and graft new one */
if (qdisc == NULL)
qdisc = &noop_qdisc;
dev_queue->qdisc_sleeping = qdisc;
rcu_assign_pointer(dev_queue->qdisc, &noop_qdisc);
spin_unlock_bh(root_lock);
return oqdisc;
}
static void attach_one_default_qdisc(struct net_device *dev,
struct netdev_queue *dev_queue,
void *_unused)
{
struct Qdisc *qdisc;
if (dev->tx_queue_len) {
qdisc = qdisc_create_dflt(dev, dev_queue,
&pfifo_fast_ops, TC_H_ROOT);
if (!qdisc) {
printk(KERN_INFO "%s: activation failed\n", dev->name);
return;
}
net: Avoid enqueuing skb for default qdiscs dev_queue_xmit enqueue's a skb and calls qdisc_run which dequeue's the skb and xmits it. In most cases, the skb that is enqueue'd is the same one that is dequeue'd (unless the queue gets stopped or multiple cpu's write to the same queue and ends in a race with qdisc_run). For default qdiscs, we can remove the redundant enqueue/dequeue and simply xmit the skb since the default qdisc is work-conserving. The patch uses a new flag - TCQ_F_CAN_BYPASS to identify the default fast queue. The controversial part of the patch is incrementing qlen when a skb is requeued - this is to avoid checks like the second line below: + } else if ((q->flags & TCQ_F_CAN_BYPASS) && !qdisc_qlen(q) && >> !q->gso_skb && + !test_and_set_bit(__QDISC_STATE_RUNNING, &q->state)) { Results of a 2 hour testing for multiple netperf sessions (1, 2, 4, 8, 12 sessions on a 4 cpu system-X). The BW numbers are aggregate Mb/s across iterations tested with this version on System-X boxes with Chelsio 10gbps cards: ---------------------------------- Size | ORG BW NEW BW | ---------------------------------- 128K | 156964 159381 | 256K | 158650 162042 | ---------------------------------- Changes from ver1: 1. Move sch_direct_xmit declaration from sch_generic.h to pkt_sched.h 2. Update qdisc basic statistics for direct xmit path. 3. Set qlen to zero in qdisc_reset. 4. Changed some function names to more meaningful ones. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-08-06 09:44:21 +08:00
/* Can by-pass the queue discipline for default qdisc */
qdisc->flags |= TCQ_F_CAN_BYPASS;
} else {
qdisc = &noqueue_qdisc;
}
dev_queue->qdisc_sleeping = qdisc;
}
static void attach_default_qdiscs(struct net_device *dev)
{
struct netdev_queue *txq;
struct Qdisc *qdisc;
txq = netdev_get_tx_queue(dev, 0);
if (!netif_is_multiqueue(dev) || dev->tx_queue_len == 0) {
netdev_for_each_tx_queue(dev, attach_one_default_qdisc, NULL);
dev->qdisc = txq->qdisc_sleeping;
atomic_inc(&dev->qdisc->refcnt);
} else {
qdisc = qdisc_create_dflt(dev, txq, &mq_qdisc_ops, TC_H_ROOT);
if (qdisc) {
qdisc->ops->attach(qdisc);
dev->qdisc = qdisc;
}
}
}
static void transition_one_qdisc(struct net_device *dev,
struct netdev_queue *dev_queue,
void *_need_watchdog)
{
struct Qdisc *new_qdisc = dev_queue->qdisc_sleeping;
int *need_watchdog_p = _need_watchdog;
if (!(new_qdisc->flags & TCQ_F_BUILTIN))
clear_bit(__QDISC_STATE_DEACTIVATED, &new_qdisc->state);
rcu_assign_pointer(dev_queue->qdisc, new_qdisc);
if (need_watchdog_p && new_qdisc != &noqueue_qdisc) {
dev_queue->trans_start = 0;
*need_watchdog_p = 1;
}
}
void dev_activate(struct net_device *dev)
{
int need_watchdog;
/* No queueing discipline is attached to device;
create default one i.e. pfifo_fast for devices,
which need queueing and noqueue_qdisc for
virtual interfaces
*/
if (dev->qdisc == &noop_qdisc)
attach_default_qdiscs(dev);
if (!netif_carrier_ok(dev))
/* Delay activation until next carrier-on event */
return;
need_watchdog = 0;
netdev_for_each_tx_queue(dev, transition_one_qdisc, &need_watchdog);
transition_one_qdisc(dev, &dev->rx_queue, NULL);
if (need_watchdog) {
dev->trans_start = jiffies;
dev_watchdog_up(dev);
}
}
static void dev_deactivate_queue(struct net_device *dev,
struct netdev_queue *dev_queue,
void *_qdisc_default)
{
struct Qdisc *qdisc_default = _qdisc_default;
struct Qdisc *qdisc;
qdisc = dev_queue->qdisc;
if (qdisc) {
spin_lock_bh(qdisc_lock(qdisc));
if (!(qdisc->flags & TCQ_F_BUILTIN))
set_bit(__QDISC_STATE_DEACTIVATED, &qdisc->state);
rcu_assign_pointer(dev_queue->qdisc, qdisc_default);
qdisc_reset(qdisc);
spin_unlock_bh(qdisc_lock(qdisc));
}
}
static bool some_qdisc_is_busy(struct net_device *dev)
{
unsigned int i;
for (i = 0; i < dev->num_tx_queues; i++) {
struct netdev_queue *dev_queue;
spinlock_t *root_lock;
struct Qdisc *q;
int val;
dev_queue = netdev_get_tx_queue(dev, i);
q = dev_queue->qdisc_sleeping;
root_lock = qdisc_lock(q);
spin_lock_bh(root_lock);
val = (test_bit(__QDISC_STATE_RUNNING, &q->state) ||
test_bit(__QDISC_STATE_SCHED, &q->state));
spin_unlock_bh(root_lock);
if (val)
return true;
}
return false;
}
void dev_deactivate(struct net_device *dev)
{
netdev_for_each_tx_queue(dev, dev_deactivate_queue, &noop_qdisc);
dev_deactivate_queue(dev, &dev->rx_queue, &noop_qdisc);
dev_watchdog_down(dev);
/* Wait for outstanding qdisc-less dev_queue_xmit calls. */
synchronize_rcu();
/* Wait for outstanding qdisc_run calls. */
while (some_qdisc_is_busy(dev))
yield();
}
static void dev_init_scheduler_queue(struct net_device *dev,
struct netdev_queue *dev_queue,
void *_qdisc)
{
struct Qdisc *qdisc = _qdisc;
dev_queue->qdisc = qdisc;
dev_queue->qdisc_sleeping = qdisc;
}
void dev_init_scheduler(struct net_device *dev)
{
dev->qdisc = &noop_qdisc;
netdev_for_each_tx_queue(dev, dev_init_scheduler_queue, &noop_qdisc);
dev_init_scheduler_queue(dev, &dev->rx_queue, &noop_qdisc);
setup_timer(&dev->watchdog_timer, dev_watchdog, (unsigned long)dev);
}
static void shutdown_scheduler_queue(struct net_device *dev,
struct netdev_queue *dev_queue,
void *_qdisc_default)
{
struct Qdisc *qdisc = dev_queue->qdisc_sleeping;
struct Qdisc *qdisc_default = _qdisc_default;
if (qdisc) {
rcu_assign_pointer(dev_queue->qdisc, qdisc_default);
dev_queue->qdisc_sleeping = qdisc_default;
qdisc_destroy(qdisc);
}
}
void dev_shutdown(struct net_device *dev)
{
netdev_for_each_tx_queue(dev, shutdown_scheduler_queue, &noop_qdisc);
shutdown_scheduler_queue(dev, &dev->rx_queue, &noop_qdisc);
qdisc_destroy(dev->qdisc);
dev->qdisc = &noop_qdisc;
WARN_ON(timer_pending(&dev->watchdog_timer));
}