linux/net/sched/sch_generic.c

1409 lines
33 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 cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 16:04:11 +08:00
#include <linux/slab.h>
#include <linux/if_vlan.h>
#include <linux/skb_array.h>
#include <linux/if_macvlan.h>
#include <net/sch_generic.h>
#include <net/pkt_sched.h>
#include <net/dst.h>
#include <trace/events/qdisc.h>
#include <net/xfrm.h>
/* Qdisc to use by default */
const struct Qdisc_ops *default_qdisc_ops = &pfifo_fast_ops;
EXPORT_SYMBOL(default_qdisc_ops);
/* 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 struct sk_buff *__skb_dequeue_bad_txq(struct Qdisc *q)
{
const struct netdev_queue *txq = q->dev_queue;
spinlock_t *lock = NULL;
struct sk_buff *skb;
if (q->flags & TCQ_F_NOLOCK) {
lock = qdisc_lock(q);
spin_lock(lock);
}
skb = skb_peek(&q->skb_bad_txq);
if (skb) {
/* check the reason of requeuing without tx lock first */
txq = skb_get_tx_queue(txq->dev, skb);
if (!netif_xmit_frozen_or_stopped(txq)) {
skb = __skb_dequeue(&q->skb_bad_txq);
if (qdisc_is_percpu_stats(q)) {
qdisc_qstats_cpu_backlog_dec(q, skb);
qdisc_qstats_cpu_qlen_dec(q);
} else {
qdisc_qstats_backlog_dec(q, skb);
q->q.qlen--;
}
} else {
skb = NULL;
}
}
if (lock)
spin_unlock(lock);
return skb;
}
static inline struct sk_buff *qdisc_dequeue_skb_bad_txq(struct Qdisc *q)
{
struct sk_buff *skb = skb_peek(&q->skb_bad_txq);
if (unlikely(skb))
skb = __skb_dequeue_bad_txq(q);
return skb;
}
static inline void qdisc_enqueue_skb_bad_txq(struct Qdisc *q,
struct sk_buff *skb)
{
spinlock_t *lock = NULL;
if (q->flags & TCQ_F_NOLOCK) {
lock = qdisc_lock(q);
spin_lock(lock);
}
__skb_queue_tail(&q->skb_bad_txq, skb);
net: sched: fix uses after free syzbot reported one use-after-free in pfifo_fast_enqueue() [1] Issue here is that we can not reuse skb after a successful skb_array_produce() since another cpu might have consumed it already. I believe a similar problem exists in try_bulk_dequeue_skb_slow() in case we put an skb into qdisc_enqueue_skb_bad_txq() for lockless qdisc. [1] BUG: KASAN: use-after-free in qdisc_pkt_len include/net/sch_generic.h:610 [inline] BUG: KASAN: use-after-free in qdisc_qstats_cpu_backlog_inc include/net/sch_generic.h:712 [inline] BUG: KASAN: use-after-free in pfifo_fast_enqueue+0x4bc/0x5e0 net/sched/sch_generic.c:639 Read of size 4 at addr ffff8801cede37e8 by task syzkaller717588/5543 CPU: 1 PID: 5543 Comm: syzkaller717588 Not tainted 4.16.0-rc4+ #265 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:17 [inline] dump_stack+0x194/0x24d lib/dump_stack.c:53 print_address_description+0x73/0x250 mm/kasan/report.c:256 kasan_report_error mm/kasan/report.c:354 [inline] kasan_report+0x23c/0x360 mm/kasan/report.c:412 __asan_report_load4_noabort+0x14/0x20 mm/kasan/report.c:432 qdisc_pkt_len include/net/sch_generic.h:610 [inline] qdisc_qstats_cpu_backlog_inc include/net/sch_generic.h:712 [inline] pfifo_fast_enqueue+0x4bc/0x5e0 net/sched/sch_generic.c:639 __dev_xmit_skb net/core/dev.c:3216 [inline] Fixes: c5ad119fb6c0 ("net: sched: pfifo_fast use skb_array") Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: syzbot+ed43b6903ab968b16f54@syzkaller.appspotmail.com Cc: John Fastabend <john.fastabend@gmail.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: Jiri Pirko <jiri@resnulli.us> Acked-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-03-15 09:53:00 +08:00
if (qdisc_is_percpu_stats(q)) {
qdisc_qstats_cpu_backlog_inc(q, skb);
qdisc_qstats_cpu_qlen_inc(q);
} else {
qdisc_qstats_backlog_inc(q, skb);
q->q.qlen++;
}
if (lock)
spin_unlock(lock);
}
static inline int __dev_requeue_skb(struct sk_buff *skb, struct Qdisc *q)
{
while (skb) {
struct sk_buff *next = skb->next;
__skb_queue_tail(&q->gso_skb, skb);
q->qstats.requeues++;
qdisc_qstats_backlog_inc(q, skb);
q->q.qlen++; /* it's still part of the queue */
skb = next;
}
__netif_schedule(q);
return 0;
}
static inline int dev_requeue_skb_locked(struct sk_buff *skb, struct Qdisc *q)
{
spinlock_t *lock = qdisc_lock(q);
spin_lock(lock);
while (skb) {
struct sk_buff *next = skb->next;
__skb_queue_tail(&q->gso_skb, skb);
qdisc_qstats_cpu_requeues_inc(q);
qdisc_qstats_cpu_backlog_inc(q, skb);
qdisc_qstats_cpu_qlen_inc(q);
skb = next;
}
spin_unlock(lock);
__netif_schedule(q);
return 0;
}
static inline int dev_requeue_skb(struct sk_buff *skb, struct Qdisc *q)
{
if (q->flags & TCQ_F_NOLOCK)
return dev_requeue_skb_locked(skb, q);
else
return __dev_requeue_skb(skb, q);
}
static void try_bulk_dequeue_skb(struct Qdisc *q,
struct sk_buff *skb,
net_sched: restore qdisc quota fairness limits after bulk dequeue Restore the quota fairness between qdisc's, that we broke with commit 5772e9a346 ("qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE"). Before that commit, the quota in __qdisc_run() were in packets as dequeue_skb() would only dequeue a single packet, that assumption broke with bulk dequeue. We choose not to account for the number of packets inside the TSO/GSO packets (accessable via "skb_gso_segs"). As the previous fairness also had this "defect". Thus, GSO/TSO packets counts as a single packet. Further more, we choose to slack on accuracy, by allowing a bulk dequeue try_bulk_dequeue_skb() to exceed the "packets" limit, only limited by the BQL bytelimit. This is done because BQL prefers to get its full budget for appropriate feedback from TX completion. In future, we might consider reworking this further and, if it allows, switch to a time-based model, as suggested by Eric. Right now, we only restore old semantics. Joint work with Eric, Hannes, Daniel and Jesper. Hannes wrote the first patch in cooperation with Daniel and Jesper. Eric rewrote the patch. Fixes: 5772e9a346 ("qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE") Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-09 18:18:10 +08:00
const struct netdev_queue *txq,
int *packets)
qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE Based on DaveM's recent API work on dev_hard_start_xmit(), that allows sending/processing an entire skb list. This patch implements qdisc bulk dequeue, by allowing multiple packets to be dequeued in dequeue_skb(). The optimization principle for this is two fold, (1) to amortize locking cost and (2) avoid expensive tailptr update for notifying HW. (1) Several packets are dequeued while holding the qdisc root_lock, amortizing locking cost over several packet. The dequeued SKB list is processed under the TXQ lock in dev_hard_start_xmit(), thus also amortizing the cost of the TXQ lock. (2) Further more, dev_hard_start_xmit() will utilize the skb->xmit_more API to delay HW tailptr update, which also reduces the cost per packet. One restriction of the new API is that every SKB must belong to the same TXQ. This patch takes the easy way out, by restricting bulk dequeue to qdisc's with the TCQ_F_ONETXQUEUE flag, that specifies the qdisc only have attached a single TXQ. Some detail about the flow; dev_hard_start_xmit() will process the skb list, and transmit packets individually towards the driver (see xmit_one()). In case the driver stops midway in the list, the remaining skb list is returned by dev_hard_start_xmit(). In sch_direct_xmit() this returned list is requeued by dev_requeue_skb(). To avoid overshooting the HW limits, which results in requeuing, the patch limits the amount of bytes dequeued, based on the drivers BQL limits. In-effect bulking will only happen for BQL enabled drivers. Small amounts for extra HoL blocking (2x MTU/0.24ms) were measured at 100Mbit/s, with bulking 8 packets, but the oscillating nature of the measurement indicate something, like sched latency might be causing this effect. More comparisons show, that this oscillation goes away occationally. Thus, we disregard this artifact completely and remove any "magic" bulking limit. For now, as a conservative approach, stop bulking when seeing TSO and segmented GSO packets. They already benefit from bulking on their own. A followup patch add this, to allow easier bisect-ability for finding regressions. Jointed work with Hannes, Daniel and Florian. Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-02 04:35:59 +08:00
{
int bytelimit = qdisc_avail_bulklimit(txq) - skb->len;
qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE Based on DaveM's recent API work on dev_hard_start_xmit(), that allows sending/processing an entire skb list. This patch implements qdisc bulk dequeue, by allowing multiple packets to be dequeued in dequeue_skb(). The optimization principle for this is two fold, (1) to amortize locking cost and (2) avoid expensive tailptr update for notifying HW. (1) Several packets are dequeued while holding the qdisc root_lock, amortizing locking cost over several packet. The dequeued SKB list is processed under the TXQ lock in dev_hard_start_xmit(), thus also amortizing the cost of the TXQ lock. (2) Further more, dev_hard_start_xmit() will utilize the skb->xmit_more API to delay HW tailptr update, which also reduces the cost per packet. One restriction of the new API is that every SKB must belong to the same TXQ. This patch takes the easy way out, by restricting bulk dequeue to qdisc's with the TCQ_F_ONETXQUEUE flag, that specifies the qdisc only have attached a single TXQ. Some detail about the flow; dev_hard_start_xmit() will process the skb list, and transmit packets individually towards the driver (see xmit_one()). In case the driver stops midway in the list, the remaining skb list is returned by dev_hard_start_xmit(). In sch_direct_xmit() this returned list is requeued by dev_requeue_skb(). To avoid overshooting the HW limits, which results in requeuing, the patch limits the amount of bytes dequeued, based on the drivers BQL limits. In-effect bulking will only happen for BQL enabled drivers. Small amounts for extra HoL blocking (2x MTU/0.24ms) were measured at 100Mbit/s, with bulking 8 packets, but the oscillating nature of the measurement indicate something, like sched latency might be causing this effect. More comparisons show, that this oscillation goes away occationally. Thus, we disregard this artifact completely and remove any "magic" bulking limit. For now, as a conservative approach, stop bulking when seeing TSO and segmented GSO packets. They already benefit from bulking on their own. A followup patch add this, to allow easier bisect-ability for finding regressions. Jointed work with Hannes, Daniel and Florian. Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-02 04:35:59 +08:00
while (bytelimit > 0) {
struct sk_buff *nskb = q->dequeue(q);
qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE Based on DaveM's recent API work on dev_hard_start_xmit(), that allows sending/processing an entire skb list. This patch implements qdisc bulk dequeue, by allowing multiple packets to be dequeued in dequeue_skb(). The optimization principle for this is two fold, (1) to amortize locking cost and (2) avoid expensive tailptr update for notifying HW. (1) Several packets are dequeued while holding the qdisc root_lock, amortizing locking cost over several packet. The dequeued SKB list is processed under the TXQ lock in dev_hard_start_xmit(), thus also amortizing the cost of the TXQ lock. (2) Further more, dev_hard_start_xmit() will utilize the skb->xmit_more API to delay HW tailptr update, which also reduces the cost per packet. One restriction of the new API is that every SKB must belong to the same TXQ. This patch takes the easy way out, by restricting bulk dequeue to qdisc's with the TCQ_F_ONETXQUEUE flag, that specifies the qdisc only have attached a single TXQ. Some detail about the flow; dev_hard_start_xmit() will process the skb list, and transmit packets individually towards the driver (see xmit_one()). In case the driver stops midway in the list, the remaining skb list is returned by dev_hard_start_xmit(). In sch_direct_xmit() this returned list is requeued by dev_requeue_skb(). To avoid overshooting the HW limits, which results in requeuing, the patch limits the amount of bytes dequeued, based on the drivers BQL limits. In-effect bulking will only happen for BQL enabled drivers. Small amounts for extra HoL blocking (2x MTU/0.24ms) were measured at 100Mbit/s, with bulking 8 packets, but the oscillating nature of the measurement indicate something, like sched latency might be causing this effect. More comparisons show, that this oscillation goes away occationally. Thus, we disregard this artifact completely and remove any "magic" bulking limit. For now, as a conservative approach, stop bulking when seeing TSO and segmented GSO packets. They already benefit from bulking on their own. A followup patch add this, to allow easier bisect-ability for finding regressions. Jointed work with Hannes, Daniel and Florian. Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-02 04:35:59 +08:00
if (!nskb)
qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE Based on DaveM's recent API work on dev_hard_start_xmit(), that allows sending/processing an entire skb list. This patch implements qdisc bulk dequeue, by allowing multiple packets to be dequeued in dequeue_skb(). The optimization principle for this is two fold, (1) to amortize locking cost and (2) avoid expensive tailptr update for notifying HW. (1) Several packets are dequeued while holding the qdisc root_lock, amortizing locking cost over several packet. The dequeued SKB list is processed under the TXQ lock in dev_hard_start_xmit(), thus also amortizing the cost of the TXQ lock. (2) Further more, dev_hard_start_xmit() will utilize the skb->xmit_more API to delay HW tailptr update, which also reduces the cost per packet. One restriction of the new API is that every SKB must belong to the same TXQ. This patch takes the easy way out, by restricting bulk dequeue to qdisc's with the TCQ_F_ONETXQUEUE flag, that specifies the qdisc only have attached a single TXQ. Some detail about the flow; dev_hard_start_xmit() will process the skb list, and transmit packets individually towards the driver (see xmit_one()). In case the driver stops midway in the list, the remaining skb list is returned by dev_hard_start_xmit(). In sch_direct_xmit() this returned list is requeued by dev_requeue_skb(). To avoid overshooting the HW limits, which results in requeuing, the patch limits the amount of bytes dequeued, based on the drivers BQL limits. In-effect bulking will only happen for BQL enabled drivers. Small amounts for extra HoL blocking (2x MTU/0.24ms) were measured at 100Mbit/s, with bulking 8 packets, but the oscillating nature of the measurement indicate something, like sched latency might be causing this effect. More comparisons show, that this oscillation goes away occationally. Thus, we disregard this artifact completely and remove any "magic" bulking limit. For now, as a conservative approach, stop bulking when seeing TSO and segmented GSO packets. They already benefit from bulking on their own. A followup patch add this, to allow easier bisect-ability for finding regressions. Jointed work with Hannes, Daniel and Florian. Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-02 04:35:59 +08:00
break;
bytelimit -= nskb->len; /* covers GSO len */
skb->next = nskb;
skb = nskb;
net_sched: restore qdisc quota fairness limits after bulk dequeue Restore the quota fairness between qdisc's, that we broke with commit 5772e9a346 ("qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE"). Before that commit, the quota in __qdisc_run() were in packets as dequeue_skb() would only dequeue a single packet, that assumption broke with bulk dequeue. We choose not to account for the number of packets inside the TSO/GSO packets (accessable via "skb_gso_segs"). As the previous fairness also had this "defect". Thus, GSO/TSO packets counts as a single packet. Further more, we choose to slack on accuracy, by allowing a bulk dequeue try_bulk_dequeue_skb() to exceed the "packets" limit, only limited by the BQL bytelimit. This is done because BQL prefers to get its full budget for appropriate feedback from TX completion. In future, we might consider reworking this further and, if it allows, switch to a time-based model, as suggested by Eric. Right now, we only restore old semantics. Joint work with Eric, Hannes, Daniel and Jesper. Hannes wrote the first patch in cooperation with Daniel and Jesper. Eric rewrote the patch. Fixes: 5772e9a346 ("qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE") Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-09 18:18:10 +08:00
(*packets)++; /* GSO counts as one pkt */
qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE Based on DaveM's recent API work on dev_hard_start_xmit(), that allows sending/processing an entire skb list. This patch implements qdisc bulk dequeue, by allowing multiple packets to be dequeued in dequeue_skb(). The optimization principle for this is two fold, (1) to amortize locking cost and (2) avoid expensive tailptr update for notifying HW. (1) Several packets are dequeued while holding the qdisc root_lock, amortizing locking cost over several packet. The dequeued SKB list is processed under the TXQ lock in dev_hard_start_xmit(), thus also amortizing the cost of the TXQ lock. (2) Further more, dev_hard_start_xmit() will utilize the skb->xmit_more API to delay HW tailptr update, which also reduces the cost per packet. One restriction of the new API is that every SKB must belong to the same TXQ. This patch takes the easy way out, by restricting bulk dequeue to qdisc's with the TCQ_F_ONETXQUEUE flag, that specifies the qdisc only have attached a single TXQ. Some detail about the flow; dev_hard_start_xmit() will process the skb list, and transmit packets individually towards the driver (see xmit_one()). In case the driver stops midway in the list, the remaining skb list is returned by dev_hard_start_xmit(). In sch_direct_xmit() this returned list is requeued by dev_requeue_skb(). To avoid overshooting the HW limits, which results in requeuing, the patch limits the amount of bytes dequeued, based on the drivers BQL limits. In-effect bulking will only happen for BQL enabled drivers. Small amounts for extra HoL blocking (2x MTU/0.24ms) were measured at 100Mbit/s, with bulking 8 packets, but the oscillating nature of the measurement indicate something, like sched latency might be causing this effect. More comparisons show, that this oscillation goes away occationally. Thus, we disregard this artifact completely and remove any "magic" bulking limit. For now, as a conservative approach, stop bulking when seeing TSO and segmented GSO packets. They already benefit from bulking on their own. A followup patch add this, to allow easier bisect-ability for finding regressions. Jointed work with Hannes, Daniel and Florian. Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-02 04:35:59 +08:00
}
skb_mark_not_on_list(skb);
qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE Based on DaveM's recent API work on dev_hard_start_xmit(), that allows sending/processing an entire skb list. This patch implements qdisc bulk dequeue, by allowing multiple packets to be dequeued in dequeue_skb(). The optimization principle for this is two fold, (1) to amortize locking cost and (2) avoid expensive tailptr update for notifying HW. (1) Several packets are dequeued while holding the qdisc root_lock, amortizing locking cost over several packet. The dequeued SKB list is processed under the TXQ lock in dev_hard_start_xmit(), thus also amortizing the cost of the TXQ lock. (2) Further more, dev_hard_start_xmit() will utilize the skb->xmit_more API to delay HW tailptr update, which also reduces the cost per packet. One restriction of the new API is that every SKB must belong to the same TXQ. This patch takes the easy way out, by restricting bulk dequeue to qdisc's with the TCQ_F_ONETXQUEUE flag, that specifies the qdisc only have attached a single TXQ. Some detail about the flow; dev_hard_start_xmit() will process the skb list, and transmit packets individually towards the driver (see xmit_one()). In case the driver stops midway in the list, the remaining skb list is returned by dev_hard_start_xmit(). In sch_direct_xmit() this returned list is requeued by dev_requeue_skb(). To avoid overshooting the HW limits, which results in requeuing, the patch limits the amount of bytes dequeued, based on the drivers BQL limits. In-effect bulking will only happen for BQL enabled drivers. Small amounts for extra HoL blocking (2x MTU/0.24ms) were measured at 100Mbit/s, with bulking 8 packets, but the oscillating nature of the measurement indicate something, like sched latency might be causing this effect. More comparisons show, that this oscillation goes away occationally. Thus, we disregard this artifact completely and remove any "magic" bulking limit. For now, as a conservative approach, stop bulking when seeing TSO and segmented GSO packets. They already benefit from bulking on their own. A followup patch add this, to allow easier bisect-ability for finding regressions. Jointed work with Hannes, Daniel and Florian. Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-02 04:35:59 +08:00
}
/* This variant of try_bulk_dequeue_skb() makes sure
* all skbs in the chain are for the same txq
*/
static void try_bulk_dequeue_skb_slow(struct Qdisc *q,
struct sk_buff *skb,
int *packets)
{
int mapping = skb_get_queue_mapping(skb);
struct sk_buff *nskb;
int cnt = 0;
do {
nskb = q->dequeue(q);
if (!nskb)
break;
if (unlikely(skb_get_queue_mapping(nskb) != mapping)) {
qdisc_enqueue_skb_bad_txq(q, nskb);
break;
}
skb->next = nskb;
skb = nskb;
} while (++cnt < 8);
(*packets) += cnt;
skb_mark_not_on_list(skb);
}
qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE Based on DaveM's recent API work on dev_hard_start_xmit(), that allows sending/processing an entire skb list. This patch implements qdisc bulk dequeue, by allowing multiple packets to be dequeued in dequeue_skb(). The optimization principle for this is two fold, (1) to amortize locking cost and (2) avoid expensive tailptr update for notifying HW. (1) Several packets are dequeued while holding the qdisc root_lock, amortizing locking cost over several packet. The dequeued SKB list is processed under the TXQ lock in dev_hard_start_xmit(), thus also amortizing the cost of the TXQ lock. (2) Further more, dev_hard_start_xmit() will utilize the skb->xmit_more API to delay HW tailptr update, which also reduces the cost per packet. One restriction of the new API is that every SKB must belong to the same TXQ. This patch takes the easy way out, by restricting bulk dequeue to qdisc's with the TCQ_F_ONETXQUEUE flag, that specifies the qdisc only have attached a single TXQ. Some detail about the flow; dev_hard_start_xmit() will process the skb list, and transmit packets individually towards the driver (see xmit_one()). In case the driver stops midway in the list, the remaining skb list is returned by dev_hard_start_xmit(). In sch_direct_xmit() this returned list is requeued by dev_requeue_skb(). To avoid overshooting the HW limits, which results in requeuing, the patch limits the amount of bytes dequeued, based on the drivers BQL limits. In-effect bulking will only happen for BQL enabled drivers. Small amounts for extra HoL blocking (2x MTU/0.24ms) were measured at 100Mbit/s, with bulking 8 packets, but the oscillating nature of the measurement indicate something, like sched latency might be causing this effect. More comparisons show, that this oscillation goes away occationally. Thus, we disregard this artifact completely and remove any "magic" bulking limit. For now, as a conservative approach, stop bulking when seeing TSO and segmented GSO packets. They already benefit from bulking on their own. A followup patch add this, to allow easier bisect-ability for finding regressions. Jointed work with Hannes, Daniel and Florian. Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-02 04:35:59 +08:00
/* Note that dequeue_skb can possibly return a SKB list (via skb->next).
* A requeued skb (via q->gso_skb) can also be a SKB list.
*/
net_sched: restore qdisc quota fairness limits after bulk dequeue Restore the quota fairness between qdisc's, that we broke with commit 5772e9a346 ("qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE"). Before that commit, the quota in __qdisc_run() were in packets as dequeue_skb() would only dequeue a single packet, that assumption broke with bulk dequeue. We choose not to account for the number of packets inside the TSO/GSO packets (accessable via "skb_gso_segs"). As the previous fairness also had this "defect". Thus, GSO/TSO packets counts as a single packet. Further more, we choose to slack on accuracy, by allowing a bulk dequeue try_bulk_dequeue_skb() to exceed the "packets" limit, only limited by the BQL bytelimit. This is done because BQL prefers to get its full budget for appropriate feedback from TX completion. In future, we might consider reworking this further and, if it allows, switch to a time-based model, as suggested by Eric. Right now, we only restore old semantics. Joint work with Eric, Hannes, Daniel and Jesper. Hannes wrote the first patch in cooperation with Daniel and Jesper. Eric rewrote the patch. Fixes: 5772e9a346 ("qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE") Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-09 18:18:10 +08:00
static struct sk_buff *dequeue_skb(struct Qdisc *q, bool *validate,
int *packets)
{
const struct netdev_queue *txq = q->dev_queue;
struct sk_buff *skb = NULL;
net_sched: restore qdisc quota fairness limits after bulk dequeue Restore the quota fairness between qdisc's, that we broke with commit 5772e9a346 ("qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE"). Before that commit, the quota in __qdisc_run() were in packets as dequeue_skb() would only dequeue a single packet, that assumption broke with bulk dequeue. We choose not to account for the number of packets inside the TSO/GSO packets (accessable via "skb_gso_segs"). As the previous fairness also had this "defect". Thus, GSO/TSO packets counts as a single packet. Further more, we choose to slack on accuracy, by allowing a bulk dequeue try_bulk_dequeue_skb() to exceed the "packets" limit, only limited by the BQL bytelimit. This is done because BQL prefers to get its full budget for appropriate feedback from TX completion. In future, we might consider reworking this further and, if it allows, switch to a time-based model, as suggested by Eric. Right now, we only restore old semantics. Joint work with Eric, Hannes, Daniel and Jesper. Hannes wrote the first patch in cooperation with Daniel and Jesper. Eric rewrote the patch. Fixes: 5772e9a346 ("qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE") Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-09 18:18:10 +08:00
*packets = 1;
if (unlikely(!skb_queue_empty(&q->gso_skb))) {
spinlock_t *lock = NULL;
if (q->flags & TCQ_F_NOLOCK) {
lock = qdisc_lock(q);
spin_lock(lock);
}
skb = skb_peek(&q->gso_skb);
/* skb may be null if another cpu pulls gso_skb off in between
* empty check and lock.
*/
if (!skb) {
if (lock)
spin_unlock(lock);
goto validate;
}
/* skb in gso_skb were already validated */
*validate = false;
if (xfrm_offload(skb))
*validate = true;
/* check the reason of requeuing without tx lock first */
txq = skb_get_tx_queue(txq->dev, skb);
if (!netif_xmit_frozen_or_stopped(txq)) {
skb = __skb_dequeue(&q->gso_skb);
if (qdisc_is_percpu_stats(q)) {
qdisc_qstats_cpu_backlog_dec(q, skb);
qdisc_qstats_cpu_qlen_dec(q);
} else {
qdisc_qstats_backlog_dec(q, skb);
q->q.qlen--;
}
} else {
skb = NULL;
}
if (lock)
spin_unlock(lock);
goto trace;
}
validate:
*validate = true;
if ((q->flags & TCQ_F_ONETXQUEUE) &&
netif_xmit_frozen_or_stopped(txq))
return skb;
skb = qdisc_dequeue_skb_bad_txq(q);
if (unlikely(skb))
goto bulk;
skb = q->dequeue(q);
if (skb) {
bulk:
if (qdisc_may_bulk(q))
try_bulk_dequeue_skb(q, skb, txq, packets);
else
try_bulk_dequeue_skb_slow(q, skb, packets);
}
trace:
trace_qdisc_dequeue(q, txq, *packets, skb);
return skb;
}
/*
* Transmit possibly several skbs, and handle the return status as
* required. Owning running seqcount 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:
* false - hardware queue frozen backoff
* true - feel free to send more pkts
[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
*/
bool sch_direct_xmit(struct sk_buff *skb, struct Qdisc *q,
struct net_device *dev, struct netdev_queue *txq,
spinlock_t *root_lock, bool validate)
{
int ret = NETDEV_TX_BUSY;
bool again = false;
/* And release qdisc */
if (root_lock)
spin_unlock(root_lock);
/* Note that we validate skb (GSO, checksum, ...) outside of locks */
if (validate)
skb = validate_xmit_skb_list(skb, dev, &again);
#ifdef CONFIG_XFRM_OFFLOAD
if (unlikely(again)) {
if (root_lock)
spin_lock(root_lock);
dev_requeue_skb(skb, q);
return false;
}
#endif
if (likely(skb)) {
HARD_TX_LOCK(dev, txq, smp_processor_id());
if (!netif_xmit_frozen_or_stopped(txq))
skb = dev_hard_start_xmit(skb, dev, txq, &ret);
HARD_TX_UNLOCK(dev, txq);
} else {
if (root_lock)
spin_lock(root_lock);
return true;
}
if (root_lock)
spin_lock(root_lock);
if (!dev_xmit_complete(ret)) {
[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
/* Driver returned NETDEV_TX_BUSY - requeue skb */
if (unlikely(ret != NETDEV_TX_BUSY))
net_warn_ratelimited("BUG %s code %d qlen %d\n",
dev->name, ret, q->q.qlen);
[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
dev_requeue_skb(skb, q);
return false;
[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 true;
}
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.
*
* running seqcount guarantees only one CPU can process
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
* 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 bool qdisc_restart(struct Qdisc *q, int *packets)
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
{
spinlock_t *root_lock = 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
struct netdev_queue *txq;
struct net_device *dev;
struct sk_buff *skb;
bool validate;
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
/* Dequeue packet */
net_sched: restore qdisc quota fairness limits after bulk dequeue Restore the quota fairness between qdisc's, that we broke with commit 5772e9a346 ("qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE"). Before that commit, the quota in __qdisc_run() were in packets as dequeue_skb() would only dequeue a single packet, that assumption broke with bulk dequeue. We choose not to account for the number of packets inside the TSO/GSO packets (accessable via "skb_gso_segs"). As the previous fairness also had this "defect". Thus, GSO/TSO packets counts as a single packet. Further more, we choose to slack on accuracy, by allowing a bulk dequeue try_bulk_dequeue_skb() to exceed the "packets" limit, only limited by the BQL bytelimit. This is done because BQL prefers to get its full budget for appropriate feedback from TX completion. In future, we might consider reworking this further and, if it allows, switch to a time-based model, as suggested by Eric. Right now, we only restore old semantics. Joint work with Eric, Hannes, Daniel and Jesper. Hannes wrote the first patch in cooperation with Daniel and Jesper. Eric rewrote the patch. Fixes: 5772e9a346 ("qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE") Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-09 18:18:10 +08:00
skb = dequeue_skb(q, &validate, packets);
if (unlikely(!skb))
return false;
if (!(q->flags & TCQ_F_NOLOCK))
root_lock = qdisc_lock(q);
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
dev = qdisc_dev(q);
txq = skb_get_tx_queue(dev, 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
return sch_direct_xmit(skb, q, dev, txq, root_lock, validate);
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
}
void __qdisc_run(struct Qdisc *q)
{
int quota = dev_tx_weight;
net_sched: restore qdisc quota fairness limits after bulk dequeue Restore the quota fairness between qdisc's, that we broke with commit 5772e9a346 ("qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE"). Before that commit, the quota in __qdisc_run() were in packets as dequeue_skb() would only dequeue a single packet, that assumption broke with bulk dequeue. We choose not to account for the number of packets inside the TSO/GSO packets (accessable via "skb_gso_segs"). As the previous fairness also had this "defect". Thus, GSO/TSO packets counts as a single packet. Further more, we choose to slack on accuracy, by allowing a bulk dequeue try_bulk_dequeue_skb() to exceed the "packets" limit, only limited by the BQL bytelimit. This is done because BQL prefers to get its full budget for appropriate feedback from TX completion. In future, we might consider reworking this further and, if it allows, switch to a time-based model, as suggested by Eric. Right now, we only restore old semantics. Joint work with Eric, Hannes, Daniel and Jesper. Hannes wrote the first patch in cooperation with Daniel and Jesper. Eric rewrote the patch. Fixes: 5772e9a346 ("qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE") Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-09 18:18:10 +08:00
int packets;
net_sched: restore qdisc quota fairness limits after bulk dequeue Restore the quota fairness between qdisc's, that we broke with commit 5772e9a346 ("qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE"). Before that commit, the quota in __qdisc_run() were in packets as dequeue_skb() would only dequeue a single packet, that assumption broke with bulk dequeue. We choose not to account for the number of packets inside the TSO/GSO packets (accessable via "skb_gso_segs"). As the previous fairness also had this "defect". Thus, GSO/TSO packets counts as a single packet. Further more, we choose to slack on accuracy, by allowing a bulk dequeue try_bulk_dequeue_skb() to exceed the "packets" limit, only limited by the BQL bytelimit. This is done because BQL prefers to get its full budget for appropriate feedback from TX completion. In future, we might consider reworking this further and, if it allows, switch to a time-based model, as suggested by Eric. Right now, we only restore old semantics. Joint work with Eric, Hannes, Daniel and Jesper. Hannes wrote the first patch in cooperation with Daniel and Jesper. Eric rewrote the patch. Fixes: 5772e9a346 ("qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE") Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-09 18:18:10 +08:00
while (qdisc_restart(q, &packets)) {
/*
net_sched: fix dequeuer fairness Results on dummy device can be seen in my netconf 2011 slides. These results are for a 10Gige IXGBE intel nic - on another i5 machine, very similar specs to the one used in the netconf2011 results. It turns out - this is a hell lot worse than dummy and so this patch is even more beneficial for 10G. Test setup: ---------- System under test sending packets out. Additional box connected directly dropping packets. Installed prio qdisc on the eth device and default netdev default length of 1000 used as is. The 3 prio bands each were set to 100 (didnt factor in the results). 5 packet runs were made and the middle 3 picked. results ------- The "cpu" column indicates the which cpu the sample was taken on, The "Pkt runx" carries the number of packets a cpu dequeued when forced to be in the "dequeuer" role. The "avg" for each run is the number of times each cpu should be a "dequeuer" if the system was fair. 3.0-rc4 (plain) cpu Pkt run1 Pkt run2 Pkt run3 ================================================ cpu0 21853354 21598183 22199900 cpu1 431058 473476 393159 cpu2 481975 477529 458466 cpu3 23261406 23412299 22894315 avg 11506948 11490372 11486460 3.0-rc4 with patch and default weight 64 cpu Pkt run1 Pkt run2 Pkt run3 ================================================ cpu0 13205312 13109359 13132333 cpu1 10189914 10159127 10122270 cpu2 10213871 10124367 10168722 cpu3 13165760 13164767 13096705 avg 11693714 11639405 11630008 As you can see the system is still not perfect but is a lot better than what it was before... At the moment we use the old backlog weight, weight_p which is 64 packets. It seems to be reasonably fine with that value. The system could be made more fair if we reduce the weight_p (as per my presentation), but we are going to affect the shared backlog weight. Unless deemed necessary, I think the default value is fine. If not we could add yet another knob. Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-06-26 16:13:54 +08:00
* Ordered by possible occurrence: Postpone processing if
* 1. we've exceeded packet quota
* 2. another process needs the CPU;
*/
net_sched: restore qdisc quota fairness limits after bulk dequeue Restore the quota fairness between qdisc's, that we broke with commit 5772e9a346 ("qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE"). Before that commit, the quota in __qdisc_run() were in packets as dequeue_skb() would only dequeue a single packet, that assumption broke with bulk dequeue. We choose not to account for the number of packets inside the TSO/GSO packets (accessable via "skb_gso_segs"). As the previous fairness also had this "defect". Thus, GSO/TSO packets counts as a single packet. Further more, we choose to slack on accuracy, by allowing a bulk dequeue try_bulk_dequeue_skb() to exceed the "packets" limit, only limited by the BQL bytelimit. This is done because BQL prefers to get its full budget for appropriate feedback from TX completion. In future, we might consider reworking this further and, if it allows, switch to a time-based model, as suggested by Eric. Right now, we only restore old semantics. Joint work with Eric, Hannes, Daniel and Jesper. Hannes wrote the first patch in cooperation with Daniel and Jesper. Eric rewrote the patch. Fixes: 5772e9a346 ("qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE") Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-09 18:18:10 +08:00
quota -= packets;
if (quota <= 0 || need_resched()) {
__netif_schedule(q);
break;
}
}
}
unsigned long dev_trans_start(struct net_device *dev)
{
unsigned long val, res;
unsigned int i;
if (is_vlan_dev(dev))
dev = vlan_dev_real_dev(dev);
else if (netif_is_macvlan(dev))
dev = macvlan_dev_real_dev(dev);
res = netdev_get_tx_queue(dev, 0)->trans_start;
for (i = 1; i < dev->num_tx_queues; i++) {
val = netdev_get_tx_queue(dev, i)->trans_start;
if (val && time_after(val, res))
res = val;
}
return res;
}
EXPORT_SYMBOL(dev_trans_start);
static void dev_watchdog(struct timer_list *t)
{
struct net_device *dev = from_timer(dev, t, watchdog_timer);
[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);
trans_start = txq->trans_start;
if (netif_xmit_stopped(txq) &&
time_after(jiffies, (trans_start +
dev->watchdog_timeo))) {
some_queue_timedout = 1;
txq->trans_timeout++;
break;
}
}
if (some_queue_timedout) {
WARN_ONCE(1, KERN_INFO "NETDEV WATCHDOG: %s (%s): transmit queue %u timed out\n",
dev->name, netdev_drivername(dev), 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;
atomic_inc(&dev->carrier_up_count);
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;
atomic_inc(&dev->carrier_down_count);
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,
struct sk_buff **to_free)
{
__qdisc_drop(skb, to_free);
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,
.q.lock = __SPIN_LOCK_UNLOCKED(noop_qdisc.q.lock),
.dev_queue = &noop_netdev_queue,
.running = SEQCNT_ZERO(noop_qdisc.running),
.busylock = __SPIN_LOCK_UNLOCKED(noop_qdisc.busylock),
.gso_skb = {
.next = (struct sk_buff *)&noop_qdisc.gso_skb,
.prev = (struct sk_buff *)&noop_qdisc.gso_skb,
.qlen = 0,
.lock = __SPIN_LOCK_UNLOCKED(noop_qdisc.gso_skb.lock),
},
.skb_bad_txq = {
.next = (struct sk_buff *)&noop_qdisc.skb_bad_txq,
.prev = (struct sk_buff *)&noop_qdisc.skb_bad_txq,
.qlen = 0,
.lock = __SPIN_LOCK_UNLOCKED(noop_qdisc.skb_bad_txq.lock),
},
};
EXPORT_SYMBOL(noop_qdisc);
static int noqueue_init(struct Qdisc *qdisc, struct nlattr *opt,
struct netlink_ext_ack *extack)
{
/* register_qdisc() assigns a default of noop_enqueue if unset,
* but __dev_queue_xmit() treats noqueue only as such
* if this is NULL - so clear it here. */
qdisc->enqueue = NULL;
return 0;
}
struct Qdisc_ops noqueue_qdisc_ops __read_mostly = {
.id = "noqueue",
.priv_size = 0,
.init = noqueue_init,
.enqueue = noop_enqueue,
.dequeue = noop_dequeue,
.peek = noop_dequeue,
.owner = THIS_MODULE,
};
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:
* - rings for priority bands
*/
struct pfifo_fast_priv {
struct skb_array q[PFIFO_FAST_BANDS];
};
static inline struct skb_array *band2list(struct pfifo_fast_priv *priv,
int band)
{
return &priv->q[band];
}
static int pfifo_fast_enqueue(struct sk_buff *skb, struct Qdisc *qdisc,
struct sk_buff **to_free)
{
int band = prio2band[skb->priority & TC_PRIO_MAX];
struct pfifo_fast_priv *priv = qdisc_priv(qdisc);
struct skb_array *q = band2list(priv, band);
net: sched: fix uses after free syzbot reported one use-after-free in pfifo_fast_enqueue() [1] Issue here is that we can not reuse skb after a successful skb_array_produce() since another cpu might have consumed it already. I believe a similar problem exists in try_bulk_dequeue_skb_slow() in case we put an skb into qdisc_enqueue_skb_bad_txq() for lockless qdisc. [1] BUG: KASAN: use-after-free in qdisc_pkt_len include/net/sch_generic.h:610 [inline] BUG: KASAN: use-after-free in qdisc_qstats_cpu_backlog_inc include/net/sch_generic.h:712 [inline] BUG: KASAN: use-after-free in pfifo_fast_enqueue+0x4bc/0x5e0 net/sched/sch_generic.c:639 Read of size 4 at addr ffff8801cede37e8 by task syzkaller717588/5543 CPU: 1 PID: 5543 Comm: syzkaller717588 Not tainted 4.16.0-rc4+ #265 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:17 [inline] dump_stack+0x194/0x24d lib/dump_stack.c:53 print_address_description+0x73/0x250 mm/kasan/report.c:256 kasan_report_error mm/kasan/report.c:354 [inline] kasan_report+0x23c/0x360 mm/kasan/report.c:412 __asan_report_load4_noabort+0x14/0x20 mm/kasan/report.c:432 qdisc_pkt_len include/net/sch_generic.h:610 [inline] qdisc_qstats_cpu_backlog_inc include/net/sch_generic.h:712 [inline] pfifo_fast_enqueue+0x4bc/0x5e0 net/sched/sch_generic.c:639 __dev_xmit_skb net/core/dev.c:3216 [inline] Fixes: c5ad119fb6c0 ("net: sched: pfifo_fast use skb_array") Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: syzbot+ed43b6903ab968b16f54@syzkaller.appspotmail.com Cc: John Fastabend <john.fastabend@gmail.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: Jiri Pirko <jiri@resnulli.us> Acked-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-03-15 09:53:00 +08:00
unsigned int pkt_len = qdisc_pkt_len(skb);
int err;
err = skb_array_produce(q, skb);
if (unlikely(err))
return qdisc_drop_cpu(skb, qdisc, to_free);
qdisc_qstats_cpu_qlen_inc(qdisc);
net: sched: fix uses after free syzbot reported one use-after-free in pfifo_fast_enqueue() [1] Issue here is that we can not reuse skb after a successful skb_array_produce() since another cpu might have consumed it already. I believe a similar problem exists in try_bulk_dequeue_skb_slow() in case we put an skb into qdisc_enqueue_skb_bad_txq() for lockless qdisc. [1] BUG: KASAN: use-after-free in qdisc_pkt_len include/net/sch_generic.h:610 [inline] BUG: KASAN: use-after-free in qdisc_qstats_cpu_backlog_inc include/net/sch_generic.h:712 [inline] BUG: KASAN: use-after-free in pfifo_fast_enqueue+0x4bc/0x5e0 net/sched/sch_generic.c:639 Read of size 4 at addr ffff8801cede37e8 by task syzkaller717588/5543 CPU: 1 PID: 5543 Comm: syzkaller717588 Not tainted 4.16.0-rc4+ #265 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:17 [inline] dump_stack+0x194/0x24d lib/dump_stack.c:53 print_address_description+0x73/0x250 mm/kasan/report.c:256 kasan_report_error mm/kasan/report.c:354 [inline] kasan_report+0x23c/0x360 mm/kasan/report.c:412 __asan_report_load4_noabort+0x14/0x20 mm/kasan/report.c:432 qdisc_pkt_len include/net/sch_generic.h:610 [inline] qdisc_qstats_cpu_backlog_inc include/net/sch_generic.h:712 [inline] pfifo_fast_enqueue+0x4bc/0x5e0 net/sched/sch_generic.c:639 __dev_xmit_skb net/core/dev.c:3216 [inline] Fixes: c5ad119fb6c0 ("net: sched: pfifo_fast use skb_array") Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: syzbot+ed43b6903ab968b16f54@syzkaller.appspotmail.com Cc: John Fastabend <john.fastabend@gmail.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: Jiri Pirko <jiri@resnulli.us> Acked-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-03-15 09:53:00 +08:00
/* Note: skb can not be used after skb_array_produce(),
* so we better not use qdisc_qstats_cpu_backlog_inc()
*/
this_cpu_add(qdisc->cpu_qstats->backlog, pkt_len);
return NET_XMIT_SUCCESS;
}
static struct sk_buff *pfifo_fast_dequeue(struct Qdisc *qdisc)
{
struct pfifo_fast_priv *priv = qdisc_priv(qdisc);
struct sk_buff *skb = NULL;
int band;
for (band = 0; band < PFIFO_FAST_BANDS && !skb; band++) {
struct skb_array *q = band2list(priv, band);
if (__skb_array_empty(q))
continue;
skb = __skb_array_consume(q);
}
if (likely(skb)) {
qdisc_qstats_cpu_backlog_dec(qdisc, skb);
qdisc_bstats_cpu_update(qdisc, skb);
qdisc_qstats_cpu_qlen_dec(qdisc);
}
return skb;
}
static struct sk_buff *pfifo_fast_peek(struct Qdisc *qdisc)
{
struct pfifo_fast_priv *priv = qdisc_priv(qdisc);
struct sk_buff *skb = NULL;
int band;
for (band = 0; band < PFIFO_FAST_BANDS && !skb; band++) {
struct skb_array *q = band2list(priv, band);
skb = __skb_array_peek(q);
}
return skb;
}
static void pfifo_fast_reset(struct Qdisc *qdisc)
{
int i, band;
struct pfifo_fast_priv *priv = qdisc_priv(qdisc);
for (band = 0; band < PFIFO_FAST_BANDS; band++) {
struct skb_array *q = band2list(priv, band);
struct sk_buff *skb;
/* NULL ring is possible if destroy path is due to a failed
* skb_array_init() in pfifo_fast_init() case.
*/
if (!q->ring.queue)
continue;
while ((skb = __skb_array_consume(q)) != NULL)
kfree_skb(skb);
}
for_each_possible_cpu(i) {
struct gnet_stats_queue *q = per_cpu_ptr(qdisc->cpu_qstats, i);
q->backlog = 0;
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);
if (nla_put(skb, TCA_OPTIONS, sizeof(opt), &opt))
goto nla_put_failure;
return skb->len;
nla_put_failure:
return -1;
}
static int pfifo_fast_init(struct Qdisc *qdisc, struct nlattr *opt,
struct netlink_ext_ack *extack)
{
unsigned int qlen = qdisc_dev(qdisc)->tx_queue_len;
struct pfifo_fast_priv *priv = qdisc_priv(qdisc);
int prio;
/* guard against zero length rings */
if (!qlen)
return -EINVAL;
for (prio = 0; prio < PFIFO_FAST_BANDS; prio++) {
struct skb_array *q = band2list(priv, prio);
int err;
err = skb_array_init(q, qlen, GFP_KERNEL);
if (err)
return -ENOMEM;
}
/* Can by-pass the queue discipline */
qdisc->flags |= TCQ_F_CAN_BYPASS;
return 0;
}
static void pfifo_fast_destroy(struct Qdisc *sch)
{
struct pfifo_fast_priv *priv = qdisc_priv(sch);
int prio;
for (prio = 0; prio < PFIFO_FAST_BANDS; prio++) {
struct skb_array *q = band2list(priv, prio);
/* NULL ring is possible if destroy path is due to a failed
* skb_array_init() in pfifo_fast_init() case.
*/
if (!q->ring.queue)
continue;
/* Destroy ring but no need to kfree_skb because a call to
* pfifo_fast_reset() has already done that work.
*/
ptr_ring_cleanup(&q->ring, NULL);
}
}
static int pfifo_fast_change_tx_queue_len(struct Qdisc *sch,
unsigned int new_len)
{
struct pfifo_fast_priv *priv = qdisc_priv(sch);
struct skb_array *bands[PFIFO_FAST_BANDS];
int prio;
for (prio = 0; prio < PFIFO_FAST_BANDS; prio++) {
struct skb_array *q = band2list(priv, prio);
bands[prio] = q;
}
return skb_array_resize_multiple(bands, PFIFO_FAST_BANDS, new_len,
GFP_KERNEL);
}
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,
.destroy = pfifo_fast_destroy,
.reset = pfifo_fast_reset,
.dump = pfifo_fast_dump,
.change_tx_queue_len = pfifo_fast_change_tx_queue_len,
.owner = THIS_MODULE,
.static_flags = TCQ_F_NOLOCK | TCQ_F_CPUSTATS,
};
EXPORT_SYMBOL(pfifo_fast_ops);
net: qdisc busylock needs lockdep annotations It seems we need to provide ability for stacked devices to use specific lock_class_key for sch->busylock We could instead default l2tpeth tx_queue_len to 0 (no qdisc), but a user might use a qdisc anyway. (So same fixes are probably needed on non LLTX stacked drivers) Noticed while stressing L2TPV3 setup : ====================================================== [ INFO: possible circular locking dependency detected ] 3.6.0-rc3+ #788 Not tainted ------------------------------------------------------- netperf/4660 is trying to acquire lock: (l2tpsock){+.-...}, at: [<ffffffffa0208db2>] l2tp_xmit_skb+0x172/0xa50 [l2tp_core] but task is already holding lock: (&(&sch->busylock)->rlock){+.-...}, at: [<ffffffff81596595>] dev_queue_xmit+0xd75/0xe00 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (&(&sch->busylock)->rlock){+.-...}: [<ffffffff810a5df0>] lock_acquire+0x90/0x200 [<ffffffff817499fc>] _raw_spin_lock_irqsave+0x4c/0x60 [<ffffffff81074872>] __wake_up+0x32/0x70 [<ffffffff8136d39e>] tty_wakeup+0x3e/0x80 [<ffffffff81378fb3>] pty_write+0x73/0x80 [<ffffffff8136cb4c>] tty_put_char+0x3c/0x40 [<ffffffff813722b2>] process_echoes+0x142/0x330 [<ffffffff813742ab>] n_tty_receive_buf+0x8fb/0x1230 [<ffffffff813777b2>] flush_to_ldisc+0x142/0x1c0 [<ffffffff81062818>] process_one_work+0x198/0x760 [<ffffffff81063236>] worker_thread+0x186/0x4b0 [<ffffffff810694d3>] kthread+0x93/0xa0 [<ffffffff81753e24>] kernel_thread_helper+0x4/0x10 -> #0 (l2tpsock){+.-...}: [<ffffffff810a5288>] __lock_acquire+0x1628/0x1b10 [<ffffffff810a5df0>] lock_acquire+0x90/0x200 [<ffffffff817498c1>] _raw_spin_lock+0x41/0x50 [<ffffffffa0208db2>] l2tp_xmit_skb+0x172/0xa50 [l2tp_core] [<ffffffffa021a802>] l2tp_eth_dev_xmit+0x32/0x60 [l2tp_eth] [<ffffffff815952b2>] dev_hard_start_xmit+0x502/0xa70 [<ffffffff815b63ce>] sch_direct_xmit+0xfe/0x290 [<ffffffff81595a05>] dev_queue_xmit+0x1e5/0xe00 [<ffffffff815d9d60>] ip_finish_output+0x3d0/0x890 [<ffffffff815db019>] ip_output+0x59/0xf0 [<ffffffff815da36d>] ip_local_out+0x2d/0xa0 [<ffffffff815da5a3>] ip_queue_xmit+0x1c3/0x680 [<ffffffff815f4192>] tcp_transmit_skb+0x402/0xa60 [<ffffffff815f4a94>] tcp_write_xmit+0x1f4/0xa30 [<ffffffff815f5300>] tcp_push_one+0x30/0x40 [<ffffffff815e6672>] tcp_sendmsg+0xe82/0x1040 [<ffffffff81614495>] inet_sendmsg+0x125/0x230 [<ffffffff81576cdc>] sock_sendmsg+0xdc/0xf0 [<ffffffff81579ece>] sys_sendto+0xfe/0x130 [<ffffffff81752c92>] system_call_fastpath+0x16/0x1b Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(&(&sch->busylock)->rlock); lock(l2tpsock); lock(&(&sch->busylock)->rlock); lock(l2tpsock); *** DEADLOCK *** 5 locks held by netperf/4660: #0: (sk_lock-AF_INET){+.+.+.}, at: [<ffffffff815e581c>] tcp_sendmsg+0x2c/0x1040 #1: (rcu_read_lock){.+.+..}, at: [<ffffffff815da3e0>] ip_queue_xmit+0x0/0x680 #2: (rcu_read_lock_bh){.+....}, at: [<ffffffff815d9ac5>] ip_finish_output+0x135/0x890 #3: (rcu_read_lock_bh){.+....}, at: [<ffffffff81595820>] dev_queue_xmit+0x0/0xe00 #4: (&(&sch->busylock)->rlock){+.-...}, at: [<ffffffff81596595>] dev_queue_xmit+0xd75/0xe00 stack backtrace: Pid: 4660, comm: netperf Not tainted 3.6.0-rc3+ #788 Call Trace: [<ffffffff8173dbf8>] print_circular_bug+0x1fb/0x20c [<ffffffff810a5288>] __lock_acquire+0x1628/0x1b10 [<ffffffff810a334b>] ? check_usage+0x9b/0x4d0 [<ffffffff810a3f44>] ? __lock_acquire+0x2e4/0x1b10 [<ffffffff810a5df0>] lock_acquire+0x90/0x200 [<ffffffffa0208db2>] ? l2tp_xmit_skb+0x172/0xa50 [l2tp_core] [<ffffffff817498c1>] _raw_spin_lock+0x41/0x50 [<ffffffffa0208db2>] ? l2tp_xmit_skb+0x172/0xa50 [l2tp_core] [<ffffffffa0208db2>] l2tp_xmit_skb+0x172/0xa50 [l2tp_core] [<ffffffffa021a802>] l2tp_eth_dev_xmit+0x32/0x60 [l2tp_eth] [<ffffffff815952b2>] dev_hard_start_xmit+0x502/0xa70 [<ffffffff81594e0e>] ? dev_hard_start_xmit+0x5e/0xa70 [<ffffffff81595961>] ? dev_queue_xmit+0x141/0xe00 [<ffffffff815b63ce>] sch_direct_xmit+0xfe/0x290 [<ffffffff81595a05>] dev_queue_xmit+0x1e5/0xe00 [<ffffffff81595820>] ? dev_hard_start_xmit+0xa70/0xa70 [<ffffffff815d9d60>] ip_finish_output+0x3d0/0x890 [<ffffffff815d9ac5>] ? ip_finish_output+0x135/0x890 [<ffffffff815db019>] ip_output+0x59/0xf0 [<ffffffff815da36d>] ip_local_out+0x2d/0xa0 [<ffffffff815da5a3>] ip_queue_xmit+0x1c3/0x680 [<ffffffff815da3e0>] ? ip_local_out+0xa0/0xa0 [<ffffffff815f4192>] tcp_transmit_skb+0x402/0xa60 [<ffffffff815fa25e>] ? tcp_md5_do_lookup+0x18e/0x1a0 [<ffffffff815f4a94>] tcp_write_xmit+0x1f4/0xa30 [<ffffffff815f5300>] tcp_push_one+0x30/0x40 [<ffffffff815e6672>] tcp_sendmsg+0xe82/0x1040 [<ffffffff81614495>] inet_sendmsg+0x125/0x230 [<ffffffff81614370>] ? inet_create+0x6b0/0x6b0 [<ffffffff8157e6e2>] ? sock_update_classid+0xc2/0x3b0 [<ffffffff8157e750>] ? sock_update_classid+0x130/0x3b0 [<ffffffff81576cdc>] sock_sendmsg+0xdc/0xf0 [<ffffffff81162579>] ? fget_light+0x3f9/0x4f0 [<ffffffff81579ece>] sys_sendto+0xfe/0x130 [<ffffffff810a69ad>] ? trace_hardirqs_on+0xd/0x10 [<ffffffff8174a0b0>] ? _raw_spin_unlock_irq+0x30/0x50 [<ffffffff810757e3>] ? finish_task_switch+0x83/0xf0 [<ffffffff810757a6>] ? finish_task_switch+0x46/0xf0 [<ffffffff81752cb7>] ? sysret_check+0x1b/0x56 [<ffffffff81752c92>] system_call_fastpath+0x16/0x1b Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-05 09:02:56 +08:00
static struct lock_class_key qdisc_tx_busylock;
static struct lock_class_key qdisc_running_key;
net: qdisc busylock needs lockdep annotations It seems we need to provide ability for stacked devices to use specific lock_class_key for sch->busylock We could instead default l2tpeth tx_queue_len to 0 (no qdisc), but a user might use a qdisc anyway. (So same fixes are probably needed on non LLTX stacked drivers) Noticed while stressing L2TPV3 setup : ====================================================== [ INFO: possible circular locking dependency detected ] 3.6.0-rc3+ #788 Not tainted ------------------------------------------------------- netperf/4660 is trying to acquire lock: (l2tpsock){+.-...}, at: [<ffffffffa0208db2>] l2tp_xmit_skb+0x172/0xa50 [l2tp_core] but task is already holding lock: (&(&sch->busylock)->rlock){+.-...}, at: [<ffffffff81596595>] dev_queue_xmit+0xd75/0xe00 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (&(&sch->busylock)->rlock){+.-...}: [<ffffffff810a5df0>] lock_acquire+0x90/0x200 [<ffffffff817499fc>] _raw_spin_lock_irqsave+0x4c/0x60 [<ffffffff81074872>] __wake_up+0x32/0x70 [<ffffffff8136d39e>] tty_wakeup+0x3e/0x80 [<ffffffff81378fb3>] pty_write+0x73/0x80 [<ffffffff8136cb4c>] tty_put_char+0x3c/0x40 [<ffffffff813722b2>] process_echoes+0x142/0x330 [<ffffffff813742ab>] n_tty_receive_buf+0x8fb/0x1230 [<ffffffff813777b2>] flush_to_ldisc+0x142/0x1c0 [<ffffffff81062818>] process_one_work+0x198/0x760 [<ffffffff81063236>] worker_thread+0x186/0x4b0 [<ffffffff810694d3>] kthread+0x93/0xa0 [<ffffffff81753e24>] kernel_thread_helper+0x4/0x10 -> #0 (l2tpsock){+.-...}: [<ffffffff810a5288>] __lock_acquire+0x1628/0x1b10 [<ffffffff810a5df0>] lock_acquire+0x90/0x200 [<ffffffff817498c1>] _raw_spin_lock+0x41/0x50 [<ffffffffa0208db2>] l2tp_xmit_skb+0x172/0xa50 [l2tp_core] [<ffffffffa021a802>] l2tp_eth_dev_xmit+0x32/0x60 [l2tp_eth] [<ffffffff815952b2>] dev_hard_start_xmit+0x502/0xa70 [<ffffffff815b63ce>] sch_direct_xmit+0xfe/0x290 [<ffffffff81595a05>] dev_queue_xmit+0x1e5/0xe00 [<ffffffff815d9d60>] ip_finish_output+0x3d0/0x890 [<ffffffff815db019>] ip_output+0x59/0xf0 [<ffffffff815da36d>] ip_local_out+0x2d/0xa0 [<ffffffff815da5a3>] ip_queue_xmit+0x1c3/0x680 [<ffffffff815f4192>] tcp_transmit_skb+0x402/0xa60 [<ffffffff815f4a94>] tcp_write_xmit+0x1f4/0xa30 [<ffffffff815f5300>] tcp_push_one+0x30/0x40 [<ffffffff815e6672>] tcp_sendmsg+0xe82/0x1040 [<ffffffff81614495>] inet_sendmsg+0x125/0x230 [<ffffffff81576cdc>] sock_sendmsg+0xdc/0xf0 [<ffffffff81579ece>] sys_sendto+0xfe/0x130 [<ffffffff81752c92>] system_call_fastpath+0x16/0x1b Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(&(&sch->busylock)->rlock); lock(l2tpsock); lock(&(&sch->busylock)->rlock); lock(l2tpsock); *** DEADLOCK *** 5 locks held by netperf/4660: #0: (sk_lock-AF_INET){+.+.+.}, at: [<ffffffff815e581c>] tcp_sendmsg+0x2c/0x1040 #1: (rcu_read_lock){.+.+..}, at: [<ffffffff815da3e0>] ip_queue_xmit+0x0/0x680 #2: (rcu_read_lock_bh){.+....}, at: [<ffffffff815d9ac5>] ip_finish_output+0x135/0x890 #3: (rcu_read_lock_bh){.+....}, at: [<ffffffff81595820>] dev_queue_xmit+0x0/0xe00 #4: (&(&sch->busylock)->rlock){+.-...}, at: [<ffffffff81596595>] dev_queue_xmit+0xd75/0xe00 stack backtrace: Pid: 4660, comm: netperf Not tainted 3.6.0-rc3+ #788 Call Trace: [<ffffffff8173dbf8>] print_circular_bug+0x1fb/0x20c [<ffffffff810a5288>] __lock_acquire+0x1628/0x1b10 [<ffffffff810a334b>] ? check_usage+0x9b/0x4d0 [<ffffffff810a3f44>] ? __lock_acquire+0x2e4/0x1b10 [<ffffffff810a5df0>] lock_acquire+0x90/0x200 [<ffffffffa0208db2>] ? l2tp_xmit_skb+0x172/0xa50 [l2tp_core] [<ffffffff817498c1>] _raw_spin_lock+0x41/0x50 [<ffffffffa0208db2>] ? l2tp_xmit_skb+0x172/0xa50 [l2tp_core] [<ffffffffa0208db2>] l2tp_xmit_skb+0x172/0xa50 [l2tp_core] [<ffffffffa021a802>] l2tp_eth_dev_xmit+0x32/0x60 [l2tp_eth] [<ffffffff815952b2>] dev_hard_start_xmit+0x502/0xa70 [<ffffffff81594e0e>] ? dev_hard_start_xmit+0x5e/0xa70 [<ffffffff81595961>] ? dev_queue_xmit+0x141/0xe00 [<ffffffff815b63ce>] sch_direct_xmit+0xfe/0x290 [<ffffffff81595a05>] dev_queue_xmit+0x1e5/0xe00 [<ffffffff81595820>] ? dev_hard_start_xmit+0xa70/0xa70 [<ffffffff815d9d60>] ip_finish_output+0x3d0/0x890 [<ffffffff815d9ac5>] ? ip_finish_output+0x135/0x890 [<ffffffff815db019>] ip_output+0x59/0xf0 [<ffffffff815da36d>] ip_local_out+0x2d/0xa0 [<ffffffff815da5a3>] ip_queue_xmit+0x1c3/0x680 [<ffffffff815da3e0>] ? ip_local_out+0xa0/0xa0 [<ffffffff815f4192>] tcp_transmit_skb+0x402/0xa60 [<ffffffff815fa25e>] ? tcp_md5_do_lookup+0x18e/0x1a0 [<ffffffff815f4a94>] tcp_write_xmit+0x1f4/0xa30 [<ffffffff815f5300>] tcp_push_one+0x30/0x40 [<ffffffff815e6672>] tcp_sendmsg+0xe82/0x1040 [<ffffffff81614495>] inet_sendmsg+0x125/0x230 [<ffffffff81614370>] ? inet_create+0x6b0/0x6b0 [<ffffffff8157e6e2>] ? sock_update_classid+0xc2/0x3b0 [<ffffffff8157e750>] ? sock_update_classid+0x130/0x3b0 [<ffffffff81576cdc>] sock_sendmsg+0xdc/0xf0 [<ffffffff81162579>] ? fget_light+0x3f9/0x4f0 [<ffffffff81579ece>] sys_sendto+0xfe/0x130 [<ffffffff810a69ad>] ? trace_hardirqs_on+0xd/0x10 [<ffffffff8174a0b0>] ? _raw_spin_unlock_irq+0x30/0x50 [<ffffffff810757e3>] ? finish_task_switch+0x83/0xf0 [<ffffffff810757a6>] ? finish_task_switch+0x46/0xf0 [<ffffffff81752cb7>] ? sysret_check+0x1b/0x56 [<ffffffff81752c92>] system_call_fastpath+0x16/0x1b Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-05 09:02:56 +08:00
struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue,
const struct Qdisc_ops *ops,
struct netlink_ext_ack *extack)
{
void *p;
struct Qdisc *sch;
unsigned int size = QDISC_ALIGN(sizeof(*sch)) + ops->priv_size;
int err = -ENOBUFS;
struct net_device *dev;
if (!dev_queue) {
NL_SET_ERR_MSG(extack, "No device queue given");
err = -EINVAL;
goto errout;
}
dev = dev_queue->dev;
p = kzalloc_node(size, GFP_KERNEL,
netdev_queue_numa_node_read(dev_queue));
if (!p)
goto errout;
sch = (struct Qdisc *) QDISC_ALIGN((unsigned long) p);
/* if we got non aligned memory, ask more and do alignment ourself */
if (sch != p) {
kfree(p);
p = kzalloc_node(size + QDISC_ALIGNTO - 1, GFP_KERNEL,
netdev_queue_numa_node_read(dev_queue));
if (!p)
goto errout;
sch = (struct Qdisc *) QDISC_ALIGN((unsigned long) p);
sch->padded = (char *) sch - (char *) p;
}
__skb_queue_head_init(&sch->gso_skb);
__skb_queue_head_init(&sch->skb_bad_txq);
qdisc_skb_head_init(&sch->q);
spin_lock_init(&sch->q.lock);
net: qdisc busylock needs lockdep annotations It seems we need to provide ability for stacked devices to use specific lock_class_key for sch->busylock We could instead default l2tpeth tx_queue_len to 0 (no qdisc), but a user might use a qdisc anyway. (So same fixes are probably needed on non LLTX stacked drivers) Noticed while stressing L2TPV3 setup : ====================================================== [ INFO: possible circular locking dependency detected ] 3.6.0-rc3+ #788 Not tainted ------------------------------------------------------- netperf/4660 is trying to acquire lock: (l2tpsock){+.-...}, at: [<ffffffffa0208db2>] l2tp_xmit_skb+0x172/0xa50 [l2tp_core] but task is already holding lock: (&(&sch->busylock)->rlock){+.-...}, at: [<ffffffff81596595>] dev_queue_xmit+0xd75/0xe00 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (&(&sch->busylock)->rlock){+.-...}: [<ffffffff810a5df0>] lock_acquire+0x90/0x200 [<ffffffff817499fc>] _raw_spin_lock_irqsave+0x4c/0x60 [<ffffffff81074872>] __wake_up+0x32/0x70 [<ffffffff8136d39e>] tty_wakeup+0x3e/0x80 [<ffffffff81378fb3>] pty_write+0x73/0x80 [<ffffffff8136cb4c>] tty_put_char+0x3c/0x40 [<ffffffff813722b2>] process_echoes+0x142/0x330 [<ffffffff813742ab>] n_tty_receive_buf+0x8fb/0x1230 [<ffffffff813777b2>] flush_to_ldisc+0x142/0x1c0 [<ffffffff81062818>] process_one_work+0x198/0x760 [<ffffffff81063236>] worker_thread+0x186/0x4b0 [<ffffffff810694d3>] kthread+0x93/0xa0 [<ffffffff81753e24>] kernel_thread_helper+0x4/0x10 -> #0 (l2tpsock){+.-...}: [<ffffffff810a5288>] __lock_acquire+0x1628/0x1b10 [<ffffffff810a5df0>] lock_acquire+0x90/0x200 [<ffffffff817498c1>] _raw_spin_lock+0x41/0x50 [<ffffffffa0208db2>] l2tp_xmit_skb+0x172/0xa50 [l2tp_core] [<ffffffffa021a802>] l2tp_eth_dev_xmit+0x32/0x60 [l2tp_eth] [<ffffffff815952b2>] dev_hard_start_xmit+0x502/0xa70 [<ffffffff815b63ce>] sch_direct_xmit+0xfe/0x290 [<ffffffff81595a05>] dev_queue_xmit+0x1e5/0xe00 [<ffffffff815d9d60>] ip_finish_output+0x3d0/0x890 [<ffffffff815db019>] ip_output+0x59/0xf0 [<ffffffff815da36d>] ip_local_out+0x2d/0xa0 [<ffffffff815da5a3>] ip_queue_xmit+0x1c3/0x680 [<ffffffff815f4192>] tcp_transmit_skb+0x402/0xa60 [<ffffffff815f4a94>] tcp_write_xmit+0x1f4/0xa30 [<ffffffff815f5300>] tcp_push_one+0x30/0x40 [<ffffffff815e6672>] tcp_sendmsg+0xe82/0x1040 [<ffffffff81614495>] inet_sendmsg+0x125/0x230 [<ffffffff81576cdc>] sock_sendmsg+0xdc/0xf0 [<ffffffff81579ece>] sys_sendto+0xfe/0x130 [<ffffffff81752c92>] system_call_fastpath+0x16/0x1b Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(&(&sch->busylock)->rlock); lock(l2tpsock); lock(&(&sch->busylock)->rlock); lock(l2tpsock); *** DEADLOCK *** 5 locks held by netperf/4660: #0: (sk_lock-AF_INET){+.+.+.}, at: [<ffffffff815e581c>] tcp_sendmsg+0x2c/0x1040 #1: (rcu_read_lock){.+.+..}, at: [<ffffffff815da3e0>] ip_queue_xmit+0x0/0x680 #2: (rcu_read_lock_bh){.+....}, at: [<ffffffff815d9ac5>] ip_finish_output+0x135/0x890 #3: (rcu_read_lock_bh){.+....}, at: [<ffffffff81595820>] dev_queue_xmit+0x0/0xe00 #4: (&(&sch->busylock)->rlock){+.-...}, at: [<ffffffff81596595>] dev_queue_xmit+0xd75/0xe00 stack backtrace: Pid: 4660, comm: netperf Not tainted 3.6.0-rc3+ #788 Call Trace: [<ffffffff8173dbf8>] print_circular_bug+0x1fb/0x20c [<ffffffff810a5288>] __lock_acquire+0x1628/0x1b10 [<ffffffff810a334b>] ? check_usage+0x9b/0x4d0 [<ffffffff810a3f44>] ? __lock_acquire+0x2e4/0x1b10 [<ffffffff810a5df0>] lock_acquire+0x90/0x200 [<ffffffffa0208db2>] ? l2tp_xmit_skb+0x172/0xa50 [l2tp_core] [<ffffffff817498c1>] _raw_spin_lock+0x41/0x50 [<ffffffffa0208db2>] ? l2tp_xmit_skb+0x172/0xa50 [l2tp_core] [<ffffffffa0208db2>] l2tp_xmit_skb+0x172/0xa50 [l2tp_core] [<ffffffffa021a802>] l2tp_eth_dev_xmit+0x32/0x60 [l2tp_eth] [<ffffffff815952b2>] dev_hard_start_xmit+0x502/0xa70 [<ffffffff81594e0e>] ? dev_hard_start_xmit+0x5e/0xa70 [<ffffffff81595961>] ? dev_queue_xmit+0x141/0xe00 [<ffffffff815b63ce>] sch_direct_xmit+0xfe/0x290 [<ffffffff81595a05>] dev_queue_xmit+0x1e5/0xe00 [<ffffffff81595820>] ? dev_hard_start_xmit+0xa70/0xa70 [<ffffffff815d9d60>] ip_finish_output+0x3d0/0x890 [<ffffffff815d9ac5>] ? ip_finish_output+0x135/0x890 [<ffffffff815db019>] ip_output+0x59/0xf0 [<ffffffff815da36d>] ip_local_out+0x2d/0xa0 [<ffffffff815da5a3>] ip_queue_xmit+0x1c3/0x680 [<ffffffff815da3e0>] ? ip_local_out+0xa0/0xa0 [<ffffffff815f4192>] tcp_transmit_skb+0x402/0xa60 [<ffffffff815fa25e>] ? tcp_md5_do_lookup+0x18e/0x1a0 [<ffffffff815f4a94>] tcp_write_xmit+0x1f4/0xa30 [<ffffffff815f5300>] tcp_push_one+0x30/0x40 [<ffffffff815e6672>] tcp_sendmsg+0xe82/0x1040 [<ffffffff81614495>] inet_sendmsg+0x125/0x230 [<ffffffff81614370>] ? inet_create+0x6b0/0x6b0 [<ffffffff8157e6e2>] ? sock_update_classid+0xc2/0x3b0 [<ffffffff8157e750>] ? sock_update_classid+0x130/0x3b0 [<ffffffff81576cdc>] sock_sendmsg+0xdc/0xf0 [<ffffffff81162579>] ? fget_light+0x3f9/0x4f0 [<ffffffff81579ece>] sys_sendto+0xfe/0x130 [<ffffffff810a69ad>] ? trace_hardirqs_on+0xd/0x10 [<ffffffff8174a0b0>] ? _raw_spin_unlock_irq+0x30/0x50 [<ffffffff810757e3>] ? finish_task_switch+0x83/0xf0 [<ffffffff810757a6>] ? finish_task_switch+0x46/0xf0 [<ffffffff81752cb7>] ? sysret_check+0x1b/0x56 [<ffffffff81752c92>] system_call_fastpath+0x16/0x1b Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-05 09:02:56 +08:00
if (ops->static_flags & TCQ_F_CPUSTATS) {
sch->cpu_bstats =
netdev_alloc_pcpu_stats(struct gnet_stats_basic_cpu);
if (!sch->cpu_bstats)
goto errout1;
sch->cpu_qstats = alloc_percpu(struct gnet_stats_queue);
if (!sch->cpu_qstats) {
free_percpu(sch->cpu_bstats);
goto errout1;
}
}
spin_lock_init(&sch->busylock);
net: qdisc busylock needs lockdep annotations It seems we need to provide ability for stacked devices to use specific lock_class_key for sch->busylock We could instead default l2tpeth tx_queue_len to 0 (no qdisc), but a user might use a qdisc anyway. (So same fixes are probably needed on non LLTX stacked drivers) Noticed while stressing L2TPV3 setup : ====================================================== [ INFO: possible circular locking dependency detected ] 3.6.0-rc3+ #788 Not tainted ------------------------------------------------------- netperf/4660 is trying to acquire lock: (l2tpsock){+.-...}, at: [<ffffffffa0208db2>] l2tp_xmit_skb+0x172/0xa50 [l2tp_core] but task is already holding lock: (&(&sch->busylock)->rlock){+.-...}, at: [<ffffffff81596595>] dev_queue_xmit+0xd75/0xe00 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (&(&sch->busylock)->rlock){+.-...}: [<ffffffff810a5df0>] lock_acquire+0x90/0x200 [<ffffffff817499fc>] _raw_spin_lock_irqsave+0x4c/0x60 [<ffffffff81074872>] __wake_up+0x32/0x70 [<ffffffff8136d39e>] tty_wakeup+0x3e/0x80 [<ffffffff81378fb3>] pty_write+0x73/0x80 [<ffffffff8136cb4c>] tty_put_char+0x3c/0x40 [<ffffffff813722b2>] process_echoes+0x142/0x330 [<ffffffff813742ab>] n_tty_receive_buf+0x8fb/0x1230 [<ffffffff813777b2>] flush_to_ldisc+0x142/0x1c0 [<ffffffff81062818>] process_one_work+0x198/0x760 [<ffffffff81063236>] worker_thread+0x186/0x4b0 [<ffffffff810694d3>] kthread+0x93/0xa0 [<ffffffff81753e24>] kernel_thread_helper+0x4/0x10 -> #0 (l2tpsock){+.-...}: [<ffffffff810a5288>] __lock_acquire+0x1628/0x1b10 [<ffffffff810a5df0>] lock_acquire+0x90/0x200 [<ffffffff817498c1>] _raw_spin_lock+0x41/0x50 [<ffffffffa0208db2>] l2tp_xmit_skb+0x172/0xa50 [l2tp_core] [<ffffffffa021a802>] l2tp_eth_dev_xmit+0x32/0x60 [l2tp_eth] [<ffffffff815952b2>] dev_hard_start_xmit+0x502/0xa70 [<ffffffff815b63ce>] sch_direct_xmit+0xfe/0x290 [<ffffffff81595a05>] dev_queue_xmit+0x1e5/0xe00 [<ffffffff815d9d60>] ip_finish_output+0x3d0/0x890 [<ffffffff815db019>] ip_output+0x59/0xf0 [<ffffffff815da36d>] ip_local_out+0x2d/0xa0 [<ffffffff815da5a3>] ip_queue_xmit+0x1c3/0x680 [<ffffffff815f4192>] tcp_transmit_skb+0x402/0xa60 [<ffffffff815f4a94>] tcp_write_xmit+0x1f4/0xa30 [<ffffffff815f5300>] tcp_push_one+0x30/0x40 [<ffffffff815e6672>] tcp_sendmsg+0xe82/0x1040 [<ffffffff81614495>] inet_sendmsg+0x125/0x230 [<ffffffff81576cdc>] sock_sendmsg+0xdc/0xf0 [<ffffffff81579ece>] sys_sendto+0xfe/0x130 [<ffffffff81752c92>] system_call_fastpath+0x16/0x1b Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(&(&sch->busylock)->rlock); lock(l2tpsock); lock(&(&sch->busylock)->rlock); lock(l2tpsock); *** DEADLOCK *** 5 locks held by netperf/4660: #0: (sk_lock-AF_INET){+.+.+.}, at: [<ffffffff815e581c>] tcp_sendmsg+0x2c/0x1040 #1: (rcu_read_lock){.+.+..}, at: [<ffffffff815da3e0>] ip_queue_xmit+0x0/0x680 #2: (rcu_read_lock_bh){.+....}, at: [<ffffffff815d9ac5>] ip_finish_output+0x135/0x890 #3: (rcu_read_lock_bh){.+....}, at: [<ffffffff81595820>] dev_queue_xmit+0x0/0xe00 #4: (&(&sch->busylock)->rlock){+.-...}, at: [<ffffffff81596595>] dev_queue_xmit+0xd75/0xe00 stack backtrace: Pid: 4660, comm: netperf Not tainted 3.6.0-rc3+ #788 Call Trace: [<ffffffff8173dbf8>] print_circular_bug+0x1fb/0x20c [<ffffffff810a5288>] __lock_acquire+0x1628/0x1b10 [<ffffffff810a334b>] ? check_usage+0x9b/0x4d0 [<ffffffff810a3f44>] ? __lock_acquire+0x2e4/0x1b10 [<ffffffff810a5df0>] lock_acquire+0x90/0x200 [<ffffffffa0208db2>] ? l2tp_xmit_skb+0x172/0xa50 [l2tp_core] [<ffffffff817498c1>] _raw_spin_lock+0x41/0x50 [<ffffffffa0208db2>] ? l2tp_xmit_skb+0x172/0xa50 [l2tp_core] [<ffffffffa0208db2>] l2tp_xmit_skb+0x172/0xa50 [l2tp_core] [<ffffffffa021a802>] l2tp_eth_dev_xmit+0x32/0x60 [l2tp_eth] [<ffffffff815952b2>] dev_hard_start_xmit+0x502/0xa70 [<ffffffff81594e0e>] ? dev_hard_start_xmit+0x5e/0xa70 [<ffffffff81595961>] ? dev_queue_xmit+0x141/0xe00 [<ffffffff815b63ce>] sch_direct_xmit+0xfe/0x290 [<ffffffff81595a05>] dev_queue_xmit+0x1e5/0xe00 [<ffffffff81595820>] ? dev_hard_start_xmit+0xa70/0xa70 [<ffffffff815d9d60>] ip_finish_output+0x3d0/0x890 [<ffffffff815d9ac5>] ? ip_finish_output+0x135/0x890 [<ffffffff815db019>] ip_output+0x59/0xf0 [<ffffffff815da36d>] ip_local_out+0x2d/0xa0 [<ffffffff815da5a3>] ip_queue_xmit+0x1c3/0x680 [<ffffffff815da3e0>] ? ip_local_out+0xa0/0xa0 [<ffffffff815f4192>] tcp_transmit_skb+0x402/0xa60 [<ffffffff815fa25e>] ? tcp_md5_do_lookup+0x18e/0x1a0 [<ffffffff815f4a94>] tcp_write_xmit+0x1f4/0xa30 [<ffffffff815f5300>] tcp_push_one+0x30/0x40 [<ffffffff815e6672>] tcp_sendmsg+0xe82/0x1040 [<ffffffff81614495>] inet_sendmsg+0x125/0x230 [<ffffffff81614370>] ? inet_create+0x6b0/0x6b0 [<ffffffff8157e6e2>] ? sock_update_classid+0xc2/0x3b0 [<ffffffff8157e750>] ? sock_update_classid+0x130/0x3b0 [<ffffffff81576cdc>] sock_sendmsg+0xdc/0xf0 [<ffffffff81162579>] ? fget_light+0x3f9/0x4f0 [<ffffffff81579ece>] sys_sendto+0xfe/0x130 [<ffffffff810a69ad>] ? trace_hardirqs_on+0xd/0x10 [<ffffffff8174a0b0>] ? _raw_spin_unlock_irq+0x30/0x50 [<ffffffff810757e3>] ? finish_task_switch+0x83/0xf0 [<ffffffff810757a6>] ? finish_task_switch+0x46/0xf0 [<ffffffff81752cb7>] ? sysret_check+0x1b/0x56 [<ffffffff81752c92>] system_call_fastpath+0x16/0x1b Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-05 09:02:56 +08:00
lockdep_set_class(&sch->busylock,
dev->qdisc_tx_busylock ?: &qdisc_tx_busylock);
/* seqlock has the same scope of busylock, for NOLOCK qdisc */
spin_lock_init(&sch->seqlock);
lockdep_set_class(&sch->busylock,
dev->qdisc_tx_busylock ?: &qdisc_tx_busylock);
seqcount_init(&sch->running);
lockdep_set_class(&sch->running,
dev->qdisc_running_key ?: &qdisc_running_key);
sch->ops = ops;
sch->flags = ops->static_flags;
sch->enqueue = ops->enqueue;
sch->dequeue = ops->dequeue;
sch->dev_queue = dev_queue;
net: qdisc busylock needs lockdep annotations It seems we need to provide ability for stacked devices to use specific lock_class_key for sch->busylock We could instead default l2tpeth tx_queue_len to 0 (no qdisc), but a user might use a qdisc anyway. (So same fixes are probably needed on non LLTX stacked drivers) Noticed while stressing L2TPV3 setup : ====================================================== [ INFO: possible circular locking dependency detected ] 3.6.0-rc3+ #788 Not tainted ------------------------------------------------------- netperf/4660 is trying to acquire lock: (l2tpsock){+.-...}, at: [<ffffffffa0208db2>] l2tp_xmit_skb+0x172/0xa50 [l2tp_core] but task is already holding lock: (&(&sch->busylock)->rlock){+.-...}, at: [<ffffffff81596595>] dev_queue_xmit+0xd75/0xe00 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (&(&sch->busylock)->rlock){+.-...}: [<ffffffff810a5df0>] lock_acquire+0x90/0x200 [<ffffffff817499fc>] _raw_spin_lock_irqsave+0x4c/0x60 [<ffffffff81074872>] __wake_up+0x32/0x70 [<ffffffff8136d39e>] tty_wakeup+0x3e/0x80 [<ffffffff81378fb3>] pty_write+0x73/0x80 [<ffffffff8136cb4c>] tty_put_char+0x3c/0x40 [<ffffffff813722b2>] process_echoes+0x142/0x330 [<ffffffff813742ab>] n_tty_receive_buf+0x8fb/0x1230 [<ffffffff813777b2>] flush_to_ldisc+0x142/0x1c0 [<ffffffff81062818>] process_one_work+0x198/0x760 [<ffffffff81063236>] worker_thread+0x186/0x4b0 [<ffffffff810694d3>] kthread+0x93/0xa0 [<ffffffff81753e24>] kernel_thread_helper+0x4/0x10 -> #0 (l2tpsock){+.-...}: [<ffffffff810a5288>] __lock_acquire+0x1628/0x1b10 [<ffffffff810a5df0>] lock_acquire+0x90/0x200 [<ffffffff817498c1>] _raw_spin_lock+0x41/0x50 [<ffffffffa0208db2>] l2tp_xmit_skb+0x172/0xa50 [l2tp_core] [<ffffffffa021a802>] l2tp_eth_dev_xmit+0x32/0x60 [l2tp_eth] [<ffffffff815952b2>] dev_hard_start_xmit+0x502/0xa70 [<ffffffff815b63ce>] sch_direct_xmit+0xfe/0x290 [<ffffffff81595a05>] dev_queue_xmit+0x1e5/0xe00 [<ffffffff815d9d60>] ip_finish_output+0x3d0/0x890 [<ffffffff815db019>] ip_output+0x59/0xf0 [<ffffffff815da36d>] ip_local_out+0x2d/0xa0 [<ffffffff815da5a3>] ip_queue_xmit+0x1c3/0x680 [<ffffffff815f4192>] tcp_transmit_skb+0x402/0xa60 [<ffffffff815f4a94>] tcp_write_xmit+0x1f4/0xa30 [<ffffffff815f5300>] tcp_push_one+0x30/0x40 [<ffffffff815e6672>] tcp_sendmsg+0xe82/0x1040 [<ffffffff81614495>] inet_sendmsg+0x125/0x230 [<ffffffff81576cdc>] sock_sendmsg+0xdc/0xf0 [<ffffffff81579ece>] sys_sendto+0xfe/0x130 [<ffffffff81752c92>] system_call_fastpath+0x16/0x1b Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(&(&sch->busylock)->rlock); lock(l2tpsock); lock(&(&sch->busylock)->rlock); lock(l2tpsock); *** DEADLOCK *** 5 locks held by netperf/4660: #0: (sk_lock-AF_INET){+.+.+.}, at: [<ffffffff815e581c>] tcp_sendmsg+0x2c/0x1040 #1: (rcu_read_lock){.+.+..}, at: [<ffffffff815da3e0>] ip_queue_xmit+0x0/0x680 #2: (rcu_read_lock_bh){.+....}, at: [<ffffffff815d9ac5>] ip_finish_output+0x135/0x890 #3: (rcu_read_lock_bh){.+....}, at: [<ffffffff81595820>] dev_queue_xmit+0x0/0xe00 #4: (&(&sch->busylock)->rlock){+.-...}, at: [<ffffffff81596595>] dev_queue_xmit+0xd75/0xe00 stack backtrace: Pid: 4660, comm: netperf Not tainted 3.6.0-rc3+ #788 Call Trace: [<ffffffff8173dbf8>] print_circular_bug+0x1fb/0x20c [<ffffffff810a5288>] __lock_acquire+0x1628/0x1b10 [<ffffffff810a334b>] ? check_usage+0x9b/0x4d0 [<ffffffff810a3f44>] ? __lock_acquire+0x2e4/0x1b10 [<ffffffff810a5df0>] lock_acquire+0x90/0x200 [<ffffffffa0208db2>] ? l2tp_xmit_skb+0x172/0xa50 [l2tp_core] [<ffffffff817498c1>] _raw_spin_lock+0x41/0x50 [<ffffffffa0208db2>] ? l2tp_xmit_skb+0x172/0xa50 [l2tp_core] [<ffffffffa0208db2>] l2tp_xmit_skb+0x172/0xa50 [l2tp_core] [<ffffffffa021a802>] l2tp_eth_dev_xmit+0x32/0x60 [l2tp_eth] [<ffffffff815952b2>] dev_hard_start_xmit+0x502/0xa70 [<ffffffff81594e0e>] ? dev_hard_start_xmit+0x5e/0xa70 [<ffffffff81595961>] ? dev_queue_xmit+0x141/0xe00 [<ffffffff815b63ce>] sch_direct_xmit+0xfe/0x290 [<ffffffff81595a05>] dev_queue_xmit+0x1e5/0xe00 [<ffffffff81595820>] ? dev_hard_start_xmit+0xa70/0xa70 [<ffffffff815d9d60>] ip_finish_output+0x3d0/0x890 [<ffffffff815d9ac5>] ? ip_finish_output+0x135/0x890 [<ffffffff815db019>] ip_output+0x59/0xf0 [<ffffffff815da36d>] ip_local_out+0x2d/0xa0 [<ffffffff815da5a3>] ip_queue_xmit+0x1c3/0x680 [<ffffffff815da3e0>] ? ip_local_out+0xa0/0xa0 [<ffffffff815f4192>] tcp_transmit_skb+0x402/0xa60 [<ffffffff815fa25e>] ? tcp_md5_do_lookup+0x18e/0x1a0 [<ffffffff815f4a94>] tcp_write_xmit+0x1f4/0xa30 [<ffffffff815f5300>] tcp_push_one+0x30/0x40 [<ffffffff815e6672>] tcp_sendmsg+0xe82/0x1040 [<ffffffff81614495>] inet_sendmsg+0x125/0x230 [<ffffffff81614370>] ? inet_create+0x6b0/0x6b0 [<ffffffff8157e6e2>] ? sock_update_classid+0xc2/0x3b0 [<ffffffff8157e750>] ? sock_update_classid+0x130/0x3b0 [<ffffffff81576cdc>] sock_sendmsg+0xdc/0xf0 [<ffffffff81162579>] ? fget_light+0x3f9/0x4f0 [<ffffffff81579ece>] sys_sendto+0xfe/0x130 [<ffffffff810a69ad>] ? trace_hardirqs_on+0xd/0x10 [<ffffffff8174a0b0>] ? _raw_spin_unlock_irq+0x30/0x50 [<ffffffff810757e3>] ? finish_task_switch+0x83/0xf0 [<ffffffff810757a6>] ? finish_task_switch+0x46/0xf0 [<ffffffff81752cb7>] ? sysret_check+0x1b/0x56 [<ffffffff81752c92>] system_call_fastpath+0x16/0x1b Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-05 09:02:56 +08:00
dev_hold(dev);
refcount_set(&sch->refcnt, 1);
return sch;
errout1:
kfree(p);
errout:
return ERR_PTR(err);
}
struct Qdisc *qdisc_create_dflt(struct netdev_queue *dev_queue,
const struct Qdisc_ops *ops,
unsigned int parentid,
struct netlink_ext_ack *extack)
{
struct Qdisc *sch;
if (!try_module_get(ops->owner)) {
NL_SET_ERR_MSG(extack, "Failed to increase module reference counter");
return NULL;
}
sch = qdisc_alloc(dev_queue, ops, extack);
if (IS_ERR(sch)) {
module_put(ops->owner);
return NULL;
}
sch->parent = parentid;
if (!ops->init || ops->init(sch, NULL, extack) == 0)
return sch;
qdisc_put(sch);
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;
struct sk_buff *skb, *tmp;
if (ops->reset)
ops->reset(qdisc);
skb_queue_walk_safe(&qdisc->gso_skb, skb, tmp) {
__skb_unlink(skb, &qdisc->gso_skb);
kfree_skb_list(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
}
skb_queue_walk_safe(&qdisc->skb_bad_txq, skb, tmp) {
__skb_unlink(skb, &qdisc->skb_bad_txq);
kfree_skb_list(skb);
}
qdisc->q.qlen = 0;
qdisc->qstats.backlog = 0;
}
EXPORT_SYMBOL(qdisc_reset);
net, sched: fix panic when updating miniq {b,q}stats While working on fixing another bug, I ran into the following panic on arm64 by simply attaching clsact qdisc, adding a filter and running traffic on ingress to it: [...] [ 178.188591] Unable to handle kernel read from unreadable memory at virtual address 810fb501f000 [ 178.197314] Mem abort info: [ 178.200121] ESR = 0x96000004 [ 178.203168] Exception class = DABT (current EL), IL = 32 bits [ 178.209095] SET = 0, FnV = 0 [ 178.212157] EA = 0, S1PTW = 0 [ 178.215288] Data abort info: [ 178.218175] ISV = 0, ISS = 0x00000004 [ 178.222019] CM = 0, WnR = 0 [ 178.224997] user pgtable: 4k pages, 48-bit VAs, pgd = 0000000023cb3f33 [ 178.231531] [0000810fb501f000] *pgd=0000000000000000 [ 178.236508] Internal error: Oops: 96000004 [#1] SMP [...] [ 178.311855] CPU: 73 PID: 2497 Comm: ping Tainted: G W 4.15.0-rc7+ #5 [ 178.319413] Hardware name: FOXCONN R2-1221R-A4/C2U4N_MB, BIOS G31FB18A 03/31/2017 [ 178.326887] pstate: 60400005 (nZCv daif +PAN -UAO) [ 178.331685] pc : __netif_receive_skb_core+0x49c/0xac8 [ 178.336728] lr : __netif_receive_skb+0x28/0x78 [ 178.341161] sp : ffff00002344b750 [ 178.344465] x29: ffff00002344b750 x28: ffff810fbdfd0580 [ 178.349769] x27: 0000000000000000 x26: ffff000009378000 [...] [ 178.418715] x1 : 0000000000000054 x0 : 0000000000000000 [ 178.424020] Process ping (pid: 2497, stack limit = 0x000000009f0a3ff4) [ 178.430537] Call trace: [ 178.432976] __netif_receive_skb_core+0x49c/0xac8 [ 178.437670] __netif_receive_skb+0x28/0x78 [ 178.441757] process_backlog+0x9c/0x160 [ 178.445584] net_rx_action+0x2f8/0x3f0 [...] Reason is that sch_ingress and sch_clsact are doing mini_qdisc_pair_init() which sets up miniq pointers to cpu_{b,q}stats from the underlying qdisc. Problem is that this cannot work since they are actually set up right after the qdisc ->init() callback in qdisc_create(), so first packet going into sch_handle_ingress() tries to call mini_qdisc_bstats_cpu_update() and we therefore panic. In order to fix this, allocation of {b,q}stats needs to happen before we call into ->init(). In net-next, there's already such option through commit d59f5ffa59d8 ("net: sched: a dflt qdisc may be used with per cpu stats"). However, the bug needs to be fixed in net still for 4.15. Thus, include these bits to reduce any merge churn and reuse the static_flags field to set TCQ_F_CPUSTATS, and remove the allocation from qdisc_create() since there is no other user left. Prashant Bhole ran into the same issue but for net-next, thus adding him below as well as co-author. Same issue was also reported by Sandipan Das when using bcc. Fixes: 46209401f8f6 ("net: core: introduce mini_Qdisc and eliminate usage of tp->q for clsact fastpath") Reference: https://lists.iovisor.org/pipermail/iovisor-dev/2018-January/001190.html Reported-by: Sandipan Das <sandipan@linux.vnet.ibm.com> Co-authored-by: Prashant Bhole <bhole_prashant_q7@lab.ntt.co.jp> Co-authored-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Cc: Jiri Pirko <jiri@resnulli.us> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-01-16 06:12:09 +08:00
void qdisc_free(struct Qdisc *qdisc)
{
if (qdisc_is_percpu_stats(qdisc)) {
free_percpu(qdisc->cpu_bstats);
free_percpu(qdisc->cpu_qstats);
}
kfree((char *) qdisc - qdisc->padded);
}
static void qdisc_free_cb(struct rcu_head *head)
{
struct Qdisc *q = container_of(head, struct Qdisc, rcu);
qdisc_free(q);
}
static void qdisc_destroy(struct Qdisc *qdisc)
{
const struct Qdisc_ops *ops = qdisc->ops;
struct sk_buff *skb, *tmp;
#ifdef CONFIG_NET_SCHED
qdisc_hash_del(qdisc);
qdisc_put_stab(rtnl_dereference(qdisc->stab));
#endif
gen_kill_estimator(&qdisc->rate_est);
if (ops->reset)
ops->reset(qdisc);
if (ops->destroy)
ops->destroy(qdisc);
module_put(ops->owner);
dev_put(qdisc_dev(qdisc));
skb_queue_walk_safe(&qdisc->gso_skb, skb, tmp) {
__skb_unlink(skb, &qdisc->gso_skb);
kfree_skb_list(skb);
}
skb_queue_walk_safe(&qdisc->skb_bad_txq, skb, tmp) {
__skb_unlink(skb, &qdisc->skb_bad_txq);
kfree_skb_list(skb);
}
call_rcu(&qdisc->rcu, qdisc_free_cb);
}
void qdisc_put(struct Qdisc *qdisc)
{
if (qdisc->flags & TCQ_F_BUILTIN ||
!refcount_dec_and_test(&qdisc->refcnt))
return;
qdisc_destroy(qdisc);
}
EXPORT_SYMBOL(qdisc_put);
/* Version of qdisc_put() that is called with rtnl mutex unlocked.
* Intended to be used as optimization, this function only takes rtnl lock if
* qdisc reference counter reached zero.
*/
void qdisc_put_unlocked(struct Qdisc *qdisc)
{
if (qdisc->flags & TCQ_F_BUILTIN ||
!refcount_dec_and_rtnl_lock(&qdisc->refcnt))
return;
qdisc_destroy(qdisc);
rtnl_unlock();
}
EXPORT_SYMBOL(qdisc_put_unlocked);
/* 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);
/* ... 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;
}
net_sched: implement a root container qdisc sch_mqprio This implements a mqprio queueing discipline that by default creates a pfifo_fast qdisc per tx queue and provides the needed configuration interface. Using the mqprio qdisc the number of tcs currently in use along with the range of queues alloted to each class can be configured. By default skbs are mapped to traffic classes using the skb priority. This mapping is configurable. Configurable parameters, struct tc_mqprio_qopt { __u8 num_tc; __u8 prio_tc_map[TC_BITMASK + 1]; __u8 hw; __u16 count[TC_MAX_QUEUE]; __u16 offset[TC_MAX_QUEUE]; }; Here the count/offset pairing give the queue alignment and the prio_tc_map gives the mapping from skb->priority to tc. The hw bit determines if the hardware should configure the count and offset values. If the hardware bit is set then the operation will fail if the hardware does not implement the ndo_setup_tc operation. This is to avoid undetermined states where the hardware may or may not control the queue mapping. Also minimal bounds checking is done on the count/offset to verify a queue does not exceed num_tx_queues and that queue ranges do not overlap. Otherwise it is left to user policy or hardware configuration to create useful mappings. It is expected that hardware QOS schemes can be implemented by creating appropriate mappings of queues in ndo_tc_setup(). One expected use case is drivers will use the ndo_setup_tc to map queue ranges onto 802.1Q traffic classes. This provides a generic mechanism to map network traffic onto these traffic classes and removes the need for lower layer drivers to know specifics about traffic types. Signed-off-by: John Fastabend <john.r.fastabend@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-01-17 16:06:09 +08:00
EXPORT_SYMBOL(dev_graft_qdisc);
static void attach_one_default_qdisc(struct net_device *dev,
struct netdev_queue *dev_queue,
void *_unused)
{
struct Qdisc *qdisc;
const struct Qdisc_ops *ops = default_qdisc_ops;
if (dev->priv_flags & IFF_NO_QUEUE)
ops = &noqueue_qdisc_ops;
qdisc = qdisc_create_dflt(dev_queue, ops, TC_H_ROOT, NULL);
if (!qdisc) {
netdev_info(dev, "activation failed\n");
return;
}
if (!netif_is_multiqueue(dev))
net_sched: fix qdisc_tree_decrease_qlen() races qdisc_tree_decrease_qlen() suffers from two problems on multiqueue devices. One problem is that it updates sch->q.qlen and sch->qstats.drops on the mq/mqprio root qdisc, while it should not : Daniele reported underflows errors : [ 681.774821] PAX: sch->q.qlen: 0 n: 1 [ 681.774825] PAX: size overflow detected in function qdisc_tree_decrease_qlen net/sched/sch_api.c:769 cicus.693_49 min, count: 72, decl: qlen; num: 0; context: sk_buff_head; [ 681.774954] CPU: 2 PID: 19 Comm: ksoftirqd/2 Tainted: G O 4.2.6.201511282239-1-grsec #1 [ 681.774955] Hardware name: ASUSTeK COMPUTER INC. X302LJ/X302LJ, BIOS X302LJ.202 03/05/2015 [ 681.774956] ffffffffa9a04863 0000000000000000 0000000000000000 ffffffffa990ff7c [ 681.774959] ffffc90000d3bc38 ffffffffa95d2810 0000000000000007 ffffffffa991002b [ 681.774960] ffffc90000d3bc68 ffffffffa91a44f4 0000000000000001 0000000000000001 [ 681.774962] Call Trace: [ 681.774967] [<ffffffffa95d2810>] dump_stack+0x4c/0x7f [ 681.774970] [<ffffffffa91a44f4>] report_size_overflow+0x34/0x50 [ 681.774972] [<ffffffffa94d17e2>] qdisc_tree_decrease_qlen+0x152/0x160 [ 681.774976] [<ffffffffc02694b1>] fq_codel_dequeue+0x7b1/0x820 [sch_fq_codel] [ 681.774978] [<ffffffffc02680a0>] ? qdisc_peek_dequeued+0xa0/0xa0 [sch_fq_codel] [ 681.774980] [<ffffffffa94cd92d>] __qdisc_run+0x4d/0x1d0 [ 681.774983] [<ffffffffa949b2b2>] net_tx_action+0xc2/0x160 [ 681.774985] [<ffffffffa90664c1>] __do_softirq+0xf1/0x200 [ 681.774987] [<ffffffffa90665ee>] run_ksoftirqd+0x1e/0x30 [ 681.774989] [<ffffffffa90896b0>] smpboot_thread_fn+0x150/0x260 [ 681.774991] [<ffffffffa9089560>] ? sort_range+0x40/0x40 [ 681.774992] [<ffffffffa9085fe4>] kthread+0xe4/0x100 [ 681.774994] [<ffffffffa9085f00>] ? kthread_worker_fn+0x170/0x170 [ 681.774995] [<ffffffffa95d8d1e>] ret_from_fork+0x3e/0x70 mq/mqprio have their own ways to report qlen/drops by folding stats on all their queues, with appropriate locking. A second problem is that qdisc_tree_decrease_qlen() calls qdisc_lookup() without proper locking : concurrent qdisc updates could corrupt the list that qdisc_match_from_root() parses to find a qdisc given its handle. Fix first problem adding a TCQ_F_NOPARENT qdisc flag that qdisc_tree_decrease_qlen() can use to abort its tree traversal, as soon as it meets a mq/mqprio qdisc children. Second problem can be fixed by RCU protection. Qdisc are already freed after RCU grace period, so qdisc_list_add() and qdisc_list_del() simply have to use appropriate rcu list variants. A future patch will add a per struct netdev_queue list anchor, so that qdisc_tree_decrease_qlen() can have more efficient lookups. Reported-by: Daniele Fucini <dfucini@gmail.com> Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Cong Wang <cwang@twopensource.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-12-02 12:08:51 +08:00
qdisc->flags |= TCQ_F_ONETXQUEUE | TCQ_F_NOPARENT;
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->priv_flags & IFF_NO_QUEUE) {
netdev_for_each_tx_queue(dev, attach_one_default_qdisc, NULL);
dev->qdisc = txq->qdisc_sleeping;
net_sched: fix a refcount_t issue with noop_qdisc syzkaller reported a refcount_t warning [1] Issue here is that noop_qdisc refcnt was never really considered as a true refcount, since qdisc_destroy() found TCQ_F_BUILTIN set : if (qdisc->flags & TCQ_F_BUILTIN || !refcount_dec_and_test(&qdisc->refcnt))) return; Meaning that all atomic_inc() we did on noop_qdisc.refcnt were not really needed, but harmless until refcount_t came. To fix this problem, we simply need to not increment noop_qdisc.refcnt, since we never decrement it. [1] refcount_t: increment on 0; use-after-free. ------------[ cut here ]------------ WARNING: CPU: 0 PID: 21754 at lib/refcount.c:152 refcount_inc+0x47/0x50 lib/refcount.c:152 Kernel panic - not syncing: panic_on_warn set ... CPU: 0 PID: 21754 Comm: syz-executor7 Not tainted 4.13.0-rc6+ #20 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:16 [inline] dump_stack+0x194/0x257 lib/dump_stack.c:52 panic+0x1e4/0x417 kernel/panic.c:180 __warn+0x1c4/0x1d9 kernel/panic.c:541 report_bug+0x211/0x2d0 lib/bug.c:183 fixup_bug+0x40/0x90 arch/x86/kernel/traps.c:190 do_trap_no_signal arch/x86/kernel/traps.c:224 [inline] do_trap+0x260/0x390 arch/x86/kernel/traps.c:273 do_error_trap+0x120/0x390 arch/x86/kernel/traps.c:310 do_invalid_op+0x1b/0x20 arch/x86/kernel/traps.c:323 invalid_op+0x1e/0x30 arch/x86/entry/entry_64.S:846 RIP: 0010:refcount_inc+0x47/0x50 lib/refcount.c:152 RSP: 0018:ffff8801c43477a0 EFLAGS: 00010282 RAX: 000000000000002b RBX: ffffffff86093c14 RCX: 0000000000000000 RDX: 000000000000002b RSI: ffffffff8159314e RDI: ffffed0038868ee8 RBP: ffff8801c43477a8 R08: 0000000000000001 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000000 R12: ffffffff86093ac0 R13: 0000000000000001 R14: ffff8801d0f3bac0 R15: dffffc0000000000 attach_default_qdiscs net/sched/sch_generic.c:792 [inline] dev_activate+0x7d3/0xaa0 net/sched/sch_generic.c:833 __dev_open+0x227/0x330 net/core/dev.c:1380 __dev_change_flags+0x695/0x990 net/core/dev.c:6726 dev_change_flags+0x88/0x140 net/core/dev.c:6792 dev_ifsioc+0x5a6/0x930 net/core/dev_ioctl.c:256 dev_ioctl+0x2bc/0xf90 net/core/dev_ioctl.c:554 sock_do_ioctl+0x94/0xb0 net/socket.c:968 sock_ioctl+0x2c2/0x440 net/socket.c:1058 vfs_ioctl fs/ioctl.c:45 [inline] do_vfs_ioctl+0x1b1/0x1520 fs/ioctl.c:685 SYSC_ioctl fs/ioctl.c:700 [inline] SyS_ioctl+0x8f/0xc0 fs/ioctl.c:691 Fixes: 7b9364050246 ("net, sched: convert Qdisc.refcnt from atomic_t to refcount_t") Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Cc: Reshetova, Elena <elena.reshetova@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-08-25 12:12:28 +08:00
qdisc_refcount_inc(dev->qdisc);
} else {
qdisc = qdisc_create_dflt(txq, &mq_qdisc_ops, TC_H_ROOT, NULL);
if (qdisc) {
dev->qdisc = qdisc;
qdisc->ops->attach(qdisc);
}
}
#ifdef CONFIG_NET_SCHED
if (dev->qdisc != &noop_qdisc)
qdisc_hash_add(dev->qdisc, false);
#endif
}
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) {
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 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);
if (dev_ingress_queue(dev))
transition_one_qdisc(dev, dev_ingress_queue(dev), NULL);
if (need_watchdog) {
netif_trans_update(dev);
dev_watchdog_up(dev);
}
}
net_sched: implement a root container qdisc sch_mqprio This implements a mqprio queueing discipline that by default creates a pfifo_fast qdisc per tx queue and provides the needed configuration interface. Using the mqprio qdisc the number of tcs currently in use along with the range of queues alloted to each class can be configured. By default skbs are mapped to traffic classes using the skb priority. This mapping is configurable. Configurable parameters, struct tc_mqprio_qopt { __u8 num_tc; __u8 prio_tc_map[TC_BITMASK + 1]; __u8 hw; __u16 count[TC_MAX_QUEUE]; __u16 offset[TC_MAX_QUEUE]; }; Here the count/offset pairing give the queue alignment and the prio_tc_map gives the mapping from skb->priority to tc. The hw bit determines if the hardware should configure the count and offset values. If the hardware bit is set then the operation will fail if the hardware does not implement the ndo_setup_tc operation. This is to avoid undetermined states where the hardware may or may not control the queue mapping. Also minimal bounds checking is done on the count/offset to verify a queue does not exceed num_tx_queues and that queue ranges do not overlap. Otherwise it is left to user policy or hardware configuration to create useful mappings. It is expected that hardware QOS schemes can be implemented by creating appropriate mappings of queues in ndo_tc_setup(). One expected use case is drivers will use the ndo_setup_tc to map queue ranges onto 802.1Q traffic classes. This provides a generic mechanism to map network traffic onto these traffic classes and removes the need for lower layer drivers to know specifics about traffic types. Signed-off-by: John Fastabend <john.r.fastabend@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-01-17 16:06:09 +08:00
EXPORT_SYMBOL(dev_activate);
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 = rtnl_dereference(dev_queue->qdisc);
if (qdisc) {
bool nolock = qdisc->flags & TCQ_F_NOLOCK;
if (nolock)
spin_lock_bh(&qdisc->seqlock);
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));
if (nolock)
spin_unlock_bh(&qdisc->seqlock);
}
}
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 = (qdisc_is_running(q) ||
test_bit(__QDISC_STATE_SCHED, &q->state));
spin_unlock_bh(root_lock);
if (val)
return true;
}
return false;
}
static void dev_qdisc_reset(struct net_device *dev,
struct netdev_queue *dev_queue,
void *none)
{
struct Qdisc *qdisc = dev_queue->qdisc_sleeping;
if (qdisc)
qdisc_reset(qdisc);
}
/**
* dev_deactivate_many - deactivate transmissions on several devices
* @head: list of devices to deactivate
*
* This function returns only when all outstanding transmissions
* have completed, unless all devices are in dismantle phase.
*/
void dev_deactivate_many(struct list_head *head)
{
struct net_device *dev;
list_for_each_entry(dev, head, close_list) {
netdev_for_each_tx_queue(dev, dev_deactivate_queue,
&noop_qdisc);
if (dev_ingress_queue(dev))
dev_deactivate_queue(dev, dev_ingress_queue(dev),
&noop_qdisc);
dev_watchdog_down(dev);
}
/* Wait for outstanding qdisc-less dev_queue_xmit calls.
* This is avoided if all devices are in dismantle phase :
* Caller will call synchronize_net() for us
*/
synchronize_net();
/* Wait for outstanding qdisc_run calls. */
list_for_each_entry(dev, head, close_list) {
while (some_qdisc_is_busy(dev))
yield();
/* The new qdisc is assigned at this point so we can safely
* unwind stale skb lists and qdisc statistics
*/
netdev_for_each_tx_queue(dev, dev_qdisc_reset, NULL);
if (dev_ingress_queue(dev))
dev_qdisc_reset(dev, dev_ingress_queue(dev), NULL);
}
}
void dev_deactivate(struct net_device *dev)
{
LIST_HEAD(single);
list_add(&dev->close_list, &single);
dev_deactivate_many(&single);
list_del(&single);
}
net_sched: implement a root container qdisc sch_mqprio This implements a mqprio queueing discipline that by default creates a pfifo_fast qdisc per tx queue and provides the needed configuration interface. Using the mqprio qdisc the number of tcs currently in use along with the range of queues alloted to each class can be configured. By default skbs are mapped to traffic classes using the skb priority. This mapping is configurable. Configurable parameters, struct tc_mqprio_qopt { __u8 num_tc; __u8 prio_tc_map[TC_BITMASK + 1]; __u8 hw; __u16 count[TC_MAX_QUEUE]; __u16 offset[TC_MAX_QUEUE]; }; Here the count/offset pairing give the queue alignment and the prio_tc_map gives the mapping from skb->priority to tc. The hw bit determines if the hardware should configure the count and offset values. If the hardware bit is set then the operation will fail if the hardware does not implement the ndo_setup_tc operation. This is to avoid undetermined states where the hardware may or may not control the queue mapping. Also minimal bounds checking is done on the count/offset to verify a queue does not exceed num_tx_queues and that queue ranges do not overlap. Otherwise it is left to user policy or hardware configuration to create useful mappings. It is expected that hardware QOS schemes can be implemented by creating appropriate mappings of queues in ndo_tc_setup(). One expected use case is drivers will use the ndo_setup_tc to map queue ranges onto 802.1Q traffic classes. This provides a generic mechanism to map network traffic onto these traffic classes and removes the need for lower layer drivers to know specifics about traffic types. Signed-off-by: John Fastabend <john.r.fastabend@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-01-17 16:06:09 +08:00
EXPORT_SYMBOL(dev_deactivate);
static int qdisc_change_tx_queue_len(struct net_device *dev,
struct netdev_queue *dev_queue)
{
struct Qdisc *qdisc = dev_queue->qdisc_sleeping;
const struct Qdisc_ops *ops = qdisc->ops;
if (ops->change_tx_queue_len)
return ops->change_tx_queue_len(qdisc, dev->tx_queue_len);
return 0;
}
int dev_qdisc_change_tx_queue_len(struct net_device *dev)
{
bool up = dev->flags & IFF_UP;
unsigned int i;
int ret = 0;
if (up)
dev_deactivate(dev);
for (i = 0; i < dev->num_tx_queues; i++) {
ret = qdisc_change_tx_queue_len(dev, &dev->_tx[i]);
/* TODO: revert changes on a partial failure */
if (ret)
break;
}
if (up)
dev_activate(dev);
return ret;
}
static void dev_init_scheduler_queue(struct net_device *dev,
struct netdev_queue *dev_queue,
void *_qdisc)
{
struct Qdisc *qdisc = _qdisc;
rcu_assign_pointer(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);
if (dev_ingress_queue(dev))
dev_init_scheduler_queue(dev, dev_ingress_queue(dev), &noop_qdisc);
timer_setup(&dev->watchdog_timer, dev_watchdog, 0);
}
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_put(qdisc);
}
}
void dev_shutdown(struct net_device *dev)
{
netdev_for_each_tx_queue(dev, shutdown_scheduler_queue, &noop_qdisc);
if (dev_ingress_queue(dev))
shutdown_scheduler_queue(dev, dev_ingress_queue(dev), &noop_qdisc);
qdisc_put(dev->qdisc);
dev->qdisc = &noop_qdisc;
WARN_ON(timer_pending(&dev->watchdog_timer));
}
void psched_ratecfg_precompute(struct psched_ratecfg *r,
const struct tc_ratespec *conf,
u64 rate64)
{
memset(r, 0, sizeof(*r));
r->overhead = conf->overhead;
r->rate_bytes_ps = max_t(u64, conf->rate, rate64);
r->linklayer = (conf->linklayer & TC_LINKLAYER_MASK);
r->mult = 1;
/*
* The deal here is to replace a divide by a reciprocal one
* in fast path (a reciprocal divide is a multiply and a shift)
*
* Normal formula would be :
* time_in_ns = (NSEC_PER_SEC * len) / rate_bps
*
* We compute mult/shift to use instead :
* time_in_ns = (len * mult) >> shift;
*
* We try to get the highest possible mult value for accuracy,
* but have to make sure no overflows will ever happen.
*/
if (r->rate_bytes_ps > 0) {
u64 factor = NSEC_PER_SEC;
for (;;) {
r->mult = div64_u64(factor, r->rate_bytes_ps);
if (r->mult & (1U << 31) || factor & (1ULL << 63))
break;
factor <<= 1;
r->shift++;
}
}
}
EXPORT_SYMBOL(psched_ratecfg_precompute);
static void mini_qdisc_rcu_func(struct rcu_head *head)
{
}
void mini_qdisc_pair_swap(struct mini_Qdisc_pair *miniqp,
struct tcf_proto *tp_head)
{
struct mini_Qdisc *miniq_old = rtnl_dereference(*miniqp->p_miniq);
struct mini_Qdisc *miniq;
if (!tp_head) {
RCU_INIT_POINTER(*miniqp->p_miniq, NULL);
/* Wait for flying RCU callback before it is freed. */
rcu_barrier();
return;
}
miniq = !miniq_old || miniq_old == &miniqp->miniq2 ?
&miniqp->miniq1 : &miniqp->miniq2;
/* We need to make sure that readers won't see the miniq
* we are about to modify. So wait until previous call_rcu callback
* is done.
*/
rcu_barrier();
miniq->filter_list = tp_head;
rcu_assign_pointer(*miniqp->p_miniq, miniq);
if (miniq_old)
/* This is counterpart of the rcu barriers above. We need to
* block potential new user of miniq_old until all readers
* are not seeing it.
*/
call_rcu(&miniq_old->rcu, mini_qdisc_rcu_func);
}
EXPORT_SYMBOL(mini_qdisc_pair_swap);
void mini_qdisc_pair_init(struct mini_Qdisc_pair *miniqp, struct Qdisc *qdisc,
struct mini_Qdisc __rcu **p_miniq)
{
miniqp->miniq1.cpu_bstats = qdisc->cpu_bstats;
miniqp->miniq1.cpu_qstats = qdisc->cpu_qstats;
miniqp->miniq2.cpu_bstats = qdisc->cpu_bstats;
miniqp->miniq2.cpu_qstats = qdisc->cpu_qstats;
miniqp->p_miniq = p_miniq;
}
EXPORT_SYMBOL(mini_qdisc_pair_init);