iv/linux
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
ab95465cde
linux/net/sched/sch_generic.c

1608 lines
38 KiB
C
Raw Normal View History

treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 152 Based on 1 normalized pattern(s): 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 extracted by the scancode license scanner the SPDX license identifier GPL-2.0-or-later has been chosen to replace the boilerplate/reference in 3029 file(s). Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Allison Randal <allison@lohutok.net> Cc: linux-spdx@vger.kernel.org Link: https://lkml.kernel.org/r/20190527070032.746973796@linutronix.de Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2019-05-27 09:55:01 +03:00
// SPDX-License-Identifier: GPL-2.0-or-later
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
/*
* net/sched/sch_generic.c Generic packet scheduler routines.
*
* 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 11:04:11 +03:00
#include <linux/slab.h>
net_sched: make dev_trans_start return vlan's real dev trans_start Vlan devices are LLTX and don't update their own trans_start, so if dev_trans_start has to be called with a vlan device then 0 or a stale value will be returned. Currently the bonding is the only such user, and it's needed for proper arp monitoring when the slaves are vlans. Fix this by extracting the vlan's real device trans_start. Suggested-by: David Miller <davem@davemloft.net> Signed-off-by: Nikolay Aleksandrov <nikolay@redhat.com> Acked-by: Veaceslav Falico <vfalico@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-08-04 00:07:47 +04:00
#include <linux/if_vlan.h>
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
#include <linux/skb_array.h>
net_sched: use macvlan real dev trans_start in dev_trans_start() Macvlan devices are similar to vlans and do not update their own trans_start. In order for arp monitoring to work for a bond device when the slaves are macvlans, obtain its real device. Signed-off-by: Chris Dion <christopher.dion@dell.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-06 18:50:28 +03:00
#include <linux/if_macvlan.h>
sch: make htb_rate_cfg and functions around that generic As it is going to be used in tbf as well, push these to generic code. Signed-off-by: Jiri Pirko <jiri@resnulli.us> Acked-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-02-12 04:12:03 +04:00
#include <net/sch_generic.h>
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
#include <net/pkt_sched.h>
net: add a noref bit on skb dst Use low order bit of skb->_skb_dst to tell dst is not refcounted. Change _skb_dst to _skb_refdst to make sure all uses are catched. skb_dst() returns the dst, regardless of noref bit set or not, but with a lockdep check to make sure a noref dst is not given if current user is not rcu protected. New skb_dst_set_noref() helper to set an notrefcounted dst on a skb. (with lockdep check) skb_dst_drop() drops a reference only if skb dst was refcounted. skb_dst_force() helper is used to force a refcount on dst, when skb is queued and not anymore RCU protected. Use skb_dst_force() in __sk_add_backlog(), __dev_xmit_skb() if !IFF_XMIT_DST_RELEASE or skb enqueued on qdisc queue, in sock_queue_rcv_skb(), in __nf_queue(). Use skb_dst_force() in dev_requeue_skb(). Note: dst_use_noref() still dirties dst, we might transform it later to do one dirtying per jiffies. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-05-12 03:19:48 +04:00
#include <net/dst.h>
qdisc: add tracepoint qdisc:qdisc_dequeue for dequeued SKBs The main purpose of this tracepoint is to monitor bulk dequeue in the network qdisc layer, as it cannot be deducted from the existing qdisc stats. The txq_state can be used for determining the reason for zero packet dequeues, see enum netdev_queue_state_t. Notice all packets doesn't necessary activate this tracepoint. As qdiscs with flag TCQ_F_CAN_BYPASS, can directly invoke sch_direct_xmit() when qdisc_qlen is zero. Remember that perf record supports filters like: perf record -e qdisc:qdisc_dequeue \ --filter 'ifindex == 4 && (packets > 1 || txq_state > 0)' Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-08-15 22:11:03 +03:00
#include <trace/events/qdisc.h>
net: add a generic tracepoint for TX queue timeout Although devlink health report does a nice job on reporting TX timeout and other NIC errors, unfortunately it requires drivers to support it but currently only mlx5 has implemented it. Before other drivers could catch up, it is useful to have a generic tracepoint to monitor this kind of TX timeout. We have been suffering TX timeout with different drivers, we plan to start to monitor it with rasdaemon which just needs a new tracepoint. Sample output: ksoftirqd/1-16 [001] ..s2 144.043173: net_dev_xmit_timeout: dev=ens3 driver=e1000 queue=0 Cc: Eran Ben Elisha <eranbe@mellanox.com> Cc: Jiri Pirko <jiri@mellanox.com> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Reviewed-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-05-02 05:56:59 +03:00
#include <trace/events/net.h>
net: Add asynchronous callbacks for xfrm on layer 2. This patch implements asynchronous crypto callbacks and a backlog handler that can be used when IPsec is done at layer 2 in the TX path. It also extends the skb validate functions so that we can update the driver transmit return codes based on async crypto operation or to indicate that we queued the packet in a backlog queue. Joint work with: Aviv Heller <avivh@mellanox.com> Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
2017-12-20 12:41:36 +03:00
#include <net/xfrm.h>
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
qdisc: fix build with !CONFIG_NET_SCHED Multiqueue scheduler refers to default_qdisc_ops; therefore the variable definition needs to be moved to handle case where net scheduler API is not available. Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-08-31 21:15:33 +04:00
/* Qdisc to use by default */
const struct Qdisc_ops *default_qdisc_ops = &pfifo_fast_ops;
EXPORT_SYMBOL(default_qdisc_ops);
net: sched: fix tx action reschedule issue with stopped queue The netdev qeueue might be stopped when byte queue limit has reached or tx hw ring is full, net_tx_action() may still be rescheduled if STATE_MISSED is set, which consumes unnecessary cpu without dequeuing and transmiting any skb because the netdev queue is stopped, see qdisc_run_end(). This patch fixes it by checking the netdev queue state before calling qdisc_run() and clearing STATE_MISSED if netdev queue is stopped during qdisc_run(), the net_tx_action() is rescheduled again when netdev qeueue is restarted, see netif_tx_wake_queue(). As there is time window between netif_xmit_frozen_or_stopped() checking and STATE_MISSED clearing, between which STATE_MISSED may set by net_tx_action() scheduled by netif_tx_wake_queue(), so set the STATE_MISSED again if netdev queue is restarted. Fixes: 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking") Reported-by: Michal Kubecek <mkubecek@suse.cz> Acked-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-05-14 06:17:01 +03:00
static void qdisc_maybe_clear_missed(struct Qdisc *q,
const struct netdev_queue *txq)
{
clear_bit(__QDISC_STATE_MISSED, &q->state);
/* Make sure the below netif_xmit_frozen_or_stopped()
* checking happens after clearing STATE_MISSED.
*/
smp_mb__after_atomic();
/* Checking netif_xmit_frozen_or_stopped() again to
* make sure STATE_MISSED is set if the STATE_MISSED
* set by netif_tx_wake_queue()'s rescheduling of
* net_tx_action() is cleared by the above clear_bit().
*/
if (!netif_xmit_frozen_or_stopped(txq))
set_bit(__QDISC_STATE_MISSED, &q->state);
net: sched: implement TCQ_F_CAN_BYPASS for lockless qdisc Currently pfifo_fast has both TCQ_F_CAN_BYPASS and TCQ_F_NOLOCK flag set, but queue discipline by-pass does not work for lockless qdisc because skb is always enqueued to qdisc even when the qdisc is empty, see __dev_xmit_skb(). This patch calls sch_direct_xmit() to transmit the skb directly to the driver for empty lockless qdisc, which aviod enqueuing and dequeuing operation. As qdisc->empty is not reliable to indicate a empty qdisc because there is a time window between enqueuing and setting qdisc->empty. So we use the MISSED state added in commit a90c57f2cedd ("net: sched: fix packet stuck problem for lockless qdisc"), which indicate there is lock contention, suggesting that it is better not to do the qdisc bypass in order to avoid packet out of order problem. In order to make MISSED state reliable to indicate a empty qdisc, we need to ensure that testing and clearing of MISSED state is within the protection of qdisc->seqlock, only setting MISSED state can be done without the protection of qdisc->seqlock. A MISSED state testing is added without the protection of qdisc->seqlock to aviod doing unnecessary spin_trylock() for contention case. As the enqueuing is not within the protection of qdisc->seqlock, there is still a potential data race as mentioned by Jakub [1]: thread1 thread2 thread3 qdisc_run_begin() # true qdisc_run_begin(q) set(MISSED) pfifo_fast_dequeue clear(MISSED) # recheck the queue qdisc_run_end() enqueue skb1 qdisc empty # true qdisc_run_begin() # true sch_direct_xmit() # skb2 qdisc_run_begin() set(MISSED) When above happens, skb1 enqueued by thread2 is transmited after skb2 is transmited by thread3 because MISSED state setting and enqueuing is not under the qdisc->seqlock. If qdisc bypass is disabled, skb1 has better chance to be transmited quicker than skb2. This patch does not take care of the above data race, because we view this as similar as below: Even at the same time CPU1 and CPU2 write the skb to two socket which both heading to the same qdisc, there is no guarantee that which skb will hit the qdisc first, because there is a lot of factor like interrupt/softirq/cache miss/scheduling afffecting that. There are below cases that need special handling: 1. When MISSED state is cleared before another round of dequeuing in pfifo_fast_dequeue(), and __qdisc_run() might not be able to dequeue all skb in one round and call __netif_schedule(), which might result in a non-empty qdisc without MISSED set. In order to avoid this, the MISSED state is set for lockless qdisc and __netif_schedule() will be called at the end of qdisc_run_end. 2. The MISSED state also need to be set for lockless qdisc instead of calling __netif_schedule() directly when requeuing a skb for a similar reason. 3. For netdev queue stopped case, the MISSED case need clearing while the netdev queue is stopped, otherwise there may be unnecessary __netif_schedule() calling. So a new DRAINING state is added to indicate this case, which also indicate a non-empty qdisc. 4. As there is already netif_xmit_frozen_or_stopped() checking in dequeue_skb() and sch_direct_xmit(), which are both within the protection of qdisc->seqlock, but the same checking in __dev_xmit_skb() is without the protection, which might cause empty indication of a lockless qdisc to be not reliable. So remove the checking in __dev_xmit_skb(), and the checking in the protection of qdisc->seqlock seems enough to avoid the cpu consumption problem for netdev queue stopped case. 1. https://lkml.org/lkml/2021/5/29/215 Acked-by: Jakub Kicinski <kuba@kernel.org> Tested-by: Vladimir Oltean <vladimir.oltean@nxp.com> # flexcan Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-06-22 09:49:56 +03:00
else
set_bit(__QDISC_STATE_DRAINING, &q->state);
net: sched: fix tx action reschedule issue with stopped queue The netdev qeueue might be stopped when byte queue limit has reached or tx hw ring is full, net_tx_action() may still be rescheduled if STATE_MISSED is set, which consumes unnecessary cpu without dequeuing and transmiting any skb because the netdev queue is stopped, see qdisc_run_end(). This patch fixes it by checking the netdev queue state before calling qdisc_run() and clearing STATE_MISSED if netdev queue is stopped during qdisc_run(), the net_tx_action() is rescheduled again when netdev qeueue is restarted, see netif_tx_wake_queue(). As there is time window between netif_xmit_frozen_or_stopped() checking and STATE_MISSED clearing, between which STATE_MISSED may set by net_tx_action() scheduled by netif_tx_wake_queue(), so set the STATE_MISSED again if netdev queue is restarted. Fixes: 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking") Reported-by: Michal Kubecek <mkubecek@suse.cz> Acked-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-05-14 06:17:01 +03:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
/* Main transmission queue. */
[NET_SCHED]: Eliminate qdisc_tree_lock Since we're now holding the rtnl during the entire dump operation, we can remove qdisc_tree_lock, whose only purpose is to protect dump callbacks from concurrent changes to the qdisc tree. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-04-17 04:02:10 +04:00
/* Modifications to data participating in scheduling must be protected with
pkt_sched: Use qdisc_lock() on already sampled root qdisc. Based upon a bug report by Jeff Kirsher. Don't use qdisc_root_lock() in these cases as the root qdisc could have been changed, and we'd thus lock the wrong object. Tested by Emil S Tantilov who confirms that this seems to fix the problem. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-08-03 07:02:43 +04:00
* qdisc_lock(qdisc) spinlock.
[NET_SCHED]: Eliminate qdisc_tree_lock Since we're now holding the rtnl during the entire dump operation, we can remove qdisc_tree_lock, whose only purpose is to protect dump callbacks from concurrent changes to the qdisc tree. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-04-17 04:02:10 +04:00
*
* The idea is the following:
pkt_sched: Kill qdisc_lock_tree and qdisc_unlock_tree. No longer used. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-16 14:22:39 +04:00
* - enqueue, dequeue are serialized via qdisc root lock
* - ingress filtering is also serialized via qdisc root lock
[NET_SCHED]: Eliminate qdisc_tree_lock Since we're now holding the rtnl during the entire dump operation, we can remove qdisc_tree_lock, whose only purpose is to protect dump callbacks from concurrent changes to the qdisc tree. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-04-17 04:02:10 +04:00
* - updates to tree and tree walking are only done under the rtnl mutex.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
*/
net: sched: use skb list for skb_bad_tx Similar to how gso is handled use skb list for skb_bad_tx this is required with lockless qdiscs because we may have multiple cores attempting to push skbs into skb_bad_tx concurrently Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:56:23 +03:00
net: sched: fix reordering issues Whenever MQ is not used on a multiqueue device, we experience serious reordering problems. Bisection found the cited commit. The issue can be described this way : - A single qdisc hierarchy is shared by all transmit queues. (eg : tc qdisc replace dev eth0 root fq_codel) - When/if try_bulk_dequeue_skb_slow() dequeues a packet targetting a different transmit queue than the one used to build a packet train, we stop building the current list and save the 'bad' skb (P1) in a special queue. (bad_txq) - When dequeue_skb() calls qdisc_dequeue_skb_bad_txq() and finds this skb (P1), it checks if the associated transmit queues is still in frozen state. If the queue is still blocked (by BQL or NIC tx ring full), we leave the skb in bad_txq and return NULL. - dequeue_skb() calls q->dequeue() to get another packet (P2) The other packet can target the problematic queue (that we found in frozen state for the bad_txq packet), but another cpu just ran TX completion and made room in the txq that is now ready to accept new packets. - Packet P2 is sent while P1 is still held in bad_txq, P1 might be sent at next round. In practice P2 is the lead of a big packet train (P2,P3,P4 ...) filling the BQL budget and delaying P1 by many packets :/ To solve this problem, we have to block the dequeue process as long as the first packet in bad_txq can not be sent. Reordering issues disappear and no side effects have been seen. Fixes: a53851e2c321 ("net: sched: explicit locking in gso_cpu fallback") Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: John Fastabend <john.fastabend@gmail.com> Acked-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-09-05 15:20:22 +03:00
#define SKB_XOFF_MAGIC ((struct sk_buff *)1UL)
net: sched: use skb list for skb_bad_tx Similar to how gso is handled use skb list for skb_bad_tx this is required with lockless qdiscs because we may have multiple cores attempting to push skbs into skb_bad_tx concurrently Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:56:23 +03:00
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);
Revert: "net: sched: put back q.qlen into a single location" This revert commit 46b1c18f9deb ("net: sched: put back q.qlen into a single location"). After the previous patch, when a NOLOCK qdisc is enslaved to a locking qdisc it switches to global stats accounting. As a consequence, when a classful qdisc accesses directly a child qdisc's qlen, such qdisc is not doing per CPU accounting and qlen value is consistent. In the control path nobody uses directly qlen since commit e5f0e8f8e45 ("net: sched: introduce and use qdisc tree flush/purge helpers"), so we can remove the contented atomic ops from the datapath. v1 -> v2: - complete the qdisc_qstats_atomic_qlen_dec() -> qdisc_qstats_cpu_qlen_dec() replacement, fix build issue - more descriptive commit message Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-10 15:32:41 +03:00
qdisc_qstats_cpu_qlen_dec(q);
net: sched: use skb list for skb_bad_tx Similar to how gso is handled use skb list for skb_bad_tx this is required with lockless qdiscs because we may have multiple cores attempting to push skbs into skb_bad_tx concurrently Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:56:23 +03:00
} else {
qdisc_qstats_backlog_dec(q, skb);
q->q.qlen--;
}
} else {
net: sched: fix reordering issues Whenever MQ is not used on a multiqueue device, we experience serious reordering problems. Bisection found the cited commit. The issue can be described this way : - A single qdisc hierarchy is shared by all transmit queues. (eg : tc qdisc replace dev eth0 root fq_codel) - When/if try_bulk_dequeue_skb_slow() dequeues a packet targetting a different transmit queue than the one used to build a packet train, we stop building the current list and save the 'bad' skb (P1) in a special queue. (bad_txq) - When dequeue_skb() calls qdisc_dequeue_skb_bad_txq() and finds this skb (P1), it checks if the associated transmit queues is still in frozen state. If the queue is still blocked (by BQL or NIC tx ring full), we leave the skb in bad_txq and return NULL. - dequeue_skb() calls q->dequeue() to get another packet (P2) The other packet can target the problematic queue (that we found in frozen state for the bad_txq packet), but another cpu just ran TX completion and made room in the txq that is now ready to accept new packets. - Packet P2 is sent while P1 is still held in bad_txq, P1 might be sent at next round. In practice P2 is the lead of a big packet train (P2,P3,P4 ...) filling the BQL budget and delaying P1 by many packets :/ To solve this problem, we have to block the dequeue process as long as the first packet in bad_txq can not be sent. Reordering issues disappear and no side effects have been seen. Fixes: a53851e2c321 ("net: sched: explicit locking in gso_cpu fallback") Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: John Fastabend <john.fastabend@gmail.com> Acked-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-09-05 15:20:22 +03:00
skb = SKB_XOFF_MAGIC;
net: sched: fix tx action reschedule issue with stopped queue The netdev qeueue might be stopped when byte queue limit has reached or tx hw ring is full, net_tx_action() may still be rescheduled if STATE_MISSED is set, which consumes unnecessary cpu without dequeuing and transmiting any skb because the netdev queue is stopped, see qdisc_run_end(). This patch fixes it by checking the netdev queue state before calling qdisc_run() and clearing STATE_MISSED if netdev queue is stopped during qdisc_run(), the net_tx_action() is rescheduled again when netdev qeueue is restarted, see netif_tx_wake_queue(). As there is time window between netif_xmit_frozen_or_stopped() checking and STATE_MISSED clearing, between which STATE_MISSED may set by net_tx_action() scheduled by netif_tx_wake_queue(), so set the STATE_MISSED again if netdev queue is restarted. Fixes: 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking") Reported-by: Michal Kubecek <mkubecek@suse.cz> Acked-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-05-14 06:17:01 +03:00
qdisc_maybe_clear_missed(q, txq);
net: sched: use skb list for skb_bad_tx Similar to how gso is handled use skb list for skb_bad_tx this is required with lockless qdiscs because we may have multiple cores attempting to push skbs into skb_bad_tx concurrently Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:56:23 +03:00
}
}
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 04:53:00 +03:00
if (qdisc_is_percpu_stats(q)) {
qdisc_qstats_cpu_backlog_inc(q, skb);
Revert: "net: sched: put back q.qlen into a single location" This revert commit 46b1c18f9deb ("net: sched: put back q.qlen into a single location"). After the previous patch, when a NOLOCK qdisc is enslaved to a locking qdisc it switches to global stats accounting. As a consequence, when a classful qdisc accesses directly a child qdisc's qlen, such qdisc is not doing per CPU accounting and qlen value is consistent. In the control path nobody uses directly qlen since commit e5f0e8f8e45 ("net: sched: introduce and use qdisc tree flush/purge helpers"), so we can remove the contented atomic ops from the datapath. v1 -> v2: - complete the qdisc_qstats_atomic_qlen_dec() -> qdisc_qstats_cpu_qlen_dec() replacement, fix build issue - more descriptive commit message Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-10 15:32:41 +03:00
qdisc_qstats_cpu_qlen_inc(q);
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 04:53:00 +03:00
} else {
qdisc_qstats_backlog_inc(q, skb);
q->q.qlen++;
}
net: sched: use skb list for skb_bad_tx Similar to how gso is handled use skb list for skb_bad_tx this is required with lockless qdiscs because we may have multiple cores attempting to push skbs into skb_bad_tx concurrently Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:56:23 +03:00
if (lock)
spin_unlock(lock);
}
net: sched: always do stats accounting according to TCQ_F_CPUSTATS The core sched implementation checks independently for NOLOCK flag to acquire/release the root spin lock and for qdisc_is_percpu_stats() to account per CPU values in many places. This change update the last few places checking the TCQ_F_NOLOCK to do per CPU stats accounting according to qdisc_is_percpu_stats() value. The above allows to clean dev_requeue_skb() implementation a bit and makes stats update always consistent with a single flag. v1 -> v2: - do not move qdisc_is_empty definition, fix build issue Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-10 15:32:39 +03:00
static inline void dev_requeue_skb(struct sk_buff *skb, struct Qdisc *q)
[NET_SCHED]: Cleanup readability of qdisc restart Over the years this code has gotten hairier. Resulting in many long discussions over long summer days and patches that get it wrong. This patch helps tame that code so normal people will understand it. Thanks to Thomas Graf, Peter J. waskiewicz Jr, and Patrick McHardy for their valuable reviews. Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-06-11 04:31:24 +04:00
{
net: sched: always do stats accounting according to TCQ_F_CPUSTATS The core sched implementation checks independently for NOLOCK flag to acquire/release the root spin lock and for qdisc_is_percpu_stats() to account per CPU values in many places. This change update the last few places checking the TCQ_F_NOLOCK to do per CPU stats accounting according to qdisc_is_percpu_stats() value. The above allows to clean dev_requeue_skb() implementation a bit and makes stats update always consistent with a single flag. v1 -> v2: - do not move qdisc_is_empty definition, fix build issue Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-10 15:32:39 +03:00
spinlock_t *lock = NULL;
net: sched: fix skb leak in dev_requeue_skb() When dev_requeue_skb() is called with bulked skb list, only the first skb of the list will be requeued to qdisc layer, and leak the others without free them. TCP is broken due to skb leak since no free skb will be considered as still in the host queue and never be retransmitted. This happend when dev_requeue_skb() called from qdisc_restart(). qdisc_restart |-- dequeue_skb |-- sch_direct_xmit() |-- dev_requeue_skb() <-- skb may bluked Fix dev_requeue_skb() to requeue the full bluked list. Also change to use __skb_queue_tail() in __dev_requeue_skb() to avoid skb out of order. Fixes: a53851e2c321 ("net: sched: explicit locking in gso_cpu fallback") Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-27 12:05:52 +03:00
net: sched: always do stats accounting according to TCQ_F_CPUSTATS The core sched implementation checks independently for NOLOCK flag to acquire/release the root spin lock and for qdisc_is_percpu_stats() to account per CPU values in many places. This change update the last few places checking the TCQ_F_NOLOCK to do per CPU stats accounting according to qdisc_is_percpu_stats() value. The above allows to clean dev_requeue_skb() implementation a bit and makes stats update always consistent with a single flag. v1 -> v2: - do not move qdisc_is_empty definition, fix build issue Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-10 15:32:39 +03:00
if (q->flags & TCQ_F_NOLOCK) {
lock = qdisc_lock(q);
spin_lock(lock);
net: sched: fix skb leak in dev_requeue_skb() When dev_requeue_skb() is called with bulked skb list, only the first skb of the list will be requeued to qdisc layer, and leak the others without free them. TCP is broken due to skb leak since no free skb will be considered as still in the host queue and never be retransmitted. This happend when dev_requeue_skb() called from qdisc_restart(). qdisc_restart |-- dequeue_skb |-- sch_direct_xmit() |-- dev_requeue_skb() <-- skb may bluked Fix dev_requeue_skb() to requeue the full bluked list. Also change to use __skb_queue_tail() in __dev_requeue_skb() to avoid skb out of order. Fixes: a53851e2c321 ("net: sched: explicit locking in gso_cpu fallback") Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-27 12:05:52 +03:00
}
[NET_SCHED]: Cleanup readability of qdisc restart Over the years this code has gotten hairier. Resulting in many long discussions over long summer days and patches that get it wrong. This patch helps tame that code so normal people will understand it. Thanks to Thomas Graf, Peter J. waskiewicz Jr, and Patrick McHardy for their valuable reviews. Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-06-11 04:31:24 +04:00
net: sched: fix skb leak in dev_requeue_skb() When dev_requeue_skb() is called with bulked skb list, only the first skb of the list will be requeued to qdisc layer, and leak the others without free them. TCP is broken due to skb leak since no free skb will be considered as still in the host queue and never be retransmitted. This happend when dev_requeue_skb() called from qdisc_restart(). qdisc_restart |-- dequeue_skb |-- sch_direct_xmit() |-- dev_requeue_skb() <-- skb may bluked Fix dev_requeue_skb() to requeue the full bluked list. Also change to use __skb_queue_tail() in __dev_requeue_skb() to avoid skb out of order. Fixes: a53851e2c321 ("net: sched: explicit locking in gso_cpu fallback") Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-27 12:05:52 +03:00
while (skb) {
struct sk_buff *next = skb->next;
__skb_queue_tail(&q->gso_skb, skb);
net: sched: always do stats accounting according to TCQ_F_CPUSTATS The core sched implementation checks independently for NOLOCK flag to acquire/release the root spin lock and for qdisc_is_percpu_stats() to account per CPU values in many places. This change update the last few places checking the TCQ_F_NOLOCK to do per CPU stats accounting according to qdisc_is_percpu_stats() value. The above allows to clean dev_requeue_skb() implementation a bit and makes stats update always consistent with a single flag. v1 -> v2: - do not move qdisc_is_empty definition, fix build issue Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-10 15:32:39 +03:00
/* it's still part of the queue */
if (qdisc_is_percpu_stats(q)) {
qdisc_qstats_cpu_requeues_inc(q);
qdisc_qstats_cpu_backlog_inc(q, skb);
Revert: "net: sched: put back q.qlen into a single location" This revert commit 46b1c18f9deb ("net: sched: put back q.qlen into a single location"). After the previous patch, when a NOLOCK qdisc is enslaved to a locking qdisc it switches to global stats accounting. As a consequence, when a classful qdisc accesses directly a child qdisc's qlen, such qdisc is not doing per CPU accounting and qlen value is consistent. In the control path nobody uses directly qlen since commit e5f0e8f8e45 ("net: sched: introduce and use qdisc tree flush/purge helpers"), so we can remove the contented atomic ops from the datapath. v1 -> v2: - complete the qdisc_qstats_atomic_qlen_dec() -> qdisc_qstats_cpu_qlen_dec() replacement, fix build issue - more descriptive commit message Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-10 15:32:41 +03:00
qdisc_qstats_cpu_qlen_inc(q);
net: sched: always do stats accounting according to TCQ_F_CPUSTATS The core sched implementation checks independently for NOLOCK flag to acquire/release the root spin lock and for qdisc_is_percpu_stats() to account per CPU values in many places. This change update the last few places checking the TCQ_F_NOLOCK to do per CPU stats accounting according to qdisc_is_percpu_stats() value. The above allows to clean dev_requeue_skb() implementation a bit and makes stats update always consistent with a single flag. v1 -> v2: - do not move qdisc_is_empty definition, fix build issue Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-10 15:32:39 +03:00
} else {
q->qstats.requeues++;
qdisc_qstats_backlog_inc(q, skb);
q->q.qlen++;
}
net: sched: fix skb leak in dev_requeue_skb() When dev_requeue_skb() is called with bulked skb list, only the first skb of the list will be requeued to qdisc layer, and leak the others without free them. TCP is broken due to skb leak since no free skb will be considered as still in the host queue and never be retransmitted. This happend when dev_requeue_skb() called from qdisc_restart(). qdisc_restart |-- dequeue_skb |-- sch_direct_xmit() |-- dev_requeue_skb() <-- skb may bluked Fix dev_requeue_skb() to requeue the full bluked list. Also change to use __skb_queue_tail() in __dev_requeue_skb() to avoid skb out of order. Fixes: a53851e2c321 ("net: sched: explicit locking in gso_cpu fallback") Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-27 12:05:52 +03:00
skb = next;
}
net: sched: implement TCQ_F_CAN_BYPASS for lockless qdisc Currently pfifo_fast has both TCQ_F_CAN_BYPASS and TCQ_F_NOLOCK flag set, but queue discipline by-pass does not work for lockless qdisc because skb is always enqueued to qdisc even when the qdisc is empty, see __dev_xmit_skb(). This patch calls sch_direct_xmit() to transmit the skb directly to the driver for empty lockless qdisc, which aviod enqueuing and dequeuing operation. As qdisc->empty is not reliable to indicate a empty qdisc because there is a time window between enqueuing and setting qdisc->empty. So we use the MISSED state added in commit a90c57f2cedd ("net: sched: fix packet stuck problem for lockless qdisc"), which indicate there is lock contention, suggesting that it is better not to do the qdisc bypass in order to avoid packet out of order problem. In order to make MISSED state reliable to indicate a empty qdisc, we need to ensure that testing and clearing of MISSED state is within the protection of qdisc->seqlock, only setting MISSED state can be done without the protection of qdisc->seqlock. A MISSED state testing is added without the protection of qdisc->seqlock to aviod doing unnecessary spin_trylock() for contention case. As the enqueuing is not within the protection of qdisc->seqlock, there is still a potential data race as mentioned by Jakub [1]: thread1 thread2 thread3 qdisc_run_begin() # true qdisc_run_begin(q) set(MISSED) pfifo_fast_dequeue clear(MISSED) # recheck the queue qdisc_run_end() enqueue skb1 qdisc empty # true qdisc_run_begin() # true sch_direct_xmit() # skb2 qdisc_run_begin() set(MISSED) When above happens, skb1 enqueued by thread2 is transmited after skb2 is transmited by thread3 because MISSED state setting and enqueuing is not under the qdisc->seqlock. If qdisc bypass is disabled, skb1 has better chance to be transmited quicker than skb2. This patch does not take care of the above data race, because we view this as similar as below: Even at the same time CPU1 and CPU2 write the skb to two socket which both heading to the same qdisc, there is no guarantee that which skb will hit the qdisc first, because there is a lot of factor like interrupt/softirq/cache miss/scheduling afffecting that. There are below cases that need special handling: 1. When MISSED state is cleared before another round of dequeuing in pfifo_fast_dequeue(), and __qdisc_run() might not be able to dequeue all skb in one round and call __netif_schedule(), which might result in a non-empty qdisc without MISSED set. In order to avoid this, the MISSED state is set for lockless qdisc and __netif_schedule() will be called at the end of qdisc_run_end. 2. The MISSED state also need to be set for lockless qdisc instead of calling __netif_schedule() directly when requeuing a skb for a similar reason. 3. For netdev queue stopped case, the MISSED case need clearing while the netdev queue is stopped, otherwise there may be unnecessary __netif_schedule() calling. So a new DRAINING state is added to indicate this case, which also indicate a non-empty qdisc. 4. As there is already netif_xmit_frozen_or_stopped() checking in dequeue_skb() and sch_direct_xmit(), which are both within the protection of qdisc->seqlock, but the same checking in __dev_xmit_skb() is without the protection, which might cause empty indication of a lockless qdisc to be not reliable. So remove the checking in __dev_xmit_skb(), and the checking in the protection of qdisc->seqlock seems enough to avoid the cpu consumption problem for netdev queue stopped case. 1. https://lkml.org/lkml/2021/5/29/215 Acked-by: Jakub Kicinski <kuba@kernel.org> Tested-by: Vladimir Oltean <vladimir.oltean@nxp.com> # flexcan Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-06-22 09:49:56 +03:00
if (lock) {
net: sched: always do stats accounting according to TCQ_F_CPUSTATS The core sched implementation checks independently for NOLOCK flag to acquire/release the root spin lock and for qdisc_is_percpu_stats() to account per CPU values in many places. This change update the last few places checking the TCQ_F_NOLOCK to do per CPU stats accounting according to qdisc_is_percpu_stats() value. The above allows to clean dev_requeue_skb() implementation a bit and makes stats update always consistent with a single flag. v1 -> v2: - do not move qdisc_is_empty definition, fix build issue Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-10 15:32:39 +03:00
spin_unlock(lock);
net: sched: implement TCQ_F_CAN_BYPASS for lockless qdisc Currently pfifo_fast has both TCQ_F_CAN_BYPASS and TCQ_F_NOLOCK flag set, but queue discipline by-pass does not work for lockless qdisc because skb is always enqueued to qdisc even when the qdisc is empty, see __dev_xmit_skb(). This patch calls sch_direct_xmit() to transmit the skb directly to the driver for empty lockless qdisc, which aviod enqueuing and dequeuing operation. As qdisc->empty is not reliable to indicate a empty qdisc because there is a time window between enqueuing and setting qdisc->empty. So we use the MISSED state added in commit a90c57f2cedd ("net: sched: fix packet stuck problem for lockless qdisc"), which indicate there is lock contention, suggesting that it is better not to do the qdisc bypass in order to avoid packet out of order problem. In order to make MISSED state reliable to indicate a empty qdisc, we need to ensure that testing and clearing of MISSED state is within the protection of qdisc->seqlock, only setting MISSED state can be done without the protection of qdisc->seqlock. A MISSED state testing is added without the protection of qdisc->seqlock to aviod doing unnecessary spin_trylock() for contention case. As the enqueuing is not within the protection of qdisc->seqlock, there is still a potential data race as mentioned by Jakub [1]: thread1 thread2 thread3 qdisc_run_begin() # true qdisc_run_begin(q) set(MISSED) pfifo_fast_dequeue clear(MISSED) # recheck the queue qdisc_run_end() enqueue skb1 qdisc empty # true qdisc_run_begin() # true sch_direct_xmit() # skb2 qdisc_run_begin() set(MISSED) When above happens, skb1 enqueued by thread2 is transmited after skb2 is transmited by thread3 because MISSED state setting and enqueuing is not under the qdisc->seqlock. If qdisc bypass is disabled, skb1 has better chance to be transmited quicker than skb2. This patch does not take care of the above data race, because we view this as similar as below: Even at the same time CPU1 and CPU2 write the skb to two socket which both heading to the same qdisc, there is no guarantee that which skb will hit the qdisc first, because there is a lot of factor like interrupt/softirq/cache miss/scheduling afffecting that. There are below cases that need special handling: 1. When MISSED state is cleared before another round of dequeuing in pfifo_fast_dequeue(), and __qdisc_run() might not be able to dequeue all skb in one round and call __netif_schedule(), which might result in a non-empty qdisc without MISSED set. In order to avoid this, the MISSED state is set for lockless qdisc and __netif_schedule() will be called at the end of qdisc_run_end. 2. The MISSED state also need to be set for lockless qdisc instead of calling __netif_schedule() directly when requeuing a skb for a similar reason. 3. For netdev queue stopped case, the MISSED case need clearing while the netdev queue is stopped, otherwise there may be unnecessary __netif_schedule() calling. So a new DRAINING state is added to indicate this case, which also indicate a non-empty qdisc. 4. As there is already netif_xmit_frozen_or_stopped() checking in dequeue_skb() and sch_direct_xmit(), which are both within the protection of qdisc->seqlock, but the same checking in __dev_xmit_skb() is without the protection, which might cause empty indication of a lockless qdisc to be not reliable. So remove the checking in __dev_xmit_skb(), and the checking in the protection of qdisc->seqlock seems enough to avoid the cpu consumption problem for netdev queue stopped case. 1. https://lkml.org/lkml/2021/5/29/215 Acked-by: Jakub Kicinski <kuba@kernel.org> Tested-by: Vladimir Oltean <vladimir.oltean@nxp.com> # flexcan Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-06-22 09:49:56 +03:00
set_bit(__QDISC_STATE_MISSED, &q->state);
} else {
__netif_schedule(q);
}
net: sched: explicit locking in gso_cpu fallback This work is preparing the qdisc layer to support egress lockless qdiscs. If we are running the egress qdisc lockless in the case we overrun the netdev, for whatever reason, the netdev returns a busy error code and the skb is parked on the gso_skb pointer. With many cores all hitting this case at once its possible to have multiple sk_buffs here so we turn gso_skb into a queue. This should be the edge case and if we see this frequently then the netdev/qdisc layer needs to back off. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:55:45 +03:00
}
qdisc: validate skb without holding lock Validation of skb can be pretty expensive : GSO segmentation and/or checksum computations. We can do this without holding qdisc lock, so that other cpus can queue additional packets. Trick is that requeued packets were already validated, so we carry a boolean so that sch_direct_xmit() can validate a fresh skb list, or directly use an old one. Tested on 40Gb NIC (8 TX queues) and 200 concurrent flows, 48 threads host. Turning TSO on or off had no effect on throughput, only few more cpu cycles. Lock contention on qdisc lock disappeared. Same if disabling TX checksum offload. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-04 02:31:07 +04:00
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 14:18:10 +04:00
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 00:35:59 +04:00
{
qdisc: validate skb without holding lock Validation of skb can be pretty expensive : GSO segmentation and/or checksum computations. We can do this without holding qdisc lock, so that other cpus can queue additional packets. Trick is that requeued packets were already validated, so we carry a boolean so that sch_direct_xmit() can validate a fresh skb list, or directly use an old one. Tested on 40Gb NIC (8 TX queues) and 200 concurrent flows, 48 threads host. Turning TSO on or off had no effect on throughput, only few more cpu cycles. Lock contention on qdisc lock disappeared. Same if disabling TX checksum offload. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-04 02:31:07 +04:00
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 00:35:59 +04:00
while (bytelimit > 0) {
qdisc: validate skb without holding lock Validation of skb can be pretty expensive : GSO segmentation and/or checksum computations. We can do this without holding qdisc lock, so that other cpus can queue additional packets. Trick is that requeued packets were already validated, so we carry a boolean so that sch_direct_xmit() can validate a fresh skb list, or directly use an old one. Tested on 40Gb NIC (8 TX queues) and 200 concurrent flows, 48 threads host. Turning TSO on or off had no effect on throughput, only few more cpu cycles. Lock contention on qdisc lock disappeared. Same if disabling TX checksum offload. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-04 02:31:07 +04:00
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 00:35:59 +04:00
qdisc: validate skb without holding lock Validation of skb can be pretty expensive : GSO segmentation and/or checksum computations. We can do this without holding qdisc lock, so that other cpus can queue additional packets. Trick is that requeued packets were already validated, so we carry a boolean so that sch_direct_xmit() can validate a fresh skb list, or directly use an old one. Tested on 40Gb NIC (8 TX queues) and 200 concurrent flows, 48 threads host. Turning TSO on or off had no effect on throughput, only few more cpu cycles. Lock contention on qdisc lock disappeared. Same if disabling TX checksum offload. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-04 02:31:07 +04:00
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 00:35:59 +04:00
break;
qdisc: validate skb without holding lock Validation of skb can be pretty expensive : GSO segmentation and/or checksum computations. We can do this without holding qdisc lock, so that other cpus can queue additional packets. Trick is that requeued packets were already validated, so we carry a boolean so that sch_direct_xmit() can validate a fresh skb list, or directly use an old one. Tested on 40Gb NIC (8 TX queues) and 200 concurrent flows, 48 threads host. Turning TSO on or off had no effect on throughput, only few more cpu cycles. Lock contention on qdisc lock disappeared. Same if disabling TX checksum offload. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-04 02:31:07 +04:00
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 14:18:10 +04:00
(*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 00:35:59 +04:00
}
net: Add and use skb_mark_not_on_list(). An SKB is not on a list if skb->next is NULL. Codify this convention into a helper function and use it where we are dequeueing an SKB and need to mark it as such. Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-30 06:42:53 +03:00
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 00:35:59 +04:00
}
net_sched: generalize bulk dequeue When qdisc bulk dequeue was added in linux-3.18 (commit 5772e9a3463b "qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE"), it was constrained to some specific qdiscs. With some extra care, we can extend this to all qdiscs, so that typical traffic shaping solutions can benefit from small batches (8 packets in this patch). For example, HTB is often used on some multi queue device. And bonding/team are multi queue devices... Idea is to bulk-dequeue packets mapping to the same transmit queue. This brings between 35 and 80 % performance increase in HTB setup under pressure on a bonding setup : 1) NUMA node contention : 610,000 pps -> 1,110,000 pps 2) No node contention : 1,380,000 pps -> 1,930,000 pps Now we should work to add batches on the enqueue() side ;) Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: John Fastabend <john.r.fastabend@intel.com> Cc: Jesper Dangaard Brouer <brouer@redhat.com> Cc: Hannes Frederic Sowa <hannes@stressinduktion.org> Cc: Florian Westphal <fw@strlen.de> Cc: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-06-22 09:16:52 +03:00
/* 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)) {
net: sched: use skb list for skb_bad_tx Similar to how gso is handled use skb list for skb_bad_tx this is required with lockless qdiscs because we may have multiple cores attempting to push skbs into skb_bad_tx concurrently Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:56:23 +03:00
qdisc_enqueue_skb_bad_txq(q, nskb);
net_sched: generalize bulk dequeue When qdisc bulk dequeue was added in linux-3.18 (commit 5772e9a3463b "qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE"), it was constrained to some specific qdiscs. With some extra care, we can extend this to all qdiscs, so that typical traffic shaping solutions can benefit from small batches (8 packets in this patch). For example, HTB is often used on some multi queue device. And bonding/team are multi queue devices... Idea is to bulk-dequeue packets mapping to the same transmit queue. This brings between 35 and 80 % performance increase in HTB setup under pressure on a bonding setup : 1) NUMA node contention : 610,000 pps -> 1,110,000 pps 2) No node contention : 1,380,000 pps -> 1,930,000 pps Now we should work to add batches on the enqueue() side ;) Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: John Fastabend <john.r.fastabend@intel.com> Cc: Jesper Dangaard Brouer <brouer@redhat.com> Cc: Hannes Frederic Sowa <hannes@stressinduktion.org> Cc: Florian Westphal <fw@strlen.de> Cc: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-06-22 09:16:52 +03:00
break;
}
skb->next = nskb;
skb = nskb;
} while (++cnt < 8);
(*packets) += cnt;
net: Add and use skb_mark_not_on_list(). An SKB is not on a list if skb->next is NULL. Codify this convention into a helper function and use it where we are dequeueing an SKB and need to mark it as such. Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-30 06:42:53 +03:00
skb_mark_not_on_list(skb);
net_sched: generalize bulk dequeue When qdisc bulk dequeue was added in linux-3.18 (commit 5772e9a3463b "qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE"), it was constrained to some specific qdiscs. With some extra care, we can extend this to all qdiscs, so that typical traffic shaping solutions can benefit from small batches (8 packets in this patch). For example, HTB is often used on some multi queue device. And bonding/team are multi queue devices... Idea is to bulk-dequeue packets mapping to the same transmit queue. This brings between 35 and 80 % performance increase in HTB setup under pressure on a bonding setup : 1) NUMA node contention : 610,000 pps -> 1,110,000 pps 2) No node contention : 1,380,000 pps -> 1,930,000 pps Now we should work to add batches on the enqueue() side ;) Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: John Fastabend <john.r.fastabend@intel.com> Cc: Jesper Dangaard Brouer <brouer@redhat.com> Cc: Hannes Frederic Sowa <hannes@stressinduktion.org> Cc: Florian Westphal <fw@strlen.de> Cc: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-06-22 09:16:52 +03:00
}
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 00:35:59 +04:00
/* 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 14:18:10 +04:00
static struct sk_buff *dequeue_skb(struct Qdisc *q, bool *validate,
int *packets)
[NET_SCHED]: Cleanup readability of qdisc restart Over the years this code has gotten hairier. Resulting in many long discussions over long summer days and patches that get it wrong. This patch helps tame that code so normal people will understand it. Thanks to Thomas Graf, Peter J. waskiewicz Jr, and Patrick McHardy for their valuable reviews. Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-06-11 04:31:24 +04:00
{
pkt_sched: avoid requeues if possible With BQL being deployed, we can more likely have following behavior : We dequeue a packet from qdisc in dequeue_skb(), then we realize target tx queue is in XOFF state in sch_direct_xmit(), and we have to hold the skb into gso_skb for later. This shows in stats (tc -s qdisc dev eth0) as requeues. Problem of these requeues is that high priority packets can not be dequeued as long as this (possibly low prio and big TSO packet) is not removed from gso_skb. At 1Gbps speed, a full size TSO packet is 500 us of extra latency. In some cases, we know that all packets dequeued from a qdisc are for a particular and known txq : - If device is non multi queue - For all MQ/MQPRIO slave qdiscs This patch introduces a new qdisc flag, TCQ_F_ONETXQUEUE to mark this capability, so that dequeue_skb() is allowed to dequeue a packet only if the associated txq is not stopped. This indeed reduce latencies for high prio packets (or improve fairness with sfq/fq_codel), and almost remove qdisc 'requeues'. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: John Fastabend <john.r.fastabend@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-12-11 19:54:33 +04:00
const struct netdev_queue *txq = q->dev_queue;
net: sched: check for frozen queue before skb_bad_txq check I can not think of any reason to pull the bad txq skb off the qdisc if the txq we plan to send this on is still frozen. So check for frozen queue first and abort before dequeuing either skb_bad_txq skb or normal qdisc dequeue() skb. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:56:42 +03:00
struct sk_buff *skb = NULL;
pkt_sched: Fix handling of gso skbs on requeuing Jay Cliburn noticed and diagnosed a bug triggered in dev_gso_skb_destructor() after last change from qdisc->gso_skb to qdisc->requeue list. Since gso_segmented skbs can't be queued to another list this patch brings back qdisc->gso_skb for them. Reported-by: Jay Cliburn <jcliburn@gmail.com> Signed-off-by: Jarek Poplawski <jarkao2@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-10-06 20:54:39 +04:00
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 14:18:10 +04:00
*packets = 1;
net: sched: explicit locking in gso_cpu fallback This work is preparing the qdisc layer to support egress lockless qdiscs. If we are running the egress qdisc lockless in the case we overrun the netdev, for whatever reason, the netdev returns a busy error code and the skb is parked on the gso_skb pointer. With many cores all hitting this case at once its possible to have multiple sk_buffs here so we turn gso_skb into a queue. This should be the edge case and if we see this frequently then the netdev/qdisc layer needs to back off. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:55:45 +03:00
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;
}
net_sched: generalize bulk dequeue When qdisc bulk dequeue was added in linux-3.18 (commit 5772e9a3463b "qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE"), it was constrained to some specific qdiscs. With some extra care, we can extend this to all qdiscs, so that typical traffic shaping solutions can benefit from small batches (8 packets in this patch). For example, HTB is often used on some multi queue device. And bonding/team are multi queue devices... Idea is to bulk-dequeue packets mapping to the same transmit queue. This brings between 35 and 80 % performance increase in HTB setup under pressure on a bonding setup : 1) NUMA node contention : 610,000 pps -> 1,110,000 pps 2) No node contention : 1,380,000 pps -> 1,930,000 pps Now we should work to add batches on the enqueue() side ;) Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: John Fastabend <john.r.fastabend@intel.com> Cc: Jesper Dangaard Brouer <brouer@redhat.com> Cc: Hannes Frederic Sowa <hannes@stressinduktion.org> Cc: Florian Westphal <fw@strlen.de> Cc: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-06-22 09:16:52 +03:00
/* skb in gso_skb were already validated */
*validate = false;
net: Add asynchronous callbacks for xfrm on layer 2. This patch implements asynchronous crypto callbacks and a backlog handler that can be used when IPsec is done at layer 2 in the TX path. It also extends the skb validate functions so that we can update the driver transmit return codes based on async crypto operation or to indicate that we queued the packet in a backlog queue. Joint work with: Aviv Heller <avivh@mellanox.com> Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
2017-12-20 12:41:36 +03:00
if (xfrm_offload(skb))
*validate = true;
pkt_sched: Check the state of tx_queue in dequeue_skb() Check in dequeue_skb() the state of tx_queue for requeued skb to save on locking and re-requeuing, and possibly remove the current check in qdisc_run(). Based on the idea of Alexander Duyck. Signed-off-by: Jarek Poplawski <jarkao2@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-09-23 09:16:23 +04:00
/* check the reason of requeuing without tx lock first */
net: add skb_get_tx_queue() helper Replace occurences of skb_get_queue_mapping() and follow-up netdev_get_tx_queue() with an actual helper function. Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-08-27 13:11:27 +04:00
txq = skb_get_tx_queue(txq->dev, skb);
net: Add queue state xoff flag for stack Create separate queue state flags so that either the stack or drivers can turn on XOFF. Added a set of functions used in the stack to determine if a queue is really stopped (either by stack or driver) Signed-off-by: Tom Herbert <therbert@google.com> Acked-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-11-28 20:32:44 +04:00
if (!netif_xmit_frozen_or_stopped(txq)) {
net: sched: explicit locking in gso_cpu fallback This work is preparing the qdisc layer to support egress lockless qdiscs. If we are running the egress qdisc lockless in the case we overrun the netdev, for whatever reason, the netdev returns a busy error code and the skb is parked on the gso_skb pointer. With many cores all hitting this case at once its possible to have multiple sk_buffs here so we turn gso_skb into a queue. This should be the edge case and if we see this frequently then the netdev/qdisc layer needs to back off. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:55:45 +03:00
skb = __skb_dequeue(&q->gso_skb);
if (qdisc_is_percpu_stats(q)) {
qdisc_qstats_cpu_backlog_dec(q, skb);
Revert: "net: sched: put back q.qlen into a single location" This revert commit 46b1c18f9deb ("net: sched: put back q.qlen into a single location"). After the previous patch, when a NOLOCK qdisc is enslaved to a locking qdisc it switches to global stats accounting. As a consequence, when a classful qdisc accesses directly a child qdisc's qlen, such qdisc is not doing per CPU accounting and qlen value is consistent. In the control path nobody uses directly qlen since commit e5f0e8f8e45 ("net: sched: introduce and use qdisc tree flush/purge helpers"), so we can remove the contented atomic ops from the datapath. v1 -> v2: - complete the qdisc_qstats_atomic_qlen_dec() -> qdisc_qstats_cpu_qlen_dec() replacement, fix build issue - more descriptive commit message Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-10 15:32:41 +03:00
qdisc_qstats_cpu_qlen_dec(q);
net: sched: explicit locking in gso_cpu fallback This work is preparing the qdisc layer to support egress lockless qdiscs. If we are running the egress qdisc lockless in the case we overrun the netdev, for whatever reason, the netdev returns a busy error code and the skb is parked on the gso_skb pointer. With many cores all hitting this case at once its possible to have multiple sk_buffs here so we turn gso_skb into a queue. This should be the edge case and if we see this frequently then the netdev/qdisc layer needs to back off. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:55:45 +03:00
} else {
qdisc_qstats_backlog_dec(q, skb);
q->q.qlen--;
}
} else {
pkt_sched: Check the state of tx_queue in dequeue_skb() Check in dequeue_skb() the state of tx_queue for requeued skb to save on locking and re-requeuing, and possibly remove the current check in qdisc_run(). Based on the idea of Alexander Duyck. Signed-off-by: Jarek Poplawski <jarkao2@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-09-23 09:16:23 +04:00
skb = NULL;
net: sched: fix tx action reschedule issue with stopped queue The netdev qeueue might be stopped when byte queue limit has reached or tx hw ring is full, net_tx_action() may still be rescheduled if STATE_MISSED is set, which consumes unnecessary cpu without dequeuing and transmiting any skb because the netdev queue is stopped, see qdisc_run_end(). This patch fixes it by checking the netdev queue state before calling qdisc_run() and clearing STATE_MISSED if netdev queue is stopped during qdisc_run(), the net_tx_action() is rescheduled again when netdev qeueue is restarted, see netif_tx_wake_queue(). As there is time window between netif_xmit_frozen_or_stopped() checking and STATE_MISSED clearing, between which STATE_MISSED may set by net_tx_action() scheduled by netif_tx_wake_queue(), so set the STATE_MISSED again if netdev queue is restarted. Fixes: 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking") Reported-by: Michal Kubecek <mkubecek@suse.cz> Acked-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-05-14 06:17:01 +03:00
qdisc_maybe_clear_missed(q, txq);
net: sched: explicit locking in gso_cpu fallback This work is preparing the qdisc layer to support egress lockless qdiscs. If we are running the egress qdisc lockless in the case we overrun the netdev, for whatever reason, the netdev returns a busy error code and the skb is parked on the gso_skb pointer. With many cores all hitting this case at once its possible to have multiple sk_buffs here so we turn gso_skb into a queue. This should be the edge case and if we see this frequently then the netdev/qdisc layer needs to back off. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:55:45 +03:00
}
if (lock)
spin_unlock(lock);
qdisc: add tracepoint qdisc:qdisc_dequeue for dequeued SKBs The main purpose of this tracepoint is to monitor bulk dequeue in the network qdisc layer, as it cannot be deducted from the existing qdisc stats. The txq_state can be used for determining the reason for zero packet dequeues, see enum netdev_queue_state_t. Notice all packets doesn't necessary activate this tracepoint. As qdiscs with flag TCQ_F_CAN_BYPASS, can directly invoke sch_direct_xmit() when qdisc_qlen is zero. Remember that perf record supports filters like: perf record -e qdisc:qdisc_dequeue \ --filter 'ifindex == 4 && (packets > 1 || txq_state > 0)' Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-08-15 22:11:03 +03:00
goto trace;
net_sched: generalize bulk dequeue When qdisc bulk dequeue was added in linux-3.18 (commit 5772e9a3463b "qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE"), it was constrained to some specific qdiscs. With some extra care, we can extend this to all qdiscs, so that typical traffic shaping solutions can benefit from small batches (8 packets in this patch). For example, HTB is often used on some multi queue device. And bonding/team are multi queue devices... Idea is to bulk-dequeue packets mapping to the same transmit queue. This brings between 35 and 80 % performance increase in HTB setup under pressure on a bonding setup : 1) NUMA node contention : 610,000 pps -> 1,110,000 pps 2) No node contention : 1,380,000 pps -> 1,930,000 pps Now we should work to add batches on the enqueue() side ;) Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: John Fastabend <john.r.fastabend@intel.com> Cc: Jesper Dangaard Brouer <brouer@redhat.com> Cc: Hannes Frederic Sowa <hannes@stressinduktion.org> Cc: Florian Westphal <fw@strlen.de> Cc: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-06-22 09:16:52 +03:00
}
net: sched: explicit locking in gso_cpu fallback This work is preparing the qdisc layer to support egress lockless qdiscs. If we are running the egress qdisc lockless in the case we overrun the netdev, for whatever reason, the netdev returns a busy error code and the skb is parked on the gso_skb pointer. With many cores all hitting this case at once its possible to have multiple sk_buffs here so we turn gso_skb into a queue. This should be the edge case and if we see this frequently then the netdev/qdisc layer needs to back off. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:55:45 +03:00
validate:
net_sched: generalize bulk dequeue When qdisc bulk dequeue was added in linux-3.18 (commit 5772e9a3463b "qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE"), it was constrained to some specific qdiscs. With some extra care, we can extend this to all qdiscs, so that typical traffic shaping solutions can benefit from small batches (8 packets in this patch). For example, HTB is often used on some multi queue device. And bonding/team are multi queue devices... Idea is to bulk-dequeue packets mapping to the same transmit queue. This brings between 35 and 80 % performance increase in HTB setup under pressure on a bonding setup : 1) NUMA node contention : 610,000 pps -> 1,110,000 pps 2) No node contention : 1,380,000 pps -> 1,930,000 pps Now we should work to add batches on the enqueue() side ;) Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: John Fastabend <john.r.fastabend@intel.com> Cc: Jesper Dangaard Brouer <brouer@redhat.com> Cc: Hannes Frederic Sowa <hannes@stressinduktion.org> Cc: Florian Westphal <fw@strlen.de> Cc: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-06-22 09:16:52 +03:00
*validate = true;
net: sched: check for frozen queue before skb_bad_txq check I can not think of any reason to pull the bad txq skb off the qdisc if the txq we plan to send this on is still frozen. So check for frozen queue first and abort before dequeuing either skb_bad_txq skb or normal qdisc dequeue() skb. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:56:42 +03:00
if ((q->flags & TCQ_F_ONETXQUEUE) &&
net: sched: fix tx action reschedule issue with stopped queue The netdev qeueue might be stopped when byte queue limit has reached or tx hw ring is full, net_tx_action() may still be rescheduled if STATE_MISSED is set, which consumes unnecessary cpu without dequeuing and transmiting any skb because the netdev queue is stopped, see qdisc_run_end(). This patch fixes it by checking the netdev queue state before calling qdisc_run() and clearing STATE_MISSED if netdev queue is stopped during qdisc_run(), the net_tx_action() is rescheduled again when netdev qeueue is restarted, see netif_tx_wake_queue(). As there is time window between netif_xmit_frozen_or_stopped() checking and STATE_MISSED clearing, between which STATE_MISSED may set by net_tx_action() scheduled by netif_tx_wake_queue(), so set the STATE_MISSED again if netdev queue is restarted. Fixes: 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking") Reported-by: Michal Kubecek <mkubecek@suse.cz> Acked-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-05-14 06:17:01 +03:00
netif_xmit_frozen_or_stopped(txq)) {
qdisc_maybe_clear_missed(q, txq);
net: sched: check for frozen queue before skb_bad_txq check I can not think of any reason to pull the bad txq skb off the qdisc if the txq we plan to send this on is still frozen. So check for frozen queue first and abort before dequeuing either skb_bad_txq skb or normal qdisc dequeue() skb. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:56:42 +03:00
return skb;
net: sched: fix tx action reschedule issue with stopped queue The netdev qeueue might be stopped when byte queue limit has reached or tx hw ring is full, net_tx_action() may still be rescheduled if STATE_MISSED is set, which consumes unnecessary cpu without dequeuing and transmiting any skb because the netdev queue is stopped, see qdisc_run_end(). This patch fixes it by checking the netdev queue state before calling qdisc_run() and clearing STATE_MISSED if netdev queue is stopped during qdisc_run(), the net_tx_action() is rescheduled again when netdev qeueue is restarted, see netif_tx_wake_queue(). As there is time window between netif_xmit_frozen_or_stopped() checking and STATE_MISSED clearing, between which STATE_MISSED may set by net_tx_action() scheduled by netif_tx_wake_queue(), so set the STATE_MISSED again if netdev queue is restarted. Fixes: 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking") Reported-by: Michal Kubecek <mkubecek@suse.cz> Acked-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-05-14 06:17:01 +03:00
}
net: sched: check for frozen queue before skb_bad_txq check I can not think of any reason to pull the bad txq skb off the qdisc if the txq we plan to send this on is still frozen. So check for frozen queue first and abort before dequeuing either skb_bad_txq skb or normal qdisc dequeue() skb. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:56:42 +03:00
net: sched: use skb list for skb_bad_tx Similar to how gso is handled use skb list for skb_bad_tx this is required with lockless qdiscs because we may have multiple cores attempting to push skbs into skb_bad_tx concurrently Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:56:23 +03:00
skb = qdisc_dequeue_skb_bad_txq(q);
net: sched: fix reordering issues Whenever MQ is not used on a multiqueue device, we experience serious reordering problems. Bisection found the cited commit. The issue can be described this way : - A single qdisc hierarchy is shared by all transmit queues. (eg : tc qdisc replace dev eth0 root fq_codel) - When/if try_bulk_dequeue_skb_slow() dequeues a packet targetting a different transmit queue than the one used to build a packet train, we stop building the current list and save the 'bad' skb (P1) in a special queue. (bad_txq) - When dequeue_skb() calls qdisc_dequeue_skb_bad_txq() and finds this skb (P1), it checks if the associated transmit queues is still in frozen state. If the queue is still blocked (by BQL or NIC tx ring full), we leave the skb in bad_txq and return NULL. - dequeue_skb() calls q->dequeue() to get another packet (P2) The other packet can target the problematic queue (that we found in frozen state for the bad_txq packet), but another cpu just ran TX completion and made room in the txq that is now ready to accept new packets. - Packet P2 is sent while P1 is still held in bad_txq, P1 might be sent at next round. In practice P2 is the lead of a big packet train (P2,P3,P4 ...) filling the BQL budget and delaying P1 by many packets :/ To solve this problem, we have to block the dequeue process as long as the first packet in bad_txq can not be sent. Reordering issues disappear and no side effects have been seen. Fixes: a53851e2c321 ("net: sched: explicit locking in gso_cpu fallback") Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: John Fastabend <john.fastabend@gmail.com> Acked-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-09-05 15:20:22 +03:00
if (unlikely(skb)) {
if (skb == SKB_XOFF_MAGIC)
return NULL;
net: sched: use skb list for skb_bad_tx Similar to how gso is handled use skb list for skb_bad_tx this is required with lockless qdiscs because we may have multiple cores attempting to push skbs into skb_bad_tx concurrently Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:56:23 +03:00
goto bulk;
net: sched: fix reordering issues Whenever MQ is not used on a multiqueue device, we experience serious reordering problems. Bisection found the cited commit. The issue can be described this way : - A single qdisc hierarchy is shared by all transmit queues. (eg : tc qdisc replace dev eth0 root fq_codel) - When/if try_bulk_dequeue_skb_slow() dequeues a packet targetting a different transmit queue than the one used to build a packet train, we stop building the current list and save the 'bad' skb (P1) in a special queue. (bad_txq) - When dequeue_skb() calls qdisc_dequeue_skb_bad_txq() and finds this skb (P1), it checks if the associated transmit queues is still in frozen state. If the queue is still blocked (by BQL or NIC tx ring full), we leave the skb in bad_txq and return NULL. - dequeue_skb() calls q->dequeue() to get another packet (P2) The other packet can target the problematic queue (that we found in frozen state for the bad_txq packet), but another cpu just ran TX completion and made room in the txq that is now ready to accept new packets. - Packet P2 is sent while P1 is still held in bad_txq, P1 might be sent at next round. In practice P2 is the lead of a big packet train (P2,P3,P4 ...) filling the BQL budget and delaying P1 by many packets :/ To solve this problem, we have to block the dequeue process as long as the first packet in bad_txq can not be sent. Reordering issues disappear and no side effects have been seen. Fixes: a53851e2c321 ("net: sched: explicit locking in gso_cpu fallback") Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: John Fastabend <john.fastabend@gmail.com> Acked-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-09-05 15:20:22 +03:00
}
net: sched: check for frozen queue before skb_bad_txq check I can not think of any reason to pull the bad txq skb off the qdisc if the txq we plan to send this on is still frozen. So check for frozen queue first and abort before dequeuing either skb_bad_txq skb or normal qdisc dequeue() skb. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:56:42 +03:00
skb = q->dequeue(q);
net_sched: generalize bulk dequeue When qdisc bulk dequeue was added in linux-3.18 (commit 5772e9a3463b "qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE"), it was constrained to some specific qdiscs. With some extra care, we can extend this to all qdiscs, so that typical traffic shaping solutions can benefit from small batches (8 packets in this patch). For example, HTB is often used on some multi queue device. And bonding/team are multi queue devices... Idea is to bulk-dequeue packets mapping to the same transmit queue. This brings between 35 and 80 % performance increase in HTB setup under pressure on a bonding setup : 1) NUMA node contention : 610,000 pps -> 1,110,000 pps 2) No node contention : 1,380,000 pps -> 1,930,000 pps Now we should work to add batches on the enqueue() side ;) Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: John Fastabend <john.r.fastabend@intel.com> Cc: Jesper Dangaard Brouer <brouer@redhat.com> Cc: Hannes Frederic Sowa <hannes@stressinduktion.org> Cc: Florian Westphal <fw@strlen.de> Cc: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-06-22 09:16:52 +03:00
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);
pkt_sched: Check the state of tx_queue in dequeue_skb() Check in dequeue_skb() the state of tx_queue for requeued skb to save on locking and re-requeuing, and possibly remove the current check in qdisc_run(). Based on the idea of Alexander Duyck. Signed-off-by: Jarek Poplawski <jarkao2@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-09-23 09:16:23 +04:00
}
qdisc: add tracepoint qdisc:qdisc_dequeue for dequeued SKBs The main purpose of this tracepoint is to monitor bulk dequeue in the network qdisc layer, as it cannot be deducted from the existing qdisc stats. The txq_state can be used for determining the reason for zero packet dequeues, see enum netdev_queue_state_t. Notice all packets doesn't necessary activate this tracepoint. As qdiscs with flag TCQ_F_CAN_BYPASS, can directly invoke sch_direct_xmit() when qdisc_qlen is zero. Remember that perf record supports filters like: perf record -e qdisc:qdisc_dequeue \ --filter 'ifindex == 4 && (packets > 1 || txq_state > 0)' Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-08-15 22:11:03 +03:00
trace:
trace_qdisc_dequeue(q, txq, *packets, skb);
[NET_SCHED]: Cleanup readability of qdisc restart Over the years this code has gotten hairier. Resulting in many long discussions over long summer days and patches that get it wrong. This patch helps tame that code so normal people will understand it. Thanks to Thomas Graf, Peter J. waskiewicz Jr, and Patrick McHardy for their valuable reviews. Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-06-11 04:31:24 +04:00
return skb;
}
[NET] SCHED: Fix whitespace errors. Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-02-09 17:25:16 +03:00
/*
qdisc: adjustments for API allowing skb list xmits Minor adjustments for merge commit 53fda7f7f9e (Merge branch 'xmit_list') that allows us to work with a list of SKBs. Update code doc to function sch_direct_xmit(). In handle_dev_cpu_collision() use kfree_skb_list() in error handling. Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-02 18:35:33 +04:00
* Transmit possibly several skbs, and handle the return status as
net: sched: Remove Qdisc::running sequence counter The Qdisc::running sequence counter has two uses: 1. Reliably reading qdisc's tc statistics while the qdisc is running (a seqcount read/retry loop at gnet_stats_add_basic()). 2. As a flag, indicating whether the qdisc in question is running (without any retry loops). For the first usage, the Qdisc::running sequence counter write section, qdisc_run_begin() => qdisc_run_end(), covers a much wider area than what is actually needed: the raw qdisc's bstats update. A u64_stats sync point was thus introduced (in previous commits) inside the bstats structure itself. A local u64_stats write section is then started and stopped for the bstats updates. Use that u64_stats sync point mechanism for the bstats read/retry loop at gnet_stats_add_basic(). For the second qdisc->running usage, a __QDISC_STATE_RUNNING bit flag, accessed with atomic bitops, is sufficient. Using a bit flag instead of a sequence counter at qdisc_run_begin/end() and qdisc_is_running() leads to the SMP barriers implicitly added through raw_read_seqcount() and write_seqcount_begin/end() getting removed. All call sites have been surveyed though, and no required ordering was identified. Now that the qdisc->running sequence counter is no longer used, remove it. Note, using u64_stats implies no sequence counter protection for 64-bit architectures. This can lead to the qdisc tc statistics "packets" vs. "bytes" values getting out of sync on rare occasions. The individual values will still be valid. Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-10-16 11:49:10 +03:00
* required. Owning qdisc running bit guarantees that only one CPU
* can execute this function.
[NET]: qdisc_restart - readability changes plus one bug fix. New changes : - Incorporated Peter Waskiewicz's comments. - Re-added back one warning message (on driver returning wrong value). Previous changes : - Converted to use switch/case code which looks neater. - "if (ret == NETDEV_TX_LOCKED && lockless)" is buggy, and the lockless check should be removed, since driver will return NETDEV_TX_LOCKED only if lockless is true and driver has to do the locking. In the original code as well as the latest code, this code can result in a bug where if LLTX is not set for a driver (lockless == 0) but the driver is written wrongly to do a trylock (despite LLTX being set), the driver returns LOCKED. But since lockless is zero, the packet is requeue'd instead of calling collision code which will issue warning and free up the skb. Instead this skb will be retried with this driver next time, and the same result will ensue. Removing this check will catch these driver bugs instead of hiding the problem. I am keeping this change to readability section since : a. it is confusing to check two things as it is; and b. it is difficult to keep this check in the changed 'switch' code. - Changed some names, like try_get_tx_pkt to dev_dequeue_skb (as that is the work being done and easier to understand) and do_dev_requeue to dev_requeue_skb, merged handle_dev_cpu_collision and tx_islocked to dev_handle_collision (handle_dev_cpu_collision is a small routine with only one caller, so there is no need to have two separate routines which also results in getting rid of two macros, etc. - Removed an XXX comment as it should never fail (I suspect this was related to batch skb WIP, Jamal ?). Converted some functions to original coding style of having the return values and the function name on same line, eg prio2list. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-06-25 06:56:09 +04:00
*
* Returns to the caller:
net: sched: remove remaining uses for qdisc_qlen in xmit path sch_direct_xmit() uses qdisc_qlen as a return value but all call sites of the routine only check if it is zero or not. Simplify the logic so that we don't need to return an actual queue length value. This introduces a case now where sch_direct_xmit would have returned a qlen of zero but now it returns true. However in this case all call sites of sch_direct_xmit will implement a dequeue() and get a null skb and abort. This trades tracking qlen in the hotpath for an extra dequeue operation. Overall this seems to be good for performance. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:54:47 +03:00
* 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 06:56:09 +04:00
*/
net: sched: remove remaining uses for qdisc_qlen in xmit path sch_direct_xmit() uses qdisc_qlen as a return value but all call sites of the routine only check if it is zero or not. Simplify the logic so that we don't need to return an actual queue length value. This introduces a case now where sch_direct_xmit would have returned a qlen of zero but now it returns true. However in this case all call sites of sch_direct_xmit will implement a dequeue() and get a null skb and abort. This trades tracking qlen in the hotpath for an extra dequeue operation. Overall this seems to be good for performance. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:54:47 +03:00
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)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
{
[PKT_SCHED]: Check subqueue status before calling hard_start_xmit The only qdiscs that check subqueue state before dequeue'ing are PRIO and RR. The other qdiscs, including the default pfifo_fast qdisc, will allow traffic bound for subqueue 0 through to hard_start_xmit. The check for netif_queue_stopped() is done above in pkt_sched.h, so it is unnecessary for qdisc_restart(). However, if the underlying driver is multiqueue capable, and only sets queue states on subqueues, this will allow packets to enter the driver when it's currently unable to process packets, resulting in expensive requeues and driver entries. This patch re-adds the check for the subqueue status before calling hard_start_xmit, so we can try and avoid the driver entry when the queues are stopped. Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-11-14 07:40:55 +03:00
int ret = NETDEV_TX_BUSY;
net: Add asynchronous callbacks for xfrm on layer 2. This patch implements asynchronous crypto callbacks and a backlog handler that can be used when IPsec is done at layer 2 in the TX path. It also extends the skb validate functions so that we can update the driver transmit return codes based on async crypto operation or to indicate that we queued the packet in a backlog queue. Joint work with: Aviv Heller <avivh@mellanox.com> Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
2017-12-20 12:41:36 +03:00
bool again = false;
pkt_sched: Add and use qdisc_root() and qdisc_root_lock(). When code wants to lock the qdisc tree state, the logic operation it's doing is locking the top-level qdisc that sits of the root of the netdev_queue. Add qdisc_root_lock() to represent this and convert the easiest cases. In order for this to work out in all cases, we have to hook up the noop_qdisc to a dummy netdev_queue. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-16 12:42:40 +04:00
/* And release qdisc */
net: sched: allow qdiscs to handle locking This patch adds a flag for queueing disciplines to indicate the stack does not need to use the qdisc lock to protect operations. This can be used to build lockless scheduling algorithms and improving performance. The flag is checked in the tx path and the qdisc lock is only taken if it is not set. For now use a conditional if statement. Later we could be more aggressive if it proves worthwhile and use a static key or wrap this in a likely(). Also the lockless case drops the TCQ_F_CAN_BYPASS logic. The reason for this is synchronizing a qlen counter across threads proves to cost more than doing the enqueue/dequeue operations when tested with pktgen. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:54:25 +03:00
if (root_lock)
spin_unlock(root_lock);
[NET_SCHED]: Cleanup readability of qdisc restart Over the years this code has gotten hairier. Resulting in many long discussions over long summer days and patches that get it wrong. This patch helps tame that code so normal people will understand it. Thanks to Thomas Graf, Peter J. waskiewicz Jr, and Patrick McHardy for their valuable reviews. Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-06-11 04:31:24 +04:00
qdisc: validate skb without holding lock Validation of skb can be pretty expensive : GSO segmentation and/or checksum computations. We can do this without holding qdisc lock, so that other cpus can queue additional packets. Trick is that requeued packets were already validated, so we carry a boolean so that sch_direct_xmit() can validate a fresh skb list, or directly use an old one. Tested on 40Gb NIC (8 TX queues) and 200 concurrent flows, 48 threads host. Turning TSO on or off had no effect on throughput, only few more cpu cycles. Lock contention on qdisc lock disappeared. Same if disabling TX checksum offload. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-04 02:31:07 +04:00
/* Note that we validate skb (GSO, checksum, ...) outside of locks */
if (validate)
net: Add asynchronous callbacks for xfrm on layer 2. This patch implements asynchronous crypto callbacks and a backlog handler that can be used when IPsec is done at layer 2 in the TX path. It also extends the skb validate functions so that we can update the driver transmit return codes based on async crypto operation or to indicate that we queued the packet in a backlog queue. Joint work with: Aviv Heller <avivh@mellanox.com> Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
2017-12-20 12:41:36 +03:00
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
net: allow to propagate errors through ->ndo_hard_start_xmit() Currently the ->ndo_hard_start_xmit() callbacks are only permitted to return one of the NETDEV_TX codes. This prevents any kind of error propagation for virtual devices, like queue congestion of the underlying device in case of layered devices, or unreachability in case of tunnels. This patches changes the NET_XMIT codes to avoid clashes with the NETDEV_TX codes and changes the two callers of dev_hard_start_xmit() to expect either errno codes, NET_XMIT codes or NETDEV_TX codes as return value. In case of qdisc_restart(), all non NETDEV_TX codes are mapped to NETDEV_TX_OK since no error propagation is possible when using qdiscs. In case of dev_queue_xmit(), the error is propagated upwards. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-10 09:14:14 +03:00
net: sched: do not requeue a NULL skb A failure in validate_xmit_skb_list() triggered an unconditional call to dev_requeue_skb with skb=NULL. This slowly grows the queue discipline's qlen count until all traffic through the queue stops. We take the optimistic approach and continue running the queue after a failure since it is unknown if later packets also will fail in the validate path. Fixes: 55a93b3ea780 ("qdisc: validate skb without holding lock") Signed-off-by: Lars Persson <larper@axis.com> Acked-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-12 09:45:52 +03:00
if (likely(skb)) {
qdisc: validate skb without holding lock Validation of skb can be pretty expensive : GSO segmentation and/or checksum computations. We can do this without holding qdisc lock, so that other cpus can queue additional packets. Trick is that requeued packets were already validated, so we carry a boolean so that sch_direct_xmit() can validate a fresh skb list, or directly use an old one. Tested on 40Gb NIC (8 TX queues) and 200 concurrent flows, 48 threads host. Turning TSO on or off had no effect on throughput, only few more cpu cycles. Lock contention on qdisc lock disappeared. Same if disabling TX checksum offload. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-04 02:31:07 +04:00
HARD_TX_LOCK(dev, txq, smp_processor_id());
if (!netif_xmit_frozen_or_stopped(txq))
skb = dev_hard_start_xmit(skb, dev, txq, &ret);
net: sched: fix tx action reschedule issue with stopped queue The netdev qeueue might be stopped when byte queue limit has reached or tx hw ring is full, net_tx_action() may still be rescheduled if STATE_MISSED is set, which consumes unnecessary cpu without dequeuing and transmiting any skb because the netdev queue is stopped, see qdisc_run_end(). This patch fixes it by checking the netdev queue state before calling qdisc_run() and clearing STATE_MISSED if netdev queue is stopped during qdisc_run(), the net_tx_action() is rescheduled again when netdev qeueue is restarted, see netif_tx_wake_queue(). As there is time window between netif_xmit_frozen_or_stopped() checking and STATE_MISSED clearing, between which STATE_MISSED may set by net_tx_action() scheduled by netif_tx_wake_queue(), so set the STATE_MISSED again if netdev queue is restarted. Fixes: 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking") Reported-by: Michal Kubecek <mkubecek@suse.cz> Acked-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-05-14 06:17:01 +03:00
else
qdisc_maybe_clear_missed(q, txq);
[NET_SCHED]: Cleanup readability of qdisc restart Over the years this code has gotten hairier. Resulting in many long discussions over long summer days and patches that get it wrong. This patch helps tame that code so normal people will understand it. Thanks to Thomas Graf, Peter J. waskiewicz Jr, and Patrick McHardy for their valuable reviews. Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-06-11 04:31:24 +04:00
qdisc: validate skb without holding lock Validation of skb can be pretty expensive : GSO segmentation and/or checksum computations. We can do this without holding qdisc lock, so that other cpus can queue additional packets. Trick is that requeued packets were already validated, so we carry a boolean so that sch_direct_xmit() can validate a fresh skb list, or directly use an old one. Tested on 40Gb NIC (8 TX queues) and 200 concurrent flows, 48 threads host. Turning TSO on or off had no effect on throughput, only few more cpu cycles. Lock contention on qdisc lock disappeared. Same if disabling TX checksum offload. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-04 02:31:07 +04:00
HARD_TX_UNLOCK(dev, txq);
net: sched: do not requeue a NULL skb A failure in validate_xmit_skb_list() triggered an unconditional call to dev_requeue_skb with skb=NULL. This slowly grows the queue discipline's qlen count until all traffic through the queue stops. We take the optimistic approach and continue running the queue after a failure since it is unknown if later packets also will fail in the validate path. Fixes: 55a93b3ea780 ("qdisc: validate skb without holding lock") Signed-off-by: Lars Persson <larper@axis.com> Acked-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-12 09:45:52 +03:00
} else {
net: sched: allow qdiscs to handle locking This patch adds a flag for queueing disciplines to indicate the stack does not need to use the qdisc lock to protect operations. This can be used to build lockless scheduling algorithms and improving performance. The flag is checked in the tx path and the qdisc lock is only taken if it is not set. For now use a conditional if statement. Later we could be more aggressive if it proves worthwhile and use a static key or wrap this in a likely(). Also the lockless case drops the TCQ_F_CAN_BYPASS logic. The reason for this is synchronizing a qlen counter across threads proves to cost more than doing the enqueue/dequeue operations when tested with pktgen. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:54:25 +03:00
if (root_lock)
spin_lock(root_lock);
net: sched: remove remaining uses for qdisc_qlen in xmit path sch_direct_xmit() uses qdisc_qlen as a return value but all call sites of the routine only check if it is zero or not. Simplify the logic so that we don't need to return an actual queue length value. This introduces a case now where sch_direct_xmit would have returned a qlen of zero but now it returns true. However in this case all call sites of sch_direct_xmit will implement a dequeue() and get a null skb and abort. This trades tracking qlen in the hotpath for an extra dequeue operation. Overall this seems to be good for performance. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:54:47 +03:00
return true;
qdisc: validate skb without holding lock Validation of skb can be pretty expensive : GSO segmentation and/or checksum computations. We can do this without holding qdisc lock, so that other cpus can queue additional packets. Trick is that requeued packets were already validated, so we carry a boolean so that sch_direct_xmit() can validate a fresh skb list, or directly use an old one. Tested on 40Gb NIC (8 TX queues) and 200 concurrent flows, 48 threads host. Turning TSO on or off had no effect on throughput, only few more cpu cycles. Lock contention on qdisc lock disappeared. Same if disabling TX checksum offload. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-04 02:31:07 +04:00
}
net: sched: allow qdiscs to handle locking This patch adds a flag for queueing disciplines to indicate the stack does not need to use the qdisc lock to protect operations. This can be used to build lockless scheduling algorithms and improving performance. The flag is checked in the tx path and the qdisc lock is only taken if it is not set. For now use a conditional if statement. Later we could be more aggressive if it proves worthwhile and use a static key or wrap this in a likely(). Also the lockless case drops the TCQ_F_CAN_BYPASS logic. The reason for this is synchronizing a qlen counter across threads proves to cost more than doing the enqueue/dequeue operations when tested with pktgen. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:54:25 +03:00
if (root_lock)
spin_lock(root_lock);
[NET_SCHED]: Cleanup readability of qdisc restart Over the years this code has gotten hairier. Resulting in many long discussions over long summer days and patches that get it wrong. This patch helps tame that code so normal people will understand it. Thanks to Thomas Graf, Peter J. waskiewicz Jr, and Patrick McHardy for their valuable reviews. Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-06-11 04:31:24 +04:00
net: sched: remove remaining uses for qdisc_qlen in xmit path sch_direct_xmit() uses qdisc_qlen as a return value but all call sites of the routine only check if it is zero or not. Simplify the logic so that we don't need to return an actual queue length value. This introduces a case now where sch_direct_xmit would have returned a qlen of zero but now it returns true. However in this case all call sites of sch_direct_xmit will implement a dequeue() and get a null skb and abort. This trades tracking qlen in the hotpath for an extra dequeue operation. Overall this seems to be good for performance. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:54:47 +03:00
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 06:56:09 +04:00
/* Driver returned NETDEV_TX_BUSY - requeue skb */
net: Convert net_ratelimit uses to net_<level>_ratelimited Standardize the net core ratelimited logging functions. Coalesce formats, align arguments. Change a printk then vprintk sequence to use printf extension %pV. Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-05-14 01:56:26 +04:00
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 06:56:09 +04:00
net: sched: remove remaining uses for qdisc_qlen in xmit path sch_direct_xmit() uses qdisc_qlen as a return value but all call sites of the routine only check if it is zero or not. Simplify the logic so that we don't need to return an actual queue length value. This introduces a case now where sch_direct_xmit would have returned a qlen of zero but now it returns true. However in this case all call sites of sch_direct_xmit will implement a dequeue() and get a null skb and abort. This trades tracking qlen in the hotpath for an extra dequeue operation. Overall this seems to be good for performance. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:54:47 +03:00
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 06:56:09 +04:00
}
[NET_SCHED]: Cleanup readability of qdisc restart Over the years this code has gotten hairier. Resulting in many long discussions over long summer days and patches that get it wrong. This patch helps tame that code so normal people will understand it. Thanks to Thomas Graf, Peter J. waskiewicz Jr, and Patrick McHardy for their valuable reviews. Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-06-11 04:31:24 +04:00
net: sched: remove remaining uses for qdisc_qlen in xmit path sch_direct_xmit() uses qdisc_qlen as a return value but all call sites of the routine only check if it is zero or not. Simplify the logic so that we don't need to return an actual queue length value. This introduces a case now where sch_direct_xmit would have returned a qlen of zero but now it returns true. However in this case all call sites of sch_direct_xmit will implement a dequeue() and get a null skb and abort. This trades tracking qlen in the hotpath for an extra dequeue operation. Overall this seems to be good for performance. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:54:47 +03:00
return true;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
}
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 05:44:21 +04:00
/*
* NOTE: Called under qdisc_lock(q) with locally disabled BH.
*
net_sched: transform qdisc running bit into a seqcount Instead of using a single bit (__QDISC___STATE_RUNNING) in sch->__state, use a seqcount. This adds lockdep support, but more importantly it will allow us to sample qdisc/class statistics without having to grab qdisc root lock. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-06-06 19:37:15 +03:00
* 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 05:44:21 +04:00
* 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.
*
*/
net: sched: remove remaining uses for qdisc_qlen in xmit path sch_direct_xmit() uses qdisc_qlen as a return value but all call sites of the routine only check if it is zero or not. Simplify the logic so that we don't need to return an actual queue length value. This introduces a case now where sch_direct_xmit would have returned a qlen of zero but now it returns true. However in this case all call sites of sch_direct_xmit will implement a dequeue() and get a null skb and abort. This trades tracking qlen in the hotpath for an extra dequeue operation. Overall this seems to be good for performance. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:54:47 +03:00
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 05:44:21 +04:00
{
net: sched: allow qdiscs to handle locking This patch adds a flag for queueing disciplines to indicate the stack does not need to use the qdisc lock to protect operations. This can be used to build lockless scheduling algorithms and improving performance. The flag is checked in the tx path and the qdisc lock is only taken if it is not set. For now use a conditional if statement. Later we could be more aggressive if it proves worthwhile and use a static key or wrap this in a likely(). Also the lockless case drops the TCQ_F_CAN_BYPASS logic. The reason for this is synchronizing a qlen counter across threads proves to cost more than doing the enqueue/dequeue operations when tested with pktgen. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:54:25 +03:00
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 05:44:21 +04:00
struct netdev_queue *txq;
struct net_device *dev;
struct sk_buff *skb;
sched: manipulate __QDISC_STATE_RUNNING in qdisc_run_* helpers Currently NOLOCK qdiscs pay a measurable overhead to atomically manipulate the __QDISC_STATE_RUNNING. Such bit is flipped twice per packet in the uncontended scenario with packet rate below the line rate: on packed dequeue and on the next, failing dequeue attempt. This changeset moves the bit manipulation into the qdisc_run_{begin,end} helpers, so that the bit is now flipped only once per packet, with measurable performance improvement in the uncontended scenario. This also allows simplifying the qdisc teardown code path - since qdisc_is_running() is now effective for each qdisc type - and avoid a possible race between qdisc_run() and dev_deactivate_many(), as now the some_qdisc_is_busy() can properly detect NOLOCK qdiscs being busy dequeuing packets. Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-05-15 11:50:31 +03:00
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 05:44:21 +04:00
/* 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 14:18:10 +04:00
skb = dequeue_skb(q, &validate, packets);
sched: manipulate __QDISC_STATE_RUNNING in qdisc_run_* helpers Currently NOLOCK qdiscs pay a measurable overhead to atomically manipulate the __QDISC_STATE_RUNNING. Such bit is flipped twice per packet in the uncontended scenario with packet rate below the line rate: on packed dequeue and on the next, failing dequeue attempt. This changeset moves the bit manipulation into the qdisc_run_{begin,end} helpers, so that the bit is now flipped only once per packet, with measurable performance improvement in the uncontended scenario. This also allows simplifying the qdisc teardown code path - since qdisc_is_running() is now effective for each qdisc type - and avoid a possible race between qdisc_run() and dev_deactivate_many(), as now the some_qdisc_is_busy() can properly detect NOLOCK qdiscs being busy dequeuing packets. Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-05-15 11:50:31 +03:00
if (unlikely(!skb))
net: sched: remove remaining uses for qdisc_qlen in xmit path sch_direct_xmit() uses qdisc_qlen as a return value but all call sites of the routine only check if it is zero or not. Simplify the logic so that we don't need to return an actual queue length value. This introduces a case now where sch_direct_xmit would have returned a qlen of zero but now it returns true. However in this case all call sites of sch_direct_xmit will implement a dequeue() and get a null skb and abort. This trades tracking qlen in the hotpath for an extra dequeue operation. Overall this seems to be good for performance. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:54:47 +03:00
return false;
net: add skb_get_tx_queue() helper Replace occurences of skb_get_queue_mapping() and follow-up netdev_get_tx_queue() with an actual helper function. Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-08-27 13:11:27 +04:00
sched: manipulate __QDISC_STATE_RUNNING in qdisc_run_* helpers Currently NOLOCK qdiscs pay a measurable overhead to atomically manipulate the __QDISC_STATE_RUNNING. Such bit is flipped twice per packet in the uncontended scenario with packet rate below the line rate: on packed dequeue and on the next, failing dequeue attempt. This changeset moves the bit manipulation into the qdisc_run_{begin,end} helpers, so that the bit is now flipped only once per packet, with measurable performance improvement in the uncontended scenario. This also allows simplifying the qdisc teardown code path - since qdisc_is_running() is now effective for each qdisc type - and avoid a possible race between qdisc_run() and dev_deactivate_many(), as now the some_qdisc_is_busy() can properly detect NOLOCK qdiscs being busy dequeuing packets. Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-05-15 11:50:31 +03:00
if (!(q->flags & TCQ_F_NOLOCK))
net: sched: allow qdiscs to handle locking This patch adds a flag for queueing disciplines to indicate the stack does not need to use the qdisc lock to protect operations. This can be used to build lockless scheduling algorithms and improving performance. The flag is checked in the tx path and the qdisc lock is only taken if it is not set. For now use a conditional if statement. Later we could be more aggressive if it proves worthwhile and use a static key or wrap this in a likely(). Also the lockless case drops the TCQ_F_CAN_BYPASS logic. The reason for this is synchronizing a qlen counter across threads proves to cost more than doing the enqueue/dequeue operations when tested with pktgen. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:54:25 +03:00
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 05:44:21 +04:00
dev = qdisc_dev(q);
net: add skb_get_tx_queue() helper Replace occurences of skb_get_queue_mapping() and follow-up netdev_get_tx_queue() with an actual helper function. Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-08-27 13:11:27 +04:00
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 05:44:21 +04:00
sched: manipulate __QDISC_STATE_RUNNING in qdisc_run_* helpers Currently NOLOCK qdiscs pay a measurable overhead to atomically manipulate the __QDISC_STATE_RUNNING. Such bit is flipped twice per packet in the uncontended scenario with packet rate below the line rate: on packed dequeue and on the next, failing dequeue attempt. This changeset moves the bit manipulation into the qdisc_run_{begin,end} helpers, so that the bit is now flipped only once per packet, with measurable performance improvement in the uncontended scenario. This also allows simplifying the qdisc teardown code path - since qdisc_is_running() is now effective for each qdisc type - and avoid a possible race between qdisc_run() and dev_deactivate_many(), as now the some_qdisc_is_busy() can properly detect NOLOCK qdiscs being busy dequeuing packets. Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-05-15 11:50:31 +03:00
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 05:44:21 +04:00
}
pkt_sched: Schedule qdiscs instead of netdev_queue. When we have shared qdiscs, packets come out of the qdiscs for multiple transmit queues. Therefore it doesn't make any sense to schedule the transmit queue when logically we cannot know ahead of time the TX queue of the SKB that the qdisc->dequeue() will give us. Just for sanity I added a BUG check to make sure we never get into a state where the noop_qdisc is scheduled. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-16 13:15:04 +04:00
void __qdisc_run(struct Qdisc *q)
[NET]: Prevent multiple qdisc runs Having two or more qdisc_run's contend against each other is bad because it can induce packet reordering if the packets have to be requeued. It appears that this is an unintended consequence of relinquinshing the queue lock while transmitting. That in turn is needed for devices that spend a lot of time in their transmit routine. There are no advantages to be had as devices with queues are inherently single-threaded (the loopback device is not but then it doesn't have a queue). Even if you were to add a queue to a parallel virtual device (e.g., bolt a tbf filter in front of an ipip tunnel device), you would still want to process the queue in sequence to ensure that the packets are ordered correctly. The solution here is to steal a bit from net_device to prevent this. BTW, as qdisc_restart is no longer used by anyone as a module inside the kernel (IIRC it used to with netif_wake_queue), I have not exported the new __qdisc_run function. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-20 10:57:59 +04:00
{
net: dev_weight: TX/RX orthogonality Oftenly, introducing side effects on packet processing on the other half of the stack by adjusting one of TX/RX via sysctl is not desirable. There are cases of demand for asymmetric, orthogonal configurability. This holds true especially for nodes where RPS for RFS usage on top is configured and therefore use the 'old dev_weight'. This is quite a common base configuration setup nowadays, even with NICs of superior processing support (e.g. aRFS). A good example use case are nodes acting as noSQL data bases with a large number of tiny requests and rather fewer but large packets as responses. It's affordable to have large budget and rx dev_weights for the requests. But as a side effect having this large a number on TX processed in one run can overwhelm drivers. This patch therefore introduces an independent configurability via sysctl to userland. Signed-off-by: Matthias Tafelmeier <matthias.tafelmeier@gmx.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-12-29 23:37:21 +03:00
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 14:18:10 +04:00
int packets;
[NET]: Add preemption point in qdisc_run The qdisc_run loop is currently unbounded and runs entirely in a softirq. This is bad as it may create an unbounded softirq run. This patch fixes this by calling need_resched and breaking out if necessary. It also adds a break out if the jiffies value changes since that would indicate we've been transmitting for too long which starves other softirqs. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-03-29 02:25:26 +03:00
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 14:18:10 +04:00
while (qdisc_restart(q, &packets)) {
quota -= packets;
net_sched: remove need_resched() from qdisc_run() The introduction of this schedule point was done in commit 2ba2506ca7ca ("[NET]: Add preemption point in qdisc_run") at a time the loop was not bounded. Then later in commit d5b8aa1d246f ("net_sched: fix dequeuer fairness") we added a limit on the number of packets. Now is the time to remove the schedule point, since the default limit of 64 packets matches the number of packets a typical NAPI poll can process in a row. This solves a latency problem for most TCP receivers under moderate load : 1) host receives a packet. NET_RX_SOFTIRQ is raised by NIC hard IRQ handler 2) __do_softirq() does its first loop, handling NET_RX_SOFTIRQ and calling the driver napi->loop() function 3) TCP stores the skb in socket receive queue: 4) TCP calls sk->sk_data_ready() and wakeups a user thread waiting for EPOLLIN (as a result, need_resched() might now be true) 5) TCP cooks an ACK and sends it. 6) qdisc_run() processes one packet from qdisc, and sees need_resched(), this raises NET_TX_SOFTIRQ (even if there are no more packets in the qdisc) Then we go back to the __do_softirq() in 2), and we see that new softirqs were raised. Since need_resched() is true, we end up waking ksoftirqd in this path : if (pending) { if (time_before(jiffies, end) && !need_resched() && --max_restart) goto restart; wakeup_softirqd(); } So we have many wakeups of ksoftirqd kernel threads, and more calls to qdisc_run() with associated lock overhead. Note that another way to solve the issue would be to change TCP to first send the ACK packet, then signal the EPOLLIN, but this changes P99 latencies, as sending the ACK packet can add a long delay. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-10-02 00:02:36 +03:00
if (quota <= 0) {
net: sched: implement TCQ_F_CAN_BYPASS for lockless qdisc Currently pfifo_fast has both TCQ_F_CAN_BYPASS and TCQ_F_NOLOCK flag set, but queue discipline by-pass does not work for lockless qdisc because skb is always enqueued to qdisc even when the qdisc is empty, see __dev_xmit_skb(). This patch calls sch_direct_xmit() to transmit the skb directly to the driver for empty lockless qdisc, which aviod enqueuing and dequeuing operation. As qdisc->empty is not reliable to indicate a empty qdisc because there is a time window between enqueuing and setting qdisc->empty. So we use the MISSED state added in commit a90c57f2cedd ("net: sched: fix packet stuck problem for lockless qdisc"), which indicate there is lock contention, suggesting that it is better not to do the qdisc bypass in order to avoid packet out of order problem. In order to make MISSED state reliable to indicate a empty qdisc, we need to ensure that testing and clearing of MISSED state is within the protection of qdisc->seqlock, only setting MISSED state can be done without the protection of qdisc->seqlock. A MISSED state testing is added without the protection of qdisc->seqlock to aviod doing unnecessary spin_trylock() for contention case. As the enqueuing is not within the protection of qdisc->seqlock, there is still a potential data race as mentioned by Jakub [1]: thread1 thread2 thread3 qdisc_run_begin() # true qdisc_run_begin(q) set(MISSED) pfifo_fast_dequeue clear(MISSED) # recheck the queue qdisc_run_end() enqueue skb1 qdisc empty # true qdisc_run_begin() # true sch_direct_xmit() # skb2 qdisc_run_begin() set(MISSED) When above happens, skb1 enqueued by thread2 is transmited after skb2 is transmited by thread3 because MISSED state setting and enqueuing is not under the qdisc->seqlock. If qdisc bypass is disabled, skb1 has better chance to be transmited quicker than skb2. This patch does not take care of the above data race, because we view this as similar as below: Even at the same time CPU1 and CPU2 write the skb to two socket which both heading to the same qdisc, there is no guarantee that which skb will hit the qdisc first, because there is a lot of factor like interrupt/softirq/cache miss/scheduling afffecting that. There are below cases that need special handling: 1. When MISSED state is cleared before another round of dequeuing in pfifo_fast_dequeue(), and __qdisc_run() might not be able to dequeue all skb in one round and call __netif_schedule(), which might result in a non-empty qdisc without MISSED set. In order to avoid this, the MISSED state is set for lockless qdisc and __netif_schedule() will be called at the end of qdisc_run_end. 2. The MISSED state also need to be set for lockless qdisc instead of calling __netif_schedule() directly when requeuing a skb for a similar reason. 3. For netdev queue stopped case, the MISSED case need clearing while the netdev queue is stopped, otherwise there may be unnecessary __netif_schedule() calling. So a new DRAINING state is added to indicate this case, which also indicate a non-empty qdisc. 4. As there is already netif_xmit_frozen_or_stopped() checking in dequeue_skb() and sch_direct_xmit(), which are both within the protection of qdisc->seqlock, but the same checking in __dev_xmit_skb() is without the protection, which might cause empty indication of a lockless qdisc to be not reliable. So remove the checking in __dev_xmit_skb(), and the checking in the protection of qdisc->seqlock seems enough to avoid the cpu consumption problem for netdev queue stopped case. 1. https://lkml.org/lkml/2021/5/29/215 Acked-by: Jakub Kicinski <kuba@kernel.org> Tested-by: Vladimir Oltean <vladimir.oltean@nxp.com> # flexcan Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-06-22 09:49:56 +03:00
if (q->flags & TCQ_F_NOLOCK)
set_bit(__QDISC_STATE_MISSED, &q->state);
else
__netif_schedule(q);
[NET_SCHED]: Rationalise return value of qdisc_restart The current return value scheme and associated comment was invented back in the 20th century when we still had that tbusy flag. Things have changed quite a bit since then (even Tony Blair is moving on now, not to mention the new French president). All we need to indicate now is whether the caller should continue processing the queue. Therefore it's sufficient if we return 0 if we want to stop and non-zero otherwise. This is based on a patch by Krishna Kumar. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-05-10 15:55:14 +04:00
break;
[NET]: Add preemption point in qdisc_run The qdisc_run loop is currently unbounded and runs entirely in a softirq. This is bad as it may create an unbounded softirq run. This patch fixes this by calling need_resched and breaking out if necessary. It also adds a break out if the jiffies value changes since that would indicate we've been transmitting for too long which starves other softirqs. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-03-29 02:25:26 +03:00
}
}
[NET]: Prevent multiple qdisc runs Having two or more qdisc_run's contend against each other is bad because it can induce packet reordering if the packets have to be requeued. It appears that this is an unintended consequence of relinquinshing the queue lock while transmitting. That in turn is needed for devices that spend a lot of time in their transmit routine. There are no advantages to be had as devices with queues are inherently single-threaded (the loopback device is not but then it doesn't have a queue). Even if you were to add a queue to a parallel virtual device (e.g., bolt a tbf filter in front of an ipip tunnel device), you would still want to process the queue in sequence to ensure that the packets are ordered correctly. The solution here is to steal a bit from net_device to prevent this. BTW, as qdisc_restart is no longer used by anyone as a module inside the kernel (IIRC it used to with netif_wake_queue), I have not exported the new __qdisc_run function. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-20 10:57:59 +04:00
}
net: tx scalability works : trans_start struct net_device trans_start field is a hot spot on SMP and high performance devices, particularly multi queues ones, because every transmitter dirties it. Is main use is tx watchdog and bonding alive checks. But as most devices dont use NETIF_F_LLTX, we have to lock a netdev_queue before calling their ndo_start_xmit(). So it makes sense to move trans_start from net_device to netdev_queue. Its update will occur on a already present (and in exclusive state) cache line, for free. We can do this transition smoothly. An old driver continue to update dev->trans_start, while an updated one updates txq->trans_start. Further patches could also put tx_bytes/tx_packets counters in netdev_queue to avoid dirtying dev->stats (vlan device comes to mind) Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-05-18 07:55:16 +04:00
unsigned long dev_trans_start(struct net_device *dev)
{
net_sched: make dev_trans_start return vlan's real dev trans_start Vlan devices are LLTX and don't update their own trans_start, so if dev_trans_start has to be called with a vlan device then 0 or a stale value will be returned. Currently the bonding is the only such user, and it's needed for proper arp monitoring when the slaves are vlans. Fix this by extracting the vlan's real device trans_start. Suggested-by: David Miller <davem@davemloft.net> Signed-off-by: Nikolay Aleksandrov <nikolay@redhat.com> Acked-by: Veaceslav Falico <vfalico@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-08-04 00:07:47 +04:00
unsigned long val, res;
net: tx scalability works : trans_start struct net_device trans_start field is a hot spot on SMP and high performance devices, particularly multi queues ones, because every transmitter dirties it. Is main use is tx watchdog and bonding alive checks. But as most devices dont use NETIF_F_LLTX, we have to lock a netdev_queue before calling their ndo_start_xmit(). So it makes sense to move trans_start from net_device to netdev_queue. Its update will occur on a already present (and in exclusive state) cache line, for free. We can do this transition smoothly. An old driver continue to update dev->trans_start, while an updated one updates txq->trans_start. Further patches could also put tx_bytes/tx_packets counters in netdev_queue to avoid dirtying dev->stats (vlan device comes to mind) Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-05-18 07:55:16 +04:00
unsigned int i;
net_sched: make dev_trans_start return vlan's real dev trans_start Vlan devices are LLTX and don't update their own trans_start, so if dev_trans_start has to be called with a vlan device then 0 or a stale value will be returned. Currently the bonding is the only such user, and it's needed for proper arp monitoring when the slaves are vlans. Fix this by extracting the vlan's real device trans_start. Suggested-by: David Miller <davem@davemloft.net> Signed-off-by: Nikolay Aleksandrov <nikolay@redhat.com> Acked-by: Veaceslav Falico <vfalico@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-08-04 00:07:47 +04:00
if (is_vlan_dev(dev))
dev = vlan_dev_real_dev(dev);
net_sched: use macvlan real dev trans_start in dev_trans_start() Macvlan devices are similar to vlans and do not update their own trans_start. In order for arp monitoring to work for a bond device when the slaves are macvlans, obtain its real device. Signed-off-by: Chris Dion <christopher.dion@dell.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-06 18:50:28 +03:00
else if (netif_is_macvlan(dev))
dev = macvlan_dev_real_dev(dev);
net: annotate accesses to queue->trans_start In following patches, dev_watchdog() will no longer stop all queues. It will read queue->trans_start locklessly. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-11-17 06:29:22 +03:00
res = READ_ONCE(netdev_get_tx_queue(dev, 0)->trans_start);
net: remove dev->trans_start previous patches removed all direct accesses to dev->trans_start, so change the netif_trans_update helper to update trans_start of netdev queue 0 instead and then remove trans_start from struct net_device. AFAICS a lot of the netif_trans_update() invocations are now useless because they occur in ndo_start_xmit and driver doesn't set LLTX (i.e. stack already took care of the update). As I can't test any of them it seems better to just leave them alone. Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-05-03 17:33:14 +03:00
for (i = 1; i < dev->num_tx_queues; i++) {
net: annotate accesses to queue->trans_start In following patches, dev_watchdog() will no longer stop all queues. It will read queue->trans_start locklessly. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-11-17 06:29:22 +03:00
val = READ_ONCE(netdev_get_tx_queue(dev, i)->trans_start);
net: tx scalability works : trans_start struct net_device trans_start field is a hot spot on SMP and high performance devices, particularly multi queues ones, because every transmitter dirties it. Is main use is tx watchdog and bonding alive checks. But as most devices dont use NETIF_F_LLTX, we have to lock a netdev_queue before calling their ndo_start_xmit(). So it makes sense to move trans_start from net_device to netdev_queue. Its update will occur on a already present (and in exclusive state) cache line, for free. We can do this transition smoothly. An old driver continue to update dev->trans_start, while an updated one updates txq->trans_start. Further patches could also put tx_bytes/tx_packets counters in netdev_queue to avoid dirtying dev->stats (vlan device comes to mind) Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-05-18 07:55:16 +04:00
if (val && time_after(val, res))
res = val;
}
net_sched: make dev_trans_start return vlan's real dev trans_start Vlan devices are LLTX and don't update their own trans_start, so if dev_trans_start has to be called with a vlan device then 0 or a stale value will be returned. Currently the bonding is the only such user, and it's needed for proper arp monitoring when the slaves are vlans. Fix this by extracting the vlan's real device trans_start. Suggested-by: David Miller <davem@davemloft.net> Signed-off-by: Nikolay Aleksandrov <nikolay@redhat.com> Acked-by: Veaceslav Falico <vfalico@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-08-04 00:07:47 +04:00
net: tx scalability works : trans_start struct net_device trans_start field is a hot spot on SMP and high performance devices, particularly multi queues ones, because every transmitter dirties it. Is main use is tx watchdog and bonding alive checks. But as most devices dont use NETIF_F_LLTX, we have to lock a netdev_queue before calling their ndo_start_xmit(). So it makes sense to move trans_start from net_device to netdev_queue. Its update will occur on a already present (and in exclusive state) cache line, for free. We can do this transition smoothly. An old driver continue to update dev->trans_start, while an updated one updates txq->trans_start. Further patches could also put tx_bytes/tx_packets counters in netdev_queue to avoid dirtying dev->stats (vlan device comes to mind) Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-05-18 07:55:16 +04:00
return res;
}
EXPORT_SYMBOL(dev_trans_start);
net: do not inline netif_tx_lock()/netif_tx_unlock() These are not fast path, there is no point in inlining them. Also provide netif_freeze_queues()/netif_unfreeze_queues() so that we can use them from dev_watchdog() in the following patch. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-11-17 06:29:23 +03:00
static void netif_freeze_queues(struct net_device *dev)
{
unsigned int i;
int cpu;
cpu = smp_processor_id();
for (i = 0; i < dev->num_tx_queues; i++) {
struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
/* We are the only thread of execution doing a
* freeze, but we have to grab the _xmit_lock in
* order to synchronize with threads which are in
* the ->hard_start_xmit() handler and already
* checked the frozen bit.
*/
__netif_tx_lock(txq, cpu);
set_bit(__QUEUE_STATE_FROZEN, &txq->state);
__netif_tx_unlock(txq);
}
}
void netif_tx_lock(struct net_device *dev)
{
spin_lock(&dev->tx_global_lock);
netif_freeze_queues(dev);
}
EXPORT_SYMBOL(netif_tx_lock);
static void netif_unfreeze_queues(struct net_device *dev)
{
unsigned int i;
for (i = 0; i < dev->num_tx_queues; i++) {
struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
/* No need to grab the _xmit_lock here. If the
* queue is not stopped for another reason, we
* force a schedule.
*/
clear_bit(__QUEUE_STATE_FROZEN, &txq->state);
netif_schedule_queue(txq);
}
}
void netif_tx_unlock(struct net_device *dev)
{
netif_unfreeze_queues(dev);
spin_unlock(&dev->tx_global_lock);
}
EXPORT_SYMBOL(netif_tx_unlock);
net: sched: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Add pointer back to Qdisc. Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: Jiri Pirko <jiri@resnulli.us> Cc: "David S. Miller" <davem@davemloft.net> Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-10-17 03:29:17 +03:00
static void dev_watchdog(struct timer_list *t)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
{
net: sched: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Add pointer back to Qdisc. Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: Jiri Pirko <jiri@resnulli.us> Cc: "David S. Miller" <davem@davemloft.net> Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-10-17 03:29:17 +03:00
struct net_device *dev = from_timer(dev, t, watchdog_timer);
net: watchdog: add net device refcount tracker Add a netdevice_tracker inside struct net_device, to track the self reference when a device has an active watchdog timer. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-12-07 04:30:30 +03:00
bool release = true;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
net: no longer stop all TX queues in dev_watchdog() There is no reason for stopping all TX queues from dev_watchdog() Not only this stops feeding the NIC, it also migrates all qdiscs to be serviced on the cpu calling netif_tx_unlock(), causing a potential latency artifact. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-11-17 06:29:24 +03:00
spin_lock(&dev->tx_global_lock);
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
if (!qdisc_tx_is_noop(dev)) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
if (netif_device_present(dev) &&
netif_running(dev) &&
netif_carrier_ok(dev)) {
net: tx scalability works : trans_start struct net_device trans_start field is a hot spot on SMP and high performance devices, particularly multi queues ones, because every transmitter dirties it. Is main use is tx watchdog and bonding alive checks. But as most devices dont use NETIF_F_LLTX, we have to lock a netdev_queue before calling their ndo_start_xmit(). So it makes sense to move trans_start from net_device to netdev_queue. Its update will occur on a already present (and in exclusive state) cache line, for free. We can do this transition smoothly. An old driver continue to update dev->trans_start, while an updated one updates txq->trans_start. Further patches could also put tx_bytes/tx_packets counters in netdev_queue to avoid dirtying dev->stats (vlan device comes to mind) Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-05-18 07:55:16 +04:00
int some_queue_timedout = 0;
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
unsigned int i;
net: tx scalability works : trans_start struct net_device trans_start field is a hot spot on SMP and high performance devices, particularly multi queues ones, because every transmitter dirties it. Is main use is tx watchdog and bonding alive checks. But as most devices dont use NETIF_F_LLTX, we have to lock a netdev_queue before calling their ndo_start_xmit(). So it makes sense to move trans_start from net_device to netdev_queue. Its update will occur on a already present (and in exclusive state) cache line, for free. We can do this transition smoothly. An old driver continue to update dev->trans_start, while an updated one updates txq->trans_start. Further patches could also put tx_bytes/tx_packets counters in netdev_queue to avoid dirtying dev->stats (vlan device comes to mind) Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-05-18 07:55:16 +04:00
unsigned long trans_start;
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
for (i = 0; i < dev->num_tx_queues; i++) {
struct netdev_queue *txq;
txq = netdev_get_tx_queue(dev, i);
net: annotate accesses to queue->trans_start In following patches, dev_watchdog() will no longer stop all queues. It will read queue->trans_start locklessly. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-11-17 06:29:22 +03:00
trans_start = READ_ONCE(txq->trans_start);
net: Add queue state xoff flag for stack Create separate queue state flags so that either the stack or drivers can turn on XOFF. Added a set of functions used in the stack to determine if a queue is really stopped (either by stack or driver) Signed-off-by: Tom Herbert <therbert@google.com> Acked-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-11-28 20:32:44 +04:00
if (netif_xmit_stopped(txq) &&
net: tx scalability works : trans_start struct net_device trans_start field is a hot spot on SMP and high performance devices, particularly multi queues ones, because every transmitter dirties it. Is main use is tx watchdog and bonding alive checks. But as most devices dont use NETIF_F_LLTX, we have to lock a netdev_queue before calling their ndo_start_xmit(). So it makes sense to move trans_start from net_device to netdev_queue. Its update will occur on a already present (and in exclusive state) cache line, for free. We can do this transition smoothly. An old driver continue to update dev->trans_start, while an updated one updates txq->trans_start. Further patches could also put tx_bytes/tx_packets counters in netdev_queue to avoid dirtying dev->stats (vlan device comes to mind) Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-05-18 07:55:16 +04:00
time_after(jiffies, (trans_start +
dev->watchdog_timeo))) {
some_queue_timedout = 1;
net: use an atomic_long_t for queue->trans_timeout tx_timeout_show() assumed dev_watchdog() would stop all the queues, to fetch queue->trans_timeout under protection of the queue->_xmit_lock. As we want to no longer disrupt transmits, we use an atomic_long_t instead. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: david decotigny <david.decotigny@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-11-17 06:29:21 +03:00
atomic_long_inc(&txq->trans_timeout);
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
break;
}
}
[PKT_SCHED]: Potential jiffy wrap bug in dev_watchdog(). There is a potential jiffy wraparound bug in the transmit watchdog that is easily avoided by using time_after(). Signed-off-by: Stephen Hemminger <shemminger@osdl.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2006-05-17 02:02:12 +04:00
net: no longer stop all TX queues in dev_watchdog() There is no reason for stopping all TX queues from dev_watchdog() Not only this stops feeding the NIC, it also migrates all qdiscs to be serviced on the cpu calling netif_tx_unlock(), causing a potential latency artifact. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-11-17 06:29:24 +03:00
if (unlikely(some_queue_timedout)) {
net: add a generic tracepoint for TX queue timeout Although devlink health report does a nice job on reporting TX timeout and other NIC errors, unfortunately it requires drivers to support it but currently only mlx5 has implemented it. Before other drivers could catch up, it is useful to have a generic tracepoint to monitor this kind of TX timeout. We have been suffering TX timeout with different drivers, we plan to start to monitor it with rasdaemon which just needs a new tracepoint. Sample output: ksoftirqd/1-16 [001] ..s2 144.043173: net_dev_xmit_timeout: dev=ens3 driver=e1000 queue=0 Cc: Eran Ben Elisha <eranbe@mellanox.com> Cc: Jiri Pirko <jiri@mellanox.com> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Reviewed-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-05-02 05:56:59 +03:00
trace_net_dev_xmit_timeout(dev, i);
net: tx scalability works : trans_start struct net_device trans_start field is a hot spot on SMP and high performance devices, particularly multi queues ones, because every transmitter dirties it. Is main use is tx watchdog and bonding alive checks. But as most devices dont use NETIF_F_LLTX, we have to lock a netdev_queue before calling their ndo_start_xmit(). So it makes sense to move trans_start from net_device to netdev_queue. Its update will occur on a already present (and in exclusive state) cache line, for free. We can do this transition smoothly. An old driver continue to update dev->trans_start, while an updated one updates txq->trans_start. Further patches could also put tx_bytes/tx_packets counters in netdev_queue to avoid dirtying dev->stats (vlan device comes to mind) Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-05-18 07:55:16 +04:00
WARN_ONCE(1, KERN_INFO "NETDEV WATCHDOG: %s (%s): transmit queue %u timed out\n",
net: Rework netdev_drivername() to avoid warning. This interface uses a temporary buffer, but for no real reason. And now can generate warnings like: net/sched/sch_generic.c: In function dev_watchdog net/sched/sch_generic.c:254:10: warning: unused variable drivername Just return driver->name directly or "". Reported-by: Connor Hansen <cmdkhh@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-06-07 03:41:33 +04:00
dev->name, netdev_drivername(dev), i);
net: no longer stop all TX queues in dev_watchdog() There is no reason for stopping all TX queues from dev_watchdog() Not only this stops feeding the NIC, it also migrates all qdiscs to be serviced on the cpu calling netif_tx_unlock(), causing a potential latency artifact. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-11-17 06:29:24 +03:00
netif_freeze_queues(dev);
netdev: pass the stuck queue to the timeout handler This allows incrementing the correct timeout statistic without any mess. Down the road, devices can learn to reset just the specific queue. The patch was generated with the following script: use strict; use warnings; our $^I = '.bak'; my @work = ( ["arch/m68k/emu/nfeth.c", "nfeth_tx_timeout"], ["arch/um/drivers/net_kern.c", "uml_net_tx_timeout"], ["arch/um/drivers/vector_kern.c", "vector_net_tx_timeout"], ["arch/xtensa/platforms/iss/network.c", "iss_net_tx_timeout"], ["drivers/char/pcmcia/synclink_cs.c", "hdlcdev_tx_timeout"], ["drivers/infiniband/ulp/ipoib/ipoib_main.c", "ipoib_timeout"], ["drivers/infiniband/ulp/ipoib/ipoib_main.c", "ipoib_timeout"], ["drivers/message/fusion/mptlan.c", "mpt_lan_tx_timeout"], ["drivers/misc/sgi-xp/xpnet.c", "xpnet_dev_tx_timeout"], ["drivers/net/appletalk/cops.c", "cops_timeout"], ["drivers/net/arcnet/arcdevice.h", "arcnet_timeout"], ["drivers/net/arcnet/arcnet.c", "arcnet_timeout"], ["drivers/net/arcnet/com20020.c", "arcnet_timeout"], ["drivers/net/ethernet/3com/3c509.c", "el3_tx_timeout"], ["drivers/net/ethernet/3com/3c515.c", "corkscrew_timeout"], ["drivers/net/ethernet/3com/3c574_cs.c", "el3_tx_timeout"], ["drivers/net/ethernet/3com/3c589_cs.c", "el3_tx_timeout"], ["drivers/net/ethernet/3com/3c59x.c", "vortex_tx_timeout"], ["drivers/net/ethernet/3com/3c59x.c", "vortex_tx_timeout"], ["drivers/net/ethernet/3com/typhoon.c", "typhoon_tx_timeout"], ["drivers/net/ethernet/8390/8390.h", "ei_tx_timeout"], ["drivers/net/ethernet/8390/8390.h", "eip_tx_timeout"], ["drivers/net/ethernet/8390/8390.c", "ei_tx_timeout"], ["drivers/net/ethernet/8390/8390p.c", "eip_tx_timeout"], ["drivers/net/ethernet/8390/ax88796.c", "ax_ei_tx_timeout"], ["drivers/net/ethernet/8390/axnet_cs.c", "axnet_tx_timeout"], ["drivers/net/ethernet/8390/etherh.c", "__ei_tx_timeout"], ["drivers/net/ethernet/8390/hydra.c", "__ei_tx_timeout"], ["drivers/net/ethernet/8390/mac8390.c", "__ei_tx_timeout"], ["drivers/net/ethernet/8390/mcf8390.c", "__ei_tx_timeout"], ["drivers/net/ethernet/8390/lib8390.c", "__ei_tx_timeout"], ["drivers/net/ethernet/8390/ne2k-pci.c", "ei_tx_timeout"], ["drivers/net/ethernet/8390/pcnet_cs.c", "ei_tx_timeout"], ["drivers/net/ethernet/8390/smc-ultra.c", "ei_tx_timeout"], ["drivers/net/ethernet/8390/wd.c", "ei_tx_timeout"], ["drivers/net/ethernet/8390/zorro8390.c", "__ei_tx_timeout"], ["drivers/net/ethernet/adaptec/starfire.c", "tx_timeout"], ["drivers/net/ethernet/agere/et131x.c", "et131x_tx_timeout"], ["drivers/net/ethernet/allwinner/sun4i-emac.c", "emac_timeout"], ["drivers/net/ethernet/alteon/acenic.c", "ace_watchdog"], ["drivers/net/ethernet/amazon/ena/ena_netdev.c", "ena_tx_timeout"], ["drivers/net/ethernet/amd/7990.h", "lance_tx_timeout"], ["drivers/net/ethernet/amd/7990.c", "lance_tx_timeout"], ["drivers/net/ethernet/amd/a2065.c", "lance_tx_timeout"], ["drivers/net/ethernet/amd/am79c961a.c", "am79c961_timeout"], ["drivers/net/ethernet/amd/amd8111e.c", "amd8111e_tx_timeout"], ["drivers/net/ethernet/amd/ariadne.c", "ariadne_tx_timeout"], ["drivers/net/ethernet/amd/atarilance.c", "lance_tx_timeout"], ["drivers/net/ethernet/amd/au1000_eth.c", "au1000_tx_timeout"], ["drivers/net/ethernet/amd/declance.c", "lance_tx_timeout"], ["drivers/net/ethernet/amd/lance.c", "lance_tx_timeout"], ["drivers/net/ethernet/amd/mvme147.c", "lance_tx_timeout"], ["drivers/net/ethernet/amd/ni65.c", "ni65_timeout"], ["drivers/net/ethernet/amd/nmclan_cs.c", "mace_tx_timeout"], ["drivers/net/ethernet/amd/pcnet32.c", "pcnet32_tx_timeout"], ["drivers/net/ethernet/amd/sunlance.c", "lance_tx_timeout"], ["drivers/net/ethernet/amd/xgbe/xgbe-drv.c", "xgbe_tx_timeout"], ["drivers/net/ethernet/apm/xgene-v2/main.c", "xge_timeout"], ["drivers/net/ethernet/apm/xgene/xgene_enet_main.c", "xgene_enet_timeout"], ["drivers/net/ethernet/apple/macmace.c", "mace_tx_timeout"], ["drivers/net/ethernet/atheros/ag71xx.c", "ag71xx_tx_timeout"], ["drivers/net/ethernet/atheros/alx/main.c", "alx_tx_timeout"], ["drivers/net/ethernet/atheros/atl1c/atl1c_main.c", "atl1c_tx_timeout"], ["drivers/net/ethernet/atheros/atl1e/atl1e_main.c", "atl1e_tx_timeout"], ["drivers/net/ethernet/atheros/atlx/atl.c", "atlx_tx_timeout"], ["drivers/net/ethernet/atheros/atlx/atl1.c", "atlx_tx_timeout"], ["drivers/net/ethernet/atheros/atlx/atl2.c", "atl2_tx_timeout"], ["drivers/net/ethernet/broadcom/b44.c", "b44_tx_timeout"], ["drivers/net/ethernet/broadcom/bcmsysport.c", "bcm_sysport_tx_timeout"], ["drivers/net/ethernet/broadcom/bnx2.c", "bnx2_tx_timeout"], ["drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h", "bnx2x_tx_timeout"], ["drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c", "bnx2x_tx_timeout"], ["drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c", "bnx2x_tx_timeout"], ["drivers/net/ethernet/broadcom/bnxt/bnxt.c", "bnxt_tx_timeout"], ["drivers/net/ethernet/broadcom/genet/bcmgenet.c", "bcmgenet_timeout"], ["drivers/net/ethernet/broadcom/sb1250-mac.c", "sbmac_tx_timeout"], ["drivers/net/ethernet/broadcom/tg3.c", "tg3_tx_timeout"], ["drivers/net/ethernet/calxeda/xgmac.c", "xgmac_tx_timeout"], ["drivers/net/ethernet/cavium/liquidio/lio_main.c", "liquidio_tx_timeout"], ["drivers/net/ethernet/cavium/liquidio/lio_vf_main.c", "liquidio_tx_timeout"], ["drivers/net/ethernet/cavium/liquidio/lio_vf_rep.c", "lio_vf_rep_tx_timeout"], ["drivers/net/ethernet/cavium/thunder/nicvf_main.c", "nicvf_tx_timeout"], ["drivers/net/ethernet/cirrus/cs89x0.c", "net_timeout"], ["drivers/net/ethernet/cisco/enic/enic_main.c", "enic_tx_timeout"], ["drivers/net/ethernet/cisco/enic/enic_main.c", "enic_tx_timeout"], ["drivers/net/ethernet/cortina/gemini.c", "gmac_tx_timeout"], ["drivers/net/ethernet/davicom/dm9000.c", "dm9000_timeout"], ["drivers/net/ethernet/dec/tulip/de2104x.c", "de_tx_timeout"], ["drivers/net/ethernet/dec/tulip/tulip_core.c", "tulip_tx_timeout"], ["drivers/net/ethernet/dec/tulip/winbond-840.c", "tx_timeout"], ["drivers/net/ethernet/dlink/dl2k.c", "rio_tx_timeout"], ["drivers/net/ethernet/dlink/sundance.c", "tx_timeout"], ["drivers/net/ethernet/emulex/benet/be_main.c", "be_tx_timeout"], ["drivers/net/ethernet/ethoc.c", "ethoc_tx_timeout"], ["drivers/net/ethernet/faraday/ftgmac100.c", "ftgmac100_tx_timeout"], ["drivers/net/ethernet/fealnx.c", "fealnx_tx_timeout"], ["drivers/net/ethernet/freescale/dpaa/dpaa_eth.c", "dpaa_tx_timeout"], ["drivers/net/ethernet/freescale/fec_main.c", "fec_timeout"], ["drivers/net/ethernet/freescale/fec_mpc52xx.c", "mpc52xx_fec_tx_timeout"], ["drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c", "fs_timeout"], ["drivers/net/ethernet/freescale/gianfar.c", "gfar_timeout"], ["drivers/net/ethernet/freescale/ucc_geth.c", "ucc_geth_timeout"], ["drivers/net/ethernet/fujitsu/fmvj18x_cs.c", "fjn_tx_timeout"], ["drivers/net/ethernet/google/gve/gve_main.c", "gve_tx_timeout"], ["drivers/net/ethernet/hisilicon/hip04_eth.c", "hip04_timeout"], ["drivers/net/ethernet/hisilicon/hix5hd2_gmac.c", "hix5hd2_net_timeout"], ["drivers/net/ethernet/hisilicon/hns/hns_enet.c", "hns_nic_net_timeout"], ["drivers/net/ethernet/hisilicon/hns3/hns3_enet.c", "hns3_nic_net_timeout"], ["drivers/net/ethernet/huawei/hinic/hinic_main.c", "hinic_tx_timeout"], ["drivers/net/ethernet/i825xx/82596.c", "i596_tx_timeout"], ["drivers/net/ethernet/i825xx/ether1.c", "ether1_timeout"], ["drivers/net/ethernet/i825xx/lib82596.c", "i596_tx_timeout"], ["drivers/net/ethernet/i825xx/sun3_82586.c", "sun3_82586_timeout"], ["drivers/net/ethernet/ibm/ehea/ehea_main.c", "ehea_tx_watchdog"], ["drivers/net/ethernet/ibm/emac/core.c", "emac_tx_timeout"], ["drivers/net/ethernet/ibm/emac/core.c", "emac_tx_timeout"], ["drivers/net/ethernet/ibm/ibmvnic.c", "ibmvnic_tx_timeout"], ["drivers/net/ethernet/intel/e100.c", "e100_tx_timeout"], ["drivers/net/ethernet/intel/e1000/e1000_main.c", "e1000_tx_timeout"], ["drivers/net/ethernet/intel/e1000e/netdev.c", "e1000_tx_timeout"], ["drivers/net/ethernet/intel/fm10k/fm10k_netdev.c", "fm10k_tx_timeout"], ["drivers/net/ethernet/intel/i40e/i40e_main.c", "i40e_tx_timeout"], ["drivers/net/ethernet/intel/iavf/iavf_main.c", "iavf_tx_timeout"], ["drivers/net/ethernet/intel/ice/ice_main.c", "ice_tx_timeout"], ["drivers/net/ethernet/intel/ice/ice_main.c", "ice_tx_timeout"], ["drivers/net/ethernet/intel/igb/igb_main.c", "igb_tx_timeout"], ["drivers/net/ethernet/intel/igbvf/netdev.c", "igbvf_tx_timeout"], ["drivers/net/ethernet/intel/ixgb/ixgb_main.c", "ixgb_tx_timeout"], ["drivers/net/ethernet/intel/ixgbe/ixgbe_debugfs.c", "adapter->netdev->netdev_ops->ndo_tx_timeout(adapter->netdev);"], ["drivers/net/ethernet/intel/ixgbe/ixgbe_main.c", "ixgbe_tx_timeout"], ["drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c", "ixgbevf_tx_timeout"], ["drivers/net/ethernet/jme.c", "jme_tx_timeout"], ["drivers/net/ethernet/korina.c", "korina_tx_timeout"], ["drivers/net/ethernet/lantiq_etop.c", "ltq_etop_tx_timeout"], ["drivers/net/ethernet/marvell/mv643xx_eth.c", "mv643xx_eth_tx_timeout"], ["drivers/net/ethernet/marvell/pxa168_eth.c", "pxa168_eth_tx_timeout"], ["drivers/net/ethernet/marvell/skge.c", "skge_tx_timeout"], ["drivers/net/ethernet/marvell/sky2.c", "sky2_tx_timeout"], ["drivers/net/ethernet/marvell/sky2.c", "sky2_tx_timeout"], ["drivers/net/ethernet/mediatek/mtk_eth_soc.c", "mtk_tx_timeout"], ["drivers/net/ethernet/mellanox/mlx4/en_netdev.c", "mlx4_en_tx_timeout"], ["drivers/net/ethernet/mellanox/mlx4/en_netdev.c", "mlx4_en_tx_timeout"], ["drivers/net/ethernet/mellanox/mlx5/core/en_main.c", "mlx5e_tx_timeout"], ["drivers/net/ethernet/micrel/ks8842.c", "ks8842_tx_timeout"], ["drivers/net/ethernet/micrel/ksz884x.c", "netdev_tx_timeout"], ["drivers/net/ethernet/microchip/enc28j60.c", "enc28j60_tx_timeout"], ["drivers/net/ethernet/microchip/encx24j600.c", "encx24j600_tx_timeout"], ["drivers/net/ethernet/natsemi/sonic.h", "sonic_tx_timeout"], ["drivers/net/ethernet/natsemi/sonic.c", "sonic_tx_timeout"], ["drivers/net/ethernet/natsemi/jazzsonic.c", "sonic_tx_timeout"], ["drivers/net/ethernet/natsemi/macsonic.c", "sonic_tx_timeout"], ["drivers/net/ethernet/natsemi/natsemi.c", "ns_tx_timeout"], ["drivers/net/ethernet/natsemi/ns83820.c", "ns83820_tx_timeout"], ["drivers/net/ethernet/natsemi/xtsonic.c", "sonic_tx_timeout"], ["drivers/net/ethernet/neterion/s2io.h", "s2io_tx_watchdog"], ["drivers/net/ethernet/neterion/s2io.c", "s2io_tx_watchdog"], ["drivers/net/ethernet/neterion/vxge/vxge-main.c", "vxge_tx_watchdog"], ["drivers/net/ethernet/netronome/nfp/nfp_net_common.c", "nfp_net_tx_timeout"], ["drivers/net/ethernet/nvidia/forcedeth.c", "nv_tx_timeout"], ["drivers/net/ethernet/nvidia/forcedeth.c", "nv_tx_timeout"], ["drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c", "pch_gbe_tx_timeout"], ["drivers/net/ethernet/packetengines/hamachi.c", "hamachi_tx_timeout"], ["drivers/net/ethernet/packetengines/yellowfin.c", "yellowfin_tx_timeout"], ["drivers/net/ethernet/pensando/ionic/ionic_lif.c", "ionic_tx_timeout"], ["drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c", "netxen_tx_timeout"], ["drivers/net/ethernet/qlogic/qla3xxx.c", "ql3xxx_tx_timeout"], ["drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c", "qlcnic_tx_timeout"], ["drivers/net/ethernet/qualcomm/emac/emac.c", "emac_tx_timeout"], ["drivers/net/ethernet/qualcomm/qca_spi.c", "qcaspi_netdev_tx_timeout"], ["drivers/net/ethernet/qualcomm/qca_uart.c", "qcauart_netdev_tx_timeout"], ["drivers/net/ethernet/rdc/r6040.c", "r6040_tx_timeout"], ["drivers/net/ethernet/realtek/8139cp.c", "cp_tx_timeout"], ["drivers/net/ethernet/realtek/8139too.c", "rtl8139_tx_timeout"], ["drivers/net/ethernet/realtek/atp.c", "tx_timeout"], ["drivers/net/ethernet/realtek/r8169_main.c", "rtl8169_tx_timeout"], ["drivers/net/ethernet/renesas/ravb_main.c", "ravb_tx_timeout"], ["drivers/net/ethernet/renesas/sh_eth.c", "sh_eth_tx_timeout"], ["drivers/net/ethernet/renesas/sh_eth.c", "sh_eth_tx_timeout"], ["drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c", "sxgbe_tx_timeout"], ["drivers/net/ethernet/seeq/ether3.c", "ether3_timeout"], ["drivers/net/ethernet/seeq/sgiseeq.c", "timeout"], ["drivers/net/ethernet/sfc/efx.c", "efx_watchdog"], ["drivers/net/ethernet/sfc/falcon/efx.c", "ef4_watchdog"], ["drivers/net/ethernet/sgi/ioc3-eth.c", "ioc3_timeout"], ["drivers/net/ethernet/sgi/meth.c", "meth_tx_timeout"], ["drivers/net/ethernet/silan/sc92031.c", "sc92031_tx_timeout"], ["drivers/net/ethernet/sis/sis190.c", "sis190_tx_timeout"], ["drivers/net/ethernet/sis/sis900.c", "sis900_tx_timeout"], ["drivers/net/ethernet/smsc/epic100.c", "epic_tx_timeout"], ["drivers/net/ethernet/smsc/smc911x.c", "smc911x_timeout"], ["drivers/net/ethernet/smsc/smc9194.c", "smc_timeout"], ["drivers/net/ethernet/smsc/smc91c92_cs.c", "smc_tx_timeout"], ["drivers/net/ethernet/smsc/smc91x.c", "smc_timeout"], ["drivers/net/ethernet/stmicro/stmmac/stmmac_main.c", "stmmac_tx_timeout"], ["drivers/net/ethernet/sun/cassini.c", "cas_tx_timeout"], ["drivers/net/ethernet/sun/ldmvsw.c", "sunvnet_tx_timeout_common"], ["drivers/net/ethernet/sun/niu.c", "niu_tx_timeout"], ["drivers/net/ethernet/sun/sunbmac.c", "bigmac_tx_timeout"], ["drivers/net/ethernet/sun/sungem.c", "gem_tx_timeout"], ["drivers/net/ethernet/sun/sunhme.c", "happy_meal_tx_timeout"], ["drivers/net/ethernet/sun/sunqe.c", "qe_tx_timeout"], ["drivers/net/ethernet/sun/sunvnet.c", "sunvnet_tx_timeout_common"], ["drivers/net/ethernet/sun/sunvnet_common.c", "sunvnet_tx_timeout_common"], ["drivers/net/ethernet/sun/sunvnet_common.h", "sunvnet_tx_timeout_common"], ["drivers/net/ethernet/synopsys/dwc-xlgmac-net.c", "xlgmac_tx_timeout"], ["drivers/net/ethernet/ti/cpmac.c", "cpmac_tx_timeout"], ["drivers/net/ethernet/ti/cpsw.c", "cpsw_ndo_tx_timeout"], ["drivers/net/ethernet/ti/cpsw_priv.c", "cpsw_ndo_tx_timeout"], ["drivers/net/ethernet/ti/cpsw_priv.h", "cpsw_ndo_tx_timeout"], ["drivers/net/ethernet/ti/davinci_emac.c", "emac_dev_tx_timeout"], ["drivers/net/ethernet/ti/netcp_core.c", "netcp_ndo_tx_timeout"], ["drivers/net/ethernet/ti/tlan.c", "tlan_tx_timeout"], ["drivers/net/ethernet/toshiba/ps3_gelic_net.h", "gelic_net_tx_timeout"], ["drivers/net/ethernet/toshiba/ps3_gelic_net.c", "gelic_net_tx_timeout"], ["drivers/net/ethernet/toshiba/ps3_gelic_wireless.c", "gelic_net_tx_timeout"], ["drivers/net/ethernet/toshiba/spider_net.c", "spider_net_tx_timeout"], ["drivers/net/ethernet/toshiba/tc35815.c", "tc35815_tx_timeout"], ["drivers/net/ethernet/via/via-rhine.c", "rhine_tx_timeout"], ["drivers/net/ethernet/wiznet/w5100.c", "w5100_tx_timeout"], ["drivers/net/ethernet/wiznet/w5300.c", "w5300_tx_timeout"], ["drivers/net/ethernet/xilinx/xilinx_emaclite.c", "xemaclite_tx_timeout"], ["drivers/net/ethernet/xircom/xirc2ps_cs.c", "xirc_tx_timeout"], ["drivers/net/fjes/fjes_main.c", "fjes_tx_retry"], ["drivers/net/slip/slip.c", "sl_tx_timeout"], ["include/linux/usb/usbnet.h", "usbnet_tx_timeout"], ["drivers/net/usb/aqc111.c", "usbnet_tx_timeout"], ["drivers/net/usb/asix_devices.c", "usbnet_tx_timeout"], ["drivers/net/usb/asix_devices.c", "usbnet_tx_timeout"], ["drivers/net/usb/asix_devices.c", "usbnet_tx_timeout"], ["drivers/net/usb/ax88172a.c", "usbnet_tx_timeout"], ["drivers/net/usb/ax88179_178a.c", "usbnet_tx_timeout"], ["drivers/net/usb/catc.c", "catc_tx_timeout"], ["drivers/net/usb/cdc_mbim.c", "usbnet_tx_timeout"], ["drivers/net/usb/cdc_ncm.c", "usbnet_tx_timeout"], ["drivers/net/usb/dm9601.c", "usbnet_tx_timeout"], ["drivers/net/usb/hso.c", "hso_net_tx_timeout"], ["drivers/net/usb/int51x1.c", "usbnet_tx_timeout"], ["drivers/net/usb/ipheth.c", "ipheth_tx_timeout"], ["drivers/net/usb/kaweth.c", "kaweth_tx_timeout"], ["drivers/net/usb/lan78xx.c", "lan78xx_tx_timeout"], ["drivers/net/usb/mcs7830.c", "usbnet_tx_timeout"], ["drivers/net/usb/pegasus.c", "pegasus_tx_timeout"], ["drivers/net/usb/qmi_wwan.c", "usbnet_tx_timeout"], ["drivers/net/usb/r8152.c", "rtl8152_tx_timeout"], ["drivers/net/usb/rndis_host.c", "usbnet_tx_timeout"], ["drivers/net/usb/rtl8150.c", "rtl8150_tx_timeout"], ["drivers/net/usb/sierra_net.c", "usbnet_tx_timeout"], ["drivers/net/usb/smsc75xx.c", "usbnet_tx_timeout"], ["drivers/net/usb/smsc95xx.c", "usbnet_tx_timeout"], ["drivers/net/usb/sr9700.c", "usbnet_tx_timeout"], ["drivers/net/usb/sr9800.c", "usbnet_tx_timeout"], ["drivers/net/usb/usbnet.c", "usbnet_tx_timeout"], ["drivers/net/vmxnet3/vmxnet3_drv.c", "vmxnet3_tx_timeout"], ["drivers/net/wan/cosa.c", "cosa_net_timeout"], ["drivers/net/wan/farsync.c", "fst_tx_timeout"], ["drivers/net/wan/fsl_ucc_hdlc.c", "uhdlc_tx_timeout"], ["drivers/net/wan/lmc/lmc_main.c", "lmc_driver_timeout"], ["drivers/net/wan/x25_asy.c", "x25_asy_timeout"], ["drivers/net/wimax/i2400m/netdev.c", "i2400m_tx_timeout"], ["drivers/net/wireless/intel/ipw2x00/ipw2100.c", "ipw2100_tx_timeout"], ["drivers/net/wireless/intersil/hostap/hostap_main.c", "prism2_tx_timeout"], ["drivers/net/wireless/intersil/hostap/hostap_main.c", "prism2_tx_timeout"], ["drivers/net/wireless/intersil/hostap/hostap_main.c", "prism2_tx_timeout"], ["drivers/net/wireless/intersil/orinoco/main.c", "orinoco_tx_timeout"], ["drivers/net/wireless/intersil/orinoco/orinoco_usb.c", "orinoco_tx_timeout"], ["drivers/net/wireless/intersil/orinoco/orinoco.h", "orinoco_tx_timeout"], ["drivers/net/wireless/intersil/prism54/islpci_dev.c", "islpci_eth_tx_timeout"], ["drivers/net/wireless/intersil/prism54/islpci_eth.c", "islpci_eth_tx_timeout"], ["drivers/net/wireless/intersil/prism54/islpci_eth.h", "islpci_eth_tx_timeout"], ["drivers/net/wireless/marvell/mwifiex/main.c", "mwifiex_tx_timeout"], ["drivers/net/wireless/quantenna/qtnfmac/core.c", "qtnf_netdev_tx_timeout"], ["drivers/net/wireless/quantenna/qtnfmac/core.h", "qtnf_netdev_tx_timeout"], ["drivers/net/wireless/rndis_wlan.c", "usbnet_tx_timeout"], ["drivers/net/wireless/wl3501_cs.c", "wl3501_tx_timeout"], ["drivers/net/wireless/zydas/zd1201.c", "zd1201_tx_timeout"], ["drivers/s390/net/qeth_core.h", "qeth_tx_timeout"], ["drivers/s390/net/qeth_core_main.c", "qeth_tx_timeout"], ["drivers/s390/net/qeth_l2_main.c", "qeth_tx_timeout"], ["drivers/s390/net/qeth_l2_main.c", "qeth_tx_timeout"], ["drivers/s390/net/qeth_l3_main.c", "qeth_tx_timeout"], ["drivers/s390/net/qeth_l3_main.c", "qeth_tx_timeout"], ["drivers/staging/ks7010/ks_wlan_net.c", "ks_wlan_tx_timeout"], ["drivers/staging/qlge/qlge_main.c", "qlge_tx_timeout"], ["drivers/staging/rtl8192e/rtl8192e/rtl_core.c", "_rtl92e_tx_timeout"], ["drivers/staging/rtl8192u/r8192U_core.c", "tx_timeout"], ["drivers/staging/unisys/visornic/visornic_main.c", "visornic_xmit_timeout"], ["drivers/staging/wlan-ng/p80211netdev.c", "p80211knetdev_tx_timeout"], ["drivers/tty/n_gsm.c", "gsm_mux_net_tx_timeout"], ["drivers/tty/synclink.c", "hdlcdev_tx_timeout"], ["drivers/tty/synclink_gt.c", "hdlcdev_tx_timeout"], ["drivers/tty/synclinkmp.c", "hdlcdev_tx_timeout"], ["net/atm/lec.c", "lec_tx_timeout"], ["net/bluetooth/bnep/netdev.c", "bnep_net_timeout"] ); for my $p (@work) { my @pair = @$p; my $file = $pair[0]; my $func = $pair[1]; print STDERR $file , ": ", $func,"\n"; our @ARGV = ($file); while (<ARGV>) { if (m/($func\s*\(struct\s+net_device\s+\*[A-Za-z_]?[A-Za-z-0-9_]*)(\))/) { print STDERR "found $1+$2 in $file\n"; } if (s/($func\s*\(struct\s+net_device\s+\*[A-Za-z_]?[A-Za-z-0-9_]*)(\))/$1, unsigned int txqueue$2/) { print STDERR "$func found in $file\n"; } print; } } where the list of files and functions is simply from: git grep ndo_tx_timeout, with manual addition of headers in the rare cases where the function is from a header, then manually changing the few places which actually call ndo_tx_timeout. Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Acked-by: Heiner Kallweit <hkallweit1@gmail.com> Acked-by: Jakub Kicinski <jakub.kicinski@netronome.com> Acked-by: Shannon Nelson <snelson@pensando.io> Reviewed-by: Martin Habets <mhabets@solarflare.com> changes from v9: fixup a forward declaration changes from v9: more leftovers from v3 change changes from v8: fix up a missing direct call to timeout rebased on net-next changes from v7: fixup leftovers from v3 change changes from v6: fix typo in rtl driver changes from v5: add missing files (allow any net device argument name) changes from v4: add a missing driver header changes from v3: change queue # to unsigned Changes from v2: added headers Changes from v1: Fix errors found by kbuild: generalize the pattern a bit, to pick up a couple of instances missed by the previous version. Signed-off-by: David S. Miller <davem@davemloft.net>
2019-12-10 17:23:51 +03:00
dev->netdev_ops->ndo_tx_timeout(dev, i);
net: no longer stop all TX queues in dev_watchdog() There is no reason for stopping all TX queues from dev_watchdog() Not only this stops feeding the NIC, it also migrates all qdiscs to be serviced on the cpu calling netif_tx_unlock(), causing a potential latency artifact. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-11-17 06:29:24 +03:00
netif_unfreeze_queues(dev);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
}
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
if (!mod_timer(&dev->watchdog_timer,
round_jiffies(jiffies +
dev->watchdog_timeo)))
net: watchdog: add net device refcount tracker Add a netdevice_tracker inside struct net_device, to track the self reference when a device has an active watchdog timer. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-12-07 04:30:30 +03:00
release = false;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
}
}
net: no longer stop all TX queues in dev_watchdog() There is no reason for stopping all TX queues from dev_watchdog() Not only this stops feeding the NIC, it also migrates all qdiscs to be serviced on the cpu calling netif_tx_unlock(), causing a potential latency artifact. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-11-17 06:29:24 +03:00
spin_unlock(&dev->tx_global_lock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
net: watchdog: add net device refcount tracker Add a netdevice_tracker inside struct net_device, to track the self reference when a device has an active watchdog timer. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-12-07 04:30:30 +03:00
if (release)
dev_put_track(dev, &dev->watchdog_dev_tracker);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
}
void __netdev_watchdog_up(struct net_device *dev)
{
netdev: network device operations infrastructure This patch changes the network device internal API to move adminstrative operations out of the network device structure and into a separate structure. This patch involves some hackery to maintain compatablity between the new and old model, so all 300+ drivers don't have to be changed at once. For drivers that aren't converted yet, the netdevice_ops virt function list still resides in the net_device structure. For old protocols, the new net_device_ops are copied out to the old net_device pointers. After the transistion is completed the nag message can be changed to an WARN_ON, and the compatiablity code can be made configurable. Some function pointers aren't moved: * destructor can't be in net_device_ops because it may need to be referenced after the module is unloaded. * neighbor setup is manipulated in a couple of places that need special consideration * hard_start_xmit is in the fast path for transmit. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-20 08:32:24 +03:00
if (dev->netdev_ops->ndo_tx_timeout) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
if (dev->watchdog_timeo <= 0)
dev->watchdog_timeo = 5*HZ;
[NET]: Make net watchdog timers 1 sec jiffy aligned. round_jiffies for net dev watchdog timer. Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-06-01 08:28:44 +04:00
if (!mod_timer(&dev->watchdog_timer,
round_jiffies(jiffies + dev->watchdog_timeo)))
net: watchdog: add net device refcount tracker Add a netdevice_tracker inside struct net_device, to track the self reference when a device has an active watchdog timer. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-12-07 04:30:30 +03:00
dev_hold_track(dev, &dev->watchdog_dev_tracker, GFP_ATOMIC);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
}
}
net: sched: export __netdev_watchdog_up() Since the quiesce/activate rework, __netdev_watchdog_up() is directly called in the ucc_geth driver. Unfortunately, this function is not available for modules and thus ucc_geth cannot be built as a module anymore. Fix it by exporting __netdev_watchdog_up(). Since the commit introducing the regression was backported to stable branches, this one should ideally be as well. Fixes: 79dde73cf9bc ("net/ethernet/freescale: rework quiesce/activate for ucc_geth") Signed-off-by: Valentin Longchamp <valentin@longchamp.me> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-09 23:11:54 +03:00
EXPORT_SYMBOL_GPL(__netdev_watchdog_up);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
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>
2006-06-09 23:20:56 +04:00
netif_tx_lock_bh(dev);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
if (del_timer(&dev->watchdog_timer))
net: watchdog: add net device refcount tracker Add a netdevice_tracker inside struct net_device, to track the self reference when a device has an active watchdog timer. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-12-07 04:30:30 +03:00
dev_put_track(dev, &dev->watchdog_dev_tracker);
[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>
2006-06-09 23:20:56 +04:00
netif_tx_unlock_bh(dev);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
}
[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 03:41:36 +04:00
/**
* netif_carrier_on - set carrier
* @dev: network device
*
Documentation: bring operstate documentation up-to-date Netlink has moved from bitmasks to group numbers long ago. Signed-off-by: Jouke Witteveen <j.witteveen@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-02-07 19:14:32 +03:00
* Device has detected acquisition of carrier.
[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 03:41:36 +04:00
*/
[NET]: Deinline netif_carrier_{on,off}(). # grep -r 'netif_carrier_o[nf]' linux-2.6.12 | wc -l 246 # size vmlinux.org vmlinux.carrier text data bss dec hex filename 4339634 1054414 259296 5653344 564360 vmlinux.org 4337710 1054414 259296 5651420 563bdc vmlinux.carrier And this ain't an allyesconfig kernel! Signed-off-by: David S. Miller <davem@davemloft.net>
2005-08-12 02:32:53 +04:00
void netif_carrier_on(struct net_device *dev)
{
[NET]: fix carrier-on bug? While looking at a net driver with the following construct, if (!netif_carrier_ok(dev)) netif_carrier_on(dev); it stuck me that the netif_carrier_ok() check was redundant, since netif_carrier_on() checks bit __LINK_STATE_NOCARRIER anyway. This is the same reason why netif_queue_stopped() need not be called prior to netif_wake_queue(). This is true, but there is however an unwanted side effect from assuming that netif_carrier_on() can be called multiple times: it touches the watchdog, regardless of pre-existing carrier state. The fix: move watchdog-up inside the bit-cleared code path. Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-18 10:26:43 +04:00
if (test_and_clear_bit(__LINK_STATE_NOCARRIER, &dev->state)) {
net: Do not fire linkwatch events until the device is registered. Several device drivers try to do things like netif_carrier_off() before register_netdev() is invoked. This is bogus, but too many drivers do this to fix them all up in one go. Reported-by: Folkert van Heusden <folkert@vanheusden.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-20 02:33:54 +03:00
if (dev->reg_state == NETREG_UNINITIALIZED)
return;
net: core: Expose number of link up/down transitions Expose the number of times the link has been going UP or DOWN, and update the "carrier_changes" counter to be the sum of these two events. While at it, also update the sysfs-class-net documentation to cover: carrier_changes (3.15), carrier_up_count (4.16) and carrier_down_count (4.16) Signed-off-by: David Decotigny <decot@googlers.com> [Florian: * rebase * add documentation * merge carrier_changes with up/down counters] Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-01-18 20:59:13 +03:00
atomic_inc(&dev->carrier_up_count);
[NET]: Deinline netif_carrier_{on,off}(). # grep -r 'netif_carrier_o[nf]' linux-2.6.12 | wc -l 246 # size vmlinux.org vmlinux.carrier text data bss dec hex filename 4339634 1054414 259296 5653344 564360 vmlinux.org 4337710 1054414 259296 5651420 563bdc vmlinux.carrier And this ain't an allyesconfig kernel! Signed-off-by: David S. Miller <davem@davemloft.net>
2005-08-12 02:32:53 +04:00
linkwatch_fire_event(dev);
[NET]: fix carrier-on bug? While looking at a net driver with the following construct, if (!netif_carrier_ok(dev)) netif_carrier_on(dev); it stuck me that the netif_carrier_ok() check was redundant, since netif_carrier_on() checks bit __LINK_STATE_NOCARRIER anyway. This is the same reason why netif_queue_stopped() need not be called prior to netif_wake_queue(). This is true, but there is however an unwanted side effect from assuming that netif_carrier_on() can be called multiple times: it touches the watchdog, regardless of pre-existing carrier state. The fix: move watchdog-up inside the bit-cleared code path. Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-18 10:26:43 +04:00
if (netif_running(dev))
__netdev_watchdog_up(dev);
}
[NET]: Deinline netif_carrier_{on,off}(). # grep -r 'netif_carrier_o[nf]' linux-2.6.12 | wc -l 246 # size vmlinux.org vmlinux.carrier text data bss dec hex filename 4339634 1054414 259296 5653344 564360 vmlinux.org 4337710 1054414 259296 5651420 563bdc vmlinux.carrier And this ain't an allyesconfig kernel! Signed-off-by: David S. Miller <davem@davemloft.net>
2005-08-12 02:32:53 +04:00
}
[NET_SCHED]: Move EXPORT_SYMBOL next to exported symbol Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-23 09:10:23 +03:00
EXPORT_SYMBOL(netif_carrier_on);
[NET]: Deinline netif_carrier_{on,off}(). # grep -r 'netif_carrier_o[nf]' linux-2.6.12 | wc -l 246 # size vmlinux.org vmlinux.carrier text data bss dec hex filename 4339634 1054414 259296 5653344 564360 vmlinux.org 4337710 1054414 259296 5651420 563bdc vmlinux.carrier And this ain't an allyesconfig kernel! Signed-off-by: David S. Miller <davem@davemloft.net>
2005-08-12 02:32:53 +04:00
[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 03:41:36 +04:00
/**
* netif_carrier_off - clear carrier
* @dev: network device
*
* Device has detected loss of carrier.
*/
[NET]: Deinline netif_carrier_{on,off}(). # grep -r 'netif_carrier_o[nf]' linux-2.6.12 | wc -l 246 # size vmlinux.org vmlinux.carrier text data bss dec hex filename 4339634 1054414 259296 5653344 564360 vmlinux.org 4337710 1054414 259296 5651420 563bdc vmlinux.carrier And this ain't an allyesconfig kernel! Signed-off-by: David S. Miller <davem@davemloft.net>
2005-08-12 02:32:53 +04:00
void netif_carrier_off(struct net_device *dev)
{
net: Do not fire linkwatch events until the device is registered. Several device drivers try to do things like netif_carrier_off() before register_netdev() is invoked. This is bogus, but too many drivers do this to fix them all up in one go. Reported-by: Folkert van Heusden <folkert@vanheusden.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-20 02:33:54 +03:00
if (!test_and_set_bit(__LINK_STATE_NOCARRIER, &dev->state)) {
if (dev->reg_state == NETREG_UNINITIALIZED)
return;
net: core: Expose number of link up/down transitions Expose the number of times the link has been going UP or DOWN, and update the "carrier_changes" counter to be the sum of these two events. While at it, also update the sysfs-class-net documentation to cover: carrier_changes (3.15), carrier_up_count (4.16) and carrier_down_count (4.16) Signed-off-by: David Decotigny <decot@googlers.com> [Florian: * rebase * add documentation * merge carrier_changes with up/down counters] Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-01-18 20:59:13 +03:00
atomic_inc(&dev->carrier_down_count);
[NET]: Deinline netif_carrier_{on,off}(). # grep -r 'netif_carrier_o[nf]' linux-2.6.12 | wc -l 246 # size vmlinux.org vmlinux.carrier text data bss dec hex filename 4339634 1054414 259296 5653344 564360 vmlinux.org 4337710 1054414 259296 5651420 563bdc vmlinux.carrier And this ain't an allyesconfig kernel! Signed-off-by: David S. Miller <davem@davemloft.net>
2005-08-12 02:32:53 +04:00
linkwatch_fire_event(dev);
net: Do not fire linkwatch events until the device is registered. Several device drivers try to do things like netif_carrier_off() before register_netdev() is invoked. This is bogus, but too many drivers do this to fix them all up in one go. Reported-by: Folkert van Heusden <folkert@vanheusden.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-20 02:33:54 +03:00
}
[NET]: Deinline netif_carrier_{on,off}(). # grep -r 'netif_carrier_o[nf]' linux-2.6.12 | wc -l 246 # size vmlinux.org vmlinux.carrier text data bss dec hex filename 4339634 1054414 259296 5653344 564360 vmlinux.org 4337710 1054414 259296 5651420 563bdc vmlinux.carrier And this ain't an allyesconfig kernel! Signed-off-by: David S. Miller <davem@davemloft.net>
2005-08-12 02:32:53 +04:00
}
[NET_SCHED]: Move EXPORT_SYMBOL next to exported symbol Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-23 09:10:23 +03:00
EXPORT_SYMBOL(netif_carrier_off);
[NET]: Deinline netif_carrier_{on,off}(). # grep -r 'netif_carrier_o[nf]' linux-2.6.12 | wc -l 246 # size vmlinux.org vmlinux.carrier text data bss dec hex filename 4339634 1054414 259296 5653344 564360 vmlinux.org 4337710 1054414 259296 5651420 563bdc vmlinux.carrier And this ain't an allyesconfig kernel! Signed-off-by: David S. Miller <davem@davemloft.net>
2005-08-12 02:32:53 +04:00
mlx5: count all link events mlx5 devices were observed generating MLX5_PORT_CHANGE_SUBTYPE_ACTIVE events without an intervening MLX5_PORT_CHANGE_SUBTYPE_DOWN. This breaks link flap detection based on Linux carrier state transition count as netif_carrier_on() does nothing if carrier is already on. Make sure we count such events. netif_carrier_event() increments the counters and fires the linkwatch events. The latter is not necessary for the use case but seems like the right thing to do. Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2021-05-19 20:18:25 +03:00
/**
* netif_carrier_event - report carrier state event
* @dev: network device
*
* Device has detected a carrier event but the carrier state wasn't changed.
* Use in drivers when querying carrier state asynchronously, to avoid missing
* events (link flaps) if link recovers before it's queried.
*/
void netif_carrier_event(struct net_device *dev)
{
if (dev->reg_state == NETREG_UNINITIALIZED)
return;
atomic_inc(&dev->carrier_up_count);
atomic_inc(&dev->carrier_down_count);
linkwatch_fire_event(dev);
}
EXPORT_SYMBOL_GPL(netif_carrier_event);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
/* "NOOP" scheduler: the best scheduler, recommended for all interfaces
under all circumstances. It is difficult to invent anything faster or
cheaper.
*/
net_sched: drop packets after root qdisc lock is released Qdisc performance suffers when packets are dropped at enqueue() time because drops (kfree_skb()) are done while qdisc lock is held, delaying a dequeue() draining the queue. Nominal throughput can be reduced by 50 % when this happens, at a time we would like the dequeue() to proceed as fast as possible. Even FQ is vulnerable to this problem, while one of FQ goals was to provide some flow isolation. This patch adds a 'struct sk_buff **to_free' parameter to all qdisc->enqueue(), and in qdisc_drop() helper. I measured a performance increase of up to 12 %, but this patch is a prereq so that future batches in enqueue() can fly. Signed-off-by: Eric Dumazet <edumazet@google.com> Acked-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-06-22 09:16:49 +03:00
static int noop_enqueue(struct sk_buff *skb, struct Qdisc *qdisc,
struct sk_buff **to_free)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
{
net_sched: drop packets after root qdisc lock is released Qdisc performance suffers when packets are dropped at enqueue() time because drops (kfree_skb()) are done while qdisc lock is held, delaying a dequeue() draining the queue. Nominal throughput can be reduced by 50 % when this happens, at a time we would like the dequeue() to proceed as fast as possible. Even FQ is vulnerable to this problem, while one of FQ goals was to provide some flow isolation. This patch adds a 'struct sk_buff **to_free' parameter to all qdisc->enqueue(), and in qdisc_drop() helper. I measured a performance increase of up to 12 %, but this patch is a prereq so that future batches in enqueue() can fly. Signed-off-by: Eric Dumazet <edumazet@google.com> Acked-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-06-22 09:16:49 +03:00
__qdisc_drop(skb, to_free);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
return NET_XMIT_CN;
}
net_sched: change "foo* bar" to "foo *bar" "foo* bar" or "foo * bar" should be "foo *bar". Signed-off-by: Yang Yingliang <yangyingliang@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-12-10 16:55:31 +04:00
static struct sk_buff *noop_dequeue(struct Qdisc *qdisc)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
{
return NULL;
}
[NET]: Move Qdisc_class_ops and Qdisc_ops in appropriate sections. Qdisc_class_ops are const, and Qdisc_ops are mostly read. Using "const" and "__read_mostly" qualifiers helps to reduce false sharing. Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-11-14 12:44:41 +03:00
struct Qdisc_ops noop_qdisc_ops __read_mostly = {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
.id = "noop",
.priv_size = 0,
.enqueue = noop_enqueue,
.dequeue = noop_dequeue,
pkt_sched: sch_generic: Add generic qdisc->ops->peek() implementation. With feedback from Patrick McHardy. Signed-off-by: Jarek Poplawski <jarkao2@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-10-31 10:45:27 +03:00
.peek = noop_dequeue,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
.owner = THIS_MODULE,
};
pkt_sched: Add and use qdisc_root() and qdisc_root_lock(). When code wants to lock the qdisc tree state, the logic operation it's doing is locking the top-level qdisc that sits of the root of the netdev_queue. Add qdisc_root_lock() to represent this and convert the easiest cases. In order for this to work out in all cases, we have to hook up the noop_qdisc to a dummy netdev_queue. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-16 12:42:40 +04:00
static struct netdev_queue noop_netdev_queue = {
net: Use RCU_POINTER_INITIALIZER() to init static variable This pointer is RCU protected, so proper primitives should be used. Signed-off-by: Zhang Yu <zhangyu31@baidu.com> Signed-off-by: Li RongQing <lirongqing@baidu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-02-25 05:43:06 +03:00
RCU_POINTER_INITIALIZER(qdisc, &noop_qdisc),
pkt_sched: sch_generic: Fix oops in sch_teql After these commands: # modprobe sch_teql # tc qdisc add dev eth0 root teql0 # tc qdisc del dev eth0 root we get an oops in teql_destroy() when spin_lock is taken from a null qdisc_sleeping pointer. It's because at the moment teql0 dev haven't been activated yet, and a qdisc_root_sleeping() is pointing to noop qdisc's netdev_queue with qdisc_sleeping uninitialized. This patch fixes this both for noop and noqueue netdev_queues to avoid similar problems in the future. Signed-off-by: Jarek Poplawski <jarkao2@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-10-20 10:37:47 +04:00
.qdisc_sleeping = &noop_qdisc,
pkt_sched: Add and use qdisc_root() and qdisc_root_lock(). When code wants to lock the qdisc tree state, the logic operation it's doing is locking the top-level qdisc that sits of the root of the netdev_queue. Add qdisc_root_lock() to represent this and convert the easiest cases. In order for this to work out in all cases, we have to hook up the noop_qdisc to a dummy netdev_queue. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-16 12:42:40 +04:00
};
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
struct Qdisc noop_qdisc = {
.enqueue = noop_enqueue,
.dequeue = noop_dequeue,
.flags = TCQ_F_BUILTIN,
[NET] SCHED: Fix whitespace errors. Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-02-09 17:25:16 +03:00
.ops = &noop_qdisc_ops,
pkt_sched: Kill netdev_queue lock. We can simply use the qdisc->q.lock for all of the qdisc tree synchronization. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:53:03 +04:00
.q.lock = __SPIN_LOCK_UNLOCKED(noop_qdisc.q.lock),
pkt_sched: Add and use qdisc_root() and qdisc_root_lock(). When code wants to lock the qdisc tree state, the logic operation it's doing is locking the top-level qdisc that sits of the root of the netdev_queue. Add qdisc_root_lock() to represent this and convert the easiest cases. In order for this to work out in all cases, we have to hook up the noop_qdisc to a dummy netdev_queue. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-16 12:42:40 +04:00
.dev_queue = &noop_netdev_queue,
net/sched: fix missing spinlock init Under network load, doing : tc qdisc del dev eth0 root triggers : [ 167.193087] BUG: spinlock bad magic on CPU#3, udpflood/4928 [ 167.193139] lock: c15bc324, .magic: 00000000, .owner: <none>/-1, .owner_cpu: -1 [ 167.193193] Pid: 4928, comm: udpflood Not tainted 2.6.36-rc7-11417-g215340c-dirty #323 [ 167.193245] Call Trace: [ 167.193292] [<c13abaa0>] ? printk+0x18/0x20 [ 167.193342] [<c11afb53>] spin_bug+0xa3/0xf0 [ 167.193389] [<c11afcdd>] do_raw_spin_lock+0x7d/0x160 [ 167.193440] [<c1313d4e>] ? __dev_xmit_skb+0x27e/0x2b0 [ 167.193496] [<c107382b>] ? trace_hardirqs_on+0xb/0x10 [ 167.193545] [<c13ae99a>] _raw_spin_lock+0x3a/0x40 [ 167.193593] [<c1313d4e>] ? __dev_xmit_skb+0x27e/0x2b0 [ 167.193641] [<c1313d4e>] __dev_xmit_skb+0x27e/0x2b0 commit 79640a4ca695 (add additional lock to qdisc to increase throughput) forgot to initialize noop_qdisc and noqueue_qdisc busylock Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-10-15 23:22:34 +04:00
.busylock = __SPIN_LOCK_UNLOCKED(noop_qdisc.busylock),
net: sched: avoid writing on noop_qdisc While noop_qdisc.gso_skb and noop_qdisc.skb_bad_txq are not used in other places, it seems not correct to overwrite their fields in dev_init_scheduler_queue(). noop_qdisc is essentially a shared and read-only object, even if it is not marked as const because of some implementation detail. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-10-10 01:20:50 +03:00
.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),
},
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
};
[NET_SCHED]: Move EXPORT_SYMBOL next to exported symbol Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-23 09:10:23 +03:00
EXPORT_SYMBOL(noop_qdisc);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
net: sched: sch: add extack for init callback This patch adds extack support for init callback to prepare per-qdisc specific changes for extack. Cc: David Ahern <dsahern@gmail.com> Acked-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: Alexander Aring <aring@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-20 20:35:13 +03:00
static int noqueue_init(struct Qdisc *qdisc, struct nlattr *opt,
struct netlink_ext_ack *extack)
net: sched: register noqueue qdisc This way users can attach noqueue just like any other qdisc using tc without having to mess with tx_queue_len first. Signed-off-by: Phil Sutter <phil@nwl.cc> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-27 22:21:38 +03:00
{
/* 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 = {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
.id = "noqueue",
.priv_size = 0,
net: sched: register noqueue qdisc This way users can attach noqueue just like any other qdisc using tc without having to mess with tx_queue_len first. Signed-off-by: Phil Sutter <phil@nwl.cc> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-27 22:21:38 +03:00
.init = noqueue_init,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
.enqueue = noop_enqueue,
.dequeue = noop_dequeue,
pkt_sched: sch_generic: Add generic qdisc->ops->peek() implementation. With feedback from Patrick McHardy. Signed-off-by: Jarek Poplawski <jarkao2@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-10-31 10:45:27 +03:00
.peek = noop_dequeue,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
.owner = THIS_MODULE,
};
net_sched: cleanups Cleanup net/sched code to current CodingStyle and practices. Reduce inline abuse Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-01-19 22:26:56 +03:00
static const u8 prio2band[TC_PRIO_MAX + 1] = {
1, 2, 2, 2, 1, 2, 0, 0 , 1, 1, 1, 1, 1, 1, 1, 1
};
Revert "pkt_sched: Make default qdisc nonshared-multiqueue safe." This reverts commit a0c80b80e0fb48129e4e9d6a9ede914f9ff1850d. After discussions with Jamal and Herbert on netdev, we should provide at least minimal prioritization at the qdisc level even in multiqueue situations. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-21 20:56:13 +04:00
/* 3-band FIFO queue: old style, but should be a bit faster than
generic prio+fifo combination.
*/
#define PFIFO_FAST_BANDS 3
Speed-up pfifo_fast lookup using a private bitmap Maintain a per-qdisc bitmap for pfifo_fast giving availability of skbs for each band. This allows faster lookup for a skb when there are no high priority skbs. Also, it helps in (rare) cases when there are no skbs on the list, where an immediate lookup is faster than iterating through the three bands. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-08-19 01:55:59 +04:00
/*
* Private data for a pfifo_fast scheduler containing:
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
* - rings for priority bands
Speed-up pfifo_fast lookup using a private bitmap Maintain a per-qdisc bitmap for pfifo_fast giving availability of skbs for each band. This allows faster lookup for a skb when there are no high priority skbs. Also, it helps in (rare) cases when there are no skbs on the list, where an immediate lookup is faster than iterating through the three bands. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-08-19 01:55:59 +04:00
*/
struct pfifo_fast_priv {
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
struct skb_array q[PFIFO_FAST_BANDS];
Speed-up pfifo_fast lookup using a private bitmap Maintain a per-qdisc bitmap for pfifo_fast giving availability of skbs for each band. This allows faster lookup for a skb when there are no high priority skbs. Also, it helps in (rare) cases when there are no skbs on the list, where an immediate lookup is faster than iterating through the three bands. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-08-19 01:55:59 +04:00
};
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
static inline struct skb_array *band2list(struct pfifo_fast_priv *priv,
int band)
Revert "pkt_sched: Make default qdisc nonshared-multiqueue safe." This reverts commit a0c80b80e0fb48129e4e9d6a9ede914f9ff1850d. After discussions with Jamal and Herbert on netdev, we should provide at least minimal prioritization at the qdisc level even in multiqueue situations. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-21 20:56:13 +04:00
{
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
return &priv->q[band];
Revert "pkt_sched: Make default qdisc nonshared-multiqueue safe." This reverts commit a0c80b80e0fb48129e4e9d6a9ede914f9ff1850d. After discussions with Jamal and Herbert on netdev, we should provide at least minimal prioritization at the qdisc level even in multiqueue situations. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-21 20:56:13 +04:00
}
net_sched: drop packets after root qdisc lock is released Qdisc performance suffers when packets are dropped at enqueue() time because drops (kfree_skb()) are done while qdisc lock is held, delaying a dequeue() draining the queue. Nominal throughput can be reduced by 50 % when this happens, at a time we would like the dequeue() to proceed as fast as possible. Even FQ is vulnerable to this problem, while one of FQ goals was to provide some flow isolation. This patch adds a 'struct sk_buff **to_free' parameter to all qdisc->enqueue(), and in qdisc_drop() helper. I measured a performance increase of up to 12 %, but this patch is a prereq so that future batches in enqueue() can fly. Signed-off-by: Eric Dumazet <edumazet@google.com> Acked-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-06-22 09:16:49 +03:00
static int pfifo_fast_enqueue(struct sk_buff *skb, struct Qdisc *qdisc,
struct sk_buff **to_free)
[PKT_SCHED]: Add and use prio2list() in the pfifo_fast qdisc prio2list() returns the relevant sk_buff_head for the band specified by the priority for a given skb. Signed-off-by: Thomas Graf <tgraf@suug.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2005-06-19 09:58:35 +04:00
{
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
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 04:53:00 +03:00
unsigned int pkt_len = qdisc_pkt_len(skb);
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
int err;
[PKT_SCHED]: Transform pfifo_fast to use generic queue management interface Gives pfifo_fast a byte based backlog. Signed-off-by: Thomas Graf <tgraf@suug.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2005-06-19 09:58:15 +04:00
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
err = skb_array_produce(q, skb);
net/sched: pfifo_fast: fix wrong dereference in pfifo_fast_enqueue Now that 'TCQ_F_CPUSTATS' bit can be cleared, depending on the value of 'TCQ_F_NOLOCK' bit in the parent qdisc, we can't assume anymore that per-cpu counters are there in the error path of skb_array_produce(). Otherwise, the following splat can be seen: Unable to handle kernel paging request at virtual address 0000600dea430008 Mem abort info: ESR = 0x96000005 Exception class = DABT (current EL), IL = 32 bits SET = 0, FnV = 0 EA = 0, S1PTW = 0 Data abort info: ISV = 0, ISS = 0x00000005 CM = 0, WnR = 0 user pgtable: 64k pages, 48-bit VAs, pgdp = 000000007b97530e [0000600dea430008] pgd=0000000000000000, pud=0000000000000000 Internal error: Oops: 96000005 [#1] SMP [...] pstate: 10000005 (nzcV daif -PAN -UAO) pc : pfifo_fast_enqueue+0x524/0x6e8 lr : pfifo_fast_enqueue+0x46c/0x6e8 sp : ffff800d39376fe0 x29: ffff800d39376fe0 x28: 1ffff001a07d1e40 x27: ffff800d03e8f188 x26: ffff800d03e8f200 x25: 0000000000000062 x24: ffff800d393772f0 x23: 0000000000000000 x22: 0000000000000403 x21: ffff800cca569a00 x20: ffff800d03e8ee00 x19: ffff800cca569a10 x18: 00000000000000bf x17: 0000000000000000 x16: 0000000000000000 x15: 0000000000000000 x14: ffff1001a726edd0 x13: 1fffe4000276a9a4 x12: 0000000000000000 x11: dfff200000000000 x10: ffff800d03e8f1a0 x9 : 0000000000000003 x8 : 0000000000000000 x7 : 00000000f1f1f1f1 x6 : ffff1001a726edea x5 : ffff800cca56a53c x4 : 1ffff001bf9a8003 x3 : 1ffff001bf9a8003 x2 : 1ffff001a07d1dcb x1 : 0000600dea430000 x0 : 0000600dea430008 Process ping (pid: 6067, stack limit = 0x00000000dc0aa557) Call trace: pfifo_fast_enqueue+0x524/0x6e8 htb_enqueue+0x660/0x10e0 [sch_htb] __dev_queue_xmit+0x123c/0x2de0 dev_queue_xmit+0x24/0x30 ip_finish_output2+0xc48/0x1720 ip_finish_output+0x548/0x9d8 ip_output+0x334/0x788 ip_local_out+0x90/0x138 ip_send_skb+0x44/0x1d0 ip_push_pending_frames+0x5c/0x78 raw_sendmsg+0xed8/0x28d0 inet_sendmsg+0xc4/0x5c0 sock_sendmsg+0xac/0x108 __sys_sendto+0x1ac/0x2a0 __arm64_sys_sendto+0xc4/0x138 el0_svc_handler+0x13c/0x298 el0_svc+0x8/0xc Code: f9402e80 d538d081 91002000 8b010000 (885f7c03) Fix this by testing the value of 'TCQ_F_CPUSTATS' bit in 'qdisc->flags', before dereferencing 'qdisc->cpu_qstats'. Fixes: 8a53e616de29 ("net: sched: when clearing NOLOCK, clear TCQ_F_CPUSTATS, too") CC: Paolo Abeni <pabeni@redhat.com> CC: Stefano Brivio <sbrivio@redhat.com> Reported-by: Li Shuang <shuali@redhat.com> Signed-off-by: Davide Caratti <dcaratti@redhat.com> Acked-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-08-28 00:18:53 +03:00
if (unlikely(err)) {
if (qdisc_is_percpu_stats(qdisc))
return qdisc_drop_cpu(skb, qdisc, to_free);
else
return qdisc_drop(skb, qdisc, to_free);
}
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
net: sched: when clearing NOLOCK, clear TCQ_F_CPUSTATS, too Since stats updating is always consistent with TCQ_F_CPUSTATS flag, we can disable it at qdisc creation time flipping such bit. In my experiments, if the NOLOCK flag is cleared, per CPU stats accounting does not give any measurable performance gain, but it waste some memory. Let's clear TCQ_F_CPUSTATS together with NOLOCK, when enslaving a NOLOCK qdisc to 'lock' one. Use stats update helper inside pfifo_fast, to cope correctly with TCQ_F_CPUSTATS flag change. As a side effect, q.qlen value for any child qdiscs is always consistent for all lock classfull qdiscs. Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-10 15:32:40 +03:00
qdisc_update_stats_at_enqueue(qdisc, pkt_len);
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
return NET_XMIT_SUCCESS;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
}
net_sched: cleanups Cleanup net/sched code to current CodingStyle and practices. Reduce inline abuse Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-01-19 22:26:56 +03:00
static struct sk_buff *pfifo_fast_dequeue(struct Qdisc *qdisc)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
{
Speed-up pfifo_fast lookup using a private bitmap Maintain a per-qdisc bitmap for pfifo_fast giving availability of skbs for each band. This allows faster lookup for a skb when there are no high priority skbs. Also, it helps in (rare) cases when there are no skbs on the list, where an immediate lookup is faster than iterating through the three bands. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-08-19 01:55:59 +04:00
struct pfifo_fast_priv *priv = qdisc_priv(qdisc);
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
struct sk_buff *skb = NULL;
net: sched: fix packet stuck problem for lockless qdisc Lockless qdisc has below concurrent problem: cpu0 cpu1 . . q->enqueue . . . qdisc_run_begin() . . . dequeue_skb() . . . sch_direct_xmit() . . . . q->enqueue . qdisc_run_begin() . return and do nothing . . qdisc_run_end() . cpu1 enqueue a skb without calling __qdisc_run() because cpu0 has not released the lock yet and spin_trylock() return false for cpu1 in qdisc_run_begin(), and cpu0 do not see the skb enqueued by cpu1 when calling dequeue_skb() because cpu1 may enqueue the skb after cpu0 calling dequeue_skb() and before cpu0 calling qdisc_run_end(). Lockless qdisc has below another concurrent problem when tx_action is involved: cpu0(serving tx_action) cpu1 cpu2 . . . . q->enqueue . . qdisc_run_begin() . . dequeue_skb() . . . q->enqueue . . . . sch_direct_xmit() . . . qdisc_run_begin() . . return and do nothing . . . clear __QDISC_STATE_SCHED . . qdisc_run_begin() . . return and do nothing . . . . . . qdisc_run_end() . This patch fixes the above data race by: 1. If the first spin_trylock() return false and STATE_MISSED is not set, set STATE_MISSED and retry another spin_trylock() in case other CPU may not see STATE_MISSED after it releases the lock. 2. reschedule if STATE_MISSED is set after the lock is released at the end of qdisc_run_end(). For tx_action case, STATE_MISSED is also set when cpu1 is at the end if qdisc_run_end(), so tx_action will be rescheduled again to dequeue the skb enqueued by cpu2. Clear STATE_MISSED before retrying a dequeuing when dequeuing returns NULL in order to reduce the overhead of the second spin_trylock() and __netif_schedule() calling. Also clear the STATE_MISSED before calling __netif_schedule() at the end of qdisc_run_end() to avoid doing another round of dequeuing in the pfifo_fast_dequeue(). The performance impact of this patch, tested using pktgen and dummy netdev with pfifo_fast qdisc attached: threads without+this_patch with+this_patch delta 1 2.61Mpps 2.60Mpps -0.3% 2 3.97Mpps 3.82Mpps -3.7% 4 5.62Mpps 5.59Mpps -0.5% 8 2.78Mpps 2.77Mpps -0.3% 16 2.22Mpps 2.22Mpps -0.0% Fixes: 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking") Acked-by: Jakub Kicinski <kuba@kernel.org> Tested-by: Juergen Gross <jgross@suse.com> Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-05-14 06:16:59 +03:00
bool need_retry = true;
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
int band;
sched: remove qdisc arg from __qdisc_dequeue_head Moves qdisc stat accouting to qdisc_dequeue_head. The only direct caller of the __qdisc_dequeue_head version open-codes this now. This allows us to later use __qdisc_dequeue_head as a replacement of __skb_dequeue() (which operates on sk_buff_head list). Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-09-18 01:57:32 +03:00
net: sched: fix packet stuck problem for lockless qdisc Lockless qdisc has below concurrent problem: cpu0 cpu1 . . q->enqueue . . . qdisc_run_begin() . . . dequeue_skb() . . . sch_direct_xmit() . . . . q->enqueue . qdisc_run_begin() . return and do nothing . . qdisc_run_end() . cpu1 enqueue a skb without calling __qdisc_run() because cpu0 has not released the lock yet and spin_trylock() return false for cpu1 in qdisc_run_begin(), and cpu0 do not see the skb enqueued by cpu1 when calling dequeue_skb() because cpu1 may enqueue the skb after cpu0 calling dequeue_skb() and before cpu0 calling qdisc_run_end(). Lockless qdisc has below another concurrent problem when tx_action is involved: cpu0(serving tx_action) cpu1 cpu2 . . . . q->enqueue . . qdisc_run_begin() . . dequeue_skb() . . . q->enqueue . . . . sch_direct_xmit() . . . qdisc_run_begin() . . return and do nothing . . . clear __QDISC_STATE_SCHED . . qdisc_run_begin() . . return and do nothing . . . . . . qdisc_run_end() . This patch fixes the above data race by: 1. If the first spin_trylock() return false and STATE_MISSED is not set, set STATE_MISSED and retry another spin_trylock() in case other CPU may not see STATE_MISSED after it releases the lock. 2. reschedule if STATE_MISSED is set after the lock is released at the end of qdisc_run_end(). For tx_action case, STATE_MISSED is also set when cpu1 is at the end if qdisc_run_end(), so tx_action will be rescheduled again to dequeue the skb enqueued by cpu2. Clear STATE_MISSED before retrying a dequeuing when dequeuing returns NULL in order to reduce the overhead of the second spin_trylock() and __netif_schedule() calling. Also clear the STATE_MISSED before calling __netif_schedule() at the end of qdisc_run_end() to avoid doing another round of dequeuing in the pfifo_fast_dequeue(). The performance impact of this patch, tested using pktgen and dummy netdev with pfifo_fast qdisc attached: threads without+this_patch with+this_patch delta 1 2.61Mpps 2.60Mpps -0.3% 2 3.97Mpps 3.82Mpps -3.7% 4 5.62Mpps 5.59Mpps -0.5% 8 2.78Mpps 2.77Mpps -0.3% 16 2.22Mpps 2.22Mpps -0.0% Fixes: 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking") Acked-by: Jakub Kicinski <kuba@kernel.org> Tested-by: Juergen Gross <jgross@suse.com> Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-05-14 06:16:59 +03:00
retry:
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
for (band = 0; band < PFIFO_FAST_BANDS && !skb; band++) {
struct skb_array *q = band2list(priv, band);
Speed-up pfifo_fast lookup using a private bitmap Maintain a per-qdisc bitmap for pfifo_fast giving availability of skbs for each band. This allows faster lookup for a skb when there are no high priority skbs. Also, it helps in (rare) cases when there are no skbs on the list, where an immediate lookup is faster than iterating through the three bands. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-08-19 01:55:59 +04:00
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
if (__skb_array_empty(q))
continue;
Speed-up pfifo_fast lookup using a private bitmap Maintain a per-qdisc bitmap for pfifo_fast giving availability of skbs for each band. This allows faster lookup for a skb when there are no high priority skbs. Also, it helps in (rare) cases when there are no skbs on the list, where an immediate lookup is faster than iterating through the three bands. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-08-19 01:55:59 +04:00
pfifo_fast: drop unneeded additional lock on dequeue After the previous patch, for NOLOCK qdiscs, q->seqlock is always held when the dequeue() is invoked, we can drop any additional locking to protect such operation. Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-05-15 17:24:37 +03:00
skb = __skb_array_consume(q);
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
}
if (likely(skb)) {
net: sched: when clearing NOLOCK, clear TCQ_F_CPUSTATS, too Since stats updating is always consistent with TCQ_F_CPUSTATS flag, we can disable it at qdisc creation time flipping such bit. In my experiments, if the NOLOCK flag is cleared, per CPU stats accounting does not give any measurable performance gain, but it waste some memory. Let's clear TCQ_F_CPUSTATS together with NOLOCK, when enslaving a NOLOCK qdisc to 'lock' one. Use stats update helper inside pfifo_fast, to cope correctly with TCQ_F_CPUSTATS flag change. As a side effect, q.qlen value for any child qdiscs is always consistent for all lock classfull qdiscs. Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-10 15:32:40 +03:00
qdisc_update_stats_at_dequeue(qdisc, skb);
net: sched: fix packet stuck problem for lockless qdisc Lockless qdisc has below concurrent problem: cpu0 cpu1 . . q->enqueue . . . qdisc_run_begin() . . . dequeue_skb() . . . sch_direct_xmit() . . . . q->enqueue . qdisc_run_begin() . return and do nothing . . qdisc_run_end() . cpu1 enqueue a skb without calling __qdisc_run() because cpu0 has not released the lock yet and spin_trylock() return false for cpu1 in qdisc_run_begin(), and cpu0 do not see the skb enqueued by cpu1 when calling dequeue_skb() because cpu1 may enqueue the skb after cpu0 calling dequeue_skb() and before cpu0 calling qdisc_run_end(). Lockless qdisc has below another concurrent problem when tx_action is involved: cpu0(serving tx_action) cpu1 cpu2 . . . . q->enqueue . . qdisc_run_begin() . . dequeue_skb() . . . q->enqueue . . . . sch_direct_xmit() . . . qdisc_run_begin() . . return and do nothing . . . clear __QDISC_STATE_SCHED . . qdisc_run_begin() . . return and do nothing . . . . . . qdisc_run_end() . This patch fixes the above data race by: 1. If the first spin_trylock() return false and STATE_MISSED is not set, set STATE_MISSED and retry another spin_trylock() in case other CPU may not see STATE_MISSED after it releases the lock. 2. reschedule if STATE_MISSED is set after the lock is released at the end of qdisc_run_end(). For tx_action case, STATE_MISSED is also set when cpu1 is at the end if qdisc_run_end(), so tx_action will be rescheduled again to dequeue the skb enqueued by cpu2. Clear STATE_MISSED before retrying a dequeuing when dequeuing returns NULL in order to reduce the overhead of the second spin_trylock() and __netif_schedule() calling. Also clear the STATE_MISSED before calling __netif_schedule() at the end of qdisc_run_end() to avoid doing another round of dequeuing in the pfifo_fast_dequeue(). The performance impact of this patch, tested using pktgen and dummy netdev with pfifo_fast qdisc attached: threads without+this_patch with+this_patch delta 1 2.61Mpps 2.60Mpps -0.3% 2 3.97Mpps 3.82Mpps -3.7% 4 5.62Mpps 5.59Mpps -0.5% 8 2.78Mpps 2.77Mpps -0.3% 16 2.22Mpps 2.22Mpps -0.0% Fixes: 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking") Acked-by: Jakub Kicinski <kuba@kernel.org> Tested-by: Juergen Gross <jgross@suse.com> Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-05-14 06:16:59 +03:00
} else if (need_retry &&
net: sched: implement TCQ_F_CAN_BYPASS for lockless qdisc Currently pfifo_fast has both TCQ_F_CAN_BYPASS and TCQ_F_NOLOCK flag set, but queue discipline by-pass does not work for lockless qdisc because skb is always enqueued to qdisc even when the qdisc is empty, see __dev_xmit_skb(). This patch calls sch_direct_xmit() to transmit the skb directly to the driver for empty lockless qdisc, which aviod enqueuing and dequeuing operation. As qdisc->empty is not reliable to indicate a empty qdisc because there is a time window between enqueuing and setting qdisc->empty. So we use the MISSED state added in commit a90c57f2cedd ("net: sched: fix packet stuck problem for lockless qdisc"), which indicate there is lock contention, suggesting that it is better not to do the qdisc bypass in order to avoid packet out of order problem. In order to make MISSED state reliable to indicate a empty qdisc, we need to ensure that testing and clearing of MISSED state is within the protection of qdisc->seqlock, only setting MISSED state can be done without the protection of qdisc->seqlock. A MISSED state testing is added without the protection of qdisc->seqlock to aviod doing unnecessary spin_trylock() for contention case. As the enqueuing is not within the protection of qdisc->seqlock, there is still a potential data race as mentioned by Jakub [1]: thread1 thread2 thread3 qdisc_run_begin() # true qdisc_run_begin(q) set(MISSED) pfifo_fast_dequeue clear(MISSED) # recheck the queue qdisc_run_end() enqueue skb1 qdisc empty # true qdisc_run_begin() # true sch_direct_xmit() # skb2 qdisc_run_begin() set(MISSED) When above happens, skb1 enqueued by thread2 is transmited after skb2 is transmited by thread3 because MISSED state setting and enqueuing is not under the qdisc->seqlock. If qdisc bypass is disabled, skb1 has better chance to be transmited quicker than skb2. This patch does not take care of the above data race, because we view this as similar as below: Even at the same time CPU1 and CPU2 write the skb to two socket which both heading to the same qdisc, there is no guarantee that which skb will hit the qdisc first, because there is a lot of factor like interrupt/softirq/cache miss/scheduling afffecting that. There are below cases that need special handling: 1. When MISSED state is cleared before another round of dequeuing in pfifo_fast_dequeue(), and __qdisc_run() might not be able to dequeue all skb in one round and call __netif_schedule(), which might result in a non-empty qdisc without MISSED set. In order to avoid this, the MISSED state is set for lockless qdisc and __netif_schedule() will be called at the end of qdisc_run_end. 2. The MISSED state also need to be set for lockless qdisc instead of calling __netif_schedule() directly when requeuing a skb for a similar reason. 3. For netdev queue stopped case, the MISSED case need clearing while the netdev queue is stopped, otherwise there may be unnecessary __netif_schedule() calling. So a new DRAINING state is added to indicate this case, which also indicate a non-empty qdisc. 4. As there is already netif_xmit_frozen_or_stopped() checking in dequeue_skb() and sch_direct_xmit(), which are both within the protection of qdisc->seqlock, but the same checking in __dev_xmit_skb() is without the protection, which might cause empty indication of a lockless qdisc to be not reliable. So remove the checking in __dev_xmit_skb(), and the checking in the protection of qdisc->seqlock seems enough to avoid the cpu consumption problem for netdev queue stopped case. 1. https://lkml.org/lkml/2021/5/29/215 Acked-by: Jakub Kicinski <kuba@kernel.org> Tested-by: Vladimir Oltean <vladimir.oltean@nxp.com> # flexcan Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-06-22 09:49:56 +03:00
READ_ONCE(qdisc->state) & QDISC_STATE_NON_EMPTY) {
net: sched: fix packet stuck problem for lockless qdisc Lockless qdisc has below concurrent problem: cpu0 cpu1 . . q->enqueue . . . qdisc_run_begin() . . . dequeue_skb() . . . sch_direct_xmit() . . . . q->enqueue . qdisc_run_begin() . return and do nothing . . qdisc_run_end() . cpu1 enqueue a skb without calling __qdisc_run() because cpu0 has not released the lock yet and spin_trylock() return false for cpu1 in qdisc_run_begin(), and cpu0 do not see the skb enqueued by cpu1 when calling dequeue_skb() because cpu1 may enqueue the skb after cpu0 calling dequeue_skb() and before cpu0 calling qdisc_run_end(). Lockless qdisc has below another concurrent problem when tx_action is involved: cpu0(serving tx_action) cpu1 cpu2 . . . . q->enqueue . . qdisc_run_begin() . . dequeue_skb() . . . q->enqueue . . . . sch_direct_xmit() . . . qdisc_run_begin() . . return and do nothing . . . clear __QDISC_STATE_SCHED . . qdisc_run_begin() . . return and do nothing . . . . . . qdisc_run_end() . This patch fixes the above data race by: 1. If the first spin_trylock() return false and STATE_MISSED is not set, set STATE_MISSED and retry another spin_trylock() in case other CPU may not see STATE_MISSED after it releases the lock. 2. reschedule if STATE_MISSED is set after the lock is released at the end of qdisc_run_end(). For tx_action case, STATE_MISSED is also set when cpu1 is at the end if qdisc_run_end(), so tx_action will be rescheduled again to dequeue the skb enqueued by cpu2. Clear STATE_MISSED before retrying a dequeuing when dequeuing returns NULL in order to reduce the overhead of the second spin_trylock() and __netif_schedule() calling. Also clear the STATE_MISSED before calling __netif_schedule() at the end of qdisc_run_end() to avoid doing another round of dequeuing in the pfifo_fast_dequeue(). The performance impact of this patch, tested using pktgen and dummy netdev with pfifo_fast qdisc attached: threads without+this_patch with+this_patch delta 1 2.61Mpps 2.60Mpps -0.3% 2 3.97Mpps 3.82Mpps -3.7% 4 5.62Mpps 5.59Mpps -0.5% 8 2.78Mpps 2.77Mpps -0.3% 16 2.22Mpps 2.22Mpps -0.0% Fixes: 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking") Acked-by: Jakub Kicinski <kuba@kernel.org> Tested-by: Juergen Gross <jgross@suse.com> Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-05-14 06:16:59 +03:00
/* Delay clearing the STATE_MISSED here to reduce
* the overhead of the second spin_trylock() in
* qdisc_run_begin() and __netif_schedule() calling
* in qdisc_run_end().
*/
clear_bit(__QDISC_STATE_MISSED, &qdisc->state);
net: sched: implement TCQ_F_CAN_BYPASS for lockless qdisc Currently pfifo_fast has both TCQ_F_CAN_BYPASS and TCQ_F_NOLOCK flag set, but queue discipline by-pass does not work for lockless qdisc because skb is always enqueued to qdisc even when the qdisc is empty, see __dev_xmit_skb(). This patch calls sch_direct_xmit() to transmit the skb directly to the driver for empty lockless qdisc, which aviod enqueuing and dequeuing operation. As qdisc->empty is not reliable to indicate a empty qdisc because there is a time window between enqueuing and setting qdisc->empty. So we use the MISSED state added in commit a90c57f2cedd ("net: sched: fix packet stuck problem for lockless qdisc"), which indicate there is lock contention, suggesting that it is better not to do the qdisc bypass in order to avoid packet out of order problem. In order to make MISSED state reliable to indicate a empty qdisc, we need to ensure that testing and clearing of MISSED state is within the protection of qdisc->seqlock, only setting MISSED state can be done without the protection of qdisc->seqlock. A MISSED state testing is added without the protection of qdisc->seqlock to aviod doing unnecessary spin_trylock() for contention case. As the enqueuing is not within the protection of qdisc->seqlock, there is still a potential data race as mentioned by Jakub [1]: thread1 thread2 thread3 qdisc_run_begin() # true qdisc_run_begin(q) set(MISSED) pfifo_fast_dequeue clear(MISSED) # recheck the queue qdisc_run_end() enqueue skb1 qdisc empty # true qdisc_run_begin() # true sch_direct_xmit() # skb2 qdisc_run_begin() set(MISSED) When above happens, skb1 enqueued by thread2 is transmited after skb2 is transmited by thread3 because MISSED state setting and enqueuing is not under the qdisc->seqlock. If qdisc bypass is disabled, skb1 has better chance to be transmited quicker than skb2. This patch does not take care of the above data race, because we view this as similar as below: Even at the same time CPU1 and CPU2 write the skb to two socket which both heading to the same qdisc, there is no guarantee that which skb will hit the qdisc first, because there is a lot of factor like interrupt/softirq/cache miss/scheduling afffecting that. There are below cases that need special handling: 1. When MISSED state is cleared before another round of dequeuing in pfifo_fast_dequeue(), and __qdisc_run() might not be able to dequeue all skb in one round and call __netif_schedule(), which might result in a non-empty qdisc without MISSED set. In order to avoid this, the MISSED state is set for lockless qdisc and __netif_schedule() will be called at the end of qdisc_run_end. 2. The MISSED state also need to be set for lockless qdisc instead of calling __netif_schedule() directly when requeuing a skb for a similar reason. 3. For netdev queue stopped case, the MISSED case need clearing while the netdev queue is stopped, otherwise there may be unnecessary __netif_schedule() calling. So a new DRAINING state is added to indicate this case, which also indicate a non-empty qdisc. 4. As there is already netif_xmit_frozen_or_stopped() checking in dequeue_skb() and sch_direct_xmit(), which are both within the protection of qdisc->seqlock, but the same checking in __dev_xmit_skb() is without the protection, which might cause empty indication of a lockless qdisc to be not reliable. So remove the checking in __dev_xmit_skb(), and the checking in the protection of qdisc->seqlock seems enough to avoid the cpu consumption problem for netdev queue stopped case. 1. https://lkml.org/lkml/2021/5/29/215 Acked-by: Jakub Kicinski <kuba@kernel.org> Tested-by: Vladimir Oltean <vladimir.oltean@nxp.com> # flexcan Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-06-22 09:49:56 +03:00
clear_bit(__QDISC_STATE_DRAINING, &qdisc->state);
net: sched: fix packet stuck problem for lockless qdisc Lockless qdisc has below concurrent problem: cpu0 cpu1 . . q->enqueue . . . qdisc_run_begin() . . . dequeue_skb() . . . sch_direct_xmit() . . . . q->enqueue . qdisc_run_begin() . return and do nothing . . qdisc_run_end() . cpu1 enqueue a skb without calling __qdisc_run() because cpu0 has not released the lock yet and spin_trylock() return false for cpu1 in qdisc_run_begin(), and cpu0 do not see the skb enqueued by cpu1 when calling dequeue_skb() because cpu1 may enqueue the skb after cpu0 calling dequeue_skb() and before cpu0 calling qdisc_run_end(). Lockless qdisc has below another concurrent problem when tx_action is involved: cpu0(serving tx_action) cpu1 cpu2 . . . . q->enqueue . . qdisc_run_begin() . . dequeue_skb() . . . q->enqueue . . . . sch_direct_xmit() . . . qdisc_run_begin() . . return and do nothing . . . clear __QDISC_STATE_SCHED . . qdisc_run_begin() . . return and do nothing . . . . . . qdisc_run_end() . This patch fixes the above data race by: 1. If the first spin_trylock() return false and STATE_MISSED is not set, set STATE_MISSED and retry another spin_trylock() in case other CPU may not see STATE_MISSED after it releases the lock. 2. reschedule if STATE_MISSED is set after the lock is released at the end of qdisc_run_end(). For tx_action case, STATE_MISSED is also set when cpu1 is at the end if qdisc_run_end(), so tx_action will be rescheduled again to dequeue the skb enqueued by cpu2. Clear STATE_MISSED before retrying a dequeuing when dequeuing returns NULL in order to reduce the overhead of the second spin_trylock() and __netif_schedule() calling. Also clear the STATE_MISSED before calling __netif_schedule() at the end of qdisc_run_end() to avoid doing another round of dequeuing in the pfifo_fast_dequeue(). The performance impact of this patch, tested using pktgen and dummy netdev with pfifo_fast qdisc attached: threads without+this_patch with+this_patch delta 1 2.61Mpps 2.60Mpps -0.3% 2 3.97Mpps 3.82Mpps -3.7% 4 5.62Mpps 5.59Mpps -0.5% 8 2.78Mpps 2.77Mpps -0.3% 16 2.22Mpps 2.22Mpps -0.0% Fixes: 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking") Acked-by: Jakub Kicinski <kuba@kernel.org> Tested-by: Juergen Gross <jgross@suse.com> Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-05-14 06:16:59 +03:00
/* Make sure dequeuing happens after clearing
* STATE_MISSED.
*/
smp_mb__after_atomic();
need_retry = false;
goto retry;
Revert "pkt_sched: Make default qdisc nonshared-multiqueue safe." This reverts commit a0c80b80e0fb48129e4e9d6a9ede914f9ff1850d. After discussions with Jamal and Herbert on netdev, we should provide at least minimal prioritization at the qdisc level even in multiqueue situations. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-21 20:56:13 +04:00
}
[PKT_SCHED]: Cleanup pfifo_fast qdisc and remove unnecessary code Removes the skb trimming code which is not needed since we never touch the skb upon failure. Removes unnecessary initializers, and simplifies the code a bit. Signed-off-by: Thomas Graf <tgraf@suug.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2005-06-19 09:58:53 +04:00
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
return skb;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
}
net_sched: cleanups Cleanup net/sched code to current CodingStyle and practices. Reduce inline abuse Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-01-19 22:26:56 +03:00
static struct sk_buff *pfifo_fast_peek(struct Qdisc *qdisc)
pkt_sched: sch_generic: Add generic qdisc->ops->peek() implementation. With feedback from Patrick McHardy. Signed-off-by: Jarek Poplawski <jarkao2@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-10-31 10:45:27 +03:00
{
Speed-up pfifo_fast lookup using a private bitmap Maintain a per-qdisc bitmap for pfifo_fast giving availability of skbs for each band. This allows faster lookup for a skb when there are no high priority skbs. Also, it helps in (rare) cases when there are no skbs on the list, where an immediate lookup is faster than iterating through the three bands. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-08-19 01:55:59 +04:00
struct pfifo_fast_priv *priv = qdisc_priv(qdisc);
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
struct sk_buff *skb = NULL;
int band;
Speed-up pfifo_fast lookup using a private bitmap Maintain a per-qdisc bitmap for pfifo_fast giving availability of skbs for each band. This allows faster lookup for a skb when there are no high priority skbs. Also, it helps in (rare) cases when there are no skbs on the list, where an immediate lookup is faster than iterating through the three bands. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-08-19 01:55:59 +04:00
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
for (band = 0; band < PFIFO_FAST_BANDS && !skb; band++) {
struct skb_array *q = band2list(priv, band);
pkt_sched: sch_generic: Add generic qdisc->ops->peek() implementation. With feedback from Patrick McHardy. Signed-off-by: Jarek Poplawski <jarkao2@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-10-31 10:45:27 +03:00
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
skb = __skb_array_peek(q);
pkt_sched: sch_generic: Add generic qdisc->ops->peek() implementation. With feedback from Patrick McHardy. Signed-off-by: Jarek Poplawski <jarkao2@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-10-31 10:45:27 +03:00
}
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
return skb;
pkt_sched: sch_generic: Add generic qdisc->ops->peek() implementation. With feedback from Patrick McHardy. Signed-off-by: Jarek Poplawski <jarkao2@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-10-31 10:45:27 +03:00
}
net_sched: cleanups Cleanup net/sched code to current CodingStyle and practices. Reduce inline abuse Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-01-19 22:26:56 +03:00
static void pfifo_fast_reset(struct Qdisc *qdisc)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
{
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
int i, band;
Speed-up pfifo_fast lookup using a private bitmap Maintain a per-qdisc bitmap for pfifo_fast giving availability of skbs for each band. This allows faster lookup for a skb when there are no high priority skbs. Also, it helps in (rare) cases when there are no skbs on the list, where an immediate lookup is faster than iterating through the three bands. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-08-19 01:55:59 +04:00
struct pfifo_fast_priv *priv = qdisc_priv(qdisc);
Revert "pkt_sched: Make default qdisc nonshared-multiqueue safe." This reverts commit a0c80b80e0fb48129e4e9d6a9ede914f9ff1850d. After discussions with Jamal and Herbert on netdev, we should provide at least minimal prioritization at the qdisc level even in multiqueue situations. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-21 20:56:13 +04:00
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
for (band = 0; band < PFIFO_FAST_BANDS; band++) {
struct skb_array *q = band2list(priv, band);
struct sk_buff *skb;
Revert "pkt_sched: Make default qdisc nonshared-multiqueue safe." This reverts commit a0c80b80e0fb48129e4e9d6a9ede914f9ff1850d. After discussions with Jamal and Herbert on netdev, we should provide at least minimal prioritization at the qdisc level even in multiqueue situations. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-21 20:56:13 +04:00
net_sched: properly check for empty skb array on error path First, the check of &q->ring.queue against NULL is wrong, it is always false. We should check the value rather than the address. Secondly, we need the same check in pfifo_fast_reset() too, as both ->reset() and ->destroy() are called in qdisc_destroy(). Fixes: c5ad119fb6c0 ("net: sched: pfifo_fast use skb_array") Reported-by: syzbot <syzkaller@googlegroups.com> Cc: John Fastabend <john.fastabend@gmail.com> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Acked-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-19 01:34:26 +03:00
/* 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;
pfifo_fast: drop unneeded additional lock on dequeue After the previous patch, for NOLOCK qdiscs, q->seqlock is always held when the dequeue() is invoked, we can drop any additional locking to protect such operation. Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-05-15 17:24:37 +03:00
while ((skb = __skb_array_consume(q)) != NULL)
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
kfree_skb(skb);
}
net/sched: pfifo_fast: fix wrong dereference when qdisc is reset Now that 'TCQ_F_CPUSTATS' bit can be cleared, depending on the value of 'TCQ_F_NOLOCK' bit in the parent qdisc, we need to be sure that per-cpu counters are present when 'reset()' is called for pfifo_fast qdiscs. Otherwise, the following script: # tc q a dev lo handle 1: root htb default 100 # tc c a dev lo parent 1: classid 1:100 htb \ > rate 95Mbit ceil 100Mbit burst 64k [...] # tc f a dev lo parent 1: protocol arp basic classid 1:100 [...] # tc q a dev lo parent 1:100 handle 100: pfifo_fast [...] # tc q d dev lo root can generate the following splat: Unable to handle kernel paging request at virtual address dfff2c01bd148000 Mem abort info: ESR = 0x96000004 Exception class = DABT (current EL), IL = 32 bits SET = 0, FnV = 0 EA = 0, S1PTW = 0 Data abort info: ISV = 0, ISS = 0x00000004 CM = 0, WnR = 0 [dfff2c01bd148000] address between user and kernel address ranges Internal error: Oops: 96000004 [#1] SMP [...] pstate: 80000005 (Nzcv daif -PAN -UAO) pc : pfifo_fast_reset+0x280/0x4d8 lr : pfifo_fast_reset+0x21c/0x4d8 sp : ffff800d09676fa0 x29: ffff800d09676fa0 x28: ffff200012ee22e4 x27: dfff200000000000 x26: 0000000000000000 x25: ffff800ca0799958 x24: ffff1001940f332b x23: 0000000000000007 x22: ffff200012ee1ab8 x21: 0000600de8a40000 x20: 0000000000000000 x19: ffff800ca0799900 x18: 0000000000000000 x17: 0000000000000002 x16: 0000000000000000 x15: 0000000000000000 x14: 0000000000000000 x13: 0000000000000000 x12: ffff1001b922e6e2 x11: 1ffff001b922e6e1 x10: 0000000000000000 x9 : 1ffff001b922e6e1 x8 : dfff200000000000 x7 : 0000000000000000 x6 : 0000000000000000 x5 : 1fffe400025dc45c x4 : 1fffe400025dc357 x3 : 00000c01bd148000 x2 : 0000600de8a40000 x1 : 0000000000000007 x0 : 0000600de8a40004 Call trace: pfifo_fast_reset+0x280/0x4d8 qdisc_reset+0x6c/0x370 htb_reset+0x150/0x3b8 [sch_htb] qdisc_reset+0x6c/0x370 dev_deactivate_queue.constprop.5+0xe0/0x1a8 dev_deactivate_many+0xd8/0x908 dev_deactivate+0xe4/0x190 qdisc_graft+0x88c/0xbd0 tc_get_qdisc+0x418/0x8a8 rtnetlink_rcv_msg+0x3a8/0xa78 netlink_rcv_skb+0x18c/0x328 rtnetlink_rcv+0x28/0x38 netlink_unicast+0x3c4/0x538 netlink_sendmsg+0x538/0x9a0 sock_sendmsg+0xac/0xf8 ___sys_sendmsg+0x53c/0x658 __sys_sendmsg+0xc8/0x140 __arm64_sys_sendmsg+0x74/0xa8 el0_svc_handler+0x164/0x468 el0_svc+0x10/0x14 Code: 910012a0 92400801 d343fc03 11000c21 (38fb6863) Fix this by testing the value of 'TCQ_F_CPUSTATS' bit in 'qdisc->flags', before dereferencing 'qdisc->cpu_qstats'. Changes since v1: - coding style improvements, thanks to Stefano Brivio Fixes: 8a53e616de29 ("net: sched: when clearing NOLOCK, clear TCQ_F_CPUSTATS, too") CC: Paolo Abeni <pabeni@redhat.com> Reported-by: Li Shuang <shuali@redhat.com> Signed-off-by: Davide Caratti <dcaratti@redhat.com> Acked-by: Paolo Abeni <pabeni@redhat.com> Reviewed-by: Stefano Brivio <sbrivio@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-08-27 13:29:09 +03:00
if (qdisc_is_percpu_stats(qdisc)) {
for_each_possible_cpu(i) {
struct gnet_stats_queue *q;
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
net/sched: pfifo_fast: fix wrong dereference when qdisc is reset Now that 'TCQ_F_CPUSTATS' bit can be cleared, depending on the value of 'TCQ_F_NOLOCK' bit in the parent qdisc, we need to be sure that per-cpu counters are present when 'reset()' is called for pfifo_fast qdiscs. Otherwise, the following script: # tc q a dev lo handle 1: root htb default 100 # tc c a dev lo parent 1: classid 1:100 htb \ > rate 95Mbit ceil 100Mbit burst 64k [...] # tc f a dev lo parent 1: protocol arp basic classid 1:100 [...] # tc q a dev lo parent 1:100 handle 100: pfifo_fast [...] # tc q d dev lo root can generate the following splat: Unable to handle kernel paging request at virtual address dfff2c01bd148000 Mem abort info: ESR = 0x96000004 Exception class = DABT (current EL), IL = 32 bits SET = 0, FnV = 0 EA = 0, S1PTW = 0 Data abort info: ISV = 0, ISS = 0x00000004 CM = 0, WnR = 0 [dfff2c01bd148000] address between user and kernel address ranges Internal error: Oops: 96000004 [#1] SMP [...] pstate: 80000005 (Nzcv daif -PAN -UAO) pc : pfifo_fast_reset+0x280/0x4d8 lr : pfifo_fast_reset+0x21c/0x4d8 sp : ffff800d09676fa0 x29: ffff800d09676fa0 x28: ffff200012ee22e4 x27: dfff200000000000 x26: 0000000000000000 x25: ffff800ca0799958 x24: ffff1001940f332b x23: 0000000000000007 x22: ffff200012ee1ab8 x21: 0000600de8a40000 x20: 0000000000000000 x19: ffff800ca0799900 x18: 0000000000000000 x17: 0000000000000002 x16: 0000000000000000 x15: 0000000000000000 x14: 0000000000000000 x13: 0000000000000000 x12: ffff1001b922e6e2 x11: 1ffff001b922e6e1 x10: 0000000000000000 x9 : 1ffff001b922e6e1 x8 : dfff200000000000 x7 : 0000000000000000 x6 : 0000000000000000 x5 : 1fffe400025dc45c x4 : 1fffe400025dc357 x3 : 00000c01bd148000 x2 : 0000600de8a40000 x1 : 0000000000000007 x0 : 0000600de8a40004 Call trace: pfifo_fast_reset+0x280/0x4d8 qdisc_reset+0x6c/0x370 htb_reset+0x150/0x3b8 [sch_htb] qdisc_reset+0x6c/0x370 dev_deactivate_queue.constprop.5+0xe0/0x1a8 dev_deactivate_many+0xd8/0x908 dev_deactivate+0xe4/0x190 qdisc_graft+0x88c/0xbd0 tc_get_qdisc+0x418/0x8a8 rtnetlink_rcv_msg+0x3a8/0xa78 netlink_rcv_skb+0x18c/0x328 rtnetlink_rcv+0x28/0x38 netlink_unicast+0x3c4/0x538 netlink_sendmsg+0x538/0x9a0 sock_sendmsg+0xac/0xf8 ___sys_sendmsg+0x53c/0x658 __sys_sendmsg+0xc8/0x140 __arm64_sys_sendmsg+0x74/0xa8 el0_svc_handler+0x164/0x468 el0_svc+0x10/0x14 Code: 910012a0 92400801 d343fc03 11000c21 (38fb6863) Fix this by testing the value of 'TCQ_F_CPUSTATS' bit in 'qdisc->flags', before dereferencing 'qdisc->cpu_qstats'. Changes since v1: - coding style improvements, thanks to Stefano Brivio Fixes: 8a53e616de29 ("net: sched: when clearing NOLOCK, clear TCQ_F_CPUSTATS, too") CC: Paolo Abeni <pabeni@redhat.com> Reported-by: Li Shuang <shuali@redhat.com> Signed-off-by: Davide Caratti <dcaratti@redhat.com> Acked-by: Paolo Abeni <pabeni@redhat.com> Reviewed-by: Stefano Brivio <sbrivio@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-08-27 13:29:09 +03:00
q = per_cpu_ptr(qdisc->cpu_qstats, i);
q->backlog = 0;
q->qlen = 0;
}
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
}
Revert "pkt_sched: Make default qdisc nonshared-multiqueue safe." This reverts commit a0c80b80e0fb48129e4e9d6a9ede914f9ff1850d. After discussions with Jamal and Herbert on netdev, we should provide at least minimal prioritization at the qdisc level even in multiqueue situations. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-21 20:56:13 +04:00
static int pfifo_fast_dump(struct Qdisc *qdisc, struct sk_buff *skb)
{
struct tc_prio_qopt opt = { .bands = PFIFO_FAST_BANDS };
net_sched: cleanups Cleanup net/sched code to current CodingStyle and practices. Reduce inline abuse Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-01-19 22:26:56 +03:00
memcpy(&opt.priomap, prio2band, TC_PRIO_MAX + 1);
pkt_sched: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 13:11:39 +04:00
if (nla_put(skb, TCA_OPTIONS, sizeof(opt), &opt))
goto nla_put_failure;
Revert "pkt_sched: Make default qdisc nonshared-multiqueue safe." This reverts commit a0c80b80e0fb48129e4e9d6a9ede914f9ff1850d. After discussions with Jamal and Herbert on netdev, we should provide at least minimal prioritization at the qdisc level even in multiqueue situations. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-21 20:56:13 +04:00
return skb->len;
nla_put_failure:
return -1;
}
net: sched: sch: add extack for init callback This patch adds extack support for init callback to prepare per-qdisc specific changes for extack. Cc: David Ahern <dsahern@gmail.com> Acked-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: Alexander Aring <aring@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-20 20:35:13 +03:00
static int pfifo_fast_init(struct Qdisc *qdisc, struct nlattr *opt,
struct netlink_ext_ack *extack)
Revert "pkt_sched: Make default qdisc nonshared-multiqueue safe." This reverts commit a0c80b80e0fb48129e4e9d6a9ede914f9ff1850d. After discussions with Jamal and Herbert on netdev, we should provide at least minimal prioritization at the qdisc level even in multiqueue situations. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-21 20:56:13 +04:00
{
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
unsigned int qlen = qdisc_dev(qdisc)->tx_queue_len;
Speed-up pfifo_fast lookup using a private bitmap Maintain a per-qdisc bitmap for pfifo_fast giving availability of skbs for each band. This allows faster lookup for a skb when there are no high priority skbs. Also, it helps in (rare) cases when there are no skbs on the list, where an immediate lookup is faster than iterating through the three bands. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-08-19 01:55:59 +04:00
struct pfifo_fast_priv *priv = qdisc_priv(qdisc);
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
int prio;
/* guard against zero length rings */
if (!qlen)
return -EINVAL;
Revert "pkt_sched: Make default qdisc nonshared-multiqueue safe." This reverts commit a0c80b80e0fb48129e4e9d6a9ede914f9ff1850d. After discussions with Jamal and Herbert on netdev, we should provide at least minimal prioritization at the qdisc level even in multiqueue situations. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-21 20:56:13 +04:00
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
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;
}
Revert "pkt_sched: Make default qdisc nonshared-multiqueue safe." This reverts commit a0c80b80e0fb48129e4e9d6a9ede914f9ff1850d. After discussions with Jamal and Herbert on netdev, we should provide at least minimal prioritization at the qdisc level even in multiqueue situations. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-21 20:56:13 +04:00
net_sched: TCQ_F_CAN_BYPASS generalization Now qdisc stab is handled before TCQ_F_CAN_BYPASS test in __dev_xmit_skb(), we can generalize TCQ_F_CAN_BYPASS to other qdiscs than pfifo_fast : pfifo, bfifo, pfifo_head_drop and sfq SFQ is special because it can have external classifiers, and in these cases, we cannot bypass queue discipline (packet could be dropped by classifier) without admin asking it, or further changes. Its worth doing this, especially for SFQ, avoiding dirtying memory in case no packets are already waiting in queue. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-01-22 03:26:09 +03:00
/* Can by-pass the queue discipline */
qdisc->flags |= TCQ_F_CAN_BYPASS;
Revert "pkt_sched: Make default qdisc nonshared-multiqueue safe." This reverts commit a0c80b80e0fb48129e4e9d6a9ede914f9ff1850d. After discussions with Jamal and Herbert on netdev, we should provide at least minimal prioritization at the qdisc level even in multiqueue situations. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-21 20:56:13 +04:00
return 0;
}
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
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.
*/
net_sched: properly check for empty skb array on error path First, the check of &q->ring.queue against NULL is wrong, it is always false. We should check the value rather than the address. Secondly, we need the same check in pfifo_fast_reset() too, as both ->reset() and ->destroy() are called in qdisc_destroy(). Fixes: c5ad119fb6c0 ("net: sched: pfifo_fast use skb_array") Reported-by: syzbot <syzkaller@googlegroups.com> Cc: John Fastabend <john.fastabend@gmail.com> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Acked-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-19 01:34:26 +03:00
if (!q->ring.queue)
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
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);
}
}
net_sched: implement ->change_tx_queue_len() for pfifo_fast pfifo_fast used to drop based on qdisc_dev(qdisc)->tx_queue_len, so we have to resize skb array when we change tx_queue_len. Other qdiscs which read tx_queue_len are fine because they all save it to sch->limit or somewhere else in qdisc during init. They don't have to implement this, it is nicer if they do so that users don't have to re-configure qdisc after changing tx_queue_len. Cc: John Fastabend <john.fastabend@gmail.com> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-01-26 05:26:24 +03:00
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);
}
net_sched: add classful multiqueue dummy scheduler This patch adds a classful dummy scheduler which can be used as root qdisc for multiqueue devices and exposes each device queue as a child class. This allows to address queues individually and graft them similar to regular classes. Additionally it presents an accumulated view of the statistics of all real root qdiscs in the dummy root. Two new callbacks are added to the qdisc_ops and qdisc_class_ops: - cl_ops->select_queue selects the tx queue number for new child classes. - qdisc_ops->attach() overrides root qdisc device grafting to attach non-shared qdiscs to the queues. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-06 12:58:51 +04:00
struct Qdisc_ops pfifo_fast_ops __read_mostly = {
Revert "pkt_sched: Make default qdisc nonshared-multiqueue safe." This reverts commit a0c80b80e0fb48129e4e9d6a9ede914f9ff1850d. After discussions with Jamal and Herbert on netdev, we should provide at least minimal prioritization at the qdisc level even in multiqueue situations. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-21 20:56:13 +04:00
.id = "pfifo_fast",
Speed-up pfifo_fast lookup using a private bitmap Maintain a per-qdisc bitmap for pfifo_fast giving availability of skbs for each band. This allows faster lookup for a skb when there are no high priority skbs. Also, it helps in (rare) cases when there are no skbs on the list, where an immediate lookup is faster than iterating through the three bands. Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-08-19 01:55:59 +04:00
.priv_size = sizeof(struct pfifo_fast_priv),
Revert "pkt_sched: Make default qdisc nonshared-multiqueue safe." This reverts commit a0c80b80e0fb48129e4e9d6a9ede914f9ff1850d. After discussions with Jamal and Herbert on netdev, we should provide at least minimal prioritization at the qdisc level even in multiqueue situations. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-21 20:56:13 +04:00
.enqueue = pfifo_fast_enqueue,
.dequeue = pfifo_fast_dequeue,
pkt_sched: sch_generic: Add generic qdisc->ops->peek() implementation. With feedback from Patrick McHardy. Signed-off-by: Jarek Poplawski <jarkao2@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-10-31 10:45:27 +03:00
.peek = pfifo_fast_peek,
Revert "pkt_sched: Make default qdisc nonshared-multiqueue safe." This reverts commit a0c80b80e0fb48129e4e9d6a9ede914f9ff1850d. After discussions with Jamal and Herbert on netdev, we should provide at least minimal prioritization at the qdisc level even in multiqueue situations. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-21 20:56:13 +04:00
.init = pfifo_fast_init,
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
.destroy = pfifo_fast_destroy,
Revert "pkt_sched: Make default qdisc nonshared-multiqueue safe." This reverts commit a0c80b80e0fb48129e4e9d6a9ede914f9ff1850d. After discussions with Jamal and Herbert on netdev, we should provide at least minimal prioritization at the qdisc level even in multiqueue situations. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-21 20:56:13 +04:00
.reset = pfifo_fast_reset,
.dump = pfifo_fast_dump,
net_sched: implement ->change_tx_queue_len() for pfifo_fast pfifo_fast used to drop based on qdisc_dev(qdisc)->tx_queue_len, so we have to resize skb array when we change tx_queue_len. Other qdiscs which read tx_queue_len are fine because they all save it to sch->limit or somewhere else in qdisc during init. They don't have to implement this, it is nicer if they do so that users don't have to re-configure qdisc after changing tx_queue_len. Cc: John Fastabend <john.fastabend@gmail.com> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-01-26 05:26:24 +03:00
.change_tx_queue_len = pfifo_fast_change_tx_queue_len,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
.owner = THIS_MODULE,
net: sched: pfifo_fast use skb_array This converts the pfifo_fast qdisc to use the skb_array data structure and set the lockless qdisc bit. pfifo_fast is the first qdisc to support the lockless bit that can be a child of a qdisc requiring locking. So we add logic to clear the lock bit on initialization in these cases when the qdisc graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:58:19 +03:00
.static_flags = TCQ_F_NOLOCK | TCQ_F_CPUSTATS,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
};
net: sched: use pfifo_fast for non real queues Some devices declare a high number of TX queues, then set a much lower real_num_tx_queues This cause setups using fq_codel, sfq or fq as the default qdisc to consume more memory than really needed. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-03-02 19:21:43 +03:00
EXPORT_SYMBOL(pfifo_fast_ops);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
net: partially revert dynamic lockdep key changes This patch reverts the folowing commits: commit 064ff66e2bef84f1153087612032b5b9eab005bd "bonding: add missing netdev_update_lockdep_key()" commit 53d374979ef147ab51f5d632dfe20b14aebeccd0 "net: avoid updating qdisc_xmit_lock_key in netdev_update_lockdep_key()" commit 1f26c0d3d24125992ab0026b0dab16c08df947c7 "net: fix kernel-doc warning in <linux/netdevice.h>" commit ab92d68fc22f9afab480153bd82a20f6e2533769 "net: core: add generic lockdep keys" but keeps the addr_list_lock_key because we still lock addr_list_lock nestedly on stack devices, unlikely xmit_lock this is safe because we don't take addr_list_lock on any fast path. Reported-and-tested-by: syzbot+aaa6fa4949cc5d9b7b25@syzkaller.appspotmail.com Cc: Dmitry Vyukov <dvyukov@google.com> Cc: Taehee Yoo <ap420073@gmail.com> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Acked-by: Taehee Yoo <ap420073@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-05-03 08:22:19 +03:00
static struct lock_class_key qdisc_tx_busylock;
pkt_sched: Remove 'dev' member of struct Qdisc. It can be obtained via the netdev_queue. So create a helper routine, qdisc_dev(), to make the transformations nicer looking. Now, qdisc_alloc() now no longer needs a net_device pointer argument. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-09 04:06:30 +04:00
struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue,
net: sch: api: add extack support in qdisc_alloc This patch adds extack support for the function qdisc_alloc which is a common used function in the tc subsystem. Callers which are interested in the receiving error can assign extack to get a more detailed information why qdisc_alloc failed. Cc: David Ahern <dsahern@gmail.com> Acked-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: Alexander Aring <aring@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-20 20:35:20 +03:00
const struct Qdisc_ops *ops,
struct netlink_ext_ack *extack)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
{
struct Qdisc *sch;
net/sched: get rid of qdisc->padded kmalloc() of sufficiently big portion of memory is cache-aligned in regular conditions. If some debugging options are used, there is no reason qdisc structures would need 64-byte alignment if most other kernel structures are not aligned. This get rid of QDISC_ALIGN and QDISC_ALIGNTO. Addition of privdata field will help implementing the reverse of qdisc_priv() and documents where the private data is. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Allen Pais <allen.lkml@gmail.com> Acked-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2020-10-07 19:51:11 +03:00
unsigned int size = sizeof(*sch) + ops->priv_size;
[PKT_SCHED]: Cleanup qdisc creation and alignment macros Adds qdisc_alloc() to share code between qdisc_create() and qdisc_create_dflt(). Hides the qdisc alignment behind macros and makes use of them. Signed-off-by: Thomas Graf <tgraf@suug.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2005-07-06 01:15:09 +04:00
int err = -ENOBUFS;
net/sched: Check for null dev_queue on create flow In qdisc_alloc() the dev_queue pointer was used without any checks being performed. If qdisc_create() gets a null dev_queue pointer, it just passes it along to qdisc_alloc(), leading to a crash. That happens if a root qdisc implements select_queue() and returns a null dev_queue pointer for an "invalid handle", for example, or if the dev_queue associated with the parent qdisc is null. This patch is in preparation for the next in this series, where select_queue() is being added to mqprio and as it may return a null dev_queue. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 04:01:23 +03:00
struct net_device *dev;
if (!dev_queue) {
net: sch: api: add extack support in qdisc_alloc This patch adds extack support for the function qdisc_alloc which is a common used function in the tc subsystem. Callers which are interested in the receiving error can assign extack to get a more detailed information why qdisc_alloc failed. Cc: David Ahern <dsahern@gmail.com> Acked-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: Alexander Aring <aring@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-20 20:35:20 +03:00
NL_SET_ERR_MSG(extack, "No device queue given");
net/sched: Check for null dev_queue on create flow In qdisc_alloc() the dev_queue pointer was used without any checks being performed. If qdisc_create() gets a null dev_queue pointer, it just passes it along to qdisc_alloc(), leading to a crash. That happens if a root qdisc implements select_queue() and returns a null dev_queue pointer for an "invalid handle", for example, or if the dev_queue associated with the parent qdisc is null. This patch is in preparation for the next in this series, where select_queue() is being added to mqprio and as it may return a null dev_queue. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 04:01:23 +03:00
err = -EINVAL;
goto errout;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
net/sched: Check for null dev_queue on create flow In qdisc_alloc() the dev_queue pointer was used without any checks being performed. If qdisc_create() gets a null dev_queue pointer, it just passes it along to qdisc_alloc(), leading to a crash. That happens if a root qdisc implements select_queue() and returns a null dev_queue pointer for an "invalid handle", for example, or if the dev_queue associated with the parent qdisc is null. This patch is in preparation for the next in this series, where select_queue() is being added to mqprio and as it may return a null dev_queue. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 04:01:23 +03:00
dev = dev_queue->dev;
net/sched: get rid of qdisc->padded kmalloc() of sufficiently big portion of memory is cache-aligned in regular conditions. If some debugging options are used, there is no reason qdisc structures would need 64-byte alignment if most other kernel structures are not aligned. This get rid of QDISC_ALIGN and QDISC_ALIGNTO. Addition of privdata field will help implementing the reverse of qdisc_priv() and documents where the private data is. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Allen Pais <allen.lkml@gmail.com> Acked-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2020-10-07 19:51:11 +03:00
sch = kzalloc_node(size, GFP_KERNEL, netdev_queue_numa_node_read(dev_queue));
net sched: use xps information for qdisc NUMA affinity Allocate qdisc memory according to NUMA properties of cpus included in xps map. To be effective, qdisc should be (re)setup after changes of /sys/class/net/eth<n>/queues/tx-<n>/xps_cpus I added a numa_node field in struct netdev_queue, containing NUMA node if all cpus included in xps_cpus share same node, else -1. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Cc: Ben Hutchings <bhutchings@solarflare.com> Cc: Tom Herbert <therbert@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-11-29 11:14:37 +03:00
net/sched: get rid of qdisc->padded kmalloc() of sufficiently big portion of memory is cache-aligned in regular conditions. If some debugging options are used, there is no reason qdisc structures would need 64-byte alignment if most other kernel structures are not aligned. This get rid of QDISC_ALIGN and QDISC_ALIGNTO. Addition of privdata field will help implementing the reverse of qdisc_priv() and documents where the private data is. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Allen Pais <allen.lkml@gmail.com> Acked-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2020-10-07 19:51:11 +03:00
if (!sch)
[PKT_SCHED]: Cleanup qdisc creation and alignment macros Adds qdisc_alloc() to share code between qdisc_create() and qdisc_create_dflt(). Hides the qdisc alignment behind macros and makes use of them. Signed-off-by: Thomas Graf <tgraf@suug.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2005-07-06 01:15:09 +04:00
goto errout;
net: sched: explicit locking in gso_cpu fallback This work is preparing the qdisc layer to support egress lockless qdiscs. If we are running the egress qdisc lockless in the case we overrun the netdev, for whatever reason, the netdev returns a busy error code and the skb is parked on the gso_skb pointer. With many cores all hitting this case at once its possible to have multiple sk_buffs here so we turn gso_skb into a queue. This should be the edge case and if we see this frequently then the netdev/qdisc layer needs to back off. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:55:45 +03:00
__skb_queue_head_init(&sch->gso_skb);
net: sched: use skb list for skb_bad_tx Similar to how gso is handled use skb list for skb_bad_tx this is required with lockless qdiscs because we may have multiple cores attempting to push skbs into skb_bad_tx concurrently Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:56:23 +03:00
__skb_queue_head_init(&sch->skb_bad_txq);
sched: add and use qdisc_skb_head helpers This change replaces sk_buff_head struct in Qdiscs with new qdisc_skb_head. Its similar to the skb_buff_head api, but does not use skb->prev pointers. Qdiscs will commonly enqueue at the tail of a list and dequeue at head. While skb_buff_head works fine for this, enqueue/dequeue needs to also adjust the prev pointer of next element. The ->prev pointer is not required for qdiscs so we can just leave it undefined and avoid one cacheline write access for en/dequeue. Suggested-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-09-18 01:57:34 +03:00
qdisc_skb_head_init(&sch->q);
net: sched: Merge Qdisc::bstats and Qdisc::cpu_bstats data types The only factor differentiating per-CPU bstats data type (struct gnet_stats_basic_cpu) from the packed non-per-CPU one (struct gnet_stats_basic_packed) was a u64_stats sync point inside the former. The two data types are now equivalent: earlier commits added a u64_stats sync point to the latter. Combine both data types into "struct gnet_stats_basic_sync". This eliminates redundancy and simplifies the bstats read/write APIs. Use u64_stats_t for bstats "packets" and "bytes" data types. On 64-bit architectures, u64_stats sync points do not use sequence counter protection. Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-10-16 11:49:09 +03:00
gnet_stats_basic_sync_init(&sch->bstats);
sched: add and use qdisc_skb_head helpers This change replaces sk_buff_head struct in Qdiscs with new qdisc_skb_head. Its similar to the skb_buff_head api, but does not use skb->prev pointers. Qdiscs will commonly enqueue at the tail of a list and dequeue at head. While skb_buff_head works fine for this, enqueue/dequeue needs to also adjust the prev pointer of next element. The ->prev pointer is not required for qdiscs so we can just leave it undefined and avoid one cacheline write access for en/dequeue. Suggested-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-09-18 01:57:34 +03:00
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 05:02:56 +04:00
net: sched: a dflt qdisc may be used with per cpu stats Enable dflt qdisc support for per cpu stats before this patch a dflt qdisc was required to use the global statistics qstats and bstats. This adds a static flags field to qdisc_ops that is propagated into qdisc->flags in qdisc allocate call. This allows the allocation block to completely allocate the qdisc object so we don't have dangling allocations after qdisc init. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:55:26 +03:00
if (ops->static_flags & TCQ_F_CPUSTATS) {
sch->cpu_bstats =
net: sched: Merge Qdisc::bstats and Qdisc::cpu_bstats data types The only factor differentiating per-CPU bstats data type (struct gnet_stats_basic_cpu) from the packed non-per-CPU one (struct gnet_stats_basic_packed) was a u64_stats sync point inside the former. The two data types are now equivalent: earlier commits added a u64_stats sync point to the latter. Combine both data types into "struct gnet_stats_basic_sync". This eliminates redundancy and simplifies the bstats read/write APIs. Use u64_stats_t for bstats "packets" and "bytes" data types. On 64-bit architectures, u64_stats sync points do not use sequence counter protection. Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-10-16 11:49:09 +03:00
netdev_alloc_pcpu_stats(struct gnet_stats_basic_sync);
net: sched: a dflt qdisc may be used with per cpu stats Enable dflt qdisc support for per cpu stats before this patch a dflt qdisc was required to use the global statistics qstats and bstats. This adds a static flags field to qdisc_ops that is propagated into qdisc->flags in qdisc allocate call. This allows the allocation block to completely allocate the qdisc object so we don't have dangling allocations after qdisc init. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:55:26 +03:00
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;
}
}
net: add additional lock to qdisc to increase throughput When many cpus compete for sending frames on a given qdisc, the qdisc spinlock suffers from very high contention. The cpu owning __QDISC_STATE_RUNNING bit has same priority to acquire the lock, and cannot dequeue packets fast enough, since it must wait for this lock for each dequeued packet. One solution to this problem is to force all cpus spinning on a second lock before trying to get the main lock, when/if they see __QDISC_STATE_RUNNING already set. The owning cpu then compete with at most one other cpu for the main lock, allowing for higher dequeueing rate. Based on a previous patch from Alexander Duyck. I added the heuristic to avoid the atomic in fast path, and put the new lock far away from the cache line used by the dequeue worker. Also try to release the busylock lock as late as possible. Tests with following script gave a boost from ~50.000 pps to ~600.000 pps on a dual quad core machine (E5450 @3.00GHz), tg3 driver. (A single netperf flow can reach ~800.000 pps on this platform) for j in `seq 0 3`; do for i in `seq 0 7`; do netperf -H 192.168.0.1 -t UDP_STREAM -l 60 -N -T $i -- -m 6 & done done Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Acked-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-06-02 16:09:29 +04:00
spin_lock_init(&sch->busylock);
net: partially revert dynamic lockdep key changes This patch reverts the folowing commits: commit 064ff66e2bef84f1153087612032b5b9eab005bd "bonding: add missing netdev_update_lockdep_key()" commit 53d374979ef147ab51f5d632dfe20b14aebeccd0 "net: avoid updating qdisc_xmit_lock_key in netdev_update_lockdep_key()" commit 1f26c0d3d24125992ab0026b0dab16c08df947c7 "net: fix kernel-doc warning in <linux/netdevice.h>" commit ab92d68fc22f9afab480153bd82a20f6e2533769 "net: core: add generic lockdep keys" but keeps the addr_list_lock_key because we still lock addr_list_lock nestedly on stack devices, unlikely xmit_lock this is safe because we don't take addr_list_lock on any fast path. Reported-and-tested-by: syzbot+aaa6fa4949cc5d9b7b25@syzkaller.appspotmail.com Cc: Dmitry Vyukov <dvyukov@google.com> Cc: Taehee Yoo <ap420073@gmail.com> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Acked-by: Taehee Yoo <ap420073@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-05-03 08:22:19 +03:00
lockdep_set_class(&sch->busylock,
dev->qdisc_tx_busylock ?: &qdisc_tx_busylock);
sched: replace __QDISC_STATE_RUNNING bit with a spin lock So that we can use lockdep on it. The newly introduced sequence lock has the same scope of busylock, so it shares the same lockdep annotation, but it's only used for NOLOCK qdiscs. With this changeset we acquire such lock in the control path around flushing operation (qdisc reset), to allow more NOLOCK qdisc perf improvement in the next patch. Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-05-15 17:24:36 +03:00
/* seqlock has the same scope of busylock, for NOLOCK qdisc */
spin_lock_init(&sch->seqlock);
net: sched: fix lockdep_set_class() typo error for sch->seqlock According to comment in qdisc_alloc(), sch->seqlock's lockdep class key should be set to qdisc_tx_busylock, due to possible type error, sch->busylock's lockdep class key is set to qdisc_tx_busylock, which is duplicated because sch->busylock's lockdep class key is already set in qdisc_alloc(). So fix it by replacing sch->busylock with sch->seqlock. Fixes: 96009c7d500e ("sched: replace __QDISC_STATE_RUNNING bit with a spin lock") Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-08-03 13:58:21 +03:00
lockdep_set_class(&sch->seqlock,
net: partially revert dynamic lockdep key changes This patch reverts the folowing commits: commit 064ff66e2bef84f1153087612032b5b9eab005bd "bonding: add missing netdev_update_lockdep_key()" commit 53d374979ef147ab51f5d632dfe20b14aebeccd0 "net: avoid updating qdisc_xmit_lock_key in netdev_update_lockdep_key()" commit 1f26c0d3d24125992ab0026b0dab16c08df947c7 "net: fix kernel-doc warning in <linux/netdevice.h>" commit ab92d68fc22f9afab480153bd82a20f6e2533769 "net: core: add generic lockdep keys" but keeps the addr_list_lock_key because we still lock addr_list_lock nestedly on stack devices, unlikely xmit_lock this is safe because we don't take addr_list_lock on any fast path. Reported-and-tested-by: syzbot+aaa6fa4949cc5d9b7b25@syzkaller.appspotmail.com Cc: Dmitry Vyukov <dvyukov@google.com> Cc: Taehee Yoo <ap420073@gmail.com> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Acked-by: Taehee Yoo <ap420073@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-05-03 08:22:19 +03:00
dev->qdisc_tx_busylock ?: &qdisc_tx_busylock);
net_sched: transform qdisc running bit into a seqcount Instead of using a single bit (__QDISC___STATE_RUNNING) in sch->__state, use a seqcount. This adds lockdep support, but more importantly it will allow us to sample qdisc/class statistics without having to grab qdisc root lock. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-06-06 19:37:15 +03:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
sch->ops = ops;
net: sched: a dflt qdisc may be used with per cpu stats Enable dflt qdisc support for per cpu stats before this patch a dflt qdisc was required to use the global statistics qstats and bstats. This adds a static flags field to qdisc_ops that is propagated into qdisc->flags in qdisc allocate call. This allows the allocation block to completely allocate the qdisc object so we don't have dangling allocations after qdisc init. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:55:26 +03:00
sch->flags = ops->static_flags;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
sch->enqueue = ops->enqueue;
sch->dequeue = ops->dequeue;
netdev: Create netdev_queue abstraction. A netdev_queue is an entity managed by a qdisc. Currently there is one RX and one TX queue, and a netdev_queue merely contains a backpointer to the net_device. The Qdisc struct is augmented with a netdev_queue pointer as well. Eventually the 'dev' Qdisc member will go away and we will have the resulting hierarchy: net_device --> netdev_queue --> Qdisc Also, qdisc_alloc() and qdisc_create_dflt() now take a netdev_queue pointer argument. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-09 03:55:56 +04:00
sch->dev_queue = dev_queue;
net/sched: add net device refcount tracker to struct Qdisc Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-12-05 07:22:13 +03:00
dev_hold_track(dev, &sch->dev_tracker, GFP_KERNEL);
net, sched: convert Qdisc.refcnt from atomic_t to refcount_t refcount_t type and corresponding API should be used instead of atomic_t when the variable is used as a reference counter. This allows to avoid accidental refcounter overflows that might lead to use-after-free situations. Signed-off-by: Elena Reshetova <elena.reshetova@intel.com> Signed-off-by: Hans Liljestrand <ishkamiel@gmail.com> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: David Windsor <dwindsor@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-07-04 15:53:07 +03:00
refcount_set(&sch->refcnt, 1);
[PKT_SCHED]: Cleanup qdisc creation and alignment macros Adds qdisc_alloc() to share code between qdisc_create() and qdisc_create_dflt(). Hides the qdisc alignment behind macros and makes use of them. Signed-off-by: Thomas Graf <tgraf@suug.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2005-07-06 01:15:09 +04:00
return sch;
net: sched: a dflt qdisc may be used with per cpu stats Enable dflt qdisc support for per cpu stats before this patch a dflt qdisc was required to use the global statistics qstats and bstats. This adds a static flags field to qdisc_ops that is propagated into qdisc->flags in qdisc allocate call. This allows the allocation block to completely allocate the qdisc object so we don't have dangling allocations after qdisc init. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:55:26 +03:00
errout1:
net/sched: get rid of qdisc->padded kmalloc() of sufficiently big portion of memory is cache-aligned in regular conditions. If some debugging options are used, there is no reason qdisc structures would need 64-byte alignment if most other kernel structures are not aligned. This get rid of QDISC_ALIGN and QDISC_ALIGNTO. Addition of privdata field will help implementing the reverse of qdisc_priv() and documents where the private data is. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Allen Pais <allen.lkml@gmail.com> Acked-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2020-10-07 19:51:11 +03:00
kfree(sch);
[PKT_SCHED]: Cleanup qdisc creation and alignment macros Adds qdisc_alloc() to share code between qdisc_create() and qdisc_create_dflt(). Hides the qdisc alignment behind macros and makes use of them. Signed-off-by: Thomas Graf <tgraf@suug.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2005-07-06 01:15:09 +04:00
errout:
pkt_sched: ERR_PTR() ususally encodes an negative errno, not positive. Note, in the following patch, 'err' is initialized as: int err = -ENOBUFS; Signed-off-by: WANG Cong <wcong@critical-links.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-06-28 06:51:35 +04:00
return ERR_PTR(err);
[PKT_SCHED]: Cleanup qdisc creation and alignment macros Adds qdisc_alloc() to share code between qdisc_create() and qdisc_create_dflt(). Hides the qdisc alignment behind macros and makes use of them. Signed-off-by: Thomas Graf <tgraf@suug.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2005-07-06 01:15:09 +04:00
}
net_sched: remove the unused parameter of qdisc_create_dflt() The first parameter dev isn't in use in qdisc_create_dflt(). Signed-off-by: Changli Gao <xiaosuo@gmail.com> Acked-by: Jamal Hadi Salim <hadi@cyberus.ca> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-10-16 17:04:08 +04:00
struct Qdisc *qdisc_create_dflt(struct netdev_queue *dev_queue,
qdisc: make args to qdisc_create_default const Fixes warnings introduced by the qdisc default patch. Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-08-31 21:15:50 +04:00
const struct Qdisc_ops *ops,
net: sch: api: add extack support in qdisc_create_dflt This patch adds extack support for the function qdisc_create_dflt which is a common used function in the tc subsystem. Callers which are interested in the receiving error can assign extack to get a more detailed information why qdisc_create_dflt failed. The function qdisc_create_dflt will also call an init callback which can fail by any per-qdisc specific handling. Cc: David Ahern <dsahern@gmail.com> Acked-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: Alexander Aring <aring@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-20 20:35:21 +03:00
unsigned int parentid,
struct netlink_ext_ack *extack)
[PKT_SCHED]: Cleanup qdisc creation and alignment macros Adds qdisc_alloc() to share code between qdisc_create() and qdisc_create_dflt(). Hides the qdisc alignment behind macros and makes use of them. Signed-off-by: Thomas Graf <tgraf@suug.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2005-07-06 01:15:09 +04:00
{
struct Qdisc *sch;
[NET] SCHED: Fix whitespace errors. Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-02-09 17:25:16 +03:00
net: sch: api: add extack support in qdisc_create_dflt This patch adds extack support for the function qdisc_create_dflt which is a common used function in the tc subsystem. Callers which are interested in the receiving error can assign extack to get a more detailed information why qdisc_create_dflt failed. The function qdisc_create_dflt will also call an init callback which can fail by any per-qdisc specific handling. Cc: David Ahern <dsahern@gmail.com> Acked-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: Alexander Aring <aring@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-20 20:35:21 +03:00
if (!try_module_get(ops->owner)) {
NL_SET_ERR_MSG(extack, "Failed to increase module reference counter");
qdisc: fix a module refcount leak in qdisc_create_dflt() Should qdisc_alloc() fail, we must release the module refcount we got right before. Fixes: 6da7c8fcbcbd ("qdisc: allow setting default queuing discipline") Signed-off-by: Eric Dumazet <edumazet@google.com> Acked-by: John Fastabend <john.r.fastabend@intel.com> Acked-by: John Fastabend <john.r.fastabend@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-08-24 19:39:02 +03:00
return NULL;
net: sch: api: add extack support in qdisc_create_dflt This patch adds extack support for the function qdisc_create_dflt which is a common used function in the tc subsystem. Callers which are interested in the receiving error can assign extack to get a more detailed information why qdisc_create_dflt failed. The function qdisc_create_dflt will also call an init callback which can fail by any per-qdisc specific handling. Cc: David Ahern <dsahern@gmail.com> Acked-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: Alexander Aring <aring@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-20 20:35:21 +03:00
}
qdisc: allow setting default queuing discipline By default, the pfifo_fast queue discipline has been used by default for all devices. But we have better choices now. This patch allow setting the default queueing discipline with sysctl. This allows easy use of better queueing disciplines on all devices without having to use tc qdisc scripts. It is intended to allow an easy path for distributions to make fq_codel or sfq the default qdisc. This patch also makes pfifo_fast more of a first class qdisc, since it is now possible to manually override the default and explicitly use pfifo_fast. The behavior for systems who do not use the sysctl is unchanged, they still get pfifo_fast Also removes leftover random # in sysctl net core. Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Acked-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-08-28 03:19:08 +04:00
net: sch: api: add extack support in qdisc_create_dflt This patch adds extack support for the function qdisc_create_dflt which is a common used function in the tc subsystem. Callers which are interested in the receiving error can assign extack to get a more detailed information why qdisc_create_dflt failed. The function qdisc_create_dflt will also call an init callback which can fail by any per-qdisc specific handling. Cc: David Ahern <dsahern@gmail.com> Acked-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: Alexander Aring <aring@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-20 20:35:21 +03:00
sch = qdisc_alloc(dev_queue, ops, extack);
qdisc: fix a module refcount leak in qdisc_create_dflt() Should qdisc_alloc() fail, we must release the module refcount we got right before. Fixes: 6da7c8fcbcbd ("qdisc: allow setting default queuing discipline") Signed-off-by: Eric Dumazet <edumazet@google.com> Acked-by: John Fastabend <john.r.fastabend@intel.com> Acked-by: John Fastabend <john.r.fastabend@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-08-24 19:39:02 +03:00
if (IS_ERR(sch)) {
module_put(ops->owner);
return NULL;
}
[NET_SCHED]: Set parent classid in default qdiscs Set parent classids in default qdiscs to allow walking up the tree from outside the qdiscs. This is needed by the next patch. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2006-11-30 04:35:18 +03:00
sch->parent = parentid;
[PKT_SCHED]: Cleanup qdisc creation and alignment macros Adds qdisc_alloc() to share code between qdisc_create() and qdisc_create_dflt(). Hides the qdisc alignment behind macros and makes use of them. Signed-off-by: Thomas Graf <tgraf@suug.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2005-07-06 01:15:09 +04:00
net_sched: add a tracepoint for qdisc creation With this tracepoint, we could know when qdisc's are created, especially those default qdisc's. Sample output: tc-736 [001] ...1 56.230107: qdisc_create: dev=ens3 kind=pfifo parent=1:0 tc-736 [001] ...1 56.230113: qdisc_create: dev=ens3 kind=hfsc parent=ffff:ffff tc-738 [001] ...1 56.256816: qdisc_create: dev=ens3 kind=pfifo parent=1:100 tc-739 [001] ...1 56.267584: qdisc_create: dev=ens3 kind=pfifo parent=1:200 tc-740 [001] ...1 56.279649: qdisc_create: dev=ens3 kind=fq_codel parent=1:100 tc-741 [001] ...1 56.289996: qdisc_create: dev=ens3 kind=pfifo_fast parent=1:200 tc-745 [000] .N.1 111.687483: qdisc_create: dev=ens3 kind=ingress parent=ffff:fff1 Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Jiri Pirko <jiri@resnulli.us> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-05-27 07:35:25 +03:00
if (!ops->init || ops->init(sch, NULL, extack) == 0) {
trace_qdisc_create(ops, dev_queue->dev, parentid);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
return sch;
net_sched: add a tracepoint for qdisc creation With this tracepoint, we could know when qdisc's are created, especially those default qdisc's. Sample output: tc-736 [001] ...1 56.230107: qdisc_create: dev=ens3 kind=pfifo parent=1:0 tc-736 [001] ...1 56.230113: qdisc_create: dev=ens3 kind=hfsc parent=ffff:ffff tc-738 [001] ...1 56.256816: qdisc_create: dev=ens3 kind=pfifo parent=1:100 tc-739 [001] ...1 56.267584: qdisc_create: dev=ens3 kind=pfifo parent=1:200 tc-740 [001] ...1 56.279649: qdisc_create: dev=ens3 kind=fq_codel parent=1:100 tc-741 [001] ...1 56.289996: qdisc_create: dev=ens3 kind=pfifo_fast parent=1:200 tc-745 [000] .N.1 111.687483: qdisc_create: dev=ens3 kind=ingress parent=ffff:fff1 Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Jiri Pirko <jiri@resnulli.us> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-05-27 07:35:25 +03:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
net: sched: rename qdisc_destroy() to qdisc_put() Current implementation of qdisc_destroy() decrements Qdisc reference counter and only actually destroy Qdisc if reference counter value reached zero. Rename qdisc_destroy() to qdisc_put() in order for it to better describe the way in which this function currently implemented and used. Extract code that deallocates Qdisc into new private qdisc_destroy() function. It is intended to be shared between regular qdisc_put() and its unlocked version that is introduced in next patch in this series. Signed-off-by: Vlad Buslov <vladbu@mellanox.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-24 19:22:50 +03:00
qdisc_put(sch);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
return NULL;
}
[NET_SCHED]: Move EXPORT_SYMBOL next to exported symbol Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-23 09:10:23 +03:00
EXPORT_SYMBOL(qdisc_create_dflt);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
pkt_sched: Use qdisc_lock() on already sampled root qdisc. Based upon a bug report by Jeff Kirsher. Don't use qdisc_root_lock() in these cases as the root qdisc could have been changed, and we'd thus lock the wrong object. Tested by Emil S Tantilov who confirms that this seems to fix the problem. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-08-03 07:02:43 +04:00
/* Under qdisc_lock(qdisc) and BH! */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
void qdisc_reset(struct Qdisc *qdisc)
{
[NET]: Move Qdisc_class_ops and Qdisc_ops in appropriate sections. Qdisc_class_ops are const, and Qdisc_ops are mostly read. Using "const" and "__read_mostly" qualifiers helps to reduce false sharing. Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-11-14 12:44:41 +03:00
const struct Qdisc_ops *ops = qdisc->ops;
net: sched: explicit locking in gso_cpu fallback This work is preparing the qdisc layer to support egress lockless qdiscs. If we are running the egress qdisc lockless in the case we overrun the netdev, for whatever reason, the netdev returns a busy error code and the skb is parked on the gso_skb pointer. With many cores all hitting this case at once its possible to have multiple sk_buffs here so we turn gso_skb into a queue. This should be the edge case and if we see this frequently then the netdev/qdisc layer needs to back off. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:55:45 +03:00
struct sk_buff *skb, *tmp;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
net_sched: add tracepoints for qdisc_reset() and qdisc_destroy() Add two tracepoints for qdisc_reset() and qdisc_destroy() to track qdisc resetting and destroying. Sample output: tc-756 [000] ...3 138.355662: qdisc_reset: dev=ens3 kind=pfifo_fast parent=ffff:ffff handle=0:0 tc-756 [000] ...1 138.355720: qdisc_reset: dev=ens3 kind=pfifo_fast parent=ffff:ffff handle=0:0 tc-756 [000] ...1 138.355867: qdisc_reset: dev=ens3 kind=pfifo_fast parent=ffff:ffff handle=0:0 tc-756 [000] ...1 138.355930: qdisc_destroy: dev=ens3 kind=pfifo_fast parent=ffff:ffff handle=0:0 tc-757 [000] ...2 143.073780: qdisc_reset: dev=ens3 kind=fq_codel parent=ffff:ffff handle=8001:0 tc-757 [000] ...1 143.073878: qdisc_reset: dev=ens3 kind=fq_codel parent=ffff:ffff handle=8001:0 tc-757 [000] ...1 143.074114: qdisc_reset: dev=ens3 kind=fq_codel parent=ffff:ffff handle=8001:0 tc-757 [000] ...1 143.074228: qdisc_destroy: dev=ens3 kind=fq_codel parent=ffff:ffff handle=8001:0 Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Jiri Pirko <jiri@resnulli.us> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-05-27 07:35:24 +03:00
trace_qdisc_reset(qdisc);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
if (ops->reset)
ops->reset(qdisc);
pkt_sched: sch_generic: Kfree gso_skb in qdisc_reset() Since gso_skb is re-used for qdisc_peek_dequeued(), and this skb is counted in the qdisc->q.qlen, it has to be kfreed during qdisc_reset() when qlen is zeroed. With help from David S. Miller <davem@davemloft.net> Signed-off-by: Jarek Poplawski <jarkao2@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-03 13:52:50 +03:00
net: sched: explicit locking in gso_cpu fallback This work is preparing the qdisc layer to support egress lockless qdiscs. If we are running the egress qdisc lockless in the case we overrun the netdev, for whatever reason, the netdev returns a busy error code and the skb is parked on the gso_skb pointer. With many cores all hitting this case at once its possible to have multiple sk_buffs here so we turn gso_skb into a queue. This should be the edge case and if we see this frequently then the netdev/qdisc layer needs to back off. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:55:45 +03:00
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 05:44:21 +04:00
}
net: sched: explicit locking in gso_cpu fallback This work is preparing the qdisc layer to support egress lockless qdiscs. If we are running the egress qdisc lockless in the case we overrun the netdev, for whatever reason, the netdev returns a busy error code and the skb is parked on the gso_skb pointer. With many cores all hitting this case at once its possible to have multiple sk_buffs here so we turn gso_skb into a queue. This should be the edge case and if we see this frequently then the netdev/qdisc layer needs to back off. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:55:45 +03:00
net: sched: use skb list for skb_bad_tx Similar to how gso is handled use skb list for skb_bad_tx this is required with lockless qdiscs because we may have multiple cores attempting to push skbs into skb_bad_tx concurrently Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:56:23 +03:00
skb_queue_walk_safe(&qdisc->skb_bad_txq, skb, tmp) {
__skb_unlink(skb, &qdisc->skb_bad_txq);
kfree_skb_list(skb);
}
net_sched: generalize bulk dequeue When qdisc bulk dequeue was added in linux-3.18 (commit 5772e9a3463b "qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE"), it was constrained to some specific qdiscs. With some extra care, we can extend this to all qdiscs, so that typical traffic shaping solutions can benefit from small batches (8 packets in this patch). For example, HTB is often used on some multi queue device. And bonding/team are multi queue devices... Idea is to bulk-dequeue packets mapping to the same transmit queue. This brings between 35 and 80 % performance increase in HTB setup under pressure on a bonding setup : 1) NUMA node contention : 610,000 pps -> 1,110,000 pps 2) No node contention : 1,380,000 pps -> 1,930,000 pps Now we should work to add batches on the enqueue() side ;) Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: John Fastabend <john.r.fastabend@intel.com> Cc: Jesper Dangaard Brouer <brouer@redhat.com> Cc: Hannes Frederic Sowa <hannes@stressinduktion.org> Cc: Florian Westphal <fw@strlen.de> Cc: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-06-22 09:16:52 +03:00
qdisc->q.qlen = 0;
net_sched: always reset qdisc backlog in qdisc_reset() SKB stored in qdisc->gso_skb also counted into backlog. Some qdiscs don't reset backlog to zero in ->reset(), for example sfq just dequeue and free all queued skb. Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru> Fixes: 2ccccf5fb43f ("net_sched: update hierarchical backlog too") Signed-off-by: David S. Miller <davem@davemloft.net>
2017-09-20 15:45:36 +03:00
qdisc->qstats.backlog = 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
}
[NET_SCHED]: Move EXPORT_SYMBOL next to exported symbol Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-23 09:10:23 +03:00
EXPORT_SYMBOL(qdisc_reset);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
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 01:12:09 +03:00
void qdisc_free(struct Qdisc *qdisc)
gen_estimator: deadlock fix One of my test machine got a deadlock during "tc" sessions, adding/deleting classes & filters, using traffic estimators. After some analysis, I believe we have a potential use after free case in est_timer() : spin_lock(e->stats_lock); << HERE >> read_lock(&est_lock); if (e->bstats == NULL) << TEST >> goto skip; Test is done a bit late, because after estimator is killed, and before rcu grace period elapsed, we might already have freed/reuse memory where e->stats_locks points to (some qdisc->q.lock) A possible fix is to respect a rcu grace period at Qdisc dismantle time. On 64bit, sizeof(struct Qdisc) is exactly 192 bytes. Adding 16 bytes to it (for struct rcu_head) is a problem because it might change performance, given QDISC_ALIGNTO is 32 bytes. This is why I also change QDISC_ALIGNTO to 64 bytes, to satisfy most current alignment requirements. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-03-31 11:06:04 +04:00
{
net: sched: fix missing free per cpu on qstats When a qdisc is using per cpu stats (currently just the ingress qdisc) only the bstats are being freed. This also free's the qstats. Fixes: b0ab6f92752b9f9d8 ("net: sched: enable per cpu qstats") Signed-off-by: John Fastabend <john.r.fastabend@intel.com> Acked-by: Eric Dumazet <edumazet@google.com> Acked-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-01-05 20:11:36 +03:00
if (qdisc_is_percpu_stats(qdisc)) {
net: sched: make bstats per cpu and estimator RCU safe In order to run qdisc's without locking statistics and estimators need to be handled correctly. To resolve bstats make the statistics per cpu. And because this is only needed for qdiscs that are running without locks which is not the case for most qdiscs in the near future only create percpu stats when qdiscs set the TCQ_F_CPUSTATS flag. Next because estimators use the bstats to calculate packets per second and bytes per second the estimator code paths are updated to use the per cpu statistics. Signed-off-by: John Fastabend <john.r.fastabend@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-28 22:52:56 +04:00
free_percpu(qdisc->cpu_bstats);
net: sched: fix missing free per cpu on qstats When a qdisc is using per cpu stats (currently just the ingress qdisc) only the bstats are being freed. This also free's the qstats. Fixes: b0ab6f92752b9f9d8 ("net: sched: enable per cpu qstats") Signed-off-by: John Fastabend <john.r.fastabend@intel.com> Acked-by: Eric Dumazet <edumazet@google.com> Acked-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-01-05 20:11:36 +03:00
free_percpu(qdisc->cpu_qstats);
}
net: sched: make bstats per cpu and estimator RCU safe In order to run qdisc's without locking statistics and estimators need to be handled correctly. To resolve bstats make the statistics per cpu. And because this is only needed for qdiscs that are running without locks which is not the case for most qdiscs in the near future only create percpu stats when qdiscs set the TCQ_F_CPUSTATS flag. Next because estimators use the bstats to calculate packets per second and bytes per second the estimator code paths are updated to use the per cpu statistics. Signed-off-by: John Fastabend <john.r.fastabend@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-28 22:52:56 +04:00
net/sched: get rid of qdisc->padded kmalloc() of sufficiently big portion of memory is cache-aligned in regular conditions. If some debugging options are used, there is no reason qdisc structures would need 64-byte alignment if most other kernel structures are not aligned. This get rid of QDISC_ALIGN and QDISC_ALIGNTO. Addition of privdata field will help implementing the reverse of qdisc_priv() and documents where the private data is. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Allen Pais <allen.lkml@gmail.com> Acked-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2020-10-07 19:51:11 +03:00
kfree(qdisc);
gen_estimator: deadlock fix One of my test machine got a deadlock during "tc" sessions, adding/deleting classes & filters, using traffic estimators. After some analysis, I believe we have a potential use after free case in est_timer() : spin_lock(e->stats_lock); << HERE >> read_lock(&est_lock); if (e->bstats == NULL) << TEST >> goto skip; Test is done a bit late, because after estimator is killed, and before rcu grace period elapsed, we might already have freed/reuse memory where e->stats_locks points to (some qdisc->q.lock) A possible fix is to respect a rcu grace period at Qdisc dismantle time. On 64bit, sizeof(struct Qdisc) is exactly 192 bytes. Adding 16 bytes to it (for struct rcu_head) is a problem because it might change performance, given QDISC_ALIGNTO is 32 bytes. This is why I also change QDISC_ALIGNTO to 64 bytes, to satisfy most current alignment requirements. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-03-31 11:06:04 +04:00
}
net: sched: make function qdisc_free_cb() static Fixes the following sparse warning: net/sched/sch_generic.c:944:6: warning: symbol 'qdisc_free_cb' was not declared. Should it be static? Fixes: 3a7d0d07a386 ("net: sched: extend Qdisc with rcu") Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-27 17:47:56 +03:00
static void qdisc_free_cb(struct rcu_head *head)
net: sched: extend Qdisc with rcu Currently, Qdisc API functions assume that users have rtnl lock taken. To implement rtnl unlocked classifiers update interface, Qdisc API must be extended with functions that do not require rtnl lock. Extend Qdisc structure with rcu. Implement special version of put function qdisc_put_unlocked() that is called without rtnl lock taken. This function only takes rtnl lock if Qdisc reference counter reached zero and is intended to be used as optimization. Signed-off-by: Vlad Buslov <vladbu@mellanox.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-24 19:22:51 +03:00
{
struct Qdisc *q = container_of(head, struct Qdisc, rcu);
qdisc_free(q);
}
net: sched: rename qdisc_destroy() to qdisc_put() Current implementation of qdisc_destroy() decrements Qdisc reference counter and only actually destroy Qdisc if reference counter value reached zero. Rename qdisc_destroy() to qdisc_put() in order for it to better describe the way in which this function currently implemented and used. Extract code that deallocates Qdisc into new private qdisc_destroy() function. It is intended to be shared between regular qdisc_put() and its unlocked version that is introduced in next patch in this series. Signed-off-by: Vlad Buslov <vladbu@mellanox.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-24 19:22:50 +03:00
static void qdisc_destroy(struct Qdisc *qdisc)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
{
pkt_sched: Perform bulk of qdisc destruction in RCU. This allows less strict control of access to the qdisc attached to a netdev_queue. It is even allowed to enqueue into a qdisc which is in the process of being destroyed. The RCU handler will toss out those packets. We will need this to handle sharing of a qdisc amongst multiple TX queues. In such a setup the lock has to be shared, so will be inside of the qdisc itself. At which point the netdev_queue lock cannot be used to hard synchronize access to the ->qdisc pointer. One operation we have to keep inside of qdisc_destroy() is the list deletion. It is the only piece of state visible after the RCU quiesce period, so we have to undo it early and under the appropriate locking. The operations in the RCU handler do not need any looking because the qdisc tree is no longer visible to anything at that point. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:47:45 +04:00
const struct Qdisc_ops *ops = qdisc->ops;
pkt_sched: Fix build with NET_SCHED disabled. The stab bits can't be referenced uniless the full packet scheduler layer is enabled. Reported by Stephen Rothwell. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-21 05:13:01 +04:00
#ifdef CONFIG_NET_SCHED
net: sched: convert qdisc linked list to hashtable Convert the per-device linked list into a hashtable. The primary motivation for this change is that currently, we're not tracking all the qdiscs in hierarchy (e.g. excluding default qdiscs), as the lookup performed over the linked list by qdisc_match_from_root() is rather expensive. The ultimate goal is to get rid of hidden qdiscs completely, which will bring much more determinism in user experience. Reviewed-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-08-10 12:05:15 +03:00
qdisc_hash_del(qdisc);
pkt_sched: Fix qdisc list locking Since some qdiscs call qdisc_tree_decrease_qlen() (so qdisc_lookup()) without rtnl_lock(), adding and deleting from a qdisc list needs additional locking. This patch adds global spinlock qdisc_list_lock and wrapper functions for modifying the list. It is considered as a temporary solution until hfsc_dequeue(), netem_dequeue() and tbf_dequeue() (or qdisc_tree_decrease_qlen()) are redone. With feedback from Herbert Xu and David S. Miller. Signed-off-by: Jarek Poplawski <jarkao2@gmail.com> Acked-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-08-22 14:24:05 +04:00
net_sched: RCU conversion of stab This patch converts stab qdisc management to RCU, so that we can perform the qdisc_calculate_pkt_len() call before getting qdisc lock. This shortens the lock's held time in __dev_xmit_skb(). This permits more qdiscs to get TCQ_F_CAN_BYPASS status, avoiding lot of cache misses and so reducing latencies. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> CC: Patrick McHardy <kaber@trash.net> CC: Jesper Dangaard Brouer <hawk@diku.dk> CC: Jarek Poplawski <jarkao2@gmail.com> CC: Jamal Hadi Salim <hadi@cyberus.ca> CC: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-01-20 06:48:19 +03:00
qdisc_put_stab(rtnl_dereference(qdisc->stab));
pkt_sched: Fix build with NET_SCHED disabled. The stab bits can't be referenced uniless the full packet scheduler layer is enabled. Reported by Stephen Rothwell. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-21 05:13:01 +04:00
#endif
net_sched: gen_estimator: complete rewrite of rate estimators 1) Old code was hard to maintain, due to complex lock chains. (We probably will be able to remove some kfree_rcu() in callers) 2) Using a single timer to update all estimators does not scale. 3) Code was buggy on 32bit kernel (WRITE_ONCE() on 64bit quantity is not supposed to work well) In this rewrite : - I removed the RB tree that had to be scanned in gen_estimator_active(). qdisc dumps should be much faster. - Each estimator has its own timer. - Estimations are maintained in net_rate_estimator structure, instead of dirtying the qdisc. Minor, but part of the simplification. - Reading the estimator uses RCU and a seqcount to provide proper support for 32bit kernels. - We reduce memory need when estimators are not used, since we store a pointer, instead of the bytes/packets counters. - xt_rateest_mt() no longer has to grab a spinlock. (In the future, xt_rateest_tg() could be switched to per cpu counters) Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-12-04 20:48:16 +03:00
gen_kill_estimator(&qdisc->rate_est);
net_sched: use qdisc_reset() in qdisc_destroy() qdisc_destroy() calls ops->reset() and cleans up qdisc->gso_skb and qdisc->skb_bad_txq, these are nearly same with qdisc_reset(), so just call it directly, and cosolidate the code for the next patch. Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Jiri Pirko <jiri@resnulli.us> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-05-27 07:35:23 +03:00
qdisc_reset(qdisc);
pkt_sched: Perform bulk of qdisc destruction in RCU. This allows less strict control of access to the qdisc attached to a netdev_queue. It is even allowed to enqueue into a qdisc which is in the process of being destroyed. The RCU handler will toss out those packets. We will need this to handle sharing of a qdisc amongst multiple TX queues. In such a setup the lock has to be shared, so will be inside of the qdisc itself. At which point the netdev_queue lock cannot be used to hard synchronize access to the ->qdisc pointer. One operation we have to keep inside of qdisc_destroy() is the list deletion. It is the only piece of state visible after the RCU quiesce period, so we have to undo it early and under the appropriate locking. The operations in the RCU handler do not need any looking because the qdisc tree is no longer visible to anything at that point. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:47:45 +04:00
if (ops->destroy)
ops->destroy(qdisc);
module_put(ops->owner);
net/sched: add net device refcount tracker to struct Qdisc Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-12-05 07:22:13 +03:00
dev_put_track(qdisc_dev(qdisc), &qdisc->dev_tracker);
pkt_sched: Perform bulk of qdisc destruction in RCU. This allows less strict control of access to the qdisc attached to a netdev_queue. It is even allowed to enqueue into a qdisc which is in the process of being destroyed. The RCU handler will toss out those packets. We will need this to handle sharing of a qdisc amongst multiple TX queues. In such a setup the lock has to be shared, so will be inside of the qdisc itself. At which point the netdev_queue lock cannot be used to hard synchronize access to the ->qdisc pointer. One operation we have to keep inside of qdisc_destroy() is the list deletion. It is the only piece of state visible after the RCU quiesce period, so we have to undo it early and under the appropriate locking. The operations in the RCU handler do not need any looking because the qdisc tree is no longer visible to anything at that point. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:47:45 +04:00
net_sched: add tracepoints for qdisc_reset() and qdisc_destroy() Add two tracepoints for qdisc_reset() and qdisc_destroy() to track qdisc resetting and destroying. Sample output: tc-756 [000] ...3 138.355662: qdisc_reset: dev=ens3 kind=pfifo_fast parent=ffff:ffff handle=0:0 tc-756 [000] ...1 138.355720: qdisc_reset: dev=ens3 kind=pfifo_fast parent=ffff:ffff handle=0:0 tc-756 [000] ...1 138.355867: qdisc_reset: dev=ens3 kind=pfifo_fast parent=ffff:ffff handle=0:0 tc-756 [000] ...1 138.355930: qdisc_destroy: dev=ens3 kind=pfifo_fast parent=ffff:ffff handle=0:0 tc-757 [000] ...2 143.073780: qdisc_reset: dev=ens3 kind=fq_codel parent=ffff:ffff handle=8001:0 tc-757 [000] ...1 143.073878: qdisc_reset: dev=ens3 kind=fq_codel parent=ffff:ffff handle=8001:0 tc-757 [000] ...1 143.074114: qdisc_reset: dev=ens3 kind=fq_codel parent=ffff:ffff handle=8001:0 tc-757 [000] ...1 143.074228: qdisc_destroy: dev=ens3 kind=fq_codel parent=ffff:ffff handle=8001:0 Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Jiri Pirko <jiri@resnulli.us> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-05-27 07:35:24 +03:00
trace_qdisc_destroy(qdisc);
net: sched: extend Qdisc with rcu Currently, Qdisc API functions assume that users have rtnl lock taken. To implement rtnl unlocked classifiers update interface, Qdisc API must be extended with functions that do not require rtnl lock. Extend Qdisc structure with rcu. Implement special version of put function qdisc_put_unlocked() that is called without rtnl lock taken. This function only takes rtnl lock if Qdisc reference counter reached zero and is intended to be used as optimization. Signed-off-by: Vlad Buslov <vladbu@mellanox.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-24 19:22:51 +03:00
call_rcu(&qdisc->rcu, qdisc_free_cb);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
}
net: sched: rename qdisc_destroy() to qdisc_put() Current implementation of qdisc_destroy() decrements Qdisc reference counter and only actually destroy Qdisc if reference counter value reached zero. Rename qdisc_destroy() to qdisc_put() in order for it to better describe the way in which this function currently implemented and used. Extract code that deallocates Qdisc into new private qdisc_destroy() function. It is intended to be shared between regular qdisc_put() and its unlocked version that is introduced in next patch in this series. Signed-off-by: Vlad Buslov <vladbu@mellanox.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-24 19:22:50 +03:00
void qdisc_put(struct Qdisc *qdisc)
{
net_sched: let qdisc_put() accept NULL pointer When tcf_block_get() fails in sfb_init(), q->qdisc is still a NULL pointer which leads to a crash in sfb_destroy(). Similar for sch_dsmark. Instead of fixing each separately, Linus suggested to just accept NULL pointer in qdisc_put(), which would make callers easier. (For sch_dsmark, the bug probably exists long before commit 6529eaba33f0.) Fixes: 6529eaba33f0 ("net: sched: introduce tcf block infractructure") Reported-by: syzbot+d5870a903591faaca4ae@syzkaller.appspotmail.com Suggested-by: Linus Torvalds <torvalds@linux-foundation.org> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Jiri Pirko <jiri@resnulli.us> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-09-12 20:22:30 +03:00
if (!qdisc)
return;
net: sched: rename qdisc_destroy() to qdisc_put() Current implementation of qdisc_destroy() decrements Qdisc reference counter and only actually destroy Qdisc if reference counter value reached zero. Rename qdisc_destroy() to qdisc_put() in order for it to better describe the way in which this function currently implemented and used. Extract code that deallocates Qdisc into new private qdisc_destroy() function. It is intended to be shared between regular qdisc_put() and its unlocked version that is introduced in next patch in this series. Signed-off-by: Vlad Buslov <vladbu@mellanox.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-24 19:22:50 +03:00
if (qdisc->flags & TCQ_F_BUILTIN ||
!refcount_dec_and_test(&qdisc->refcnt))
return;
qdisc_destroy(qdisc);
}
EXPORT_SYMBOL(qdisc_put);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
net: sched: extend Qdisc with rcu Currently, Qdisc API functions assume that users have rtnl lock taken. To implement rtnl unlocked classifiers update interface, Qdisc API must be extended with functions that do not require rtnl lock. Extend Qdisc structure with rcu. Implement special version of put function qdisc_put_unlocked() that is called without rtnl lock taken. This function only takes rtnl lock if Qdisc reference counter reached zero and is intended to be used as optimization. Signed-off-by: Vlad Buslov <vladbu@mellanox.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-24 19:22:51 +03:00
/* 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);
net_sched: move dev_graft_qdisc() to sch_generic.c It will be used in a following patch by the multiqueue qdisc. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-04 10:41:20 +04:00
/* 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 11:06:09 +03:00
EXPORT_SYMBOL(dev_graft_qdisc);
net_sched: move dev_graft_qdisc() to sch_generic.c It will be used in a following patch by the multiqueue qdisc. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-04 10:41:20 +04:00
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
static void attach_one_default_qdisc(struct net_device *dev,
struct netdev_queue *dev_queue,
void *_unused)
{
net: sched: simplify attach_one_default_qdisc() Now that noqueue qdisc can be attached just like any other qdisc, no special treatment is necessary anymore when attaching it as default qdisc. This change has the added benefit that 'tc qdisc show' prints noqueue instead of nothing for devices defaulting to noqueue. Signed-off-by: Phil Sutter <phil@nwl.cc> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-27 22:21:39 +03:00
struct Qdisc *qdisc;
const struct Qdisc_ops *ops = default_qdisc_ops;
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
net: sched: simplify attach_one_default_qdisc() Now that noqueue qdisc can be attached just like any other qdisc, no special treatment is necessary anymore when attaching it as default qdisc. This change has the added benefit that 'tc qdisc show' prints noqueue instead of nothing for devices defaulting to noqueue. Signed-off-by: Phil Sutter <phil@nwl.cc> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-27 22:21:39 +03:00
if (dev->priv_flags & IFF_NO_QUEUE)
ops = &noqueue_qdisc_ops;
net: sch_generic: Use pfifo_fast as fallback scheduler for CAN hardware There is networking hardware that isn't based on Ethernet for layers 1 and 2. For example CAN. CAN is a multi-master serial bus standard for connecting Electronic Control Units [ECUs] also known as nodes. A frame on the CAN bus carries up to 8 bytes of payload. Frame corruption is detected by a CRC. However frame loss due to corruption is possible, but a quite unusual phenomenon. While fq_codel works great for TCP/IP, it doesn't for CAN. There are a lot of legacy protocols on top of CAN, which are not build with flow control or high CAN frame drop rates in mind. When using fq_codel, as soon as the queue reaches a certain delay based length, skbs from the head of the queue are silently dropped. Silently meaning that the user space using a send() or similar syscall doesn't get an error. However TCP's flow control algorithm will detect dropped packages and adjust the bandwidth accordingly. When using fq_codel and sending raw frames over CAN, which is the common use case, the user space thinks the package has been sent without problems, because send() returned without an error. pfifo_fast will drop skbs, if the queue length exceeds the maximum. But with this scheduler the skbs at the tail are dropped, an error (-ENOBUFS) is propagated to user space. So that the user space can slow down the package generation. On distributions, where fq_codel is made default via CONFIG_DEFAULT_NET_SCH during compile time, or set default during runtime with sysctl net.core.default_qdisc (see [1]), we get a bad user experience. In my test case with pfifo_fast, I can transfer thousands of million CAN frames without a frame drop. On the other hand with fq_codel there is more then one lost CAN frame per thousand frames. As pointed out fq_codel is not suited for CAN hardware, so this patch changes attach_one_default_qdisc() to use pfifo_fast for "ARPHRD_CAN" network devices. During transition of a netdev from down to up state the default queuing discipline is attached by attach_default_qdiscs() with the help of attach_one_default_qdisc(). This patch modifies attach_one_default_qdisc() to attach the pfifo_fast (pfifo_fast_ops) if the network device type is "ARPHRD_CAN". [1] https://github.com/systemd/systemd/issues/9194 Suggested-by: Marc Kleine-Budde <mkl@pengutronix.de> Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de> Signed-off-by: Vincent Prince <vincent.prince.fr@gmail.com> Acked-by: Dave Taht <dave.taht@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-10-23 16:44:20 +03:00
else if(dev->type == ARPHRD_CAN)
ops = &pfifo_fast_ops;
net: sched: simplify attach_one_default_qdisc() Now that noqueue qdisc can be attached just like any other qdisc, no special treatment is necessary anymore when attaching it as default qdisc. This change has the added benefit that 'tc qdisc show' prints noqueue instead of nothing for devices defaulting to noqueue. Signed-off-by: Phil Sutter <phil@nwl.cc> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-27 22:21:39 +03:00
net: sch: api: add extack support in qdisc_create_dflt This patch adds extack support for the function qdisc_create_dflt which is a common used function in the tc subsystem. Callers which are interested in the receiving error can assign extack to get a more detailed information why qdisc_create_dflt failed. The function qdisc_create_dflt will also call an init callback which can fail by any per-qdisc specific handling. Cc: David Ahern <dsahern@gmail.com> Acked-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: Alexander Aring <aring@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-20 20:35:21 +03:00
qdisc = qdisc_create_dflt(dev_queue, ops, TC_H_ROOT, NULL);
net: sched: fallback to qdisc noqueue if default qdisc setup fail Currently if the default qdisc setup/init fails, the device ends up with qdisc "noop", which causes all TX packets to get dropped. With the introduction of sysctl net/core/default_qdisc it is possible to change the default qdisc to be more advanced, which opens for the possibility that Qdisc_ops->init() can fail. This patch detect these kind of failures, and choose to fallback to qdisc "noqueue", which is so simple that its init call will not fail. This allows the interface to continue functioning. V2: As this also captures memory failures, which are transient, the device is not kept in IFF_NO_QUEUE state. This allows the net_device to retry to default qdisc assignment. Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-04-30 14:42:22 +03:00
if (!qdisc)
net: sched: simplify attach_one_default_qdisc() Now that noqueue qdisc can be attached just like any other qdisc, no special treatment is necessary anymore when attaching it as default qdisc. This change has the added benefit that 'tc qdisc show' prints noqueue instead of nothing for devices defaulting to noqueue. Signed-off-by: Phil Sutter <phil@nwl.cc> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-27 22:21:39 +03:00
return;
net: sched: fallback to qdisc noqueue if default qdisc setup fail Currently if the default qdisc setup/init fails, the device ends up with qdisc "noop", which causes all TX packets to get dropped. With the introduction of sysctl net/core/default_qdisc it is possible to change the default qdisc to be more advanced, which opens for the possibility that Qdisc_ops->init() can fail. This patch detect these kind of failures, and choose to fallback to qdisc "noqueue", which is so simple that its init call will not fail. This allows the interface to continue functioning. V2: As this also captures memory failures, which are transient, the device is not kept in IFF_NO_QUEUE state. This allows the net_device to retry to default qdisc assignment. Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-04-30 14:42:22 +03:00
net: sched: simplify attach_one_default_qdisc() Now that noqueue qdisc can be attached just like any other qdisc, no special treatment is necessary anymore when attaching it as default qdisc. This change has the added benefit that 'tc qdisc show' prints noqueue instead of nothing for devices defaulting to noqueue. Signed-off-by: Phil Sutter <phil@nwl.cc> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-27 22:21:39 +03:00
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 07:08:51 +03:00
qdisc->flags |= TCQ_F_ONETXQUEUE | TCQ_F_NOPARENT;
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
dev_queue->qdisc_sleeping = qdisc;
}
net_sched: add classful multiqueue dummy scheduler This patch adds a classful dummy scheduler which can be used as root qdisc for multiqueue devices and exposes each device queue as a child class. This allows to address queues individually and graft them similar to regular classes. Additionally it presents an accumulated view of the statistics of all real root qdiscs in the dummy root. Two new callbacks are added to the qdisc_ops and qdisc_class_ops: - cl_ops->select_queue selects the tx queue number for new child classes. - qdisc_ops->attach() overrides root qdisc device grafting to attach non-shared qdiscs to the queues. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-06 12:58:51 +04:00
static void attach_default_qdiscs(struct net_device *dev)
{
struct netdev_queue *txq;
struct Qdisc *qdisc;
txq = netdev_get_tx_queue(dev, 0);
net: sch_generic: react upon IFF_NO_QUEUE flag Handle IFF_NO_QUEUE as alternative to tx_queue_len being zero. Signed-off-by: Phil Sutter <phil@nwl.cc> Acked-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-13 20:01:07 +03:00
if (!netif_is_multiqueue(dev) ||
dev->priv_flags & IFF_NO_QUEUE) {
net_sched: add classful multiqueue dummy scheduler This patch adds a classful dummy scheduler which can be used as root qdisc for multiqueue devices and exposes each device queue as a child class. This allows to address queues individually and graft them similar to regular classes. Additionally it presents an accumulated view of the statistics of all real root qdiscs in the dummy root. Two new callbacks are added to the qdisc_ops and qdisc_class_ops: - cl_ops->select_queue selects the tx queue number for new child classes. - qdisc_ops->attach() overrides root qdisc device grafting to attach non-shared qdiscs to the queues. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-06 12:58:51 +04:00
netdev_for_each_tx_queue(dev, attach_one_default_qdisc, NULL);
net_sched: add __rcu annotation to netdev->qdisc syzbot found a data-race [1] which lead me to add __rcu annotations to netdev->qdisc, and proper accessors to get LOCKDEP support. [1] BUG: KCSAN: data-race in dev_activate / qdisc_lookup_rcu write to 0xffff888168ad6410 of 8 bytes by task 13559 on cpu 1: attach_default_qdiscs net/sched/sch_generic.c:1167 [inline] dev_activate+0x2ed/0x8f0 net/sched/sch_generic.c:1221 __dev_open+0x2e9/0x3a0 net/core/dev.c:1416 __dev_change_flags+0x167/0x3f0 net/core/dev.c:8139 rtnl_configure_link+0xc2/0x150 net/core/rtnetlink.c:3150 __rtnl_newlink net/core/rtnetlink.c:3489 [inline] rtnl_newlink+0xf4d/0x13e0 net/core/rtnetlink.c:3529 rtnetlink_rcv_msg+0x745/0x7e0 net/core/rtnetlink.c:5594 netlink_rcv_skb+0x14e/0x250 net/netlink/af_netlink.c:2494 rtnetlink_rcv+0x18/0x20 net/core/rtnetlink.c:5612 netlink_unicast_kernel net/netlink/af_netlink.c:1317 [inline] netlink_unicast+0x602/0x6d0 net/netlink/af_netlink.c:1343 netlink_sendmsg+0x728/0x850 net/netlink/af_netlink.c:1919 sock_sendmsg_nosec net/socket.c:705 [inline] sock_sendmsg net/socket.c:725 [inline] ____sys_sendmsg+0x39a/0x510 net/socket.c:2413 ___sys_sendmsg net/socket.c:2467 [inline] __sys_sendmsg+0x195/0x230 net/socket.c:2496 __do_sys_sendmsg net/socket.c:2505 [inline] __se_sys_sendmsg net/socket.c:2503 [inline] __x64_sys_sendmsg+0x42/0x50 net/socket.c:2503 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae read to 0xffff888168ad6410 of 8 bytes by task 13560 on cpu 0: qdisc_lookup_rcu+0x30/0x2e0 net/sched/sch_api.c:323 __tcf_qdisc_find+0x74/0x3a0 net/sched/cls_api.c:1050 tc_del_tfilter+0x1c7/0x1350 net/sched/cls_api.c:2211 rtnetlink_rcv_msg+0x5ba/0x7e0 net/core/rtnetlink.c:5585 netlink_rcv_skb+0x14e/0x250 net/netlink/af_netlink.c:2494 rtnetlink_rcv+0x18/0x20 net/core/rtnetlink.c:5612 netlink_unicast_kernel net/netlink/af_netlink.c:1317 [inline] netlink_unicast+0x602/0x6d0 net/netlink/af_netlink.c:1343 netlink_sendmsg+0x728/0x850 net/netlink/af_netlink.c:1919 sock_sendmsg_nosec net/socket.c:705 [inline] sock_sendmsg net/socket.c:725 [inline] ____sys_sendmsg+0x39a/0x510 net/socket.c:2413 ___sys_sendmsg net/socket.c:2467 [inline] __sys_sendmsg+0x195/0x230 net/socket.c:2496 __do_sys_sendmsg net/socket.c:2505 [inline] __se_sys_sendmsg net/socket.c:2503 [inline] __x64_sys_sendmsg+0x42/0x50 net/socket.c:2503 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae value changed: 0xffffffff85dee080 -> 0xffff88815d96ec00 Reported by Kernel Concurrency Sanitizer on: CPU: 0 PID: 13560 Comm: syz-executor.2 Not tainted 5.17.0-rc3-syzkaller-00116-gf1baf68e1383-dirty #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Fixes: 470502de5bdb ("net: sched: unlock rules update API") Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Vlad Buslov <vladbu@mellanox.com> Reported-by: syzbot <syzkaller@googlegroups.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: Jiri Pirko <jiri@resnulli.us> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-02-11 23:06:23 +03:00
qdisc = txq->qdisc_sleeping;
rcu_assign_pointer(dev->qdisc, qdisc);
qdisc_refcount_inc(qdisc);
net_sched: add classful multiqueue dummy scheduler This patch adds a classful dummy scheduler which can be used as root qdisc for multiqueue devices and exposes each device queue as a child class. This allows to address queues individually and graft them similar to regular classes. Additionally it presents an accumulated view of the statistics of all real root qdiscs in the dummy root. Two new callbacks are added to the qdisc_ops and qdisc_class_ops: - cl_ops->select_queue selects the tx queue number for new child classes. - qdisc_ops->attach() overrides root qdisc device grafting to attach non-shared qdiscs to the queues. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-06 12:58:51 +04:00
} else {
net: sch: api: add extack support in qdisc_create_dflt This patch adds extack support for the function qdisc_create_dflt which is a common used function in the tc subsystem. Callers which are interested in the receiving error can assign extack to get a more detailed information why qdisc_create_dflt failed. The function qdisc_create_dflt will also call an init callback which can fail by any per-qdisc specific handling. Cc: David Ahern <dsahern@gmail.com> Acked-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: Alexander Aring <aring@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-20 20:35:21 +03:00
qdisc = qdisc_create_dflt(txq, &mq_qdisc_ops, TC_H_ROOT, NULL);
net_sched: add classful multiqueue dummy scheduler This patch adds a classful dummy scheduler which can be used as root qdisc for multiqueue devices and exposes each device queue as a child class. This allows to address queues individually and graft them similar to regular classes. Additionally it presents an accumulated view of the statistics of all real root qdiscs in the dummy root. Two new callbacks are added to the qdisc_ops and qdisc_class_ops: - cl_ops->select_queue selects the tx queue number for new child classes. - qdisc_ops->attach() overrides root qdisc device grafting to attach non-shared qdiscs to the queues. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-06 12:58:51 +04:00
if (qdisc) {
net_sched: add __rcu annotation to netdev->qdisc syzbot found a data-race [1] which lead me to add __rcu annotations to netdev->qdisc, and proper accessors to get LOCKDEP support. [1] BUG: KCSAN: data-race in dev_activate / qdisc_lookup_rcu write to 0xffff888168ad6410 of 8 bytes by task 13559 on cpu 1: attach_default_qdiscs net/sched/sch_generic.c:1167 [inline] dev_activate+0x2ed/0x8f0 net/sched/sch_generic.c:1221 __dev_open+0x2e9/0x3a0 net/core/dev.c:1416 __dev_change_flags+0x167/0x3f0 net/core/dev.c:8139 rtnl_configure_link+0xc2/0x150 net/core/rtnetlink.c:3150 __rtnl_newlink net/core/rtnetlink.c:3489 [inline] rtnl_newlink+0xf4d/0x13e0 net/core/rtnetlink.c:3529 rtnetlink_rcv_msg+0x745/0x7e0 net/core/rtnetlink.c:5594 netlink_rcv_skb+0x14e/0x250 net/netlink/af_netlink.c:2494 rtnetlink_rcv+0x18/0x20 net/core/rtnetlink.c:5612 netlink_unicast_kernel net/netlink/af_netlink.c:1317 [inline] netlink_unicast+0x602/0x6d0 net/netlink/af_netlink.c:1343 netlink_sendmsg+0x728/0x850 net/netlink/af_netlink.c:1919 sock_sendmsg_nosec net/socket.c:705 [inline] sock_sendmsg net/socket.c:725 [inline] ____sys_sendmsg+0x39a/0x510 net/socket.c:2413 ___sys_sendmsg net/socket.c:2467 [inline] __sys_sendmsg+0x195/0x230 net/socket.c:2496 __do_sys_sendmsg net/socket.c:2505 [inline] __se_sys_sendmsg net/socket.c:2503 [inline] __x64_sys_sendmsg+0x42/0x50 net/socket.c:2503 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae read to 0xffff888168ad6410 of 8 bytes by task 13560 on cpu 0: qdisc_lookup_rcu+0x30/0x2e0 net/sched/sch_api.c:323 __tcf_qdisc_find+0x74/0x3a0 net/sched/cls_api.c:1050 tc_del_tfilter+0x1c7/0x1350 net/sched/cls_api.c:2211 rtnetlink_rcv_msg+0x5ba/0x7e0 net/core/rtnetlink.c:5585 netlink_rcv_skb+0x14e/0x250 net/netlink/af_netlink.c:2494 rtnetlink_rcv+0x18/0x20 net/core/rtnetlink.c:5612 netlink_unicast_kernel net/netlink/af_netlink.c:1317 [inline] netlink_unicast+0x602/0x6d0 net/netlink/af_netlink.c:1343 netlink_sendmsg+0x728/0x850 net/netlink/af_netlink.c:1919 sock_sendmsg_nosec net/socket.c:705 [inline] sock_sendmsg net/socket.c:725 [inline] ____sys_sendmsg+0x39a/0x510 net/socket.c:2413 ___sys_sendmsg net/socket.c:2467 [inline] __sys_sendmsg+0x195/0x230 net/socket.c:2496 __do_sys_sendmsg net/socket.c:2505 [inline] __se_sys_sendmsg net/socket.c:2503 [inline] __x64_sys_sendmsg+0x42/0x50 net/socket.c:2503 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae value changed: 0xffffffff85dee080 -> 0xffff88815d96ec00 Reported by Kernel Concurrency Sanitizer on: CPU: 0 PID: 13560 Comm: syz-executor.2 Not tainted 5.17.0-rc3-syzkaller-00116-gf1baf68e1383-dirty #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Fixes: 470502de5bdb ("net: sched: unlock rules update API") Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Vlad Buslov <vladbu@mellanox.com> Reported-by: syzbot <syzkaller@googlegroups.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: Jiri Pirko <jiri@resnulli.us> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-02-11 23:06:23 +03:00
rcu_assign_pointer(dev->qdisc, qdisc);
pkt_sched: set root qdisc before change() in attach_default_qdiscs() After commit 95dc19299f74 ("pkt_sched: give visibility to mq slave qdiscs") we call disc_list_add() while the device qdisc might be the noop_qdisc one. This shows up as duplicates in "tc qdisc show", as all inactive devices point to noop_qdisc. Fix this by setting dev->qdisc to the new qdisc before calling ops->change() in attach_default_qdiscs() Add a WARN_ON_ONCE() to catch any future similar problem. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-12-13 03:41:56 +04:00
qdisc->ops->attach(qdisc);
net_sched: add classful multiqueue dummy scheduler This patch adds a classful dummy scheduler which can be used as root qdisc for multiqueue devices and exposes each device queue as a child class. This allows to address queues individually and graft them similar to regular classes. Additionally it presents an accumulated view of the statistics of all real root qdiscs in the dummy root. Two new callbacks are added to the qdisc_ops and qdisc_class_ops: - cl_ops->select_queue selects the tx queue number for new child classes. - qdisc_ops->attach() overrides root qdisc device grafting to attach non-shared qdiscs to the queues. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-06 12:58:51 +04:00
}
}
net_sched: add __rcu annotation to netdev->qdisc syzbot found a data-race [1] which lead me to add __rcu annotations to netdev->qdisc, and proper accessors to get LOCKDEP support. [1] BUG: KCSAN: data-race in dev_activate / qdisc_lookup_rcu write to 0xffff888168ad6410 of 8 bytes by task 13559 on cpu 1: attach_default_qdiscs net/sched/sch_generic.c:1167 [inline] dev_activate+0x2ed/0x8f0 net/sched/sch_generic.c:1221 __dev_open+0x2e9/0x3a0 net/core/dev.c:1416 __dev_change_flags+0x167/0x3f0 net/core/dev.c:8139 rtnl_configure_link+0xc2/0x150 net/core/rtnetlink.c:3150 __rtnl_newlink net/core/rtnetlink.c:3489 [inline] rtnl_newlink+0xf4d/0x13e0 net/core/rtnetlink.c:3529 rtnetlink_rcv_msg+0x745/0x7e0 net/core/rtnetlink.c:5594 netlink_rcv_skb+0x14e/0x250 net/netlink/af_netlink.c:2494 rtnetlink_rcv+0x18/0x20 net/core/rtnetlink.c:5612 netlink_unicast_kernel net/netlink/af_netlink.c:1317 [inline] netlink_unicast+0x602/0x6d0 net/netlink/af_netlink.c:1343 netlink_sendmsg+0x728/0x850 net/netlink/af_netlink.c:1919 sock_sendmsg_nosec net/socket.c:705 [inline] sock_sendmsg net/socket.c:725 [inline] ____sys_sendmsg+0x39a/0x510 net/socket.c:2413 ___sys_sendmsg net/socket.c:2467 [inline] __sys_sendmsg+0x195/0x230 net/socket.c:2496 __do_sys_sendmsg net/socket.c:2505 [inline] __se_sys_sendmsg net/socket.c:2503 [inline] __x64_sys_sendmsg+0x42/0x50 net/socket.c:2503 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae read to 0xffff888168ad6410 of 8 bytes by task 13560 on cpu 0: qdisc_lookup_rcu+0x30/0x2e0 net/sched/sch_api.c:323 __tcf_qdisc_find+0x74/0x3a0 net/sched/cls_api.c:1050 tc_del_tfilter+0x1c7/0x1350 net/sched/cls_api.c:2211 rtnetlink_rcv_msg+0x5ba/0x7e0 net/core/rtnetlink.c:5585 netlink_rcv_skb+0x14e/0x250 net/netlink/af_netlink.c:2494 rtnetlink_rcv+0x18/0x20 net/core/rtnetlink.c:5612 netlink_unicast_kernel net/netlink/af_netlink.c:1317 [inline] netlink_unicast+0x602/0x6d0 net/netlink/af_netlink.c:1343 netlink_sendmsg+0x728/0x850 net/netlink/af_netlink.c:1919 sock_sendmsg_nosec net/socket.c:705 [inline] sock_sendmsg net/socket.c:725 [inline] ____sys_sendmsg+0x39a/0x510 net/socket.c:2413 ___sys_sendmsg net/socket.c:2467 [inline] __sys_sendmsg+0x195/0x230 net/socket.c:2496 __do_sys_sendmsg net/socket.c:2505 [inline] __se_sys_sendmsg net/socket.c:2503 [inline] __x64_sys_sendmsg+0x42/0x50 net/socket.c:2503 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae value changed: 0xffffffff85dee080 -> 0xffff88815d96ec00 Reported by Kernel Concurrency Sanitizer on: CPU: 0 PID: 13560 Comm: syz-executor.2 Not tainted 5.17.0-rc3-syzkaller-00116-gf1baf68e1383-dirty #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Fixes: 470502de5bdb ("net: sched: unlock rules update API") Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Vlad Buslov <vladbu@mellanox.com> Reported-by: syzbot <syzkaller@googlegroups.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: Jiri Pirko <jiri@resnulli.us> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-02-11 23:06:23 +03:00
qdisc = rtnl_dereference(dev->qdisc);
net: sched: fallback to qdisc noqueue if default qdisc setup fail Currently if the default qdisc setup/init fails, the device ends up with qdisc "noop", which causes all TX packets to get dropped. With the introduction of sysctl net/core/default_qdisc it is possible to change the default qdisc to be more advanced, which opens for the possibility that Qdisc_ops->init() can fail. This patch detect these kind of failures, and choose to fallback to qdisc "noqueue", which is so simple that its init call will not fail. This allows the interface to continue functioning. V2: As this also captures memory failures, which are transient, the device is not kept in IFF_NO_QUEUE state. This allows the net_device to retry to default qdisc assignment. Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-04-30 14:42:22 +03:00
/* Detect default qdisc setup/init failed and fallback to "noqueue" */
net_sched: add __rcu annotation to netdev->qdisc syzbot found a data-race [1] which lead me to add __rcu annotations to netdev->qdisc, and proper accessors to get LOCKDEP support. [1] BUG: KCSAN: data-race in dev_activate / qdisc_lookup_rcu write to 0xffff888168ad6410 of 8 bytes by task 13559 on cpu 1: attach_default_qdiscs net/sched/sch_generic.c:1167 [inline] dev_activate+0x2ed/0x8f0 net/sched/sch_generic.c:1221 __dev_open+0x2e9/0x3a0 net/core/dev.c:1416 __dev_change_flags+0x167/0x3f0 net/core/dev.c:8139 rtnl_configure_link+0xc2/0x150 net/core/rtnetlink.c:3150 __rtnl_newlink net/core/rtnetlink.c:3489 [inline] rtnl_newlink+0xf4d/0x13e0 net/core/rtnetlink.c:3529 rtnetlink_rcv_msg+0x745/0x7e0 net/core/rtnetlink.c:5594 netlink_rcv_skb+0x14e/0x250 net/netlink/af_netlink.c:2494 rtnetlink_rcv+0x18/0x20 net/core/rtnetlink.c:5612 netlink_unicast_kernel net/netlink/af_netlink.c:1317 [inline] netlink_unicast+0x602/0x6d0 net/netlink/af_netlink.c:1343 netlink_sendmsg+0x728/0x850 net/netlink/af_netlink.c:1919 sock_sendmsg_nosec net/socket.c:705 [inline] sock_sendmsg net/socket.c:725 [inline] ____sys_sendmsg+0x39a/0x510 net/socket.c:2413 ___sys_sendmsg net/socket.c:2467 [inline] __sys_sendmsg+0x195/0x230 net/socket.c:2496 __do_sys_sendmsg net/socket.c:2505 [inline] __se_sys_sendmsg net/socket.c:2503 [inline] __x64_sys_sendmsg+0x42/0x50 net/socket.c:2503 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae read to 0xffff888168ad6410 of 8 bytes by task 13560 on cpu 0: qdisc_lookup_rcu+0x30/0x2e0 net/sched/sch_api.c:323 __tcf_qdisc_find+0x74/0x3a0 net/sched/cls_api.c:1050 tc_del_tfilter+0x1c7/0x1350 net/sched/cls_api.c:2211 rtnetlink_rcv_msg+0x5ba/0x7e0 net/core/rtnetlink.c:5585 netlink_rcv_skb+0x14e/0x250 net/netlink/af_netlink.c:2494 rtnetlink_rcv+0x18/0x20 net/core/rtnetlink.c:5612 netlink_unicast_kernel net/netlink/af_netlink.c:1317 [inline] netlink_unicast+0x602/0x6d0 net/netlink/af_netlink.c:1343 netlink_sendmsg+0x728/0x850 net/netlink/af_netlink.c:1919 sock_sendmsg_nosec net/socket.c:705 [inline] sock_sendmsg net/socket.c:725 [inline] ____sys_sendmsg+0x39a/0x510 net/socket.c:2413 ___sys_sendmsg net/socket.c:2467 [inline] __sys_sendmsg+0x195/0x230 net/socket.c:2496 __do_sys_sendmsg net/socket.c:2505 [inline] __se_sys_sendmsg net/socket.c:2503 [inline] __x64_sys_sendmsg+0x42/0x50 net/socket.c:2503 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae value changed: 0xffffffff85dee080 -> 0xffff88815d96ec00 Reported by Kernel Concurrency Sanitizer on: CPU: 0 PID: 13560 Comm: syz-executor.2 Not tainted 5.17.0-rc3-syzkaller-00116-gf1baf68e1383-dirty #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Fixes: 470502de5bdb ("net: sched: unlock rules update API") Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Vlad Buslov <vladbu@mellanox.com> Reported-by: syzbot <syzkaller@googlegroups.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: Jiri Pirko <jiri@resnulli.us> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-02-11 23:06:23 +03:00
if (qdisc == &noop_qdisc) {
net: sched: fallback to qdisc noqueue if default qdisc setup fail Currently if the default qdisc setup/init fails, the device ends up with qdisc "noop", which causes all TX packets to get dropped. With the introduction of sysctl net/core/default_qdisc it is possible to change the default qdisc to be more advanced, which opens for the possibility that Qdisc_ops->init() can fail. This patch detect these kind of failures, and choose to fallback to qdisc "noqueue", which is so simple that its init call will not fail. This allows the interface to continue functioning. V2: As this also captures memory failures, which are transient, the device is not kept in IFF_NO_QUEUE state. This allows the net_device to retry to default qdisc assignment. Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-04-30 14:42:22 +03:00
netdev_warn(dev, "default qdisc (%s) fail, fallback to %s\n",
default_qdisc_ops->id, noqueue_qdisc_ops.id);
dev->priv_flags |= IFF_NO_QUEUE;
netdev_for_each_tx_queue(dev, attach_one_default_qdisc, NULL);
net_sched: add __rcu annotation to netdev->qdisc syzbot found a data-race [1] which lead me to add __rcu annotations to netdev->qdisc, and proper accessors to get LOCKDEP support. [1] BUG: KCSAN: data-race in dev_activate / qdisc_lookup_rcu write to 0xffff888168ad6410 of 8 bytes by task 13559 on cpu 1: attach_default_qdiscs net/sched/sch_generic.c:1167 [inline] dev_activate+0x2ed/0x8f0 net/sched/sch_generic.c:1221 __dev_open+0x2e9/0x3a0 net/core/dev.c:1416 __dev_change_flags+0x167/0x3f0 net/core/dev.c:8139 rtnl_configure_link+0xc2/0x150 net/core/rtnetlink.c:3150 __rtnl_newlink net/core/rtnetlink.c:3489 [inline] rtnl_newlink+0xf4d/0x13e0 net/core/rtnetlink.c:3529 rtnetlink_rcv_msg+0x745/0x7e0 net/core/rtnetlink.c:5594 netlink_rcv_skb+0x14e/0x250 net/netlink/af_netlink.c:2494 rtnetlink_rcv+0x18/0x20 net/core/rtnetlink.c:5612 netlink_unicast_kernel net/netlink/af_netlink.c:1317 [inline] netlink_unicast+0x602/0x6d0 net/netlink/af_netlink.c:1343 netlink_sendmsg+0x728/0x850 net/netlink/af_netlink.c:1919 sock_sendmsg_nosec net/socket.c:705 [inline] sock_sendmsg net/socket.c:725 [inline] ____sys_sendmsg+0x39a/0x510 net/socket.c:2413 ___sys_sendmsg net/socket.c:2467 [inline] __sys_sendmsg+0x195/0x230 net/socket.c:2496 __do_sys_sendmsg net/socket.c:2505 [inline] __se_sys_sendmsg net/socket.c:2503 [inline] __x64_sys_sendmsg+0x42/0x50 net/socket.c:2503 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae read to 0xffff888168ad6410 of 8 bytes by task 13560 on cpu 0: qdisc_lookup_rcu+0x30/0x2e0 net/sched/sch_api.c:323 __tcf_qdisc_find+0x74/0x3a0 net/sched/cls_api.c:1050 tc_del_tfilter+0x1c7/0x1350 net/sched/cls_api.c:2211 rtnetlink_rcv_msg+0x5ba/0x7e0 net/core/rtnetlink.c:5585 netlink_rcv_skb+0x14e/0x250 net/netlink/af_netlink.c:2494 rtnetlink_rcv+0x18/0x20 net/core/rtnetlink.c:5612 netlink_unicast_kernel net/netlink/af_netlink.c:1317 [inline] netlink_unicast+0x602/0x6d0 net/netlink/af_netlink.c:1343 netlink_sendmsg+0x728/0x850 net/netlink/af_netlink.c:1919 sock_sendmsg_nosec net/socket.c:705 [inline] sock_sendmsg net/socket.c:725 [inline] ____sys_sendmsg+0x39a/0x510 net/socket.c:2413 ___sys_sendmsg net/socket.c:2467 [inline] __sys_sendmsg+0x195/0x230 net/socket.c:2496 __do_sys_sendmsg net/socket.c:2505 [inline] __se_sys_sendmsg net/socket.c:2503 [inline] __x64_sys_sendmsg+0x42/0x50 net/socket.c:2503 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae value changed: 0xffffffff85dee080 -> 0xffff88815d96ec00 Reported by Kernel Concurrency Sanitizer on: CPU: 0 PID: 13560 Comm: syz-executor.2 Not tainted 5.17.0-rc3-syzkaller-00116-gf1baf68e1383-dirty #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Fixes: 470502de5bdb ("net: sched: unlock rules update API") Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Vlad Buslov <vladbu@mellanox.com> Reported-by: syzbot <syzkaller@googlegroups.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: Jiri Pirko <jiri@resnulli.us> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-02-11 23:06:23 +03:00
qdisc = txq->qdisc_sleeping;
rcu_assign_pointer(dev->qdisc, qdisc);
qdisc_refcount_inc(qdisc);
net: sched: fallback to qdisc noqueue if default qdisc setup fail Currently if the default qdisc setup/init fails, the device ends up with qdisc "noop", which causes all TX packets to get dropped. With the introduction of sysctl net/core/default_qdisc it is possible to change the default qdisc to be more advanced, which opens for the possibility that Qdisc_ops->init() can fail. This patch detect these kind of failures, and choose to fallback to qdisc "noqueue", which is so simple that its init call will not fail. This allows the interface to continue functioning. V2: As this also captures memory failures, which are transient, the device is not kept in IFF_NO_QUEUE state. This allows the net_device to retry to default qdisc assignment. Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-04-30 14:42:22 +03:00
dev->priv_flags ^= IFF_NO_QUEUE;
}
net: sched: convert qdisc linked list to hashtable Convert the per-device linked list into a hashtable. The primary motivation for this change is that currently, we're not tracking all the qdiscs in hierarchy (e.g. excluding default qdiscs), as the lookup performed over the linked list by qdisc_match_from_root() is rather expensive. The ultimate goal is to get rid of hidden qdiscs completely, which will bring much more determinism in user experience. Reviewed-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-08-10 12:05:15 +03:00
#ifdef CONFIG_NET_SCHED
net_sched: add __rcu annotation to netdev->qdisc syzbot found a data-race [1] which lead me to add __rcu annotations to netdev->qdisc, and proper accessors to get LOCKDEP support. [1] BUG: KCSAN: data-race in dev_activate / qdisc_lookup_rcu write to 0xffff888168ad6410 of 8 bytes by task 13559 on cpu 1: attach_default_qdiscs net/sched/sch_generic.c:1167 [inline] dev_activate+0x2ed/0x8f0 net/sched/sch_generic.c:1221 __dev_open+0x2e9/0x3a0 net/core/dev.c:1416 __dev_change_flags+0x167/0x3f0 net/core/dev.c:8139 rtnl_configure_link+0xc2/0x150 net/core/rtnetlink.c:3150 __rtnl_newlink net/core/rtnetlink.c:3489 [inline] rtnl_newlink+0xf4d/0x13e0 net/core/rtnetlink.c:3529 rtnetlink_rcv_msg+0x745/0x7e0 net/core/rtnetlink.c:5594 netlink_rcv_skb+0x14e/0x250 net/netlink/af_netlink.c:2494 rtnetlink_rcv+0x18/0x20 net/core/rtnetlink.c:5612 netlink_unicast_kernel net/netlink/af_netlink.c:1317 [inline] netlink_unicast+0x602/0x6d0 net/netlink/af_netlink.c:1343 netlink_sendmsg+0x728/0x850 net/netlink/af_netlink.c:1919 sock_sendmsg_nosec net/socket.c:705 [inline] sock_sendmsg net/socket.c:725 [inline] ____sys_sendmsg+0x39a/0x510 net/socket.c:2413 ___sys_sendmsg net/socket.c:2467 [inline] __sys_sendmsg+0x195/0x230 net/socket.c:2496 __do_sys_sendmsg net/socket.c:2505 [inline] __se_sys_sendmsg net/socket.c:2503 [inline] __x64_sys_sendmsg+0x42/0x50 net/socket.c:2503 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae read to 0xffff888168ad6410 of 8 bytes by task 13560 on cpu 0: qdisc_lookup_rcu+0x30/0x2e0 net/sched/sch_api.c:323 __tcf_qdisc_find+0x74/0x3a0 net/sched/cls_api.c:1050 tc_del_tfilter+0x1c7/0x1350 net/sched/cls_api.c:2211 rtnetlink_rcv_msg+0x5ba/0x7e0 net/core/rtnetlink.c:5585 netlink_rcv_skb+0x14e/0x250 net/netlink/af_netlink.c:2494 rtnetlink_rcv+0x18/0x20 net/core/rtnetlink.c:5612 netlink_unicast_kernel net/netlink/af_netlink.c:1317 [inline] netlink_unicast+0x602/0x6d0 net/netlink/af_netlink.c:1343 netlink_sendmsg+0x728/0x850 net/netlink/af_netlink.c:1919 sock_sendmsg_nosec net/socket.c:705 [inline] sock_sendmsg net/socket.c:725 [inline] ____sys_sendmsg+0x39a/0x510 net/socket.c:2413 ___sys_sendmsg net/socket.c:2467 [inline] __sys_sendmsg+0x195/0x230 net/socket.c:2496 __do_sys_sendmsg net/socket.c:2505 [inline] __se_sys_sendmsg net/socket.c:2503 [inline] __x64_sys_sendmsg+0x42/0x50 net/socket.c:2503 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae value changed: 0xffffffff85dee080 -> 0xffff88815d96ec00 Reported by Kernel Concurrency Sanitizer on: CPU: 0 PID: 13560 Comm: syz-executor.2 Not tainted 5.17.0-rc3-syzkaller-00116-gf1baf68e1383-dirty #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Fixes: 470502de5bdb ("net: sched: unlock rules update API") Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Vlad Buslov <vladbu@mellanox.com> Reported-by: syzbot <syzkaller@googlegroups.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: Jiri Pirko <jiri@resnulli.us> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-02-11 23:06:23 +03:00
if (qdisc != &noop_qdisc)
qdisc_hash_add(qdisc, false);
net: sched: convert qdisc linked list to hashtable Convert the per-device linked list into a hashtable. The primary motivation for this change is that currently, we're not tracking all the qdiscs in hierarchy (e.g. excluding default qdiscs), as the lookup performed over the linked list by qdisc_match_from_root() is rather expensive. The ultimate goal is to get rid of hidden qdiscs completely, which will bring much more determinism in user experience. Reviewed-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-08-10 12:05:15 +03:00
#endif
net_sched: add classful multiqueue dummy scheduler This patch adds a classful dummy scheduler which can be used as root qdisc for multiqueue devices and exposes each device queue as a child class. This allows to address queues individually and graft them similar to regular classes. Additionally it presents an accumulated view of the statistics of all real root qdiscs in the dummy root. Two new callbacks are added to the qdisc_ops and qdisc_class_ops: - cl_ops->select_queue selects the tx queue number for new child classes. - qdisc_ops->attach() overrides root qdisc device grafting to attach non-shared qdiscs to the queues. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-06 12:58:51 +04:00
}
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
static void transition_one_qdisc(struct net_device *dev,
struct netdev_queue *dev_queue,
void *_need_watchdog)
{
pkt_sched: Kill netdev_queue lock. We can simply use the qdisc->q.lock for all of the qdisc tree synchronization. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:53:03 +04:00
struct Qdisc *new_qdisc = dev_queue->qdisc_sleeping;
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
int *need_watchdog_p = _need_watchdog;
pkt_sched: Add 'deactivated' state. This new state lets dev_deactivate() mark a qdisc as having been deactivated. dev_queue_xmit() and ing_filter() check for this bit and do not try to process the qdisc if the bit is set. dev_deactivate() polls the qdisc after setting the bit, waiting for both __QDISC_STATE_RUNNING and __QDISC_STATE_SCHED to clear. This isn't perfect yet, but subsequent changesets will make it so. This part is just one piece of the puzzle. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-08-18 08:51:03 +04:00
if (!(new_qdisc->flags & TCQ_F_BUILTIN))
clear_bit(__QDISC_STATE_DEACTIVATED, &new_qdisc->state);
pkt_sched: Kill netdev_queue lock. We can simply use the qdisc->q.lock for all of the qdisc tree synchronization. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:53:03 +04:00
rcu_assign_pointer(dev_queue->qdisc, new_qdisc);
net: sched: simplify attach_one_default_qdisc() Now that noqueue qdisc can be attached just like any other qdisc, no special treatment is necessary anymore when attaching it as default qdisc. This change has the added benefit that 'tc qdisc show' prints noqueue instead of nothing for devices defaulting to noqueue. Signed-off-by: Phil Sutter <phil@nwl.cc> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-27 22:21:39 +03:00
if (need_watchdog_p) {
net: annotate accesses to queue->trans_start In following patches, dev_watchdog() will no longer stop all queues. It will read queue->trans_start locklessly. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-11-17 06:29:22 +03:00
WRITE_ONCE(dev_queue->trans_start, 0);
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
*need_watchdog_p = 1;
net: tx scalability works : trans_start struct net_device trans_start field is a hot spot on SMP and high performance devices, particularly multi queues ones, because every transmitter dirties it. Is main use is tx watchdog and bonding alive checks. But as most devices dont use NETIF_F_LLTX, we have to lock a netdev_queue before calling their ndo_start_xmit(). So it makes sense to move trans_start from net_device to netdev_queue. Its update will occur on a already present (and in exclusive state) cache line, for free. We can do this transition smoothly. An old driver continue to update dev->trans_start, while an updated one updates txq->trans_start. Further patches could also put tx_bytes/tx_packets counters in netdev_queue to avoid dirtying dev->stats (vlan device comes to mind) Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-05-18 07:55:16 +04:00
}
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
void dev_activate(struct net_device *dev)
{
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
int need_watchdog;
netdev: Move rest of qdisc state into struct netdev_queue Now qdisc, qdisc_sleeping, and qdisc_list also live there. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-09 04:42:10 +04:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
/* No queueing discipline is attached to device;
qdisc: allow setting default queuing discipline By default, the pfifo_fast queue discipline has been used by default for all devices. But we have better choices now. This patch allow setting the default queueing discipline with sysctl. This allows easy use of better queueing disciplines on all devices without having to use tc qdisc scripts. It is intended to allow an easy path for distributions to make fq_codel or sfq the default qdisc. This patch also makes pfifo_fast more of a first class qdisc, since it is now possible to manually override the default and explicitly use pfifo_fast. The behavior for systems who do not use the sysctl is unchanged, they still get pfifo_fast Also removes leftover random # in sysctl net core. Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Acked-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-08-28 03:19:08 +04:00
* create default one for devices, which need queueing
* and noqueue_qdisc for virtual interfaces
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
*/
net_sched: add __rcu annotation to netdev->qdisc syzbot found a data-race [1] which lead me to add __rcu annotations to netdev->qdisc, and proper accessors to get LOCKDEP support. [1] BUG: KCSAN: data-race in dev_activate / qdisc_lookup_rcu write to 0xffff888168ad6410 of 8 bytes by task 13559 on cpu 1: attach_default_qdiscs net/sched/sch_generic.c:1167 [inline] dev_activate+0x2ed/0x8f0 net/sched/sch_generic.c:1221 __dev_open+0x2e9/0x3a0 net/core/dev.c:1416 __dev_change_flags+0x167/0x3f0 net/core/dev.c:8139 rtnl_configure_link+0xc2/0x150 net/core/rtnetlink.c:3150 __rtnl_newlink net/core/rtnetlink.c:3489 [inline] rtnl_newlink+0xf4d/0x13e0 net/core/rtnetlink.c:3529 rtnetlink_rcv_msg+0x745/0x7e0 net/core/rtnetlink.c:5594 netlink_rcv_skb+0x14e/0x250 net/netlink/af_netlink.c:2494 rtnetlink_rcv+0x18/0x20 net/core/rtnetlink.c:5612 netlink_unicast_kernel net/netlink/af_netlink.c:1317 [inline] netlink_unicast+0x602/0x6d0 net/netlink/af_netlink.c:1343 netlink_sendmsg+0x728/0x850 net/netlink/af_netlink.c:1919 sock_sendmsg_nosec net/socket.c:705 [inline] sock_sendmsg net/socket.c:725 [inline] ____sys_sendmsg+0x39a/0x510 net/socket.c:2413 ___sys_sendmsg net/socket.c:2467 [inline] __sys_sendmsg+0x195/0x230 net/socket.c:2496 __do_sys_sendmsg net/socket.c:2505 [inline] __se_sys_sendmsg net/socket.c:2503 [inline] __x64_sys_sendmsg+0x42/0x50 net/socket.c:2503 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae read to 0xffff888168ad6410 of 8 bytes by task 13560 on cpu 0: qdisc_lookup_rcu+0x30/0x2e0 net/sched/sch_api.c:323 __tcf_qdisc_find+0x74/0x3a0 net/sched/cls_api.c:1050 tc_del_tfilter+0x1c7/0x1350 net/sched/cls_api.c:2211 rtnetlink_rcv_msg+0x5ba/0x7e0 net/core/rtnetlink.c:5585 netlink_rcv_skb+0x14e/0x250 net/netlink/af_netlink.c:2494 rtnetlink_rcv+0x18/0x20 net/core/rtnetlink.c:5612 netlink_unicast_kernel net/netlink/af_netlink.c:1317 [inline] netlink_unicast+0x602/0x6d0 net/netlink/af_netlink.c:1343 netlink_sendmsg+0x728/0x850 net/netlink/af_netlink.c:1919 sock_sendmsg_nosec net/socket.c:705 [inline] sock_sendmsg net/socket.c:725 [inline] ____sys_sendmsg+0x39a/0x510 net/socket.c:2413 ___sys_sendmsg net/socket.c:2467 [inline] __sys_sendmsg+0x195/0x230 net/socket.c:2496 __do_sys_sendmsg net/socket.c:2505 [inline] __se_sys_sendmsg net/socket.c:2503 [inline] __x64_sys_sendmsg+0x42/0x50 net/socket.c:2503 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae value changed: 0xffffffff85dee080 -> 0xffff88815d96ec00 Reported by Kernel Concurrency Sanitizer on: CPU: 0 PID: 13560 Comm: syz-executor.2 Not tainted 5.17.0-rc3-syzkaller-00116-gf1baf68e1383-dirty #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Fixes: 470502de5bdb ("net: sched: unlock rules update API") Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Vlad Buslov <vladbu@mellanox.com> Reported-by: syzbot <syzkaller@googlegroups.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: Jiri Pirko <jiri@resnulli.us> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-02-11 23:06:23 +03:00
if (rtnl_dereference(dev->qdisc) == &noop_qdisc)
net_sched: add classful multiqueue dummy scheduler This patch adds a classful dummy scheduler which can be used as root qdisc for multiqueue devices and exposes each device queue as a child class. This allows to address queues individually and graft them similar to regular classes. Additionally it presents an accumulated view of the statistics of all real root qdiscs in the dummy root. Two new callbacks are added to the qdisc_ops and qdisc_class_ops: - cl_ops->select_queue selects the tx queue number for new child classes. - qdisc_ops->attach() overrides root qdisc device grafting to attach non-shared qdiscs to the queues. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-06 12:58:51 +04:00
attach_default_qdiscs(dev);
net_sched: reintroduce dev->qdisc for use by sch_api Currently the multiqueue integration with the qdisc API suffers from a few problems: - with multiple queues, all root qdiscs use the same handle. This means they can't be exposed to userspace in a backwards compatible fashion. - all API operations always refer to queue number 0. Newly created qdiscs are automatically shared between all queues, its not possible to address individual queues or restore multiqueue behaviour once a shared qdisc has been attached. - Dumps only contain the root qdisc of queue 0, in case of non-shared qdiscs this means the statistics are incomplete. This patch reintroduces dev->qdisc, which points to the (single) root qdisc from userspace's point of view. Currently it either points to the first (non-shared) default qdisc, or a qdisc shared between all queues. The following patches will introduce a classful dummy qdisc, which will be used as root qdisc and contain the per-queue qdiscs as children. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-04 10:41:18 +04:00
[NET]: Disable queueing when carrier is lost. Some network drivers call netif_stop_queue() when detecting loss of carrier. This leads to packets being queued up at the qdisc level for an unbound period of time. In order to prevent this effect, the core networking stack will now cease to queue packets for any device, that is operationally down (i.e. the queue is flushed and disabled). Signed-off-by: Tommy S. Christensen <tommy.christensen@tpack.net> Acked-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2005-05-04 03:18:52 +04:00
if (!netif_carrier_ok(dev))
/* Delay activation until next carrier-on event */
return;
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
need_watchdog = 0;
netdev_for_each_tx_queue(dev, transition_one_qdisc, &need_watchdog);
net: dynamic ingress_queue allocation ingress being not used very much, and net_device->ingress_queue being quite a big object (128 or 256 bytes), use a dynamic allocation if needed (tc qdisc add dev eth0 ingress ...) dev_ingress_queue(dev) helper should be used only with RTNL taken. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-10-02 10:11:55 +04:00
if (dev_ingress_queue(dev))
transition_one_qdisc(dev, dev_ingress_queue(dev), NULL);
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
if (need_watchdog) {
treewide: replace dev->trans_start update with helper Replace all trans_start updates with netif_trans_update helper. change was done via spatch: struct net_device *d; @@ - d->trans_start = jiffies + netif_trans_update(d) Compile tested only. Cc: user-mode-linux-devel@lists.sourceforge.net Cc: linux-xtensa@linux-xtensa.org Cc: linux1394-devel@lists.sourceforge.net Cc: linux-rdma@vger.kernel.org Cc: netdev@vger.kernel.org Cc: MPT-FusionLinux.pdl@broadcom.com Cc: linux-scsi@vger.kernel.org Cc: linux-can@vger.kernel.org Cc: linux-parisc@vger.kernel.org Cc: linux-omap@vger.kernel.org Cc: linux-hams@vger.kernel.org Cc: linux-usb@vger.kernel.org Cc: linux-wireless@vger.kernel.org Cc: linux-s390@vger.kernel.org Cc: devel@driverdev.osuosl.org Cc: b.a.t.m.a.n@lists.open-mesh.org Cc: linux-bluetooth@vger.kernel.org Signed-off-by: Florian Westphal <fw@strlen.de> Acked-by: Felipe Balbi <felipe.balbi@linux.intel.com> Acked-by: Mugunthan V N <mugunthanvnm@ti.com> Acked-by: Antonio Quartulli <a@unstable.cc> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-05-03 17:33:13 +03:00
netif_trans_update(dev);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
dev_watchdog_up(dev);
}
netdev: Move rest of qdisc state into struct netdev_queue Now qdisc, qdisc_sleeping, and qdisc_list also live there. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-09 04:42:10 +04:00
}
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 11:06:09 +03:00
EXPORT_SYMBOL(dev_activate);
netdev: Move rest of qdisc state into struct netdev_queue Now qdisc, qdisc_sleeping, and qdisc_list also live there. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-09 04:42:10 +04:00
net_sched: avoid resetting active qdisc for multiple times Except for sch_mq and sch_mqprio, each dev queue points to the same root qdisc, so when we reset the dev queues with netdev_for_each_tx_queue() we end up resetting the same instance of the root qdisc for multiple times. Avoid this by checking the __QDISC_STATE_DEACTIVATED bit in each iteration, so for sch_mq/sch_mqprio, we still reset all of them like before, for the rest, we only reset it once. Reported-by: Václav Zindulka <vaclav.zindulka@tlapnet.cz> Tested-by: Václav Zindulka <vaclav.zindulka@tlapnet.cz> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Jiri Pirko <jiri@resnulli.us> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-05-27 07:35:26 +03:00
static void qdisc_deactivate(struct Qdisc *qdisc)
{
if (qdisc->flags & TCQ_F_BUILTIN)
return;
set_bit(__QDISC_STATE_DEACTIVATED, &qdisc->state);
}
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
static void dev_deactivate_queue(struct net_device *dev,
struct netdev_queue *dev_queue,
void *_qdisc_default)
netdev: Move rest of qdisc state into struct netdev_queue Now qdisc, qdisc_sleeping, and qdisc_list also live there. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-09 04:42:10 +04:00
{
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
struct Qdisc *qdisc_default = _qdisc_default;
netdev: Move gso_skb into netdev_queue. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-09 10:10:33 +04:00
struct Qdisc *qdisc;
net: qdisc: use rcu prefix and silence sparse warnings Add __rcu notation to qdisc handling by doing this we can make smatch output more legible. And anyways some of the cases should be using rcu_dereference() see qdisc_all_tx_empty(), qdisc_tx_chainging(), and so on. Also *wake_queue() API is commonly called from driver timer routines without rcu lock or rtnl lock. So I added rcu_read_lock() blocks around netif_wake_subqueue and netif_tx_wake_queue. Signed-off-by: John Fastabend <john.r.fastabend@intel.com> Acked-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-13 07:04:52 +04:00
qdisc = rtnl_dereference(dev_queue->qdisc);
netdev: Move rest of qdisc state into struct netdev_queue Now qdisc, qdisc_sleeping, and qdisc_list also live there. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-09 04:42:10 +04:00
if (qdisc) {
net_sched: avoid resetting active qdisc for multiple times Except for sch_mq and sch_mqprio, each dev queue points to the same root qdisc, so when we reset the dev queues with netdev_for_each_tx_queue() we end up resetting the same instance of the root qdisc for multiple times. Avoid this by checking the __QDISC_STATE_DEACTIVATED bit in each iteration, so for sch_mq/sch_mqprio, we still reset all of them like before, for the rest, we only reset it once. Reported-by: Václav Zindulka <vaclav.zindulka@tlapnet.cz> Tested-by: Václav Zindulka <vaclav.zindulka@tlapnet.cz> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Jiri Pirko <jiri@resnulli.us> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-05-27 07:35:26 +03:00
qdisc_deactivate(qdisc);
pkt_sched: Use rcu_assign_pointer() to change dev_queue->qdisc These pointers are RCU protected, so proper primitives should be used. Signed-off-by: Jarek Poplawski <jarkao2@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-08-27 13:22:07 +04:00
rcu_assign_pointer(dev_queue->qdisc, qdisc_default);
netdev: Move rest of qdisc state into struct netdev_queue Now qdisc, qdisc_sleeping, and qdisc_list also live there. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-09 04:42:10 +04:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
}
net: sch_generic: aviod concurrent reset and enqueue op for lockless qdisc Currently there is concurrent reset and enqueue operation for the same lockless qdisc when there is no lock to synchronize the q->enqueue() in __dev_xmit_skb() with the qdisc reset operation in qdisc_deactivate() called by dev_deactivate_queue(), which may cause out-of-bounds access for priv->ring[] in hns3 driver if user has requested a smaller queue num when __dev_xmit_skb() still enqueue a skb with a larger queue_mapping after the corresponding qdisc is reset, and call hns3_nic_net_xmit() with that skb later. Reused the existing synchronize_net() in dev_deactivate_many() to make sure skb with larger queue_mapping enqueued to old qdisc(which is saved in dev_queue->qdisc_sleeping) will always be reset when dev_reset_queue() is called. Fixes: 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking") Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-09-08 14:02:34 +03:00
static void dev_reset_queue(struct net_device *dev,
struct netdev_queue *dev_queue,
void *_unused)
{
struct Qdisc *qdisc;
bool nolock;
qdisc = dev_queue->qdisc_sleeping;
if (!qdisc)
return;
nolock = qdisc->flags & TCQ_F_NOLOCK;
if (nolock)
spin_lock_bh(&qdisc->seqlock);
spin_lock_bh(qdisc_lock(qdisc));
qdisc_reset(qdisc);
spin_unlock_bh(qdisc_lock(qdisc));
net: sched: fix tx action rescheduling issue during deactivation Currently qdisc_run() checks the STATE_DEACTIVATED of lockless qdisc before calling __qdisc_run(), which ultimately clear the STATE_MISSED when all the skb is dequeued. If STATE_DEACTIVATED is set before clearing STATE_MISSED, there may be rescheduling of net_tx_action() at the end of qdisc_run_end(), see below: CPU0(net_tx_atcion) CPU1(__dev_xmit_skb) CPU2(dev_deactivate) . . . . set STATE_MISSED . . __netif_schedule() . . . set STATE_DEACTIVATED . . qdisc_reset() . . . .<--------------- . synchronize_net() clear __QDISC_STATE_SCHED | . . . | . . . | . some_qdisc_is_busy() . | . return *false* . | . . test STATE_DEACTIVATED | . . __qdisc_run() *not* called | . . . | . . test STATE_MISS | . . __netif_schedule()--------| . . . . . . . . __qdisc_run() is not called by net_tx_atcion() in CPU0 because CPU2 has set STATE_DEACTIVATED flag during dev_deactivate(), and STATE_MISSED is only cleared in __qdisc_run(), __netif_schedule is called at the end of qdisc_run_end(), causing tx action rescheduling problem. qdisc_run() called by net_tx_action() runs in the softirq context, which should has the same semantic as the qdisc_run() called by __dev_xmit_skb() protected by rcu_read_lock_bh(). And there is a synchronize_net() between STATE_DEACTIVATED flag being set and qdisc_reset()/some_qdisc_is_busy in dev_deactivate(), we can safely bail out for the deactived lockless qdisc in net_tx_action(), and qdisc_reset() will reset all skb not dequeued yet. So add the rcu_read_lock() explicitly to protect the qdisc_run() and do the STATE_DEACTIVATED checking in net_tx_action() before calling qdisc_run_begin(). Another option is to do the checking in the qdisc_run_end(), but it will add unnecessary overhead for non-tx_action case, because __dev_queue_xmit() will not see qdisc with STATE_DEACTIVATED after synchronize_net(), the qdisc with STATE_DEACTIVATED can only be seen by net_tx_action() because of __netif_schedule(). The STATE_DEACTIVATED checking in qdisc_run() is to avoid race between net_tx_action() and qdisc_reset(), see: commit d518d2ed8640 ("net/sched: fix race between deactivation and dequeue for NOLOCK qdisc"). As the bailout added above for deactived lockless qdisc in net_tx_action() provides better protection for the race without calling qdisc_run() at all, so remove the STATE_DEACTIVATED checking in qdisc_run(). After qdisc_reset(), there is no skb in qdisc to be dequeued, so clear the STATE_MISSED in dev_reset_queue() too. Fixes: 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking") Acked-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> V8: Clearing STATE_MISSED before calling __netif_schedule() has avoid the endless rescheduling problem, but there may still be a unnecessary rescheduling, so adjust the commit log. Signed-off-by: David S. Miller <davem@davemloft.net>
2021-05-14 06:17:00 +03:00
if (nolock) {
clear_bit(__QDISC_STATE_MISSED, &qdisc->state);
net: sched: implement TCQ_F_CAN_BYPASS for lockless qdisc Currently pfifo_fast has both TCQ_F_CAN_BYPASS and TCQ_F_NOLOCK flag set, but queue discipline by-pass does not work for lockless qdisc because skb is always enqueued to qdisc even when the qdisc is empty, see __dev_xmit_skb(). This patch calls sch_direct_xmit() to transmit the skb directly to the driver for empty lockless qdisc, which aviod enqueuing and dequeuing operation. As qdisc->empty is not reliable to indicate a empty qdisc because there is a time window between enqueuing and setting qdisc->empty. So we use the MISSED state added in commit a90c57f2cedd ("net: sched: fix packet stuck problem for lockless qdisc"), which indicate there is lock contention, suggesting that it is better not to do the qdisc bypass in order to avoid packet out of order problem. In order to make MISSED state reliable to indicate a empty qdisc, we need to ensure that testing and clearing of MISSED state is within the protection of qdisc->seqlock, only setting MISSED state can be done without the protection of qdisc->seqlock. A MISSED state testing is added without the protection of qdisc->seqlock to aviod doing unnecessary spin_trylock() for contention case. As the enqueuing is not within the protection of qdisc->seqlock, there is still a potential data race as mentioned by Jakub [1]: thread1 thread2 thread3 qdisc_run_begin() # true qdisc_run_begin(q) set(MISSED) pfifo_fast_dequeue clear(MISSED) # recheck the queue qdisc_run_end() enqueue skb1 qdisc empty # true qdisc_run_begin() # true sch_direct_xmit() # skb2 qdisc_run_begin() set(MISSED) When above happens, skb1 enqueued by thread2 is transmited after skb2 is transmited by thread3 because MISSED state setting and enqueuing is not under the qdisc->seqlock. If qdisc bypass is disabled, skb1 has better chance to be transmited quicker than skb2. This patch does not take care of the above data race, because we view this as similar as below: Even at the same time CPU1 and CPU2 write the skb to two socket which both heading to the same qdisc, there is no guarantee that which skb will hit the qdisc first, because there is a lot of factor like interrupt/softirq/cache miss/scheduling afffecting that. There are below cases that need special handling: 1. When MISSED state is cleared before another round of dequeuing in pfifo_fast_dequeue(), and __qdisc_run() might not be able to dequeue all skb in one round and call __netif_schedule(), which might result in a non-empty qdisc without MISSED set. In order to avoid this, the MISSED state is set for lockless qdisc and __netif_schedule() will be called at the end of qdisc_run_end. 2. The MISSED state also need to be set for lockless qdisc instead of calling __netif_schedule() directly when requeuing a skb for a similar reason. 3. For netdev queue stopped case, the MISSED case need clearing while the netdev queue is stopped, otherwise there may be unnecessary __netif_schedule() calling. So a new DRAINING state is added to indicate this case, which also indicate a non-empty qdisc. 4. As there is already netif_xmit_frozen_or_stopped() checking in dequeue_skb() and sch_direct_xmit(), which are both within the protection of qdisc->seqlock, but the same checking in __dev_xmit_skb() is without the protection, which might cause empty indication of a lockless qdisc to be not reliable. So remove the checking in __dev_xmit_skb(), and the checking in the protection of qdisc->seqlock seems enough to avoid the cpu consumption problem for netdev queue stopped case. 1. https://lkml.org/lkml/2021/5/29/215 Acked-by: Jakub Kicinski <kuba@kernel.org> Tested-by: Vladimir Oltean <vladimir.oltean@nxp.com> # flexcan Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-06-22 09:49:56 +03:00
clear_bit(__QDISC_STATE_DRAINING, &qdisc->state);
net: sch_generic: aviod concurrent reset and enqueue op for lockless qdisc Currently there is concurrent reset and enqueue operation for the same lockless qdisc when there is no lock to synchronize the q->enqueue() in __dev_xmit_skb() with the qdisc reset operation in qdisc_deactivate() called by dev_deactivate_queue(), which may cause out-of-bounds access for priv->ring[] in hns3 driver if user has requested a smaller queue num when __dev_xmit_skb() still enqueue a skb with a larger queue_mapping after the corresponding qdisc is reset, and call hns3_nic_net_xmit() with that skb later. Reused the existing synchronize_net() in dev_deactivate_many() to make sure skb with larger queue_mapping enqueued to old qdisc(which is saved in dev_queue->qdisc_sleeping) will always be reset when dev_reset_queue() is called. Fixes: 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking") Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-09-08 14:02:34 +03:00
spin_unlock_bh(&qdisc->seqlock);
net: sched: fix tx action rescheduling issue during deactivation Currently qdisc_run() checks the STATE_DEACTIVATED of lockless qdisc before calling __qdisc_run(), which ultimately clear the STATE_MISSED when all the skb is dequeued. If STATE_DEACTIVATED is set before clearing STATE_MISSED, there may be rescheduling of net_tx_action() at the end of qdisc_run_end(), see below: CPU0(net_tx_atcion) CPU1(__dev_xmit_skb) CPU2(dev_deactivate) . . . . set STATE_MISSED . . __netif_schedule() . . . set STATE_DEACTIVATED . . qdisc_reset() . . . .<--------------- . synchronize_net() clear __QDISC_STATE_SCHED | . . . | . . . | . some_qdisc_is_busy() . | . return *false* . | . . test STATE_DEACTIVATED | . . __qdisc_run() *not* called | . . . | . . test STATE_MISS | . . __netif_schedule()--------| . . . . . . . . __qdisc_run() is not called by net_tx_atcion() in CPU0 because CPU2 has set STATE_DEACTIVATED flag during dev_deactivate(), and STATE_MISSED is only cleared in __qdisc_run(), __netif_schedule is called at the end of qdisc_run_end(), causing tx action rescheduling problem. qdisc_run() called by net_tx_action() runs in the softirq context, which should has the same semantic as the qdisc_run() called by __dev_xmit_skb() protected by rcu_read_lock_bh(). And there is a synchronize_net() between STATE_DEACTIVATED flag being set and qdisc_reset()/some_qdisc_is_busy in dev_deactivate(), we can safely bail out for the deactived lockless qdisc in net_tx_action(), and qdisc_reset() will reset all skb not dequeued yet. So add the rcu_read_lock() explicitly to protect the qdisc_run() and do the STATE_DEACTIVATED checking in net_tx_action() before calling qdisc_run_begin(). Another option is to do the checking in the qdisc_run_end(), but it will add unnecessary overhead for non-tx_action case, because __dev_queue_xmit() will not see qdisc with STATE_DEACTIVATED after synchronize_net(), the qdisc with STATE_DEACTIVATED can only be seen by net_tx_action() because of __netif_schedule(). The STATE_DEACTIVATED checking in qdisc_run() is to avoid race between net_tx_action() and qdisc_reset(), see: commit d518d2ed8640 ("net/sched: fix race between deactivation and dequeue for NOLOCK qdisc"). As the bailout added above for deactived lockless qdisc in net_tx_action() provides better protection for the race without calling qdisc_run() at all, so remove the STATE_DEACTIVATED checking in qdisc_run(). After qdisc_reset(), there is no skb in qdisc to be dequeued, so clear the STATE_MISSED in dev_reset_queue() too. Fixes: 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking") Acked-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> V8: Clearing STATE_MISSED before calling __netif_schedule() has avoid the endless rescheduling problem, but there may still be a unnecessary rescheduling, so adjust the commit log. Signed-off-by: David S. Miller <davem@davemloft.net>
2021-05-14 06:17:00 +03:00
}
net: sch_generic: aviod concurrent reset and enqueue op for lockless qdisc Currently there is concurrent reset and enqueue operation for the same lockless qdisc when there is no lock to synchronize the q->enqueue() in __dev_xmit_skb() with the qdisc reset operation in qdisc_deactivate() called by dev_deactivate_queue(), which may cause out-of-bounds access for priv->ring[] in hns3 driver if user has requested a smaller queue num when __dev_xmit_skb() still enqueue a skb with a larger queue_mapping after the corresponding qdisc is reset, and call hns3_nic_net_xmit() with that skb later. Reused the existing synchronize_net() in dev_deactivate_many() to make sure skb with larger queue_mapping enqueued to old qdisc(which is saved in dev_queue->qdisc_sleeping) will always be reset when dev_reset_queue() is called. Fixes: 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking") Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-09-08 14:02:34 +03:00
}
pkt_sched: Simplify dev_deactivate() polling loop. The condition under which the previous qdisc has no more references after we've attached &noop_qdisc is that both RUNNING and SCHED are both seen clear while holding the root lock. So just make specifically that check in the polling loop, instead of this overly complex "check without then check with lock held" sequence. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-08-18 08:58:07 +04:00
static bool some_qdisc_is_busy(struct net_device *dev)
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
{
unsigned int i;
for (i = 0; i < dev->num_tx_queues; i++) {
struct netdev_queue *dev_queue;
pkt_sched: Add and use qdisc_root() and qdisc_root_lock(). When code wants to lock the qdisc tree state, the logic operation it's doing is locking the top-level qdisc that sits of the root of the netdev_queue. Add qdisc_root_lock() to represent this and convert the easiest cases. In order for this to work out in all cases, we have to hook up the noop_qdisc to a dummy netdev_queue. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-16 12:42:40 +04:00
spinlock_t *root_lock;
pkt_sched: Make QDISC_RUNNING a qdisc state. Currently it is associated with a netdev_queue, but when we have qdisc sharing that no longer makes any sense. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-16 11:56:32 +04:00
struct Qdisc *q;
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
int val;
dev_queue = netdev_get_tx_queue(dev, i);
pkt_sched: Fix queue quiescence testing in dev_deactivate(). Based upon discussions with Jarek P. and Herbert Xu. First, we're testing the wrong qdisc. We just reset the device queue qdiscs to &noop_qdisc and checking it's state is completely pointless here. We want to wait until the previous qdisc that was sitting at the ->qdisc pointer is not busy any more. And that would be ->qdisc_sleeping. Because of how we propagate the samples qdisc pointer down into qdisc_run and friends via per-cpu ->output_queue and netif_schedule, we have to wait also for the __QDISC_STATE_SCHED bit to clear as well. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-08-14 02:18:38 +04:00
q = dev_queue->qdisc_sleeping;
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
sched: manipulate __QDISC_STATE_RUNNING in qdisc_run_* helpers Currently NOLOCK qdiscs pay a measurable overhead to atomically manipulate the __QDISC_STATE_RUNNING. Such bit is flipped twice per packet in the uncontended scenario with packet rate below the line rate: on packed dequeue and on the next, failing dequeue attempt. This changeset moves the bit manipulation into the qdisc_run_{begin,end} helpers, so that the bit is now flipped only once per packet, with measurable performance improvement in the uncontended scenario. This also allows simplifying the qdisc teardown code path - since qdisc_is_running() is now effective for each qdisc type - and avoid a possible race between qdisc_run() and dev_deactivate_many(), as now the some_qdisc_is_busy() can properly detect NOLOCK qdiscs being busy dequeuing packets. Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-05-15 11:50:31 +03:00
root_lock = qdisc_lock(q);
spin_lock_bh(root_lock);
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
sched: manipulate __QDISC_STATE_RUNNING in qdisc_run_* helpers Currently NOLOCK qdiscs pay a measurable overhead to atomically manipulate the __QDISC_STATE_RUNNING. Such bit is flipped twice per packet in the uncontended scenario with packet rate below the line rate: on packed dequeue and on the next, failing dequeue attempt. This changeset moves the bit manipulation into the qdisc_run_{begin,end} helpers, so that the bit is now flipped only once per packet, with measurable performance improvement in the uncontended scenario. This also allows simplifying the qdisc teardown code path - since qdisc_is_running() is now effective for each qdisc type - and avoid a possible race between qdisc_run() and dev_deactivate_many(), as now the some_qdisc_is_busy() can properly detect NOLOCK qdiscs being busy dequeuing packets. Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-05-15 11:50:31 +03:00
val = (qdisc_is_running(q) ||
test_bit(__QDISC_STATE_SCHED, &q->state));
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
sched: manipulate __QDISC_STATE_RUNNING in qdisc_run_* helpers Currently NOLOCK qdiscs pay a measurable overhead to atomically manipulate the __QDISC_STATE_RUNNING. Such bit is flipped twice per packet in the uncontended scenario with packet rate below the line rate: on packed dequeue and on the next, failing dequeue attempt. This changeset moves the bit manipulation into the qdisc_run_{begin,end} helpers, so that the bit is now flipped only once per packet, with measurable performance improvement in the uncontended scenario. This also allows simplifying the qdisc teardown code path - since qdisc_is_running() is now effective for each qdisc type - and avoid a possible race between qdisc_run() and dev_deactivate_many(), as now the some_qdisc_is_busy() can properly detect NOLOCK qdiscs being busy dequeuing packets. Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-05-15 11:50:31 +03:00
spin_unlock_bh(root_lock);
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
if (val)
return true;
}
return false;
}
net: avoid synchronize_rcu() in dev_deactivate_many dev_deactivate_many() issues one synchronize_rcu() call after qdiscs set to noop_qdisc. This call is here to make sure they are no outstanding qdisc-less dev_queue_xmit calls before returning to caller. But in dismantle phase, we dont have to wait, because we wont activate again the device, and we are going to wait one rcu grace period later in rollback_registered_many(). After this patch, device dismantle uses one synchronize_net() and one rcu_barrier() call only, so we have a ~30% speedup and a smaller RTNL latency. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> CC: Patrick McHardy <kaber@trash.net>, CC: Ben Greear <greearb@candelatech.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-05-20 03:42:09 +04:00
/**
* 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.
*/
net: factorize sync-rcu call in unregister_netdevice_many Add dev_close_many and dev_deactivate_many to factorize another sync-rcu operation on the netdevice unregister path. $ modprobe dummy numdummies=10000 $ ip link set dev dummy* up $ time rmmod dummy Without the patch With the patch real 0m 24.63s real 0m 5.15s user 0m 0.00s user 0m 0.00s sys 0m 6.05s sys 0m 5.14s Signed-off-by: Octavian Purdila <opurdila@ixiacom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-12-13 15:44:07 +03:00
void dev_deactivate_many(struct list_head *head)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
{
net: factorize sync-rcu call in unregister_netdevice_many Add dev_close_many and dev_deactivate_many to factorize another sync-rcu operation on the netdevice unregister path. $ modprobe dummy numdummies=10000 $ ip link set dev dummy* up $ time rmmod dummy Without the patch With the patch real 0m 24.63s real 0m 5.15s user 0m 0.00s user 0m 0.00s sys 0m 6.05s sys 0m 5.14s Signed-off-by: Octavian Purdila <opurdila@ixiacom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-12-13 15:44:07 +03:00
struct net_device *dev;
[NET_SCHED]: Avoid requeue warning on dev_deactivate When we relinquish queue_lock in qdisc_restart and then retake it for requeueing, we might race against dev_deactivate and end up requeueing onto noop_qdisc. This causes a warning to be printed. This patch fixes this by checking this before we requeue. As an added bonus, we can remove the same check in __qdisc_run which was added to prevent dev->gso_skb from being requeued when we're shutting down. Even though we've had to add a new conditional in its place, it's better because it only happens on requeues rather than every single time that qdisc_run is called. For this to work we also need to move the clearing of gso_skb up in dev_deactivate as now qdisc_restart can occur even after we wait for __LINK_STATE_QDISC_RUNNING to clear (but it won't do anything as long as the queue and gso_skb is already clear). Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-05-11 01:12:47 +04:00
net: Separate the close_list and the unreg_list v2 Separate the unreg_list and the close_list in dev_close_many preventing dev_close_many from permuting the unreg_list. The permutations of the unreg_list have resulted in cases where the loopback device is accessed it has been freed in code such as dst_ifdown. Resulting in subtle memory corruption. This is the second bug from sharing the storage between the close_list and the unreg_list. The issues that crop up with sharing are apparently too subtle to show up in normal testing or usage, so let's forget about being clever and use two separate lists. v2: Make all callers pass in a close_list to dev_close_many Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-10-06 06:26:05 +04:00
list_for_each_entry(dev, head, close_list) {
net: factorize sync-rcu call in unregister_netdevice_many Add dev_close_many and dev_deactivate_many to factorize another sync-rcu operation on the netdevice unregister path. $ modprobe dummy numdummies=10000 $ ip link set dev dummy* up $ time rmmod dummy Without the patch With the patch real 0m 24.63s real 0m 5.15s user 0m 0.00s user 0m 0.00s sys 0m 6.05s sys 0m 5.14s Signed-off-by: Octavian Purdila <opurdila@ixiacom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-12-13 15:44:07 +03:00
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);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
net: sch_generic: aviod concurrent reset and enqueue op for lockless qdisc Currently there is concurrent reset and enqueue operation for the same lockless qdisc when there is no lock to synchronize the q->enqueue() in __dev_xmit_skb() with the qdisc reset operation in qdisc_deactivate() called by dev_deactivate_queue(), which may cause out-of-bounds access for priv->ring[] in hns3 driver if user has requested a smaller queue num when __dev_xmit_skb() still enqueue a skb with a larger queue_mapping after the corresponding qdisc is reset, and call hns3_nic_net_xmit() with that skb later. Reused the existing synchronize_net() in dev_deactivate_many() to make sure skb with larger queue_mapping enqueued to old qdisc(which is saved in dev_queue->qdisc_sleeping) will always be reset when dev_reset_queue() is called. Fixes: 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking") Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-09-08 14:02:34 +03:00
/* Wait for outstanding qdisc-less dev_queue_xmit calls or
* outstanding qdisc enqueuing calls.
net: avoid synchronize_rcu() in dev_deactivate_many dev_deactivate_many() issues one synchronize_rcu() call after qdiscs set to noop_qdisc. This call is here to make sure they are no outstanding qdisc-less dev_queue_xmit calls before returning to caller. But in dismantle phase, we dont have to wait, because we wont activate again the device, and we are going to wait one rcu grace period later in rollback_registered_many(). After this patch, device dismantle uses one synchronize_net() and one rcu_barrier() call only, so we have a ~30% speedup and a smaller RTNL latency. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> CC: Patrick McHardy <kaber@trash.net>, CC: Ben Greear <greearb@candelatech.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-05-20 03:42:09 +04:00
* This is avoided if all devices are in dismantle phase :
* Caller will call synchronize_net() for us
*/
net: sched: drop qdisc_reset from dev_graft_qdisc In qdisc_graft_qdisc a "new" qdisc is attached and the 'qdisc_destroy' operation is called on the old qdisc. The destroy operation will wait a rcu grace period and call qdisc_rcu_free(). At which point gso_cpu_skb is free'd along with all stats so no need to zero stats and gso_cpu_skb from the graft operation itself. Further after dropping the qdisc locks we can not continue to call qdisc_reset before waiting an rcu grace period so that the qdisc is detached from all cpus. By removing the qdisc_reset() here we get the correct property of waiting an rcu grace period and letting the qdisc_destroy operation clean up the qdisc correctly. Note, a refcnt greater than 1 would cause the destroy operation to be aborted however if this ever happened the reference to the qdisc would be lost and we would have a memory leak. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:56:04 +03:00
synchronize_net();
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
net: sch_generic: aviod concurrent reset and enqueue op for lockless qdisc Currently there is concurrent reset and enqueue operation for the same lockless qdisc when there is no lock to synchronize the q->enqueue() in __dev_xmit_skb() with the qdisc reset operation in qdisc_deactivate() called by dev_deactivate_queue(), which may cause out-of-bounds access for priv->ring[] in hns3 driver if user has requested a smaller queue num when __dev_xmit_skb() still enqueue a skb with a larger queue_mapping after the corresponding qdisc is reset, and call hns3_nic_net_xmit() with that skb later. Reused the existing synchronize_net() in dev_deactivate_many() to make sure skb with larger queue_mapping enqueued to old qdisc(which is saved in dev_queue->qdisc_sleeping) will always be reset when dev_reset_queue() is called. Fixes: 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking") Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-09-08 14:02:34 +03:00
list_for_each_entry(dev, head, close_list) {
netdev_for_each_tx_queue(dev, dev_reset_queue, NULL);
if (dev_ingress_queue(dev))
dev_reset_queue(dev, dev_ingress_queue(dev), NULL);
}
[NET]: Prevent transmission after dev_deactivate The dev_deactivate function has bit-rotted since the introduction of lockless drivers. In particular, the spin_unlock_wait call at the end has no effect on the xmit routine of lockless drivers. With a little bit of work, we can make it much more useful by providing the guarantee that when it returns, no more calls to the xmit routine of the underlying driver will be made. The idea is simple. There are two entry points in to the xmit routine. The first comes from dev_queue_xmit. That one is easily stopped by using synchronize_rcu. This works because we set the qdisc to noop_qdisc before the synchronize_rcu call. That in turn causes all subsequent packets sent to dev_queue_xmit to be dropped. The synchronize_rcu call also ensures all outstanding calls leave their critical section. The other entry point is from qdisc_run. Since we now have a bit that indicates whether it's running, all we have to do is to wait until the bit is off. I've removed the loop to wait for __LINK_STATE_SCHED to clear. This is useless because netif_wake_queue can cause it to be set again. It is also harmless because we've disarmed qdisc_run. I've also removed the spin_unlock_wait on xmit_lock because its only purpose of making sure that all outstanding xmit_lock holders have exited is also given by dev_watchdog_down. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-22 13:28:18 +04:00
/* Wait for outstanding qdisc_run calls. */
net: sched: drop qdisc_reset from dev_graft_qdisc In qdisc_graft_qdisc a "new" qdisc is attached and the 'qdisc_destroy' operation is called on the old qdisc. The destroy operation will wait a rcu grace period and call qdisc_rcu_free(). At which point gso_cpu_skb is free'd along with all stats so no need to zero stats and gso_cpu_skb from the graft operation itself. Further after dropping the qdisc locks we can not continue to call qdisc_reset before waiting an rcu grace period so that the qdisc is detached from all cpus. By removing the qdisc_reset() here we get the correct property of waiting an rcu grace period and letting the qdisc_destroy operation clean up the qdisc correctly. Note, a refcnt greater than 1 would cause the destroy operation to be aborted however if this ever happened the reference to the qdisc would be lost and we would have a memory leak. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:56:04 +03:00
list_for_each_entry(dev, head, close_list) {
net: sched: Avoid using yield() in a busy waiting loop With threaded interrupts enabled, the interrupt thread runs as SCHED_RR with priority 50. If a user application with a higher priority preempts the interrupt thread and tries to shutdown the network interface then it will loop forever. The kernel will spin in the loop waiting for the device to become idle and the scheduler will never consider the interrupt thread because its priority is lower. Avoid the problem by sleeping for a jiffy giving other tasks, including the interrupt thread, a chance to run and make progress. In the original thread it has been suggested to use wait_event() and properly waiting for the state to occur. DaveM explained that this would require to add expensive checks in the fast paths of packet processing. Link: https://lkml.kernel.org/r/1393976987-23555-1-git-send-email-mkl@pengutronix.de Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de> [bigeasy: Rewrite commit message, add comment, use schedule_timeout_uninterruptible()] Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Acked-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-10-16 11:28:33 +03:00
while (some_qdisc_is_busy(dev)) {
/* wait_event() would avoid this sleep-loop but would
* require expensive checks in the fast paths of packet
* processing which isn't worth it.
*/
schedule_timeout_uninterruptible(1);
}
net: sched: drop qdisc_reset from dev_graft_qdisc In qdisc_graft_qdisc a "new" qdisc is attached and the 'qdisc_destroy' operation is called on the old qdisc. The destroy operation will wait a rcu grace period and call qdisc_rcu_free(). At which point gso_cpu_skb is free'd along with all stats so no need to zero stats and gso_cpu_skb from the graft operation itself. Further after dropping the qdisc locks we can not continue to call qdisc_reset before waiting an rcu grace period so that the qdisc is detached from all cpus. By removing the qdisc_reset() here we get the correct property of waiting an rcu grace period and letting the qdisc_destroy operation clean up the qdisc correctly. Note, a refcnt greater than 1 would cause the destroy operation to be aborted however if this ever happened the reference to the qdisc would be lost and we would have a memory leak. Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-07 20:56:04 +03:00
}
net: factorize sync-rcu call in unregister_netdevice_many Add dev_close_many and dev_deactivate_many to factorize another sync-rcu operation on the netdevice unregister path. $ modprobe dummy numdummies=10000 $ ip link set dev dummy* up $ time rmmod dummy Without the patch With the patch real 0m 24.63s real 0m 5.15s user 0m 0.00s user 0m 0.00s sys 0m 6.05s sys 0m 5.14s Signed-off-by: Octavian Purdila <opurdila@ixiacom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-12-13 15:44:07 +03:00
}
void dev_deactivate(struct net_device *dev)
{
LIST_HEAD(single);
net: Separate the close_list and the unreg_list v2 Separate the unreg_list and the close_list in dev_close_many preventing dev_close_many from permuting the unreg_list. The permutations of the unreg_list have resulted in cases where the loopback device is accessed it has been freed in code such as dst_ifdown. Resulting in subtle memory corruption. This is the second bug from sharing the storage between the close_list and the unreg_list. The issues that crop up with sharing are apparently too subtle to show up in normal testing or usage, so let's forget about being clever and use two separate lists. v2: Make all callers pass in a close_list to dev_close_many Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-10-06 06:26:05 +04:00
list_add(&dev->close_list, &single);
net: factorize sync-rcu call in unregister_netdevice_many Add dev_close_many and dev_deactivate_many to factorize another sync-rcu operation on the netdevice unregister path. $ modprobe dummy numdummies=10000 $ ip link set dev dummy* up $ time rmmod dummy Without the patch With the patch real 0m 24.63s real 0m 5.15s user 0m 0.00s user 0m 0.00s sys 0m 6.05s sys 0m 5.14s Signed-off-by: Octavian Purdila <opurdila@ixiacom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-12-13 15:44:07 +03:00
dev_deactivate_many(&single);
net: Fix more stale on-stack list_head objects. From: Eric W. Biederman <ebiederm@xmission.com> In the beginning with batching unreg_list was a list that was used only once in the lifetime of a network device (I think). Now we have calls using the unreg_list that can happen multiple times in the life of a network device like dev_deactivate and dev_close that are also using the unreg_list. In addition in unregister_netdevice_queue we also do a list_move because for devices like veth pairs it is possible that unregister_netdevice_queue will be called multiple times. So I think the change below to fix dev_deactivate which Eric D. missed will fix this problem. Now to go test that. Signed-off-by: David S. Miller <davem@davemloft.net>
2011-02-20 22:49:45 +03:00
list_del(&single);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
}
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 11:06:09 +03:00
EXPORT_SYMBOL(dev_deactivate);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
net_sched: plug in qdisc ops change_tx_queue_len Introduce a new qdisc ops ->change_tx_queue_len() so that each qdisc could decide how to implement this if it wants. Previously we simply read dev->tx_queue_len, after pfifo_fast switches to skb array, we need this API to resize the skb array when we change dev->tx_queue_len. To avoid handling race conditions with TX BH, we need to deactivate all TX queues before change the value and bring them back after we are done, this also makes implementation easier. Cc: John Fastabend <john.fastabend@gmail.com> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-01-26 05:26:23 +03:00
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;
}
net: sched: update default qdisc visibility after Tx queue cnt changes mq / mqprio make the default child qdiscs visible. They only do so for the qdiscs which are within real_num_tx_queues when the device is registered. Depending on order of calls in the driver, or if user space changes config via ethtool -L the number of qdiscs visible under tc qdisc show will differ from the number of queues. This is confusing to users and potentially to system configuration scripts which try to make sure qdiscs have the right parameters. Add a new Qdisc_ops callback and make relevant qdiscs TTRT. Note that this uncovers the "shortcut" created by commit 1f27cde313d7 ("net: sched: use pfifo_fast for non real queues") The default child qdiscs beyond initial real_num_tx are always pfifo_fast, no matter what the sysfs setting is. Fixing this gets a little tricky because we'd need to keep a reference on whatever the default qdisc was at the time of creation. In practice this is likely an non-issue the qdiscs likely have to be configured to non-default settings, so whatever user space is doing such configuration can replace the pfifos... now that it will see them. Reported-by: Matthew Massey <matthewmassey@fb.com> Reviewed-by: Dave Taht <dave.taht@gmail.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-09-14 01:53:30 +03:00
void dev_qdisc_change_real_num_tx(struct net_device *dev,
unsigned int new_real_tx)
{
net_sched: add __rcu annotation to netdev->qdisc syzbot found a data-race [1] which lead me to add __rcu annotations to netdev->qdisc, and proper accessors to get LOCKDEP support. [1] BUG: KCSAN: data-race in dev_activate / qdisc_lookup_rcu write to 0xffff888168ad6410 of 8 bytes by task 13559 on cpu 1: attach_default_qdiscs net/sched/sch_generic.c:1167 [inline] dev_activate+0x2ed/0x8f0 net/sched/sch_generic.c:1221 __dev_open+0x2e9/0x3a0 net/core/dev.c:1416 __dev_change_flags+0x167/0x3f0 net/core/dev.c:8139 rtnl_configure_link+0xc2/0x150 net/core/rtnetlink.c:3150 __rtnl_newlink net/core/rtnetlink.c:3489 [inline] rtnl_newlink+0xf4d/0x13e0 net/core/rtnetlink.c:3529 rtnetlink_rcv_msg+0x745/0x7e0 net/core/rtnetlink.c:5594 netlink_rcv_skb+0x14e/0x250 net/netlink/af_netlink.c:2494 rtnetlink_rcv+0x18/0x20 net/core/rtnetlink.c:5612 netlink_unicast_kernel net/netlink/af_netlink.c:1317 [inline] netlink_unicast+0x602/0x6d0 net/netlink/af_netlink.c:1343 netlink_sendmsg+0x728/0x850 net/netlink/af_netlink.c:1919 sock_sendmsg_nosec net/socket.c:705 [inline] sock_sendmsg net/socket.c:725 [inline] ____sys_sendmsg+0x39a/0x510 net/socket.c:2413 ___sys_sendmsg net/socket.c:2467 [inline] __sys_sendmsg+0x195/0x230 net/socket.c:2496 __do_sys_sendmsg net/socket.c:2505 [inline] __se_sys_sendmsg net/socket.c:2503 [inline] __x64_sys_sendmsg+0x42/0x50 net/socket.c:2503 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae read to 0xffff888168ad6410 of 8 bytes by task 13560 on cpu 0: qdisc_lookup_rcu+0x30/0x2e0 net/sched/sch_api.c:323 __tcf_qdisc_find+0x74/0x3a0 net/sched/cls_api.c:1050 tc_del_tfilter+0x1c7/0x1350 net/sched/cls_api.c:2211 rtnetlink_rcv_msg+0x5ba/0x7e0 net/core/rtnetlink.c:5585 netlink_rcv_skb+0x14e/0x250 net/netlink/af_netlink.c:2494 rtnetlink_rcv+0x18/0x20 net/core/rtnetlink.c:5612 netlink_unicast_kernel net/netlink/af_netlink.c:1317 [inline] netlink_unicast+0x602/0x6d0 net/netlink/af_netlink.c:1343 netlink_sendmsg+0x728/0x850 net/netlink/af_netlink.c:1919 sock_sendmsg_nosec net/socket.c:705 [inline] sock_sendmsg net/socket.c:725 [inline] ____sys_sendmsg+0x39a/0x510 net/socket.c:2413 ___sys_sendmsg net/socket.c:2467 [inline] __sys_sendmsg+0x195/0x230 net/socket.c:2496 __do_sys_sendmsg net/socket.c:2505 [inline] __se_sys_sendmsg net/socket.c:2503 [inline] __x64_sys_sendmsg+0x42/0x50 net/socket.c:2503 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae value changed: 0xffffffff85dee080 -> 0xffff88815d96ec00 Reported by Kernel Concurrency Sanitizer on: CPU: 0 PID: 13560 Comm: syz-executor.2 Not tainted 5.17.0-rc3-syzkaller-00116-gf1baf68e1383-dirty #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Fixes: 470502de5bdb ("net: sched: unlock rules update API") Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Vlad Buslov <vladbu@mellanox.com> Reported-by: syzbot <syzkaller@googlegroups.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: Jiri Pirko <jiri@resnulli.us> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-02-11 23:06:23 +03:00
struct Qdisc *qdisc = rtnl_dereference(dev->qdisc);
net: sched: update default qdisc visibility after Tx queue cnt changes mq / mqprio make the default child qdiscs visible. They only do so for the qdiscs which are within real_num_tx_queues when the device is registered. Depending on order of calls in the driver, or if user space changes config via ethtool -L the number of qdiscs visible under tc qdisc show will differ from the number of queues. This is confusing to users and potentially to system configuration scripts which try to make sure qdiscs have the right parameters. Add a new Qdisc_ops callback and make relevant qdiscs TTRT. Note that this uncovers the "shortcut" created by commit 1f27cde313d7 ("net: sched: use pfifo_fast for non real queues") The default child qdiscs beyond initial real_num_tx are always pfifo_fast, no matter what the sysfs setting is. Fixing this gets a little tricky because we'd need to keep a reference on whatever the default qdisc was at the time of creation. In practice this is likely an non-issue the qdiscs likely have to be configured to non-default settings, so whatever user space is doing such configuration can replace the pfifos... now that it will see them. Reported-by: Matthew Massey <matthewmassey@fb.com> Reviewed-by: Dave Taht <dave.taht@gmail.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-09-14 01:53:30 +03:00
if (qdisc->ops->change_real_num_tx)
qdisc->ops->change_real_num_tx(qdisc, new_real_tx);
}
net: sched: move and reuse mq_change_real_num_tx() The code for handling active queue changes is identical between mq and mqprio, reuse it. Suggested-by: Cong Wang <cong.wang@bytedance.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-09-17 16:55:06 +03:00
void mq_change_real_num_tx(struct Qdisc *sch, unsigned int new_real_tx)
{
#ifdef CONFIG_NET_SCHED
struct net_device *dev = qdisc_dev(sch);
struct Qdisc *qdisc;
unsigned int i;
for (i = new_real_tx; i < dev->real_num_tx_queues; i++) {
qdisc = netdev_get_tx_queue(dev, i)->qdisc_sleeping;
/* Only update the default qdiscs we created,
* qdiscs with handles are always hashed.
*/
if (qdisc != &noop_qdisc && !qdisc->handle)
qdisc_hash_del(qdisc);
}
for (i = dev->real_num_tx_queues; i < new_real_tx; i++) {
qdisc = netdev_get_tx_queue(dev, i)->qdisc_sleeping;
if (qdisc != &noop_qdisc && !qdisc->handle)
qdisc_hash_add(qdisc, false);
}
#endif
}
EXPORT_SYMBOL(mq_change_real_num_tx);
net_sched: plug in qdisc ops change_tx_queue_len Introduce a new qdisc ops ->change_tx_queue_len() so that each qdisc could decide how to implement this if it wants. Previously we simply read dev->tx_queue_len, after pfifo_fast switches to skb array, we need this API to resize the skb array when we change dev->tx_queue_len. To avoid handling race conditions with TX BH, we need to deactivate all TX queues before change the value and bring them back after we are done, this also makes implementation easier. Cc: John Fastabend <john.fastabend@gmail.com> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-01-26 05:26:23 +03:00
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;
}
netdev: Move rest of qdisc state into struct netdev_queue Now qdisc, qdisc_sleeping, and qdisc_list also live there. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-09 04:42:10 +04:00
static void dev_init_scheduler_queue(struct net_device *dev,
struct netdev_queue *dev_queue,
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
void *_qdisc)
netdev: Move rest of qdisc state into struct netdev_queue Now qdisc, qdisc_sleeping, and qdisc_list also live there. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-09 04:42:10 +04:00
{
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
struct Qdisc *qdisc = _qdisc;
net: qdisc: use rcu prefix and silence sparse warnings Add __rcu notation to qdisc handling by doing this we can make smatch output more legible. And anyways some of the cases should be using rcu_dereference() see qdisc_all_tx_empty(), qdisc_tx_chainging(), and so on. Also *wake_queue() API is commonly called from driver timer routines without rcu lock or rtnl lock. So I added rcu_read_lock() blocks around netif_wake_subqueue and netif_tx_wake_queue. Signed-off-by: John Fastabend <john.r.fastabend@intel.com> Acked-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-13 07:04:52 +04:00
rcu_assign_pointer(dev_queue->qdisc, qdisc);
netdev: Move rest of qdisc state into struct netdev_queue Now qdisc, qdisc_sleeping, and qdisc_list also live there. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-09 04:42:10 +04:00
dev_queue->qdisc_sleeping = qdisc;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
void dev_init_scheduler(struct net_device *dev)
{
net_sched: add __rcu annotation to netdev->qdisc syzbot found a data-race [1] which lead me to add __rcu annotations to netdev->qdisc, and proper accessors to get LOCKDEP support. [1] BUG: KCSAN: data-race in dev_activate / qdisc_lookup_rcu write to 0xffff888168ad6410 of 8 bytes by task 13559 on cpu 1: attach_default_qdiscs net/sched/sch_generic.c:1167 [inline] dev_activate+0x2ed/0x8f0 net/sched/sch_generic.c:1221 __dev_open+0x2e9/0x3a0 net/core/dev.c:1416 __dev_change_flags+0x167/0x3f0 net/core/dev.c:8139 rtnl_configure_link+0xc2/0x150 net/core/rtnetlink.c:3150 __rtnl_newlink net/core/rtnetlink.c:3489 [inline] rtnl_newlink+0xf4d/0x13e0 net/core/rtnetlink.c:3529 rtnetlink_rcv_msg+0x745/0x7e0 net/core/rtnetlink.c:5594 netlink_rcv_skb+0x14e/0x250 net/netlink/af_netlink.c:2494 rtnetlink_rcv+0x18/0x20 net/core/rtnetlink.c:5612 netlink_unicast_kernel net/netlink/af_netlink.c:1317 [inline] netlink_unicast+0x602/0x6d0 net/netlink/af_netlink.c:1343 netlink_sendmsg+0x728/0x850 net/netlink/af_netlink.c:1919 sock_sendmsg_nosec net/socket.c:705 [inline] sock_sendmsg net/socket.c:725 [inline] ____sys_sendmsg+0x39a/0x510 net/socket.c:2413 ___sys_sendmsg net/socket.c:2467 [inline] __sys_sendmsg+0x195/0x230 net/socket.c:2496 __do_sys_sendmsg net/socket.c:2505 [inline] __se_sys_sendmsg net/socket.c:2503 [inline] __x64_sys_sendmsg+0x42/0x50 net/socket.c:2503 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae read to 0xffff888168ad6410 of 8 bytes by task 13560 on cpu 0: qdisc_lookup_rcu+0x30/0x2e0 net/sched/sch_api.c:323 __tcf_qdisc_find+0x74/0x3a0 net/sched/cls_api.c:1050 tc_del_tfilter+0x1c7/0x1350 net/sched/cls_api.c:2211 rtnetlink_rcv_msg+0x5ba/0x7e0 net/core/rtnetlink.c:5585 netlink_rcv_skb+0x14e/0x250 net/netlink/af_netlink.c:2494 rtnetlink_rcv+0x18/0x20 net/core/rtnetlink.c:5612 netlink_unicast_kernel net/netlink/af_netlink.c:1317 [inline] netlink_unicast+0x602/0x6d0 net/netlink/af_netlink.c:1343 netlink_sendmsg+0x728/0x850 net/netlink/af_netlink.c:1919 sock_sendmsg_nosec net/socket.c:705 [inline] sock_sendmsg net/socket.c:725 [inline] ____sys_sendmsg+0x39a/0x510 net/socket.c:2413 ___sys_sendmsg net/socket.c:2467 [inline] __sys_sendmsg+0x195/0x230 net/socket.c:2496 __do_sys_sendmsg net/socket.c:2505 [inline] __se_sys_sendmsg net/socket.c:2503 [inline] __x64_sys_sendmsg+0x42/0x50 net/socket.c:2503 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae value changed: 0xffffffff85dee080 -> 0xffff88815d96ec00 Reported by Kernel Concurrency Sanitizer on: CPU: 0 PID: 13560 Comm: syz-executor.2 Not tainted 5.17.0-rc3-syzkaller-00116-gf1baf68e1383-dirty #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Fixes: 470502de5bdb ("net: sched: unlock rules update API") Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Vlad Buslov <vladbu@mellanox.com> Reported-by: syzbot <syzkaller@googlegroups.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: Jiri Pirko <jiri@resnulli.us> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-02-11 23:06:23 +03:00
rcu_assign_pointer(dev->qdisc, &noop_qdisc);
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
netdev_for_each_tx_queue(dev, dev_init_scheduler_queue, &noop_qdisc);
net: dynamic ingress_queue allocation ingress being not used very much, and net_device->ingress_queue being quite a big object (128 or 256 bytes), use a dynamic allocation if needed (tc qdisc add dev eth0 ingress ...) dev_ingress_queue(dev) helper should be used only with RTNL taken. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-10-02 10:11:55 +04:00
if (dev_ingress_queue(dev))
dev_init_scheduler_queue(dev, dev_ingress_queue(dev), &noop_qdisc);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
net: sched: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Add pointer back to Qdisc. Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: Jiri Pirko <jiri@resnulli.us> Cc: "David S. Miller" <davem@davemloft.net> Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-10-17 03:29:17 +03:00
timer_setup(&dev->watchdog_timer, dev_watchdog, 0);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
}
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
static void shutdown_scheduler_queue(struct net_device *dev,
struct netdev_queue *dev_queue,
void *_qdisc_default)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
{
netdev: Move rest of qdisc state into struct netdev_queue Now qdisc, qdisc_sleeping, and qdisc_list also live there. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-09 04:42:10 +04:00
struct Qdisc *qdisc = dev_queue->qdisc_sleeping;
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
struct Qdisc *qdisc_default = _qdisc_default;
netdev: Move rest of qdisc state into struct netdev_queue Now qdisc, qdisc_sleeping, and qdisc_list also live there. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-09 04:42:10 +04:00
if (qdisc) {
pkt_sched: Use rcu_assign_pointer() to change dev_queue->qdisc These pointers are RCU protected, so proper primitives should be used. Signed-off-by: Jarek Poplawski <jarkao2@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-08-27 13:22:07 +04:00
rcu_assign_pointer(dev_queue->qdisc, qdisc_default);
netdev: Move rest of qdisc state into struct netdev_queue Now qdisc, qdisc_sleeping, and qdisc_list also live there. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-09 04:42:10 +04:00
dev_queue->qdisc_sleeping = qdisc_default;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
net: sched: rename qdisc_destroy() to qdisc_put() Current implementation of qdisc_destroy() decrements Qdisc reference counter and only actually destroy Qdisc if reference counter value reached zero. Rename qdisc_destroy() to qdisc_put() in order for it to better describe the way in which this function currently implemented and used. Extract code that deallocates Qdisc into new private qdisc_destroy() function. It is intended to be shared between regular qdisc_put() and its unlocked version that is introduced in next patch in this series. Signed-off-by: Vlad Buslov <vladbu@mellanox.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-24 19:22:50 +03:00
qdisc_put(qdisc);
[NET] SCHED: Fix whitespace errors. Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-02-09 17:25:16 +03:00
}
netdev: Move rest of qdisc state into struct netdev_queue Now qdisc, qdisc_sleeping, and qdisc_list also live there. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-09 04:42:10 +04:00
}
void dev_shutdown(struct net_device *dev)
{
netdev: Allocate multiple queues for TX. alloc_netdev_mq() now allocates an array of netdev_queue structures for TX, based upon the queue_count argument. Furthermore, all accesses to the TX queues are now vectored through the netdev_get_tx_queue() and netdev_for_each_tx_queue() interfaces. This makes it easy to grep the tree for all things that want to get to a TX queue of a net device. Problem spots which are not really multiqueue aware yet, and only work with one queue, can easily be spotted by grepping for all netdev_get_tx_queue() calls that pass in a zero index. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 11:34:19 +04:00
netdev_for_each_tx_queue(dev, shutdown_scheduler_queue, &noop_qdisc);
net: dynamic ingress_queue allocation ingress being not used very much, and net_device->ingress_queue being quite a big object (128 or 256 bytes), use a dynamic allocation if needed (tc qdisc add dev eth0 ingress ...) dev_ingress_queue(dev) helper should be used only with RTNL taken. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-10-02 10:11:55 +04:00
if (dev_ingress_queue(dev))
shutdown_scheduler_queue(dev, dev_ingress_queue(dev), &noop_qdisc);
net_sched: add __rcu annotation to netdev->qdisc syzbot found a data-race [1] which lead me to add __rcu annotations to netdev->qdisc, and proper accessors to get LOCKDEP support. [1] BUG: KCSAN: data-race in dev_activate / qdisc_lookup_rcu write to 0xffff888168ad6410 of 8 bytes by task 13559 on cpu 1: attach_default_qdiscs net/sched/sch_generic.c:1167 [inline] dev_activate+0x2ed/0x8f0 net/sched/sch_generic.c:1221 __dev_open+0x2e9/0x3a0 net/core/dev.c:1416 __dev_change_flags+0x167/0x3f0 net/core/dev.c:8139 rtnl_configure_link+0xc2/0x150 net/core/rtnetlink.c:3150 __rtnl_newlink net/core/rtnetlink.c:3489 [inline] rtnl_newlink+0xf4d/0x13e0 net/core/rtnetlink.c:3529 rtnetlink_rcv_msg+0x745/0x7e0 net/core/rtnetlink.c:5594 netlink_rcv_skb+0x14e/0x250 net/netlink/af_netlink.c:2494 rtnetlink_rcv+0x18/0x20 net/core/rtnetlink.c:5612 netlink_unicast_kernel net/netlink/af_netlink.c:1317 [inline] netlink_unicast+0x602/0x6d0 net/netlink/af_netlink.c:1343 netlink_sendmsg+0x728/0x850 net/netlink/af_netlink.c:1919 sock_sendmsg_nosec net/socket.c:705 [inline] sock_sendmsg net/socket.c:725 [inline] ____sys_sendmsg+0x39a/0x510 net/socket.c:2413 ___sys_sendmsg net/socket.c:2467 [inline] __sys_sendmsg+0x195/0x230 net/socket.c:2496 __do_sys_sendmsg net/socket.c:2505 [inline] __se_sys_sendmsg net/socket.c:2503 [inline] __x64_sys_sendmsg+0x42/0x50 net/socket.c:2503 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae read to 0xffff888168ad6410 of 8 bytes by task 13560 on cpu 0: qdisc_lookup_rcu+0x30/0x2e0 net/sched/sch_api.c:323 __tcf_qdisc_find+0x74/0x3a0 net/sched/cls_api.c:1050 tc_del_tfilter+0x1c7/0x1350 net/sched/cls_api.c:2211 rtnetlink_rcv_msg+0x5ba/0x7e0 net/core/rtnetlink.c:5585 netlink_rcv_skb+0x14e/0x250 net/netlink/af_netlink.c:2494 rtnetlink_rcv+0x18/0x20 net/core/rtnetlink.c:5612 netlink_unicast_kernel net/netlink/af_netlink.c:1317 [inline] netlink_unicast+0x602/0x6d0 net/netlink/af_netlink.c:1343 netlink_sendmsg+0x728/0x850 net/netlink/af_netlink.c:1919 sock_sendmsg_nosec net/socket.c:705 [inline] sock_sendmsg net/socket.c:725 [inline] ____sys_sendmsg+0x39a/0x510 net/socket.c:2413 ___sys_sendmsg net/socket.c:2467 [inline] __sys_sendmsg+0x195/0x230 net/socket.c:2496 __do_sys_sendmsg net/socket.c:2505 [inline] __se_sys_sendmsg net/socket.c:2503 [inline] __x64_sys_sendmsg+0x42/0x50 net/socket.c:2503 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae value changed: 0xffffffff85dee080 -> 0xffff88815d96ec00 Reported by Kernel Concurrency Sanitizer on: CPU: 0 PID: 13560 Comm: syz-executor.2 Not tainted 5.17.0-rc3-syzkaller-00116-gf1baf68e1383-dirty #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Fixes: 470502de5bdb ("net: sched: unlock rules update API") Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Vlad Buslov <vladbu@mellanox.com> Reported-by: syzbot <syzkaller@googlegroups.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: Jiri Pirko <jiri@resnulli.us> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-02-11 23:06:23 +03:00
qdisc_put(rtnl_dereference(dev->qdisc));
rcu_assign_pointer(dev->qdisc, &noop_qdisc);
net_sched: reintroduce dev->qdisc for use by sch_api Currently the multiqueue integration with the qdisc API suffers from a few problems: - with multiple queues, all root qdiscs use the same handle. This means they can't be exposed to userspace in a backwards compatible fashion. - all API operations always refer to queue number 0. Newly created qdiscs are automatically shared between all queues, its not possible to address individual queues or restore multiqueue behaviour once a shared qdisc has been attached. - Dumps only contain the root qdisc of queue 0, in case of non-shared qdiscs this means the statistics are incomplete. This patch reintroduces dev->qdisc, which points to the (single) root qdisc from userspace's point of view. Currently it either points to the first (non-shared) default qdisc, or a qdisc shared between all queues. The following patches will introduce a classful dummy qdisc, which will be used as root qdisc and contain the per-queue qdiscs as children. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-04 10:41:18 +04:00
net: convert BUG_TRAP to generic WARN_ON Removes legacy reinvent-the-wheel type thing. The generic machinery integrates much better to automated debugging aids such as kerneloops.org (and others), and is unambiguous due to better naming. Non-intuively BUG_TRAP() is actually equal to WARN_ON() rather than BUG_ON() though some might actually be promoted to BUG_ON() but I left that to future. I could make at least one BUILD_BUG_ON conversion. Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-26 08:43:18 +04:00
WARN_ON(timer_pending(&dev->watchdog_timer));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 02:20:36 +04:00
}
sch: make htb_rate_cfg and functions around that generic As it is going to be used in tbf as well, push these to generic code. Signed-off-by: Jiri Pirko <jiri@resnulli.us> Acked-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-02-12 04:12:03 +04:00
net/sched: act_police: add support for packet-per-second policing Allow a policer action to enforce a rate-limit based on packets-per-second, configurable using a packet-per-second rate and burst parameters. e.g. tc filter add dev tap1 parent ffff: u32 match \ u32 0 0 police pkts_rate 3000 pkts_burst 1000 Testing was unable to uncover a performance impact of this change on existing features. Signed-off-by: Baowen Zheng <baowen.zheng@corigine.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: Louis Peens <louis.peens@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-03-12 17:08:31 +03:00
/**
* psched_ratecfg_precompute__() - Pre-compute values for reciprocal division
* @rate: Rate to compute reciprocal division values of
* @mult: Multiplier for reciprocal division
* @shift: Shift for reciprocal division
*
* The multiplier and shift for reciprocal division by rate are stored
* in mult and shift.
*
* 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.
*
* reciprocal_value() is not used here it doesn't handle 64-bit values.
*/
static void psched_ratecfg_precompute__(u64 rate, u32 *mult, u8 *shift)
{
u64 factor = NSEC_PER_SEC;
*mult = 1;
*shift = 0;
if (rate <= 0)
return;
for (;;) {
*mult = div64_u64(factor, rate);
if (*mult & (1U << 31) || factor & (1ULL << 63))
break;
factor <<= 1;
(*shift)++;
}
}
net_sched: restore "overhead xxx" handling commit 56b765b79 ("htb: improved accuracy at high rates") broke the "overhead xxx" handling, as well as the "linklayer atm" attribute. tc class add ... htb rate X ceil Y linklayer atm overhead 10 This patch restores the "overhead xxx" handling, for htb, tbf and act_police The "linklayer atm" thing needs a separate fix. Reported-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Vimalkumar <j.vimal@gmail.com> Cc: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-06-02 17:55:05 +04:00
void psched_ratecfg_precompute(struct psched_ratecfg *r,
net_sched: add u64 rate to psched_ratecfg_precompute() Add an extra u64 rate parameter to psched_ratecfg_precompute() so that some qdisc can opt-in for 64bit rates in the future, to overcome the ~34 Gbits limit. psched_ratecfg_getrate() reports a legacy structure to tc utility, so if actual rate is above the 32bit rate field, cap it to the 34Gbit limit. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-09-19 20:10:03 +04:00
const struct tc_ratespec *conf,
u64 rate64)
sch: make htb_rate_cfg and functions around that generic As it is going to be used in tbf as well, push these to generic code. Signed-off-by: Jiri Pirko <jiri@resnulli.us> Acked-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-02-12 04:12:03 +04:00
{
net_sched: restore "overhead xxx" handling commit 56b765b79 ("htb: improved accuracy at high rates") broke the "overhead xxx" handling, as well as the "linklayer atm" attribute. tc class add ... htb rate X ceil Y linklayer atm overhead 10 This patch restores the "overhead xxx" handling, for htb, tbf and act_police The "linklayer atm" thing needs a separate fix. Reported-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Vimalkumar <j.vimal@gmail.com> Cc: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-06-02 17:55:05 +04:00
memset(r, 0, sizeof(*r));
r->overhead = conf->overhead;
net_sched: restore "mpu xxx" handling commit 56b765b79e9a ("htb: improved accuracy at high rates") broke "overhead X", "linklayer atm" and "mpu X" attributes. "overhead X" and "linklayer atm" have already been fixed. This restores the "mpu X" handling, as might be used by DOCSIS or Ethernet shaping: tc class add ... htb rate X overhead 4 mpu 64 The code being fixed is used by htb, tbf and act_police. Cake has its own mpu handling. qdisc_calculate_pkt_len still uses the size table containing values adjusted for mpu by user space. iproute2 tc has always passed mpu into the kernel via a tc_ratespec structure, but the kernel never directly acted on it, merely stored it so that it could be read back by `tc class show`. Rather, tc would generate length-to-time tables that included the mpu (and linklayer) in their construction, and the kernel used those tables. Since v3.7, the tables were no longer used. Along with "mpu", this also broke "overhead" and "linklayer" which were fixed in 01cb71d2d47b ("net_sched: restore "overhead xxx" handling", v3.10) and 8a8e3d84b171 ("net_sched: restore "linklayer atm" handling", v3.11). "overhead" was fixed by simply restoring use of tc_ratespec::overhead - this had originally been used by the kernel but was initially omitted from the new non-table-based calculations. "linklayer" had been handled in the table like "mpu", but the mode was not originally passed in tc_ratespec. The new implementation was made to handle it by getting new versions of tc to pass the mode in an extended tc_ratespec, and for older versions of tc the table contents were analysed at load time to deduce linklayer. As "mpu" has always been given to the kernel in tc_ratespec, accompanying the mpu-based table, we can restore system functionality with no userspace change by making the kernel act on the tc_ratespec value. Fixes: 56b765b79e9a ("htb: improved accuracy at high rates") Signed-off-by: Kevin Bracey <kevin@bracey.fi> Cc: Eric Dumazet <edumazet@google.com> Cc: Jiri Pirko <jiri@resnulli.us> Cc: Vimalkumar <j.vimal@gmail.com> Link: https://lore.kernel.org/r/20220112170210.1014351-1-kevin@bracey.fi Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-01-12 20:02:10 +03:00
r->mpu = conf->mpu;
net_sched: add u64 rate to psched_ratecfg_precompute() Add an extra u64 rate parameter to psched_ratecfg_precompute() so that some qdisc can opt-in for 64bit rates in the future, to overcome the ~34 Gbits limit. psched_ratecfg_getrate() reports a legacy structure to tc utility, so if actual rate is above the 32bit rate field, cap it to the 34Gbit limit. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-09-19 20:10:03 +04:00
r->rate_bytes_ps = max_t(u64, conf->rate, rate64);
net_sched: restore "linklayer atm" handling commit 56b765b79 ("htb: improved accuracy at high rates") broke the "linklayer atm" handling. tc class add ... htb rate X ceil Y linklayer atm The linklayer setting is implemented by modifying the rate table which is send to the kernel. No direct parameter were transferred to the kernel indicating the linklayer setting. The commit 56b765b79 ("htb: improved accuracy at high rates") removed the use of the rate table system. To keep compatible with older iproute2 utils, this patch detects the linklayer by parsing the rate table. It also supports future versions of iproute2 to send this linklayer parameter to the kernel directly. This is done by using the __reserved field in struct tc_ratespec, to convey the choosen linklayer option, but only using the lower 4 bits of this field. Linklayer detection is limited to speeds below 100Mbit/s, because at high rates the rtab is gets too inaccurate, so bad that several fields contain the same values, this resembling the ATM detect. Fields even start to contain "0" time to send, e.g. at 1000Mbit/s sending a 96 bytes packet cost "0", thus the rtab have been more broken than we first realized. Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-08-15 01:47:11 +04:00
r->linklayer = (conf->linklayer & TC_LINKLAYER_MASK);
net/sched: act_police: add support for packet-per-second policing Allow a policer action to enforce a rate-limit based on packets-per-second, configurable using a packet-per-second rate and burst parameters. e.g. tc filter add dev tap1 parent ffff: u32 match \ u32 0 0 police pkts_rate 3000 pkts_burst 1000 Testing was unable to uncover a performance impact of this change on existing features. Signed-off-by: Baowen Zheng <baowen.zheng@corigine.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: Louis Peens <louis.peens@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-03-12 17:08:31 +03:00
psched_ratecfg_precompute__(r->rate_bytes_ps, &r->mult, &r->shift);
sch: make htb_rate_cfg and functions around that generic As it is going to be used in tbf as well, push these to generic code. Signed-off-by: Jiri Pirko <jiri@resnulli.us> Acked-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-02-12 04:12:03 +04:00
}
EXPORT_SYMBOL(psched_ratecfg_precompute);
net: core: introduce mini_Qdisc and eliminate usage of tp->q for clsact fastpath In sch_handle_egress and sch_handle_ingress tp->q is used only in order to update stats. So stats and filter list are the only things that are needed in clsact qdisc fastpath processing. Introduce new mini_Qdisc struct to hold those items. Also, introduce a helper to swap the mini_Qdisc structures in case filter list head changes. This removes need for tp->q usage without added overhead. Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-11-03 13:46:25 +03:00
net/sched: act_police: add support for packet-per-second policing Allow a policer action to enforce a rate-limit based on packets-per-second, configurable using a packet-per-second rate and burst parameters. e.g. tc filter add dev tap1 parent ffff: u32 match \ u32 0 0 police pkts_rate 3000 pkts_burst 1000 Testing was unable to uncover a performance impact of this change on existing features. Signed-off-by: Baowen Zheng <baowen.zheng@corigine.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: Louis Peens <louis.peens@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-03-12 17:08:31 +03:00
void psched_ppscfg_precompute(struct psched_pktrate *r, u64 pktrate64)
{
r->rate_pkts_ps = pktrate64;
psched_ratecfg_precompute__(r->rate_pkts_ps, &r->mult, &r->shift);
}
EXPORT_SYMBOL(psched_ppscfg_precompute);
net: core: introduce mini_Qdisc and eliminate usage of tp->q for clsact fastpath In sch_handle_egress and sch_handle_ingress tp->q is used only in order to update stats. So stats and filter list are the only things that are needed in clsact qdisc fastpath processing. Introduce new mini_Qdisc struct to hold those items. Also, introduce a helper to swap the mini_Qdisc structures in case filter list head changes. This removes need for tp->q usage without added overhead. Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-11-03 13:46:25 +03:00
void mini_qdisc_pair_swap(struct mini_Qdisc_pair *miniqp,
struct tcf_proto *tp_head)
{
net: sched: protect filter_chain list with filter_chain_lock mutex Extend tcf_chain with new filter_chain_lock mutex. Always lock the chain when accessing filter_chain list, instead of relying on rtnl lock. Dereference filter_chain with tcf_chain_dereference() lockdep macro to verify that all users of chain_list have the lock taken. Rearrange tp insert/remove code in tc_new_tfilter/tc_del_tfilter to execute all necessary code while holding chain lock in order to prevent invalidation of chain_info structure by potential concurrent change. This also serializes calls to tcf_chain0_head_change(), which allows head change callbacks to rely on filter_chain_lock for synchronization instead of rtnl mutex. Signed-off-by: Vlad Buslov <vladbu@mellanox.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-02-11 11:55:38 +03:00
/* Protected with chain0->filter_chain_lock.
* Can't access chain directly because tp_head can be NULL.
*/
struct mini_Qdisc *miniq_old =
rcu_dereference_protected(*miniqp->p_miniq, 1);
net: core: introduce mini_Qdisc and eliminate usage of tp->q for clsact fastpath In sch_handle_egress and sch_handle_ingress tp->q is used only in order to update stats. So stats and filter list are the only things that are needed in clsact qdisc fastpath processing. Introduce new mini_Qdisc struct to hold those items. Also, introduce a helper to swap the mini_Qdisc structures in case filter list head changes. This removes need for tp->q usage without added overhead. Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-11-03 13:46:25 +03:00
struct mini_Qdisc *miniq;
if (!tp_head) {
RCU_INIT_POINTER(*miniqp->p_miniq, NULL);
net: sch: eliminate unnecessary RCU waits in mini_qdisc_pair_swap() Currently rcu_barrier() is used to ensure that no readers of the inactive mini_Qdisc buffer remain before it is reused. This waits for any pending RCU callbacks to complete, when all that is actually required is to wait for one RCU grace period to elapse after the buffer was made inactive. This means that using rcu_barrier() may result in unnecessary waits. To improve this, store the current RCU state when a buffer is made inactive and use poll_state_synchronize_rcu() to check whether a full grace period has elapsed before reusing it. If a full grace period has not elapsed, wait for a grace period to elapse, and in the non-RT case use synchronize_rcu_expedited() to hasten it. Since this approach eliminates the RCU callback it is no longer necessary to synchronize_rcu() in the tp_head==NULL case. However, the RCU state should still be saved for the previously active buffer. Before this change I would typically see mini_qdisc_pair_swap() take tens of milliseconds to complete. After this change it typcially finishes in less than 1 ms, and often it takes just a few microseconds. Thanks to Paul for walking me through the options for improving this. Cc: "Paul E. McKenney" <paulmck@kernel.org> Signed-off-by: Seth Forshee <sforshee@digitalocean.com> Link: https://lore.kernel.org/r/20211026130700.121189-1-seth@forshee.me Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-10-26 16:06:59 +03:00
} else {
net: sch: simplify condtion for selecting mini_Qdisc_pair buffer The only valid values for a miniq pointer are NULL or a pointer to miniq1 or miniq2, so testing for miniq_old != &miniq1 is functionally equivalent to testing that it is NULL or equal to &miniq2. Suggested-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Seth Forshee <sforshee@digitalocean.com> Link: https://lore.kernel.org/r/20211026183721.137930-1-seth@forshee.me Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-10-26 21:37:21 +03:00
miniq = miniq_old != &miniqp->miniq1 ?
net: sch: eliminate unnecessary RCU waits in mini_qdisc_pair_swap() Currently rcu_barrier() is used to ensure that no readers of the inactive mini_Qdisc buffer remain before it is reused. This waits for any pending RCU callbacks to complete, when all that is actually required is to wait for one RCU grace period to elapse after the buffer was made inactive. This means that using rcu_barrier() may result in unnecessary waits. To improve this, store the current RCU state when a buffer is made inactive and use poll_state_synchronize_rcu() to check whether a full grace period has elapsed before reusing it. If a full grace period has not elapsed, wait for a grace period to elapse, and in the non-RT case use synchronize_rcu_expedited() to hasten it. Since this approach eliminates the RCU callback it is no longer necessary to synchronize_rcu() in the tp_head==NULL case. However, the RCU state should still be saved for the previously active buffer. Before this change I would typically see mini_qdisc_pair_swap() take tens of milliseconds to complete. After this change it typcially finishes in less than 1 ms, and often it takes just a few microseconds. Thanks to Paul for walking me through the options for improving this. Cc: "Paul E. McKenney" <paulmck@kernel.org> Signed-off-by: Seth Forshee <sforshee@digitalocean.com> Link: https://lore.kernel.org/r/20211026130700.121189-1-seth@forshee.me Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-10-26 16:06:59 +03:00
&miniqp->miniq1 : &miniqp->miniq2;
net: core: introduce mini_Qdisc and eliminate usage of tp->q for clsact fastpath In sch_handle_egress and sch_handle_ingress tp->q is used only in order to update stats. So stats and filter list are the only things that are needed in clsact qdisc fastpath processing. Introduce new mini_Qdisc struct to hold those items. Also, introduce a helper to swap the mini_Qdisc structures in case filter list head changes. This removes need for tp->q usage without added overhead. Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-11-03 13:46:25 +03:00
net: sch: eliminate unnecessary RCU waits in mini_qdisc_pair_swap() Currently rcu_barrier() is used to ensure that no readers of the inactive mini_Qdisc buffer remain before it is reused. This waits for any pending RCU callbacks to complete, when all that is actually required is to wait for one RCU grace period to elapse after the buffer was made inactive. This means that using rcu_barrier() may result in unnecessary waits. To improve this, store the current RCU state when a buffer is made inactive and use poll_state_synchronize_rcu() to check whether a full grace period has elapsed before reusing it. If a full grace period has not elapsed, wait for a grace period to elapse, and in the non-RT case use synchronize_rcu_expedited() to hasten it. Since this approach eliminates the RCU callback it is no longer necessary to synchronize_rcu() in the tp_head==NULL case. However, the RCU state should still be saved for the previously active buffer. Before this change I would typically see mini_qdisc_pair_swap() take tens of milliseconds to complete. After this change it typcially finishes in less than 1 ms, and often it takes just a few microseconds. Thanks to Paul for walking me through the options for improving this. Cc: "Paul E. McKenney" <paulmck@kernel.org> Signed-off-by: Seth Forshee <sforshee@digitalocean.com> Link: https://lore.kernel.org/r/20211026130700.121189-1-seth@forshee.me Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-10-26 16:06:59 +03:00
/* We need to make sure that readers won't see the miniq
* we are about to modify. So ensure that at least one RCU
* grace period has elapsed since the miniq was made
* inactive.
*/
if (IS_ENABLED(CONFIG_PREEMPT_RT))
cond_synchronize_rcu(miniq->rcu_state);
else if (!poll_state_synchronize_rcu(miniq->rcu_state))
synchronize_rcu_expedited();
net: core: introduce mini_Qdisc and eliminate usage of tp->q for clsact fastpath In sch_handle_egress and sch_handle_ingress tp->q is used only in order to update stats. So stats and filter list are the only things that are needed in clsact qdisc fastpath processing. Introduce new mini_Qdisc struct to hold those items. Also, introduce a helper to swap the mini_Qdisc structures in case filter list head changes. This removes need for tp->q usage without added overhead. Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-11-03 13:46:25 +03:00
net: sch: eliminate unnecessary RCU waits in mini_qdisc_pair_swap() Currently rcu_barrier() is used to ensure that no readers of the inactive mini_Qdisc buffer remain before it is reused. This waits for any pending RCU callbacks to complete, when all that is actually required is to wait for one RCU grace period to elapse after the buffer was made inactive. This means that using rcu_barrier() may result in unnecessary waits. To improve this, store the current RCU state when a buffer is made inactive and use poll_state_synchronize_rcu() to check whether a full grace period has elapsed before reusing it. If a full grace period has not elapsed, wait for a grace period to elapse, and in the non-RT case use synchronize_rcu_expedited() to hasten it. Since this approach eliminates the RCU callback it is no longer necessary to synchronize_rcu() in the tp_head==NULL case. However, the RCU state should still be saved for the previously active buffer. Before this change I would typically see mini_qdisc_pair_swap() take tens of milliseconds to complete. After this change it typcially finishes in less than 1 ms, and often it takes just a few microseconds. Thanks to Paul for walking me through the options for improving this. Cc: "Paul E. McKenney" <paulmck@kernel.org> Signed-off-by: Seth Forshee <sforshee@digitalocean.com> Link: https://lore.kernel.org/r/20211026130700.121189-1-seth@forshee.me Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-10-26 16:06:59 +03:00
miniq->filter_list = tp_head;
rcu_assign_pointer(*miniqp->p_miniq, miniq);
}
net: core: introduce mini_Qdisc and eliminate usage of tp->q for clsact fastpath In sch_handle_egress and sch_handle_ingress tp->q is used only in order to update stats. So stats and filter list are the only things that are needed in clsact qdisc fastpath processing. Introduce new mini_Qdisc struct to hold those items. Also, introduce a helper to swap the mini_Qdisc structures in case filter list head changes. This removes need for tp->q usage without added overhead. Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-11-03 13:46:25 +03:00
if (miniq_old)
net: sch: eliminate unnecessary RCU waits in mini_qdisc_pair_swap() Currently rcu_barrier() is used to ensure that no readers of the inactive mini_Qdisc buffer remain before it is reused. This waits for any pending RCU callbacks to complete, when all that is actually required is to wait for one RCU grace period to elapse after the buffer was made inactive. This means that using rcu_barrier() may result in unnecessary waits. To improve this, store the current RCU state when a buffer is made inactive and use poll_state_synchronize_rcu() to check whether a full grace period has elapsed before reusing it. If a full grace period has not elapsed, wait for a grace period to elapse, and in the non-RT case use synchronize_rcu_expedited() to hasten it. Since this approach eliminates the RCU callback it is no longer necessary to synchronize_rcu() in the tp_head==NULL case. However, the RCU state should still be saved for the previously active buffer. Before this change I would typically see mini_qdisc_pair_swap() take tens of milliseconds to complete. After this change it typcially finishes in less than 1 ms, and often it takes just a few microseconds. Thanks to Paul for walking me through the options for improving this. Cc: "Paul E. McKenney" <paulmck@kernel.org> Signed-off-by: Seth Forshee <sforshee@digitalocean.com> Link: https://lore.kernel.org/r/20211026130700.121189-1-seth@forshee.me Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-10-26 16:06:59 +03:00
/* This is counterpart of the rcu sync above. We need to
net: core: introduce mini_Qdisc and eliminate usage of tp->q for clsact fastpath In sch_handle_egress and sch_handle_ingress tp->q is used only in order to update stats. So stats and filter list are the only things that are needed in clsact qdisc fastpath processing. Introduce new mini_Qdisc struct to hold those items. Also, introduce a helper to swap the mini_Qdisc structures in case filter list head changes. This removes need for tp->q usage without added overhead. Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-11-03 13:46:25 +03:00
* block potential new user of miniq_old until all readers
* are not seeing it.
*/
net: sch: eliminate unnecessary RCU waits in mini_qdisc_pair_swap() Currently rcu_barrier() is used to ensure that no readers of the inactive mini_Qdisc buffer remain before it is reused. This waits for any pending RCU callbacks to complete, when all that is actually required is to wait for one RCU grace period to elapse after the buffer was made inactive. This means that using rcu_barrier() may result in unnecessary waits. To improve this, store the current RCU state when a buffer is made inactive and use poll_state_synchronize_rcu() to check whether a full grace period has elapsed before reusing it. If a full grace period has not elapsed, wait for a grace period to elapse, and in the non-RT case use synchronize_rcu_expedited() to hasten it. Since this approach eliminates the RCU callback it is no longer necessary to synchronize_rcu() in the tp_head==NULL case. However, the RCU state should still be saved for the previously active buffer. Before this change I would typically see mini_qdisc_pair_swap() take tens of milliseconds to complete. After this change it typcially finishes in less than 1 ms, and often it takes just a few microseconds. Thanks to Paul for walking me through the options for improving this. Cc: "Paul E. McKenney" <paulmck@kernel.org> Signed-off-by: Seth Forshee <sforshee@digitalocean.com> Link: https://lore.kernel.org/r/20211026130700.121189-1-seth@forshee.me Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-10-26 16:06:59 +03:00
miniq_old->rcu_state = start_poll_synchronize_rcu();
net: core: introduce mini_Qdisc and eliminate usage of tp->q for clsact fastpath In sch_handle_egress and sch_handle_ingress tp->q is used only in order to update stats. So stats and filter list are the only things that are needed in clsact qdisc fastpath processing. Introduce new mini_Qdisc struct to hold those items. Also, introduce a helper to swap the mini_Qdisc structures in case filter list head changes. This removes need for tp->q usage without added overhead. Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-11-03 13:46:25 +03:00
}
EXPORT_SYMBOL(mini_qdisc_pair_swap);
net: sched: Pass ingress block to tcf_classify_ingress On ingress and cls_act qdiscs init, save the block on ingress mini_Qdisc and and pass it on to ingress classification, so it can be used for the looking up a specified chain index. Co-developed-by: Vlad Buslov <vladbu@mellanox.com> Signed-off-by: Vlad Buslov <vladbu@mellanox.com> Signed-off-by: Paul Blakey <paulb@mellanox.com> Reviewed-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
2020-02-16 13:01:22 +03:00
void mini_qdisc_pair_block_init(struct mini_Qdisc_pair *miniqp,
struct tcf_block *block)
{
miniqp->miniq1.block = block;
miniqp->miniq2.block = block;
}
EXPORT_SYMBOL(mini_qdisc_pair_block_init);
net: core: introduce mini_Qdisc and eliminate usage of tp->q for clsact fastpath In sch_handle_egress and sch_handle_ingress tp->q is used only in order to update stats. So stats and filter list are the only things that are needed in clsact qdisc fastpath processing. Introduce new mini_Qdisc struct to hold those items. Also, introduce a helper to swap the mini_Qdisc structures in case filter list head changes. This removes need for tp->q usage without added overhead. Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-11-03 13:46:25 +03:00
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;
net: sch: eliminate unnecessary RCU waits in mini_qdisc_pair_swap() Currently rcu_barrier() is used to ensure that no readers of the inactive mini_Qdisc buffer remain before it is reused. This waits for any pending RCU callbacks to complete, when all that is actually required is to wait for one RCU grace period to elapse after the buffer was made inactive. This means that using rcu_barrier() may result in unnecessary waits. To improve this, store the current RCU state when a buffer is made inactive and use poll_state_synchronize_rcu() to check whether a full grace period has elapsed before reusing it. If a full grace period has not elapsed, wait for a grace period to elapse, and in the non-RT case use synchronize_rcu_expedited() to hasten it. Since this approach eliminates the RCU callback it is no longer necessary to synchronize_rcu() in the tp_head==NULL case. However, the RCU state should still be saved for the previously active buffer. Before this change I would typically see mini_qdisc_pair_swap() take tens of milliseconds to complete. After this change it typcially finishes in less than 1 ms, and often it takes just a few microseconds. Thanks to Paul for walking me through the options for improving this. Cc: "Paul E. McKenney" <paulmck@kernel.org> Signed-off-by: Seth Forshee <sforshee@digitalocean.com> Link: https://lore.kernel.org/r/20211026130700.121189-1-seth@forshee.me Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-10-26 16:06:59 +03:00
miniqp->miniq1.rcu_state = get_state_synchronize_rcu();
miniqp->miniq2.rcu_state = miniqp->miniq1.rcu_state;
net: core: introduce mini_Qdisc and eliminate usage of tp->q for clsact fastpath In sch_handle_egress and sch_handle_ingress tp->q is used only in order to update stats. So stats and filter list are the only things that are needed in clsact qdisc fastpath processing. Introduce new mini_Qdisc struct to hold those items. Also, introduce a helper to swap the mini_Qdisc structures in case filter list head changes. This removes need for tp->q usage without added overhead. Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-11-03 13:46:25 +03:00
miniqp->p_miniq = p_miniq;
}
EXPORT_SYMBOL(mini_qdisc_pair_init);
Reference in New Issue Copy Permalink