Commit Graph

165 Commits

Author SHA1 Message Date
Alexey Kodanev 35d889d10b sch_netem: fix skb leak in netem_enqueue()
When we exceed current packets limit and we have more than one
segment in the list returned by skb_gso_segment(), netem drops
only the first one, skipping the rest, hence kmemleak reports:

unreferenced object 0xffff880b5d23b600 (size 1024):
  comm "softirq", pid 0, jiffies 4384527763 (age 2770.629s)
  hex dump (first 32 bytes):
    00 80 23 5d 0b 88 ff ff 00 00 00 00 00 00 00 00  ..#]............
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
  backtrace:
    [<00000000d8a19b9d>] __alloc_skb+0xc9/0x520
    [<000000001709b32f>] skb_segment+0x8c8/0x3710
    [<00000000c7b9bb88>] tcp_gso_segment+0x331/0x1830
    [<00000000c921cba1>] inet_gso_segment+0x476/0x1370
    [<000000008b762dd4>] skb_mac_gso_segment+0x1f9/0x510
    [<000000002182660a>] __skb_gso_segment+0x1dd/0x620
    [<00000000412651b9>] netem_enqueue+0x1536/0x2590 [sch_netem]
    [<0000000005d3b2a9>] __dev_queue_xmit+0x1167/0x2120
    [<00000000fc5f7327>] ip_finish_output2+0x998/0xf00
    [<00000000d309e9d3>] ip_output+0x1aa/0x2c0
    [<000000007ecbd3a4>] tcp_transmit_skb+0x18db/0x3670
    [<0000000042d2a45f>] tcp_write_xmit+0x4d4/0x58c0
    [<0000000056a44199>] tcp_tasklet_func+0x3d9/0x540
    [<0000000013d06d02>] tasklet_action+0x1ca/0x250
    [<00000000fcde0b8b>] __do_softirq+0x1b4/0x5a3
    [<00000000e7ed027c>] irq_exit+0x1e2/0x210

Fix it by adding the rest of the segments, if any, to skb 'to_free'
list. Add new __qdisc_drop_all() and qdisc_drop_all() functions
because they can be useful in the future if we need to drop segmented
GSO packets in other places.

Fixes: 6071bd1aa1 ("netem: Segment GSO packets on enqueue")
Signed-off-by: Alexey Kodanev <alexey.kodanev@oracle.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-03-07 11:18:14 -05:00
Md. Islam 043e337f55 sch_netem: Bug fixing in calculating Netem interval
In Kernel 4.15.0+, Netem does not work properly.

Netem setup:

tc qdisc add dev h1-eth0 root handle 1: netem delay 10ms 2ms

Result:

PING 172.16.101.2 (172.16.101.2) 56(84) bytes of data.
64 bytes from 172.16.101.2: icmp_seq=1 ttl=64 time=22.8 ms
64 bytes from 172.16.101.2: icmp_seq=2 ttl=64 time=10.9 ms
64 bytes from 172.16.101.2: icmp_seq=3 ttl=64 time=10.9 ms
64 bytes from 172.16.101.2: icmp_seq=5 ttl=64 time=11.4 ms
64 bytes from 172.16.101.2: icmp_seq=6 ttl=64 time=11.8 ms
64 bytes from 172.16.101.2: icmp_seq=4 ttl=64 time=4303 ms
64 bytes from 172.16.101.2: icmp_seq=10 ttl=64 time=11.2 ms
64 bytes from 172.16.101.2: icmp_seq=11 ttl=64 time=10.3 ms
64 bytes from 172.16.101.2: icmp_seq=7 ttl=64 time=4304 ms
64 bytes from 172.16.101.2: icmp_seq=8 ttl=64 time=4303 ms

Patch:

(rnd % (2 * sigma)) - sigma was overflowing s32. After applying the
patch, I found following output which is desirable.

PING 172.16.101.2 (172.16.101.2) 56(84) bytes of data.
64 bytes from 172.16.101.2: icmp_seq=1 ttl=64 time=21.1 ms
64 bytes from 172.16.101.2: icmp_seq=2 ttl=64 time=8.46 ms
64 bytes from 172.16.101.2: icmp_seq=3 ttl=64 time=9.00 ms
64 bytes from 172.16.101.2: icmp_seq=4 ttl=64 time=11.8 ms
64 bytes from 172.16.101.2: icmp_seq=5 ttl=64 time=8.36 ms
64 bytes from 172.16.101.2: icmp_seq=6 ttl=64 time=11.8 ms
64 bytes from 172.16.101.2: icmp_seq=7 ttl=64 time=8.11 ms
64 bytes from 172.16.101.2: icmp_seq=8 ttl=64 time=10.0 ms
64 bytes from 172.16.101.2: icmp_seq=9 ttl=64 time=11.3 ms
64 bytes from 172.16.101.2: icmp_seq=10 ttl=64 time=11.5 ms
64 bytes from 172.16.101.2: icmp_seq=11 ttl=64 time=10.2 ms

Reviewed-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-07 21:59:12 -05:00
Alexander Aring 653d6fd68d net: sched: sch: add extack for graft callback
This patch adds extack support for graft 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>
2017-12-21 12:32:50 -05:00
Alexander Aring 2030721cc0 net: sched: sch: add extack for change qdisc ops
This patch adds extack support for change callback for qdisc ops
structtur 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>
2017-12-21 12:32:50 -05:00
Alexander Aring e63d7dfd2d 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>
2017-12-21 12:32:50 -05:00
Stephen Hemminger 9b0ed89172 netem: remove unnecessary 64 bit modulus
Fix compilation on 32 bit platforms (where doing modulus operation
with 64 bit requires extra glibc functions) by truncation.
The jitter for table distribution is limited to a 32 bit value
because random numbers are scaled as 32 bit value.

Also fix some whitespace.

Fixes: 99803171ef ("netem: add uapi to express delay and jitter in nanoseconds")
Reported-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-11-15 14:14:16 +09:00
Stephen Hemminger bce552fd6f netem: use 64 bit divide by rate
Since times are now expressed in nanosecond, need to now do
true 64 bit divide. Old code would truncate rate at 32 bits.
Rename function to better express current usage.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-11-15 14:14:16 +09:00
Dave Taht 836af83b54 netem: support delivering packets in delayed time slots
Slotting is a crude approximation of the behaviors of shared media such
as cable, wifi, and LTE, which gather up a bunch of packets within a
varying delay window and deliver them, relative to that, nearly all at
once.

It works within the existing loss, duplication, jitter and delay
parameters of netem. Some amount of inherent latency must be specified,
regardless.

The new "slot" parameter specifies a minimum and maximum delay between
transmission attempts.

The "bytes" and "packets" parameters can be used to limit the amount of
information transferred per slot.

Examples of use:

tc qdisc add dev eth0 root netem delay 200us \
         slot 800us 10ms bytes 64k packets 42

A more correct example, using stacked netem instances and a packet limit
to emulate a tail drop wifi queue with slots and variable packet
delivery, with a 200Mbit isochronous underlying rate, and 20ms path
delay:

tc qdisc add dev eth0 root handle 1: netem delay 20ms rate 200mbit \
         limit 10000
tc qdisc add dev eth0 parent 1:1 handle 10:1 netem delay 200us \
         slot 800us 10ms bytes 64k packets 42 limit 512

Signed-off-by: Dave Taht <dave.taht@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-11-13 10:15:47 +09:00
Dave Taht 99803171ef netem: add uapi to express delay and jitter in nanoseconds
netem userspace has long relied on a horrible /proc/net/psched hack
to translate the current notion of "ticks" to nanoseconds.

Expressing latency and jitter instead, in well defined nanoseconds,
increases the dynamic range of emulated delays and jitter in netem.

It will also ease a transition where reducing a tick to nsec
equivalence would constrain the max delay in prior versions of
netem to only 4.3 seconds.

Signed-off-by: Dave Taht <dave.taht@gmail.com>
Suggested-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-11-13 10:15:47 +09:00
Dave Taht 112f9cb656 netem: convert to qdisc_watchdog_schedule_ns
Upgrade the internal netem scheduler to use nanoseconds rather than
ticks throughout.

Convert to and from the std "ticks" userspace api automatically,
while allowing for finer grained scheduling to take place.

Signed-off-by: Dave Taht <dave.taht@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-11-13 10:15:47 +09:00
Eric Dumazet 18a4c0eab2 net: add rb_to_skb() and other rb tree helpers
Geeralize private netem_rb_to_skb()

TCP rtx queue will soon be converted to rb-tree,
so we will need skb_rbtree_walk() helpers.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-10-07 00:28:53 +01:00
Eric Dumazet 3aa605f28b sch_netem: faster rb tree removal
While running TCP tests involving netem storing millions of packets,
I had the idea to speed up tfifo_reset() and did experiments.

I tried the rbtree_postorder_for_each_entry_safe() method that is
used in skb_rbtree_purge() but discovered it was slower than the
current tfifo_reset() method.

I measured time taken to release skbs with three occupation levels :
10^4, 10^5 and 10^6 skbs with three methods :

1) (current 'naive' method)

	while ((p = rb_first(&q->t_root))) {
		struct sk_buff *skb = netem_rb_to_skb(p);

		rb_erase(p, &q->t_root);
		rtnl_kfree_skbs(skb, skb);
	}

2) Use rb_next() instead of rb_first() in the loop :

	p = rb_first(&q->t_root);
	while (p) {
		struct sk_buff *skb = netem_rb_to_skb(p);

		p = rb_next(p);
		rb_erase(&skb->rbnode, &q->t_root);
		rtnl_kfree_skbs(skb, skb);
	}

3) "optimized" method using rbtree_postorder_for_each_entry_safe()

	struct sk_buff *skb, *next;

	rbtree_postorder_for_each_entry_safe(skb, next,
					     &q->t_root, rbnode) {
               rtnl_kfree_skbs(skb, skb);
	}
	q->t_root = RB_ROOT;

Results :

method_1:while (rb_first()) rb_erase() 10000 skbs in 690378 ns (69 ns per skb)
method_2:rb_first; while (p) { p = rb_next(p); ...}  10000 skbs in 541846 ns (54 ns per skb)
method_3:rbtree_postorder_for_each_entry_safe() 10000 skbs in 868307 ns (86 ns per skb)

method_1:while (rb_first()) rb_erase() 99996 skbs in 7804021 ns (78 ns per skb)
method_2:rb_first; while (p) { p = rb_next(p); ...}  100000 skbs in 5942456 ns (59 ns per skb)
method_3:rbtree_postorder_for_each_entry_safe() 100000 skbs in 11584940 ns (115 ns per skb)

method_1:while (rb_first()) rb_erase() 1000000 skbs in 108577838 ns (108 ns per skb)
method_2:rb_first; while (p) { p = rb_next(p); ...}  1000000 skbs in 82619635 ns (82 ns per skb)
method_3:rbtree_postorder_for_each_entry_safe() 1000000 skbs in 127328743 ns (127 ns per skb)

Method 2) is simply faster, probably because it maintains a smaller
working size set.

Note that this is the method we use in tcp_ofo_queue() already.

I will also change skb_rbtree_purge() in a second patch.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-09-25 20:31:32 -07:00
Eric Dumazet bffa72cf7f net: sk_buff rbnode reorg
skb->rbnode shares space with skb->next, skb->prev and skb->tstamp

Current uses (TCP receive ofo queue and netem) need to save/restore
tstamp, while skb->dev is either NULL (TCP) or a constant for a given
queue (netem).

Since we plan using an RB tree for TCP retransmit queue to speedup SACK
processing with large BDP, this patch exchanges skb->dev and
skb->tstamp.

This saves some overhead in both TCP and netem.

v2: removes the swtstamp field from struct tcp_skb_cb

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Soheil Hassas Yeganeh <soheil@google.com>
Cc: Wei Wang <weiwan@google.com>
Cc: Willem de Bruijn <willemb@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-09-19 15:20:22 -07:00
David S. Miller 6026e043d0 Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
Three cases of simple overlapping changes.

Signed-off-by: David S. Miller <davem@davemloft.net>
2017-09-01 17:42:05 -07:00
Nikolay Aleksandrov 634576a184 sch_netem: avoid null pointer deref on init failure
netem can fail in ->init due to missing options (either not supplied by
user-space or used as a default qdisc) causing a timer->base null
pointer deref in its ->destroy() and ->reset() callbacks.

Reproduce:
$ sysctl net.core.default_qdisc=netem
$ ip l set ethX up

Crash log:
[ 1814.846943] BUG: unable to handle kernel NULL pointer dereference at (null)
[ 1814.847181] IP: hrtimer_active+0x17/0x8a
[ 1814.847270] PGD 59c34067
[ 1814.847271] P4D 59c34067
[ 1814.847337] PUD 37374067
[ 1814.847403] PMD 0
[ 1814.847468]
[ 1814.847582] Oops: 0000 [#1] SMP
[ 1814.847655] Modules linked in: sch_netem(O) sch_fq_codel(O)
[ 1814.847761] CPU: 3 PID: 1573 Comm: ip Tainted: G           O 4.13.0-rc6+ #62
[ 1814.847884] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.7.5-20140531_083030-gandalf 04/01/2014
[ 1814.848043] task: ffff88003723a700 task.stack: ffff88005adc8000
[ 1814.848235] RIP: 0010:hrtimer_active+0x17/0x8a
[ 1814.848407] RSP: 0018:ffff88005adcb590 EFLAGS: 00010246
[ 1814.848590] RAX: 0000000000000000 RBX: ffff880058e359d8 RCX: 0000000000000000
[ 1814.848793] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff880058e359d8
[ 1814.848998] RBP: ffff88005adcb5b0 R08: 00000000014080c0 R09: 00000000ffffffff
[ 1814.849204] R10: ffff88005adcb660 R11: 0000000000000020 R12: 0000000000000000
[ 1814.849410] R13: ffff880058e359d8 R14: 00000000ffffffff R15: 0000000000000001
[ 1814.849616] FS:  00007f733bbca740(0000) GS:ffff88005d980000(0000) knlGS:0000000000000000
[ 1814.849919] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 1814.850107] CR2: 0000000000000000 CR3: 0000000059f0d000 CR4: 00000000000406e0
[ 1814.850313] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 1814.850518] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 1814.850723] Call Trace:
[ 1814.850875]  hrtimer_try_to_cancel+0x1a/0x93
[ 1814.851047]  hrtimer_cancel+0x15/0x20
[ 1814.851211]  qdisc_watchdog_cancel+0x12/0x14
[ 1814.851383]  netem_reset+0xe6/0xed [sch_netem]
[ 1814.851561]  qdisc_destroy+0x8b/0xe5
[ 1814.851723]  qdisc_create_dflt+0x86/0x94
[ 1814.851890]  ? dev_activate+0x129/0x129
[ 1814.852057]  attach_one_default_qdisc+0x36/0x63
[ 1814.852232]  netdev_for_each_tx_queue+0x3d/0x48
[ 1814.852406]  dev_activate+0x4b/0x129
[ 1814.852569]  __dev_open+0xe7/0x104
[ 1814.852730]  __dev_change_flags+0xc6/0x15c
[ 1814.852899]  dev_change_flags+0x25/0x59
[ 1814.853064]  do_setlink+0x30c/0xb3f
[ 1814.853228]  ? check_chain_key+0xb0/0xfd
[ 1814.853396]  ? check_chain_key+0xb0/0xfd
[ 1814.853565]  rtnl_newlink+0x3a4/0x729
[ 1814.853728]  ? rtnl_newlink+0x117/0x729
[ 1814.853905]  ? ns_capable_common+0xd/0xb1
[ 1814.854072]  ? ns_capable+0x13/0x15
[ 1814.854234]  rtnetlink_rcv_msg+0x188/0x197
[ 1814.854404]  ? rcu_read_unlock+0x3e/0x5f
[ 1814.854572]  ? rtnl_newlink+0x729/0x729
[ 1814.854737]  netlink_rcv_skb+0x6c/0xce
[ 1814.854902]  rtnetlink_rcv+0x23/0x2a
[ 1814.855064]  netlink_unicast+0x103/0x181
[ 1814.855230]  netlink_sendmsg+0x326/0x337
[ 1814.855398]  sock_sendmsg_nosec+0x14/0x3f
[ 1814.855584]  sock_sendmsg+0x29/0x2e
[ 1814.855747]  ___sys_sendmsg+0x209/0x28b
[ 1814.855912]  ? do_raw_spin_unlock+0xcd/0xf8
[ 1814.856082]  ? _raw_spin_unlock+0x27/0x31
[ 1814.856251]  ? __handle_mm_fault+0x651/0xdb1
[ 1814.856421]  ? check_chain_key+0xb0/0xfd
[ 1814.856592]  __sys_sendmsg+0x45/0x63
[ 1814.856755]  ? __sys_sendmsg+0x45/0x63
[ 1814.856923]  SyS_sendmsg+0x19/0x1b
[ 1814.857083]  entry_SYSCALL_64_fastpath+0x23/0xc2
[ 1814.857256] RIP: 0033:0x7f733b2dd690
[ 1814.857419] RSP: 002b:00007ffe1d3387d8 EFLAGS: 00000246 ORIG_RAX: 000000000000002e
[ 1814.858238] RAX: ffffffffffffffda RBX: ffffffff810d278c RCX: 00007f733b2dd690
[ 1814.858445] RDX: 0000000000000000 RSI: 00007ffe1d338820 RDI: 0000000000000003
[ 1814.858651] RBP: ffff88005adcbf98 R08: 0000000000000001 R09: 0000000000000003
[ 1814.858856] R10: 00007ffe1d3385a0 R11: 0000000000000246 R12: 0000000000000002
[ 1814.859060] R13: 000000000066f1a0 R14: 00007ffe1d3408d0 R15: 0000000000000000
[ 1814.859267]  ? trace_hardirqs_off_caller+0xa7/0xcf
[ 1814.859446] Code: 10 55 48 89 c7 48 89 e5 e8 45 a1 fb ff 31 c0 5d c3
31 c0 c3 66 66 66 66 90 55 48 89 e5 41 56 41 55 41 54 53 49 89 fd 49 8b
45 30 <4c> 8b 20 41 8b 5c 24 38 31 c9 31 d2 48 c7 c7 50 8e 1d 82 41 89
[ 1814.860022] RIP: hrtimer_active+0x17/0x8a RSP: ffff88005adcb590
[ 1814.860214] CR2: 0000000000000000

Fixes: 87b60cfacf ("net_sched: fix error recovery at qdisc creation")
Fixes: 0fbbeb1ba4 ("[PKT_SCHED]: Fix missing qdisc_destroy() in qdisc_create_dflt()")
Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-08-30 15:26:11 -07:00
WANG Cong 143976ce99 net_sched: remove tc class reference counting
For TC classes, their ->get() and ->put() are always paired, and the
reference counting is completely useless, because:

1) For class modification and dumping paths, we already hold RTNL lock,
   so all of these ->get(),->change(),->put() are atomic.

2) For filter bindiing/unbinding, we use other reference counter than
   this one, and they should have RTNL lock too.

3) For ->qlen_notify(), it is special because it is called on ->enqueue()
   path, but we already hold qdisc tree lock there, and we hold this
   tree lock when graft or delete the class too, so it should not be gone
   or changed until we release the tree lock.

Therefore, this patch removes ->get() and ->put(), but:

1) Adds a new ->find() to find the pointer to a class by classid, no
   refcnt.

2) Move the original class destroy upon the last refcnt into ->delete(),
   right after releasing tree lock. This is fine because the class is
   already removed from hash when holding the lock.

For those who also use ->put() as ->unbind(), just rename them to reflect
this change.

Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-08-25 17:19:10 -07:00
Michal Hocko 752ade68cb treewide: use kv[mz]alloc* rather than opencoded variants
There are many code paths opencoding kvmalloc.  Let's use the helper
instead.  The main difference to kvmalloc is that those users are
usually not considering all the aspects of the memory allocator.  E.g.
allocation requests <= 32kB (with 4kB pages) are basically never failing
and invoke OOM killer to satisfy the allocation.  This sounds too
disruptive for something that has a reasonable fallback - the vmalloc.
On the other hand those requests might fallback to vmalloc even when the
memory allocator would succeed after several more reclaim/compaction
attempts previously.  There is no guarantee something like that happens
though.

This patch converts many of those places to kv[mz]alloc* helpers because
they are more conservative.

Link: http://lkml.kernel.org/r/20170306103327.2766-2-mhocko@kernel.org
Signed-off-by: Michal Hocko <mhocko@suse.com>
Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> # Xen bits
Acked-by: Kees Cook <keescook@chromium.org>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Andreas Dilger <andreas.dilger@intel.com> # Lustre
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com> # KVM/s390
Acked-by: Dan Williams <dan.j.williams@intel.com> # nvdim
Acked-by: David Sterba <dsterba@suse.com> # btrfs
Acked-by: Ilya Dryomov <idryomov@gmail.com> # Ceph
Acked-by: Tariq Toukan <tariqt@mellanox.com> # mlx4
Acked-by: Leon Romanovsky <leonro@mellanox.com> # mlx5
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: Anton Vorontsov <anton@enomsg.org>
Cc: Colin Cross <ccross@android.com>
Cc: Tony Luck <tony.luck@intel.com>
Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
Cc: Ben Skeggs <bskeggs@redhat.com>
Cc: Kent Overstreet <kent.overstreet@gmail.com>
Cc: Santosh Raspatur <santosh@chelsio.com>
Cc: Hariprasad S <hariprasad@chelsio.com>
Cc: Yishai Hadas <yishaih@mellanox.com>
Cc: Oleg Drokin <oleg.drokin@intel.com>
Cc: "Yan, Zheng" <zyan@redhat.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: David Miller <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-05-08 17:15:13 -07:00
Johannes Berg fceb6435e8 netlink: pass extended ACK struct to parsing functions
Pass the new extended ACK reporting struct to all of the generic
netlink parsing functions. For now, pass NULL in almost all callers
(except for some in the core.)

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-13 13:58:22 -04:00
Nik Unger 5080f39e8c netem: apply correct delay when rate throttling
I recently reported on the netem list that iperf network benchmarks
show unexpected results when a bandwidth throttling rate has been
configured for netem. Specifically:

1) The measured link bandwidth *increases* when a higher delay is added
2) The measured link bandwidth appears higher than the specified limit
3) The measured link bandwidth for the same very slow settings varies significantly across
  machines

The issue can be reproduced by using tc to configure netem with a
512kbit rate and various (none, 1us, 50ms, 100ms, 200ms) delays on a
veth pair between network namespaces, and then using iperf (or any
other network benchmarking tool) to test throughput. Complete detailed
instructions are in the original email chain here:
https://lists.linuxfoundation.org/pipermail/netem/2017-February/001672.html

There appear to be two underlying bugs causing these effects:

- The first issue causes long delays when the rate is slow and no
  delay is configured (e.g., "rate 512kbit"). This is because SKBs are
  not orphaned when no delay is configured, so orphaning does not
  occur until *after* the rate-induced delay has been applied. For
  this reason, adding a tiny delay (e.g., "rate 512kbit delay 1us")
  dramatically increases the measured bandwidth.

- The second issue is that rate-induced delays are not correctly
  applied, allowing SKB delays to occur in parallel. The indended
  approach is to compute the delay for an SKB and to add this delay to
  the end of the current queue. However, the code does not detect
  existing SKBs in the queue due to improperly testing sch->q.qlen,
  which is nonzero even when packets exist only in the
  rbtree. Consequently, new SKBs do not wait for the current queue to
  empty. When packet delays vary significantly (e.g., if packet sizes
  are different), then this also causes unintended reordering.

I modified the code to expect a delay (and orphan the SKB) when a rate
is configured. I also added some defensive tests that correctly find
the latest scheduled delivery time, even if it is (unexpectedly) for a
packet in sch->q. I have tested these changes on the latest kernel
(4.11.0-rc1+) and the iperf / ping test results are as expected.

Signed-off-by: Nik Unger <njunger@uwaterloo.ca>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-03-16 20:14:06 -07:00
Willem de Bruijn bc31c905e9 net-tc: convert tc_from to tc_from_ingress and tc_redirected
The tc_from field fulfills two roles. It encodes whether a packet was
redirected by an act_mirred device and, if so, whether act_mirred was
called on ingress or egress. Split it into separate fields.

The information is needed by the special IFB loop, where packets are
taken out of the normal path by act_mirred, forwarded to IFB, then
reinjected at their original location (ingress or egress) by IFB.

The IFB device cannot use skb->tc_at_ingress, because that may have
been overwritten as the packet travels from act_mirred to ifb_xmit,
when it passes through tc_classify on the IFB egress path. Cache this
value in skb->tc_from_ingress.

That field is valid only if a packet arriving at ifb_xmit came from
act_mirred. Other packets can be crafted to reach ifb_xmit. These
must be dropped. Set tc_redirected on redirection and drop all packets
that do not have this bit set.

Both fields are set only on cloned skbs in tc actions, so original
packet sources do not have to clear the bit when reusing packets
(notably, pktgen and octeon).

Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-01-08 20:58:52 -05:00
Willem de Bruijn a5135bcfba net-tc: convert tc_verd to integer bitfields
Extract the remaining two fields from tc_verd and remove the __u16
completely. TC_AT and TC_FROM are converted to equivalent two-bit
integer fields tc_at and tc_from. Where possible, use existing
helper skb_at_tc_ingress when reading tc_at. Introduce helper
skb_reset_tc to clear fields.

Not documenting tc_from and tc_at, because they will be replaced
with single bit fields in follow-on patches.

Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-01-08 20:58:52 -05:00
Thomas Gleixner 2456e85535 ktime: Get rid of the union
ktime is a union because the initial implementation stored the time in
scalar nanoseconds on 64 bit machine and in a endianess optimized timespec
variant for 32bit machines. The Y2038 cleanup removed the timespec variant
and switched everything to scalar nanoseconds. The union remained, but
become completely pointless.

Get rid of the union and just keep ktime_t as simple typedef of type s64.

The conversion was done with coccinelle and some manual mopping up.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Peter Zijlstra <peterz@infradead.org>
2016-12-25 17:21:22 +01:00
Geliang Tang 7f7cd56c33 net_sched: sch_netem: use rb_entry()
To make the code clearer, use rb_entry() instead of container_of() to
deal with rbtree.

Signed-off-by: Geliang Tang <geliangtang@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2016-12-20 14:22:48 -05:00
Florian Westphal 48da34b7a7 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>
2016-09-19 01:47:18 -04:00
Florian Westphal ed760cb8aa sched: replace __skb_dequeue with __qdisc_dequeue_head
After previous patch these functions are identical.
Replace __skb_dequeue in qdiscs with __qdisc_dequeue_head.

Next patch will then make __qdisc_dequeue_head handle
single-linked list instead of strcut sk_buff_head argument.

Doesn't change generated code.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
2016-09-19 01:47:18 -04:00
Florian Westphal 97d0678f91 sched: don't use skb queue helpers
A followup change will replace the sk_buff_head in the qdisc
struct with a slightly different list.

Use of the sk_buff_head helpers will thus cause compiler
warnings.

Open-code these accesses in an extra change to ease review.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
2016-09-19 01:47:18 -04:00
David S. Miller ee58b57100 Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
Several cases of overlapping changes, except the packet scheduler
conflicts which deal with the addition of the free list parameter
to qdisc_enqueue().

Signed-off-by: David S. Miller <davem@davemloft.net>
2016-06-30 05:03:36 -04:00
Eric Dumazet 8a6e9c6703 net_sched: netem: do not call qdisc_drop() with a NULL skb
If skb_unshare() fails, we call qdisc_drop() with a NULL skb, which
is no longer supported.

Fixes: 520ac30f45 ("net_sched: drop packets after root qdisc lock is released")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2016-06-29 08:02:24 -04:00
Eric Dumazet 520ac30f45 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>
2016-06-25 12:19:35 -04:00
Eric Dumazet 21de12ee55 netem: fix a use after free
If the packet was dropped by lower qdisc, then we must not
access it later.

Save qdisc_pkt_len(skb) in a temp variable.

Fixes: 2ccccf5fb4 ("net_sched: update hierarchical backlog too")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: WANG Cong <xiyou.wangcong@gmail.com>
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2016-06-23 15:07:44 -04:00
Eric Dumazet 2f08a9a162 net_sched: sch_netem: defer skb freeing
rtnl_kfree_skbs() can be used in tfifo_reset()

It would be nice if we could iterate through rb tree instead
of removing one skb at a time, and build a single skb chain.
But this is left for a future patch.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2016-06-15 14:08:35 -07:00
Eric Dumazet 45f50bed1d net_sched: remove generic throttled management
__QDISC_STATE_THROTTLED bit manipulation is rather expensive
for HTB and few others.

I already removed it for sch_fq in commit f2600cf02b
("net: sched: avoid costly atomic operation in fq_dequeue()")
and so far nobody complained.

When one ore more packets are stuck in one or more throttled
HTB class, a htb dequeue() performs two atomic operations
to clear/set __QDISC_STATE_THROTTLED bit, while root qdisc
lock is held.

Removing this pair of atomic operations bring me a 8 % performance
increase on 200 TCP_RR tests, in presence of throttled classes.

This patch has no side effect, since nothing actually uses
disc_is_throttled() anymore.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2016-06-10 23:58:21 -07:00
Eric Dumazet 42117927ca net_sched: netem: remove qdisc_is_throttled() use
Looks like it is only there as some optimization attempt.

Since __QDISC_STATE_THROTTLED set/unset is way too expensive,
and netem is the last user, just remove this check.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2016-06-10 23:58:21 -07:00
Florian Westphal a09ceb0e08 sched: remove qdisc->drop
after removal of TCA_CBQ_OVL_STRATEGY from cbq scheduler, there are no
more callers of ->drop() outside of other ->drop functions, i.e.
nothing calls them.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
2016-06-08 23:58:52 -07:00
Florian Westphal c3a173d7db sched: remove qdisc_rehape_fail
After the removal of TCA_CBQ_POLICE in cbq scheduler qdisc->reshape_fail
is always NULL, i.e. qdisc_rehape_fail is now the same as qdisc_drop.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
2016-06-08 23:58:51 -07:00
David S. Miller cba6532100 Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
Conflicts:
	net/ipv4/ip_gre.c

Minor conflicts between tunnel bug fixes in net and
ipv6 tunnel cleanups in net-next.

Signed-off-by: David S. Miller <davem@davemloft.net>
2016-05-04 00:52:29 -04:00
Neil Horman 6071bd1aa1 netem: Segment GSO packets on enqueue
This was recently reported to me, and reproduced on the latest net kernel,
when attempting to run netperf from a host that had a netem qdisc attached
to the egress interface:

[  788.073771] ---------------------[ cut here ]---------------------------
[  788.096716] WARNING: at net/core/dev.c:2253 skb_warn_bad_offload+0xcd/0xda()
[  788.129521] bnx2: caps=(0x00000001801949b3, 0x0000000000000000) len=2962
data_len=0 gso_size=1448 gso_type=1 ip_summed=3
[  788.182150] Modules linked in: sch_netem kvm_amd kvm crc32_pclmul ipmi_ssif
ghash_clmulni_intel sp5100_tco amd64_edac_mod aesni_intel lrw gf128mul
glue_helper ablk_helper edac_mce_amd cryptd pcspkr sg edac_core hpilo ipmi_si
i2c_piix4 k10temp fam15h_power hpwdt ipmi_msghandler shpchp acpi_power_meter
pcc_cpufreq nfsd auth_rpcgss nfs_acl lockd grace sunrpc ip_tables xfs libcrc32c
sd_mod crc_t10dif crct10dif_generic mgag200 syscopyarea sysfillrect sysimgblt
i2c_algo_bit drm_kms_helper ahci ata_generic pata_acpi ttm libahci
crct10dif_pclmul pata_atiixp tg3 libata crct10dif_common drm crc32c_intel ptp
serio_raw bnx2 r8169 hpsa pps_core i2c_core mii dm_mirror dm_region_hash dm_log
dm_mod
[  788.465294] CPU: 16 PID: 0 Comm: swapper/16 Tainted: G        W
------------   3.10.0-327.el7.x86_64 #1
[  788.511521] Hardware name: HP ProLiant DL385p Gen8, BIOS A28 12/17/2012
[  788.542260]  ffff880437c036b8 f7afc56532a53db9 ffff880437c03670
ffffffff816351f1
[  788.576332]  ffff880437c036a8 ffffffff8107b200 ffff880633e74200
ffff880231674000
[  788.611943]  0000000000000001 0000000000000003 0000000000000000
ffff880437c03710
[  788.647241] Call Trace:
[  788.658817]  <IRQ>  [<ffffffff816351f1>] dump_stack+0x19/0x1b
[  788.686193]  [<ffffffff8107b200>] warn_slowpath_common+0x70/0xb0
[  788.713803]  [<ffffffff8107b29c>] warn_slowpath_fmt+0x5c/0x80
[  788.741314]  [<ffffffff812f92f3>] ? ___ratelimit+0x93/0x100
[  788.767018]  [<ffffffff81637f49>] skb_warn_bad_offload+0xcd/0xda
[  788.796117]  [<ffffffff8152950c>] skb_checksum_help+0x17c/0x190
[  788.823392]  [<ffffffffa01463a1>] netem_enqueue+0x741/0x7c0 [sch_netem]
[  788.854487]  [<ffffffff8152cb58>] dev_queue_xmit+0x2a8/0x570
[  788.880870]  [<ffffffff8156ae1d>] ip_finish_output+0x53d/0x7d0
...

The problem occurs because netem is not prepared to handle GSO packets (as it
uses skb_checksum_help in its enqueue path, which cannot manipulate these
frames).

The solution I think is to simply segment the skb in a simmilar fashion to the
way we do in __dev_queue_xmit (via validate_xmit_skb), with some minor changes.
When we decide to corrupt an skb, if the frame is GSO, we segment it, corrupt
the first segment, and enqueue the remaining ones.

tested successfully by myself on the latest net kernel, to which this applies

Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
CC: Jamal Hadi Salim <jhs@mojatatu.com>
CC: "David S. Miller" <davem@davemloft.net>
CC: netem@lists.linux-foundation.org
CC: eric.dumazet@gmail.com
CC: stephen@networkplumber.org
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2016-05-03 00:33:14 -04:00
Nicolas Dichtel 2a51c1e8ec sched: use nla_put_u64_64bit()
Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 15:09:09 -04:00
WANG Cong 2ccccf5fb4 net_sched: update hierarchical backlog too
When the bottom qdisc decides to, for example, drop some packet,
it calls qdisc_tree_decrease_qlen() to update the queue length
for all its ancestors, we need to update the backlog too to
keep the stats on root qdisc accurate.

Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2016-02-29 17:02:33 -05:00
WANG Cong 86a7996cc8 net_sched: introduce qdisc_replace() helper
Remove nearly duplicated code and prepare for the following patch.

Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2016-02-29 17:02:33 -05:00
Eric Dumazet b396cca6fa net: sched: deprecate enqueue_root()
Only left enqueue_root() user is netem, and it looks not necessary :

qdisc_skb_cb(skb)->pkt_len is preserved after one skb_clone()

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-11 14:17:32 -04:00
Beshay, Joseph 0ad2a83659 netem: Fixes byte backlog accounting for the first of two chained netem instances
Fixes byte backlog accounting for the first of two chained netem instances.
Bytes backlog reported now corresponds to the number of queued packets.

When two netem instances are chained, for instance to apply rate and queue
limitation followed by packet delay, the number of backlogged bytes reported
by the first netem instance is wrong. It reports the sum of bytes in the queues
of the first and second netem. The first netem reports the correct number of
backlogged packets but not bytes. This is shown in the example below.

Consider a chain of two netem schedulers created using the following commands:

$ tc -s qdisc replace dev veth2 root handle 1:0 netem rate 10000kbit limit 100
$ tc -s qdisc add dev veth2 parent 1:0 handle 2: netem delay 50ms

Start an iperf session to send packets out on the specified interface and
monitor the backlog using tc:

$ tc -s qdisc show dev veth2

Output using unpatched netem:
	qdisc netem 1: root refcnt 2 limit 100 rate 10000Kbit
	 Sent 98422639 bytes 65434 pkt (dropped 123, overlimits 0 requeues 0)
	 backlog 172694b 73p requeues 0
	qdisc netem 2: parent 1: limit 1000 delay 50.0ms
	 Sent 98422639 bytes 65434 pkt (dropped 0, overlimits 0 requeues 0)
	 backlog 63588b 42p requeues 0

The interface used to produce this output has an MTU of 1500. The output for
backlogged bytes behind netem 1 is 172694b. This value is not correct. Consider
the total number of sent bytes and packets. By dividing the number of sent
bytes by the number of sent packets, we get an average packet size of ~=1504.
If we divide the number of backlogged bytes by packets, we get ~=2365. This is
due to the first netem incorrectly counting the 63588b which are in netem 2's
queue as being in its own queue. To verify this is the case, we subtract them
from the reported value and divide by the number of packets as follows:
	172694 - 63588 = 109106 bytes actualled backlogged in netem 1
	109106 / 73 packets ~= 1494 bytes (which matches our MTU)

The root cause is that the byte accounting is not done at the
same time with packet accounting. The solution is to update the backlog value
every time the packet queue is updated.

Signed-off-by: Joseph D Beshay <joseph.beshay@utdallas.edu>
Acked-by: Hagen Paul Pfeifer <hagen@jauu.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-07 18:34:24 -04:00
Eric Dumazet 56b174256b net: add rbnode to struct sk_buff
Yaogong replaces TCP out of order receive queue by an RB tree.

As netem already does a private skb->{next/prev/tstamp} union
with a 'struct rb_node', lets do this in a cleaner way.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Yaogong Wang <wygivan@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-03 16:13:03 -05:00
John Fastabend 25331d6ce4 net: sched: implement qstat helper routines
This adds helpers to manipulate qstats logic and replaces locations
that touch the counters directly. This simplifies future patches
to push qstats onto per cpu counters.

Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 01:02:26 -04:00
WANG Cong 4cb28970a2 net: use the new API kvfree()
It is available since v3.15-rc5.

Cc: Pablo Neira Ayuso <pablo@netfilter.org>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2014-06-05 00:49:51 -07:00
Yang Yingliang 3fbac2a87e sch_netem: replace magic numbers with enumerate in get_loss_clg
Replace two magic numbers which intialize clgstate::state.

Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2014-02-17 16:55:54 -05:00
Yang Yingliang c045a734da sch_netem: replace magic numbers with enumerate in GE model
Replace some magic numbers which describe states of GE model
loss generator with enumerate.

Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2014-02-14 00:15:20 -05:00
Yang Yingliang 49545a7775 sch_netem: change some func's param from "struct Qdisc *" to "struct netem_sched_data *"
In netem_change(), we have already get "struct netem_sched_data *q".
Replace params of get_correlation() and other similar functions with
"struct netem_sched_data *q".

Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2014-02-14 00:15:20 -05:00
Yang Yingliang 54a4b05cd2 sch_netem: return errcode before setting params
get_dist_table() and get_loss_clg() may be failed. These
two functions should be called after setting the members
of qdisc_priv(sch), or it will break the old settings while
either of them is failed.

Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2014-02-14 00:15:20 -05:00
Hannes Frederic Sowa 809fa972fd reciprocal_divide: update/correction of the algorithm
Jakub Zawadzki noticed that some divisions by reciprocal_divide()
were not correct [1][2], which he could also show with BPF code
after divisions are transformed into reciprocal_value() for runtime
invariance which can be passed to reciprocal_divide() later on;
reverse in BPF dump ended up with a different, off-by-one K in
some situations.

This has been fixed by Eric Dumazet in commit aee636c480
("bpf: do not use reciprocal divide"). This follow-up patch
improves reciprocal_value() and reciprocal_divide() to work in
all cases by using Granlund and Montgomery method, so that also
future use is safe and without any non-obvious side-effects.
Known problems with the old implementation were that division by 1
always returned 0 and some off-by-ones when the dividend and divisor
where very large. This seemed to not be problematic with its
current users, as far as we can tell. Eric Dumazet checked for
the slab usage, we cannot surely say so in the case of flex_array.
Still, in order to fix that, we propose an extension from the
original implementation from commit 6a2d7a955d resp. [3][4],
by using the algorithm proposed in "Division by Invariant Integers
Using Multiplication" [5], Torbjörn Granlund and Peter L.
Montgomery, that is, pseudocode for q = n/d where q, n, d is in
u32 universe:

1) Initialization:

  int l = ceil(log_2 d)
  uword m' = floor((1<<32)*((1<<l)-d)/d)+1
  int sh_1 = min(l,1)
  int sh_2 = max(l-1,0)

2) For q = n/d, all uword:

  uword t = (n*m')>>32
  q = (t+((n-t)>>sh_1))>>sh_2

The assembler implementation from Agner Fog [6] also helped a lot
while implementing. We have tested the implementation on x86_64,
ppc64, i686, s390x; on x86_64/haswell we're still half the latency
compared to normal divide.

Joint work with Daniel Borkmann.

  [1] http://www.wireshark.org/~darkjames/reciprocal-buggy.c
  [2] http://www.wireshark.org/~darkjames/set-and-dump-filter-k-bug.c
  [3] https://gmplib.org/~tege/division-paper.pdf
  [4] http://homepage.cs.uiowa.edu/~jones/bcd/divide.html
  [5] http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.1.2556
  [6] http://www.agner.org/optimize/asmlib.zip

Reported-by: Jakub Zawadzki <darkjames-ws@darkjames.pl>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Austin S Hemmelgarn <ahferroin7@gmail.com>
Cc: linux-kernel@vger.kernel.org
Cc: Jesse Gross <jesse@nicira.com>
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: Stephen Hemminger <stephen@networkplumber.org>
Cc: Matt Mackall <mpm@selenic.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Andy Gospodarek <andy@greyhouse.net>
Cc: Veaceslav Falico <vfalico@redhat.com>
Cc: Jay Vosburgh <fubar@us.ibm.com>
Cc: Jakub Zawadzki <darkjames-ws@darkjames.pl>
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-21 23:17:20 -08:00