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>
After previous patches to simplify qstats the qstats can be
made per cpu with a packed union in Qdisc struct.
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
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>
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>
Conflicts:
arch/mips/net/bpf_jit.c
drivers/net/can/flexcan.c
Both the flexcan and MIPS bpf_jit conflicts were cases of simple
overlapping changes.
Signed-off-by: David S. Miller <davem@davemloft.net>
We cannot make struct qdisc_skb_cb bigger without impacting IPoIB,
or increasing skb->cb[] size.
Commit e0f31d8498 ("flow_keys: Record IP layer protocol in
skb_flow_dissect()") broke IPoIB.
Only current offender is sch_choke, and this one do not need an
absolutely precise flow key.
If we store 17 bytes of flow key, its more than enough. (Its the actual
size of flow_keys if it was a packed structure, but we might add new
fields at the end of it later)
Signed-off-by: Eric Dumazet <edumazet@google.com>
Fixes: e0f31d8498 ("flow_keys: Record IP layer protocol in skb_flow_dissect()")
Signed-off-by: David S. Miller <davem@davemloft.net>
rcu'ify tcf_proto this allows calling tc_classify() without holding
any locks. Updaters are protected by RTNL.
This patch prepares the core net_sched infrastracture for running
the classifier/action chains without holding the qdisc lock however
it does nothing to ensure cls_xxx and act_xxx types also work without
locking. Additional patches are required to address the fall out.
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>
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>
skb_flow_dissect() dissects only transport header type in ip_proto. It dose not
give any information about IPv4 or IPv6.
This patch adds new member, n_proto, to struct flow_keys. Which records the
IP layer type. i.e IPv4 or IPv6.
This can be used in netdev->ndo_rx_flow_steer driver function to dissect flow.
Adding new member to flow_keys increases the struct size by around 4 bytes.
This causes BUILD_BUG_ON(sizeof(qcb->data) < sz); to fail in
qdisc_cb_private_validate()
So increase data size by 4
Signed-off-by: Govindarajulu Varadarajan <_govind@gmx.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When actions are attached to a filter, they are a part of the filter
itself, so when changing a filter we should allow to overwrite the actions
inside as well.
In my specific case, when I tried to _append_ a new action to an existing
filter which already has an action, I got EEXIST since kernel refused
to overwrite the existing one in kernel.
This patch checks if we are changing the filter checking NLM_F_CREATE flag
(Sigh, filters don't use NLM_F_REPLACE...) and then passes the boolean down
to actions. This fixes the problem above.
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: Cong Wang <cwang@twopensource.com>
Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
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>
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>
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>
There are a mix of function prototypes with and without extern
in the kernel sources. Standardize on not using extern for
function prototypes.
Function prototypes don't need to be written with extern.
extern is assumed by the compiler. Its use is as unnecessary as
using auto to declare automatic/local variables in a block.
Reflow modified prototypes to 80 columns.
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Before allowing 64bits bytes rates, refactor
psched_ratecfg_precompute() to get better comments
and increased accuracy.
rate_bps field is renamed to rate_bytes_ps, as we only
have to worry about bytes per second.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Ben Greear <greearb@candelatech.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
struct gnet_stats_rate_est contains u32 fields, so the bytes per second
field can wrap at 34360Mbit.
Add a new gnet_stats_rate_est64 structure to get 64bit bps/pps fields,
and switch the kernel to use this structure natively.
This structure is dumped to user space as a new attribute :
TCA_STATS_RATE_EST64
Old tc command will now display the capped bps (to 34360Mbit), instead
of wrapped values, and updated tc command will display correct
information.
Old tc command output, after patch :
eric:~# tc -s -d qd sh dev lo
qdisc pfifo 8001: root refcnt 2 limit 1000p
Sent 80868245400 bytes 1978837 pkt (dropped 0, overlimits 0 requeues 0)
rate 34360Mbit 189696pps backlog 0b 0p requeues 0
This patch carefully reorganizes "struct Qdisc" layout to get optimal
performance on SMP.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
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>
I'm not sure why, but the hlist for each entry iterators were conceived
list_for_each_entry(pos, head, member)
The hlist ones were greedy and wanted an extra parameter:
hlist_for_each_entry(tpos, pos, head, member)
Why did they need an extra pos parameter? I'm not quite sure. Not only
they don't really need it, it also prevents the iterator from looking
exactly like the list iterator, which is unfortunate.
Besides the semantic patch, there was some manual work required:
- Fix up the actual hlist iterators in linux/list.h
- Fix up the declaration of other iterators based on the hlist ones.
- A very small amount of places were using the 'node' parameter, this
was modified to use 'obj->member' instead.
- Coccinelle didn't handle the hlist_for_each_entry_safe iterator
properly, so those had to be fixed up manually.
The semantic patch which is mostly the work of Peter Senna Tschudin is here:
@@
iterator name hlist_for_each_entry, hlist_for_each_entry_continue, hlist_for_each_entry_from, hlist_for_each_entry_rcu, hlist_for_each_entry_rcu_bh, hlist_for_each_entry_continue_rcu_bh, for_each_busy_worker, ax25_uid_for_each, ax25_for_each, inet_bind_bucket_for_each, sctp_for_each_hentry, sk_for_each, sk_for_each_rcu, sk_for_each_from, sk_for_each_safe, sk_for_each_bound, hlist_for_each_entry_safe, hlist_for_each_entry_continue_rcu, nr_neigh_for_each, nr_neigh_for_each_safe, nr_node_for_each, nr_node_for_each_safe, for_each_gfn_indirect_valid_sp, for_each_gfn_sp, for_each_host;
type T;
expression a,c,d,e;
identifier b;
statement S;
@@
-T b;
<+... when != b
(
hlist_for_each_entry(a,
- b,
c, d) S
|
hlist_for_each_entry_continue(a,
- b,
c) S
|
hlist_for_each_entry_from(a,
- b,
c) S
|
hlist_for_each_entry_rcu(a,
- b,
c, d) S
|
hlist_for_each_entry_rcu_bh(a,
- b,
c, d) S
|
hlist_for_each_entry_continue_rcu_bh(a,
- b,
c) S
|
for_each_busy_worker(a, c,
- b,
d) S
|
ax25_uid_for_each(a,
- b,
c) S
|
ax25_for_each(a,
- b,
c) S
|
inet_bind_bucket_for_each(a,
- b,
c) S
|
sctp_for_each_hentry(a,
- b,
c) S
|
sk_for_each(a,
- b,
c) S
|
sk_for_each_rcu(a,
- b,
c) S
|
sk_for_each_from
-(a, b)
+(a)
S
+ sk_for_each_from(a) S
|
sk_for_each_safe(a,
- b,
c, d) S
|
sk_for_each_bound(a,
- b,
c) S
|
hlist_for_each_entry_safe(a,
- b,
c, d, e) S
|
hlist_for_each_entry_continue_rcu(a,
- b,
c) S
|
nr_neigh_for_each(a,
- b,
c) S
|
nr_neigh_for_each_safe(a,
- b,
c, d) S
|
nr_node_for_each(a,
- b,
c) S
|
nr_node_for_each_safe(a,
- b,
c, d) S
|
- for_each_gfn_sp(a, c, d, b) S
+ for_each_gfn_sp(a, c, d) S
|
- for_each_gfn_indirect_valid_sp(a, c, d, b) S
+ for_each_gfn_indirect_valid_sp(a, c, d) S
|
for_each_host(a,
- b,
c) S
|
for_each_host_safe(a,
- b,
c, d) S
|
for_each_mesh_entry(a,
- b,
c, d) S
)
...+>
[akpm@linux-foundation.org: drop bogus change from net/ipv4/raw.c]
[akpm@linux-foundation.org: drop bogus hunk from net/ipv6/raw.c]
[akpm@linux-foundation.org: checkpatch fixes]
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foudnation.org: redo intrusive kvm changes]
Tested-by: Peter Senna Tschudin <peter.senna@gmail.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Eric Dumazet pointed out that act_mirred needs to find the current net_ns,
and struct net pointer is not provided in the call chain. His original
patch made use of current->nsproxy->net_ns to find the network namespace,
but this fails to work correctly for userspace code that makes use of
netlink sockets in different network namespaces. Instead, pass the
"struct net *" down along the call chain to where it is needed.
This version removes the ifb changes as Eric has submitted that patch
separately, but is otherwise identical to the previous version.
Signed-off-by: Benjamin LaHaise <bcrl@kvack.org>
Tested-by: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
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>
