This adds support for DiffServ-based priority queueing to CAKE. If the
shaper is in use, each priority tier gets its own virtual clock, which
limits that tier's rate to a fraction of the overall shaped rate, to
discourage trying to game the priority mechanism.
CAKE defaults to a simple, three-tier mode that interprets most code points
as "best effort", but places CS1 traffic into a low-priority "bulk" tier
which is assigned 1/16 of the total rate, and a few code points indicating
latency-sensitive or control traffic (specifically TOS4, VA, EF, CS6, CS7)
into a "latency sensitive" high-priority tier, which is assigned 1/4 rate.
The other supported DiffServ modes are a 4-tier mode matching the 802.11e
precedence rules, as well as two 8-tier modes, one of which implements
strict precedence of the eight priority levels.
This commit also adds an optional DiffServ 'wash' mode, which will zero out
the DSCP fields of any packet passing through CAKE. While this can
technically be done with other mechanisms in the kernel, having the feature
available in CAKE significantly decreases configuration complexity; and the
implementation cost is low on top of the other DiffServ-handling code.
Filters and applications can set the skb->priority field to override the
DSCP-based classification into tiers. If TC_H_MAJ(skb->priority) matches
CAKE's qdisc handle, the minor number will be interpreted as a priority
tier if it is less than or equal to the number of configured priority
tiers.
Signed-off-by: Toke Høiland-Jørgensen <toke@toke.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
When CAKE is deployed on a gateway that also performs NAT (which is a
common deployment mode), the host fairness mechanism cannot distinguish
internal hosts from each other, and so fails to work correctly.
To fix this, we add an optional NAT awareness mode, which will query the
kernel conntrack mechanism to obtain the pre-NAT addresses for each packet
and use that in the flow and host hashing.
When the shaper is enabled and the host is already performing NAT, the cost
of this lookup is negligible. However, in unlimited mode with no NAT being
performed, there is a significant CPU cost at higher bandwidths. For this
reason, the feature is turned off by default.
Cc: netfilter-devel@vger.kernel.org
Signed-off-by: Toke Høiland-Jørgensen <toke@toke.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
The ACK filter is an optional feature of CAKE which is designed to improve
performance on links with very asymmetrical rate limits. On such links
(which are unfortunately quite prevalent, especially for DSL and cable
subscribers), the downstream throughput can be limited by the number of
ACKs capable of being transmitted in the *upstream* direction.
Filtering ACKs can, in general, have adverse effects on TCP performance
because it interferes with ACK clocking (especially in slow start), and it
reduces the flow's resiliency to ACKs being dropped further along the path.
To alleviate these drawbacks, the ACK filter in CAKE tries its best to
always keep enough ACKs queued to ensure forward progress in the TCP flow
being filtered. It does this by only filtering redundant ACKs. In its
default 'conservative' mode, the filter will always keep at least two
redundant ACKs in the queue, while in 'aggressive' mode, it will filter
down to a single ACK.
The ACK filter works by inspecting the per-flow queue on every packet
enqueue. Starting at the head of the queue, the filter looks for another
eligible packet to drop (so the ACK being dropped is always closer to the
head of the queue than the packet being enqueued). An ACK is eligible only
if it ACKs *fewer* bytes than the new packet being enqueued, including any
SACK options. This prevents duplicate ACKs from being filtered, to avoid
interfering with retransmission logic. In addition, we check TCP header
options and only drop those that are known to not interfere with sender
state. In particular, packets with unknown option codes are never dropped.
In aggressive mode, an eligible packet is always dropped, while in
conservative mode, at least two ACKs are kept in the queue. Only pure ACKs
(with no data segments) are considered eligible for dropping, but when an
ACK with data segments is enqueued, this can cause another pure ACK to
become eligible for dropping.
The approach described above ensures that this ACK filter avoids most of
the drawbacks of a naive filtering mechanism that only keeps flow state but
does not inspect the queue. This is the rationale for including the ACK
filter in CAKE itself rather than as separate module (as the TC filter, for
instance).
Our performance evaluation has shown that on a 30/1 Mbps link with a
bidirectional traffic test (RRUL), turning on the ACK filter on the
upstream link improves downstream throughput by ~20% (both modes) and
upstream throughput by ~12% in conservative mode and ~40% in aggressive
mode, at the cost of ~5ms of inter-flow latency due to the increased
congestion.
In *really* pathological cases, the effect can be a lot more; for instance,
the ACK filter increases the achievable downstream throughput on a link
with 100 Kbps in the upstream direction by an order of magnitude (from ~2.5
Mbps to ~25 Mbps).
Finally, even though we consider the ACK filter to be safer than most, we
do not recommend turning it on everywhere: on more symmetrical link
bandwidths the effect is negligible at best.
Cc: Yuchung Cheng <ycheng@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Toke Høiland-Jørgensen <toke@toke.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
The ingress mode is meant to be enabled when CAKE runs downlink of the
actual bottleneck (such as on an IFB device). The mode changes the shaper
to also account dropped packets to the shaped rate, as these have already
traversed the bottleneck.
Enabling ingress mode will also tune the AQM to always keep at least two
packets queued *for each flow*. This is done by scaling the minimum queue
occupancy level that will disable the AQM by the number of active bulk
flows. The rationale for this is that retransmits are more expensive in
ingress mode, since dropped packets have to traverse the bottleneck again
when they are retransmitted; thus, being more lenient and keeping a minimum
number of packets queued will improve throughput in cases where the number
of active flows are so large that they saturate the bottleneck even at
their minimum window size.
This commit also adds a separate switch to enable ingress mode rate
autoscaling. If enabled, the autoscaling code will observe the actual
traffic rate and adjust the shaper rate to match it. This can help avoid
latency increases in the case where the actual bottleneck rate decreases
below the shaped rate. The scaling filters out spikes by an EWMA filter.
Signed-off-by: Toke Høiland-Jørgensen <toke@toke.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
sch_cake targets the home router use case and is intended to squeeze the
most bandwidth and latency out of even the slowest ISP links and routers,
while presenting an API simple enough that even an ISP can configure it.
Example of use on a cable ISP uplink:
tc qdisc add dev eth0 cake bandwidth 20Mbit nat docsis ack-filter
To shape a cable download link (ifb and tc-mirred setup elided)
tc qdisc add dev ifb0 cake bandwidth 200mbit nat docsis ingress wash
CAKE is filled with:
* A hybrid Codel/Blue AQM algorithm, "Cobalt", tied to an FQ_Codel
derived Flow Queuing system, which autoconfigures based on the bandwidth.
* A novel "triple-isolate" mode (the default) which balances per-host
and per-flow FQ even through NAT.
* An deficit based shaper, that can also be used in an unlimited mode.
* 8 way set associative hashing to reduce flow collisions to a minimum.
* A reasonable interpretation of various diffserv latency/loss tradeoffs.
* Support for zeroing diffserv markings for entering and exiting traffic.
* Support for interacting well with Docsis 3.0 shaper framing.
* Extensive support for DSL framing types.
* Support for ack filtering.
* Extensive statistics for measuring, loss, ecn markings, latency
variation.
A paper describing the design of CAKE is available at
https://arxiv.org/abs/1804.07617, and will be published at the 2018 IEEE
International Symposium on Local and Metropolitan Area Networks (LANMAN).
This patch adds the base shaper and packet scheduler, while subsequent
commits add the optional (configurable) features. The full userspace API
and most data structures are included in this commit, but options not
understood in the base version will be ignored.
Various versions baking have been available as an out of tree build for
kernel versions going back to 3.10, as the embedded router world has been
running a few years behind mainline Linux. A stable version has been
generally available on lede-17.01 and later.
sch_cake replaces a combination of iptables, tc filter, htb and fq_codel
in the sqm-scripts, with sane defaults and vastly simpler configuration.
CAKE's principal author is Jonathan Morton, with contributions from
Kevin Darbyshire-Bryant, Toke Høiland-Jørgensen, Sebastian Moeller,
Ryan Mounce, Tony Ambardar, Dean Scarff, Nils Andreas Svee, Dave Täht,
and Loganaden Velvindron.
Testing from Pete Heist, Georgios Amanakis, and the many other members of
the cake@lists.bufferbloat.net mailing list.
tc -s qdisc show dev eth2
qdisc cake 8017: root refcnt 2 bandwidth 1Gbit diffserv3 triple-isolate split-gso rtt 100.0ms noatm overhead 38 mpu 84
Sent 51504294511 bytes 37724591 pkt (dropped 6, overlimits 64958695 requeues 12)
backlog 0b 0p requeues 12
memory used: 1053008b of 15140Kb
capacity estimate: 970Mbit
min/max network layer size: 28 / 1500
min/max overhead-adjusted size: 84 / 1538
average network hdr offset: 14
Bulk Best Effort Voice
thresh 62500Kbit 1Gbit 250Mbit
target 5.0ms 5.0ms 5.0ms
interval 100.0ms 100.0ms 100.0ms
pk_delay 5us 5us 6us
av_delay 3us 2us 2us
sp_delay 2us 1us 1us
backlog 0b 0b 0b
pkts 3164050 25030267 9530280
bytes 3227519915 35396974782 12879808898
way_inds 0 8 0
way_miss 21 366 25
way_cols 0 0 0
drops 5 0 1
marks 0 0 0
ack_drop 0 0 0
sp_flows 1 3 0
bk_flows 0 1 1
un_flows 0 0 0
max_len 68130 68130 68130
Tested-by: Pete Heist <peteheist@gmail.com>
Tested-by: Georgios Amanakis <gamanakis@gmail.com>
Signed-off-by: Dave Taht <dave.taht@gmail.com>
Signed-off-by: Toke Høiland-Jørgensen <toke@toke.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
The kbuild test robot reports:
>> net/sched/act_api.c:71:15: sparse: incorrect type in initializer (different address spaces) @@ expected struct tc_cookie [noderef] <asn:4>*__ret @@ got [noderef] <asn:4>*__ret @@
net/sched/act_api.c:71:15: expected struct tc_cookie [noderef] <asn:4>*__ret
net/sched/act_api.c:71:15: got struct tc_cookie *new_cookie
>> net/sched/act_api.c:71:13: sparse: incorrect type in assignment (different address spaces) @@ expected struct tc_cookie *old @@ got struct tc_cookie [noderef] <struct tc_cookie *old @@
net/sched/act_api.c:71:13: expected struct tc_cookie *old
net/sched/act_api.c:71:13: got struct tc_cookie [noderef] <asn:4>*[assigned] __ret
>> net/sched/act_api.c:132:48: sparse: dereference of noderef expression
Handle this in the usual way by force casting away the __rcu annotation
when we are using xchg() on it.
Fixes: eec94fdb04 ("net: sched: use rcu for action cookie update")
Reported-by: kbuild test robot <lkp@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Act API used linked list to pass set of actions to functions. It is
intrusive data structure that stores list nodes inside action structure
itself, which means it is not safe to modify such list concurrently.
However, action API doesn't use any linked list specific operations on this
set of actions, so it can be safely refactored into plain pointer array.
Refactor action API to use array of pointers to tc_actions instead of
linked list. Change argument 'actions' type of exported action init,
destroy and dump functions.
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Vlad Buslov <vladbu@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Implement function that atomically checks if action exists and either takes
reference to it, or allocates idr slot for action index to prevent
concurrent allocations of actions with same index. Use EBUSY error pointer
to indicate that idr slot is reserved.
Implement cleanup helper function that removes temporary error pointer from
idr. (in case of error between idr allocation and insertion of newly
created action to specified index)
Refactor all action init functions to insert new action to idr using this
API.
Reviewed-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: Vlad Buslov <vladbu@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Change action API to assume that action init function always takes
reference to action, even when overwriting existing action. This is
necessary because action API continues to use action pointer after init
function is done. At this point action becomes accessible for concurrent
modifications, so user must always hold reference to it.
Implement helper put list function to atomically release list of actions
after action API init code is done using them.
Reviewed-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: Vlad Buslov <vladbu@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Return from action init function with reference to action taken,
even when overwriting existing action.
Action init API initializes its fourth argument (pointer to pointer to tc
action) to either existing action with same index or newly created action.
In case of existing index(and bind argument is zero), init function returns
without incrementing action reference counter. Caller of action init then
proceeds working with action, without actually holding reference to it.
This means that action could be deleted concurrently.
Change action init behavior to always take reference to action before
returning successfully, in order to protect from concurrent deletion.
Reviewed-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: Vlad Buslov <vladbu@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Implement helper delete function that uses new action ops 'delete', instead
of destroying action directly. This is required so act API could delete
actions by index, without holding any references to action that is being
deleted.
Implement function __tcf_action_put() that releases reference to action and
frees it, if necessary. Refactor action deletion code to use new put
function and not to rely on rtnl lock. Remove rtnl lock assertions that are
no longer needed.
Reviewed-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: Vlad Buslov <vladbu@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Extend action ops with 'delete' function. Each action type to implements
its own delete function that doesn't depend on rtnl lock.
Implement delete function that is required to delete actions without
holding rtnl lock. Use action API function that atomically deletes action
only if it is still in action idr. This implementation prevents concurrent
threads from deleting same action twice.
Reviewed-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: Vlad Buslov <vladbu@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Implement new action API function that atomically finds and deletes action
from idr by index. Intended to be used by lockless actions that do not rely
on rtnl lock.
Reviewed-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: Vlad Buslov <vladbu@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Without rtnl lock protection it is no longer safe to use pointer to tc
action without holding reference to it. (it can be destroyed concurrently)
Remove unsafe action idr lookup function. Instead of it, implement safe tcf
idr check function that atomically looks up action in idr and increments
its reference and bind counters. Implement both action search and check
using new safe function
Reference taken by idr check is temporal and should not be accounted by
userspace clients (both logically and to preserver current API behavior).
Subtract temporal reference when dumping action to userspace using existing
tca_get_fill function arguments.
Reviewed-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: Vlad Buslov <vladbu@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add additional 'rtnl_held' argument to act API init functions. It is
required to implement actions that need to release rtnl lock before loading
kernel module and reacquire if afterwards.
Reviewed-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: Vlad Buslov <vladbu@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Change type of action reference counter to refcount_t.
Change type of action bind counter to atomic_t.
This type is used to allow decrementing bind counter without testing
for 0 result.
Reviewed-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: Vlad Buslov <vladbu@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Implement functions to atomically update and free action cookie
using rcu mechanism.
Reviewed-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: Vlad Buslov <vladbu@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As support dissecting of QinQ inner and outer vlan headers, user can
add rules to match on QinQ vlan headers.
Signed-off-by: Jianbo Liu <jianbol@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently the encapsulated ethertype is not dumped as it's the same as
TCA_FLOWER_KEY_ETH_TYPE keyvalue. But the dumping result is inconsistent
with input, we add dumping it with TCA_FLOWER_KEY_VLAN_ETH_TYPE.
Signed-off-by: Jianbo Liu <jianbol@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As flow dissector stores vlan ethertype, tc flower now can match on that.
It is to make preparation for supporting QinQ.
Signed-off-by: Jianbo Liu <jianbol@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use the socket error queue for reporting dropped packets if the
socket has enabled that feature through the SO_TXTIME API.
Packets are dropped either on enqueue() if they aren't accepted by the
qdisc or on dequeue() if the system misses their deadline. Those are
reported as different errors so applications can react accordingly.
Userspace can retrieve the errors through the socket error queue and the
corresponding cmsg interfaces. A struct sock_extended_err* is used for
returning the error data, and the packet's timestamp can be retrieved by
adding both ee_data and ee_info fields as e.g.:
((__u64) serr->ee_data << 32) + serr->ee_info
This feature is disabled by default and must be explicitly enabled by
applications. Enabling it can bring some overhead for the Tx cycles
of the application.
Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add infra so etf qdisc supports HW offload of time-based transmission.
For hw offload, the time sorted list is still used, so packets are
dequeued always in order of txtime.
Example:
$ tc qdisc replace dev enp2s0 parent root handle 100 mqprio num_tc 3 \
map 2 2 1 0 2 2 2 2 2 2 2 2 2 2 2 2 queues 1@0 1@1 2@2 hw 0
$ tc qdisc add dev enp2s0 parent 100:1 etf offload delta 100000 \
clockid CLOCK_REALTIME
In this example, the Qdisc will use HW offload for the control of the
transmission time through the network adapter. The hrtimer used for
packets scheduling inside the qdisc will use the clockid CLOCK_REALTIME
as reference and packets leave the Qdisc "delta" (100000) nanoseconds
before their transmission time. Because this will be using HW offload and
since dynamic clocks are not supported by the hrtimer, the system clock
and the PHC clock must be synchronized for this mode to behave as
expected.
Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The ETF (Earliest TxTime First) qdisc uses the information added
earlier in this series (the socket option SO_TXTIME and the new
role of sk_buff->tstamp) to schedule packets transmission based
on absolute time.
For some workloads, just bandwidth enforcement is not enough, and
precise control of the transmission of packets is necessary.
Example:
$ tc qdisc replace dev enp2s0 parent root handle 100 mqprio num_tc 3 \
map 2 2 1 0 2 2 2 2 2 2 2 2 2 2 2 2 queues 1@0 1@1 2@2 hw 0
$ tc qdisc add dev enp2s0 parent 100:1 etf delta 100000 \
clockid CLOCK_TAI
In this example, the Qdisc will provide SW best-effort for the control
of the transmission time to the network adapter, the time stamp in the
socket will be in reference to the clockid CLOCK_TAI and packets
will leave the qdisc "delta" (100000) nanoseconds before its transmission
time.
The ETF qdisc will buffer packets sorted by their txtime. It will drop
packets on enqueue() if their skbuff clockid does not match the clock
reference of the Qdisc. Moreover, on dequeue(), a packet will be dropped
if it expires while being enqueued.
The qdisc also supports the SO_TXTIME deadline mode. For this mode, it
will dequeue a packet as soon as possible and change the skb timestamp
to 'now' during etf_dequeue().
Note that both the qdisc's and the SO_TXTIME ABIs allow for a clockid
to be configured, but it's been decided that usage of CLOCK_TAI should
be enforced until we decide to allow for other clockids to be used.
The rationale here is that PTP times are usually in the TAI scale, thus
no other clocks should be necessary. For now, the qdisc will return
EINVAL if any clocks other than CLOCK_TAI are used.
Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com>
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This adds 'qdisc_watchdog_init_clockid()' that allows a clockid to be
passed, this allows other time references to be used when scheduling
the Qdisc to run.
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
'keys_ex' is malloced by tcf_pedit_keys_ex_parse() in tcf_pedit_init()
but not all of the error handle path free it, this may cause memory
leak. This patch fix it.
Fixes: 71d0ed7079 ("net/act_pedit: Support using offset relative to the conventional network headers")
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Acked-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The new action inheritdsfield copies the field DS of
IPv4 and IPv6 packets into skb->priority. This enables
later classification of packets based on the DS field.
v5:
*Update the drop counter for TC_ACT_SHOT
v4:
*Not allow setting flags other than the expected ones.
*Allow dumping the pure flags.
v3:
*Use optional flags, so that it won't break old versions of tc.
*Allow users to set both SKBEDIT_F_PRIORITY and SKBEDIT_F_INHERITDSFIELD flags.
v2:
*Fix the style issue
*Move the code from skbmod to skbedit
Original idea by Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Qiaobin Fu <qiaobinf@bu.edu>
Reviewed-by: Michel Machado <michel@digirati.com.br>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Reviewed-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Acked-by: Davide Caratti <dcaratti@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Allow setting tunnel options using the act_tunnel_key action.
Options are expressed as class:type:data and multiple options
may be listed using a comma delimiter.
# ip link add name geneve0 type geneve dstport 0 external
# tc qdisc add dev eth0 ingress
# tc filter add dev eth0 protocol ip parent ffff: \
flower indev eth0 \
ip_proto udp \
action tunnel_key \
set src_ip 10.0.99.192 \
dst_ip 10.0.99.193 \
dst_port 6081 \
id 11 \
geneve_opts 0102:80:00800022,0102:80:00800022 \
action mirred egress redirect dev geneve0
Signed-off-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add extended ack support for the tunnel key action by using NL_SET_ERR_MSG
during validation of user input.
Cc: Alexander Aring <aring@mojatatu.com>
Signed-off-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Metadata may be NULL for one of two reasons:
* Missing user input
* Failure to allocate the metadata dst
Disambiguate these case by returning -EINVAL for the former and -ENOMEM
for the latter rather than -EINVAL for both cases.
This is in preparation for using extended ack to provide more information
to users when parsing their input.
Signed-off-by: Simon Horman <simon.horman@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since char can be unsigned or signed, and bitwise operators may have
implementation-dependent results when performed on signed operands,
declare 'u8 *' operand instead.
Suggested-by: Davide Caratti <dcaratti@redhat.com>
Signed-off-by: Roman Mashak <mrv@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Change "tc filter pedit .." to "tc actions pedit .." in error
messages to clearly refer to pedit action.
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Roman Mashak <mrv@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Replace constant integer with sizeof() to clearly indicate
the destination buffer length in skb_header_pointer() calls.
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Roman Mashak <mrv@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The variable _data in include/asm-generic/sections.h defines sections,
this causes sparse warning in pedit:
net/sched/act_pedit.c:293:35: warning: symbol '_data' shadows an earlier one
./include/asm-generic/sections.h:36:13: originally declared here
Therefore rename the variable.
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Roman Mashak <mrv@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix coding style issues in tc pedit action detected by the
checkpatch script.
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Roman Mashak <mrv@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Extend slotting with support for non-uniform distributions. This is
similar to netem's non-uniform distribution delay feature.
Commit f043efeae2f1 ("netem: support delivering packets in delayed
time slots") added the slotting feature to approximate the behaviors
of media with packet aggregation but only supported a uniform
distribution for delays between transmission attempts. Tests with TCP
BBR with emulated wifi links with non-uniform distributions produced
more useful results.
Syntax:
slot dist DISTRIBUTION DELAY JITTER [packets MAX_PACKETS] \
[bytes MAX_BYTES]
The syntax and use of the distribution table is the same as in the
non-uniform distribution delay feature. A file DISTRIBUTION must be
present in TC_LIB_DIR (e.g. /usr/lib/tc) containing numbers scaled by
NETEM_DIST_SCALE. A random value x is selected from the table and it
takes DELAY + ( x * JITTER ) as delay. Correlation between values is not
supported.
Examples:
Normal distribution delay with mean = 800us and stdev = 100us.
> tc qdisc add dev eth0 root netem slot dist normal 800us 100us
Optionally set the max slot size in bytes and/or packets.
> tc qdisc add dev eth0 root netem slot dist normal 800us 100us \
bytes 64k packets 42
Signed-off-by: Yousuk Seung <ysseung@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Call the reoffload tcf_proto_op on all tcf_proto nodes in all chains of a
block when a callback tries to register to a block that already has
offloaded rules. If all existing rules cannot be offloaded then the
registration is rejected. This replaces the previous policy of rejecting
such callback registration outright.
On unregistration of a callback, the rules are flushed for that given cb.
The implementation of block sharing in the NFP driver, for example,
duplicates shared rules to all devs bound to a block. This meant that
rules could still exist in hw even after a device is unbound from a block
(assuming the block still remains active).
Signed-off-by: John Hurley <john.hurley@netronome.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add the offload tcf_proto_op in cls_bpf to generate an offload message for
each bpf prog in the given tcf_proto. Call the specified callback with
this new offload message. The function only returns an error if the
callback rejects adding a 'hardware only' prog.
A prog contains a flag to indicate if it is in hardware or not. To
ensure the offload function properly maintains this flag, keep a reference
counter for the number of instances of the prog that are in hardware. Only
update the flag when this counter changes from or to 0.
Signed-off-by: John Hurley <john.hurley@netronome.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add the offload tcf_proto_op in cls_u32 to generate an offload message for
each filter and the hashtable in the given tcf_proto. Call the specified
callback with this new offload message. The function only returns an error
if the callback rejects adding a 'hardware only' rule.
A filter contains a flag to indicate if it is in hardware or not. To
ensure the offload function properly maintains this flag, keep a reference
counter for the number of instances of the filter that are in hardware.
Only update the flag when this counter changes from or to 0.
Signed-off-by: John Hurley <john.hurley@netronome.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add the reoffload tcf_proto_op in matchall to generate an offload message
for each filter in the given tcf_proto. Call the specified callback with
this new offload message. The function only returns an error if the
callback rejects adding a 'hardware only' rule.
Ensure matchall flags correctly report if the rule is in hw by keeping a
reference counter for the number of instances of the rule offloaded. Only
update the flag when this counter changes from or to 0.
Signed-off-by: John Hurley <john.hurley@netronome.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add the reoffload tcf_proto_op in flower to generate an offload message
for each filter in the given tcf_proto. Call the specified callback with
this new offload message. The function only returns an error if the
callback rejects adding a 'hardware only' rule.
A filter contains a flag to indicate if it is in hardware or not. To
ensure the reoffload function properly maintains this flag, keep a
reference counter for the number of instances of the filter that are in
hardware. Only update the flag when this counter changes from or to 0. Add
a generic helper function to implement this behaviour.
Signed-off-by: John Hurley <john.hurley@netronome.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pass the extact struct from a tc qdisc add to the block bind function and,
in turn, to the setup_tc ndo of binding device via the tc_block_offload
struct. Pass this back to any block callback registrations to allow
netlink logging of fails in the bind process.
Signed-off-by: John Hurley <john.hurley@netronome.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
After commit a09ceb0e08 ("sched: remove qdisc->drop"),
it is no longer used.
Cc: Florian Westphal <fw@strlen.de>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In update_vf():
cftree_remove(cl);
update_cfmin(cl->cl_parent);
the cl_cfmin of cl->cl_parent is intentionally updated to 0
when that parent only has one child. And if this parent is
root qdisc, we could end up, in hfsc_schedule_watchdog(),
that we can't decide the next schedule time for qdisc watchdog.
But it seems safe that we can just skip it, as this watchdog is
not always scheduled anyway.
Thanks to Marco for testing all the cases, nothing is broken.
Reported-by: Marco Berizzi <pupilla@libero.it>
Tested-by: Marco Berizzi <pupilla@libero.it>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If flower filter is created without the skip_sw flag, fl_mask_put()
can race with fl_classify() and we can destroy the mask rhashtable
while a lookup operation is accessing it.
BUG: unable to handle kernel paging request at 00000000000911d1
PGD 0 P4D 0
SMP PTI
CPU: 3 PID: 5582 Comm: vhost-5541 Not tainted 4.18.0-rc1.vanilla+ #1950
Hardware name: Dell Inc. PowerEdge R730/072T6D, BIOS 2.1.7 06/16/2016
RIP: 0010:rht_bucket_nested+0x20/0x60
Code: 31 c8 c1 c1 18 29 c8 c3 66 90 8b 4f 04 ba 01 00 00 00 8b 07 48 8b bf 80 00 00 0
RSP: 0018:ffffafc5cfbb7a48 EFLAGS: 00010206
RAX: 0000000000001978 RBX: ffff9f12dff88a00 RCX: 00000000ffff9f12
RDX: 00000000000911d1 RSI: 0000000000000148 RDI: 0000000000000001
RBP: ffff9f12dff88a00 R08: 000000005f1cc119 R09: 00000000a715fae2
R10: ffffafc5cfbb7aa8 R11: ffff9f1cb4be804e R12: ffff9f1265e13000
R13: 0000000000000000 R14: ffffafc5cfbb7b48 R15: ffff9f12dff88b68
FS: 0000000000000000(0000) GS:ffff9f1d3f0c0000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00000000000911d1 CR3: 0000001575a94006 CR4: 00000000001626e0
Call Trace:
fl_lookup+0x134/0x140 [cls_flower]
fl_classify+0xf3/0x180 [cls_flower]
tcf_classify+0x78/0x150
__netif_receive_skb_core+0x69e/0xa50
netif_receive_skb_internal+0x42/0xf0
tun_get_user+0xdd5/0xfd0 [tun]
tun_sendmsg+0x52/0x70 [tun]
handle_tx+0x2b3/0x5f0 [vhost_net]
vhost_worker+0xab/0x100 [vhost]
kthread+0xf8/0x130
ret_from_fork+0x35/0x40
Modules linked in: act_mirred act_gact cls_flower vhost_net vhost tap sch_ingress
CR2: 00000000000911d1
Fix the above waiting for a RCU grace period before destroying the
rhashtable: we need to use tcf_queue_work(), as rhashtable_destroy()
must run in process context, as pointed out by Cong Wang.
v1 -> v2: use tcf_queue_work to run rhashtable_destroy().
Fixes: 05cd271fd6 ("cls_flower: Support multiple masks per priority")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
in the following script
# tc actions add action ife encode allow prio pass index 42
# tc actions replace action ife encode allow tcindex drop index 42
the action control should remain equal to 'pass', if the kernel failed
to replace the TC action. Pospone the assignment of the action control,
to ensure it is not overwritten in the error path of tcf_ife_init().
Fixes: ef6980b6be ("introduce IFE action")
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Acked-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
a recursive lock warning [1] can be observed with the following script,
# $TC actions add action ife encode allow prio pass index 42
IFE type 0xED3E
# $TC actions replace action ife encode allow tcindex pass index 42
in case the kernel was unable to run the last command (e.g. because of
the impossibility to load 'act_meta_skbtcindex'). For a similar reason,
the kernel can leak idr in the error path of tcf_ife_init(), because
tcf_idr_release() is not called after successful idr reservation:
# $TC actions add action ife encode allow tcindex index 47
IFE type 0xED3E
RTNETLINK answers: No such file or directory
We have an error talking to the kernel
# $TC actions add action ife encode allow tcindex index 47
IFE type 0xED3E
RTNETLINK answers: No space left on device
We have an error talking to the kernel
# $TC actions add action ife encode use mark 7 type 0xfefe pass index 47
IFE type 0xFEFE
RTNETLINK answers: No space left on device
We have an error talking to the kernel
Since tcfa_lock is already taken when the action is being edited, a call
to tcf_idr_release() wrongly makes tcf_idr_cleanup() take the same lock
again. On the other hand, tcf_idr_release() needs to be called in the
error path of tcf_ife_init(), to undo the last tcf_idr_create() invocation.
Fix both problems in tcf_ife_init().
Since the cleanup() routine can now be called when ife->params is NULL,
also add a NULL pointer check to avoid calling kfree_rcu(NULL, rcu).
[1]
============================================
WARNING: possible recursive locking detected
4.17.0-rc4.kasan+ #417 Tainted: G E
--------------------------------------------
tc/3932 is trying to acquire lock:
000000005097c9a6 (&(&p->tcfa_lock)->rlock){+...}, at: tcf_ife_cleanup+0x19/0x80 [act_ife]
but task is already holding lock:
000000005097c9a6 (&(&p->tcfa_lock)->rlock){+...}, at: tcf_ife_init+0xf6d/0x13c0 [act_ife]
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(&(&p->tcfa_lock)->rlock);
lock(&(&p->tcfa_lock)->rlock);
*** DEADLOCK ***
May be due to missing lock nesting notation
2 locks held by tc/3932:
#0: 000000007ca8e990 (rtnl_mutex){+.+.}, at: tcf_ife_init+0xf61/0x13c0 [act_ife]
#1: 000000005097c9a6 (&(&p->tcfa_lock)->rlock){+...}, at: tcf_ife_init+0xf6d/0x13c0 [act_ife]
stack backtrace:
CPU: 3 PID: 3932 Comm: tc Tainted: G E 4.17.0-rc4.kasan+ #417
Hardware name: Red Hat KVM, BIOS 0.5.1 01/01/2011
Call Trace:
dump_stack+0x9a/0xeb
__lock_acquire+0xf43/0x34a0
? debug_check_no_locks_freed+0x2b0/0x2b0
? debug_check_no_locks_freed+0x2b0/0x2b0
? debug_check_no_locks_freed+0x2b0/0x2b0
? __mutex_lock+0x62f/0x1240
? kvm_sched_clock_read+0x1a/0x30
? sched_clock+0x5/0x10
? sched_clock_cpu+0x18/0x170
? find_held_lock+0x39/0x1d0
? lock_acquire+0x10b/0x330
lock_acquire+0x10b/0x330
? tcf_ife_cleanup+0x19/0x80 [act_ife]
_raw_spin_lock_bh+0x38/0x70
? tcf_ife_cleanup+0x19/0x80 [act_ife]
tcf_ife_cleanup+0x19/0x80 [act_ife]
__tcf_idr_release+0xff/0x350
tcf_ife_init+0xdde/0x13c0 [act_ife]
? ife_exit_net+0x290/0x290 [act_ife]
? __lock_is_held+0xb4/0x140
tcf_action_init_1+0x67b/0xad0
? tcf_action_dump_old+0xa0/0xa0
? sched_clock+0x5/0x10
? sched_clock_cpu+0x18/0x170
? kvm_sched_clock_read+0x1a/0x30
? sched_clock+0x5/0x10
? sched_clock_cpu+0x18/0x170
? memset+0x1f/0x40
tcf_action_init+0x30f/0x590
? tcf_action_init_1+0xad0/0xad0
? memset+0x1f/0x40
tc_ctl_action+0x48e/0x5e0
? mutex_lock_io_nested+0x1160/0x1160
? tca_action_gd+0x990/0x990
? sched_clock+0x5/0x10
? find_held_lock+0x39/0x1d0
rtnetlink_rcv_msg+0x4da/0x990
? validate_linkmsg+0x680/0x680
? sched_clock_cpu+0x18/0x170
? find_held_lock+0x39/0x1d0
netlink_rcv_skb+0x127/0x350
? validate_linkmsg+0x680/0x680
? netlink_ack+0x970/0x970
? __kmalloc_node_track_caller+0x304/0x3a0
netlink_unicast+0x40f/0x5d0
? netlink_attachskb+0x580/0x580
? _copy_from_iter_full+0x187/0x760
? import_iovec+0x90/0x390
netlink_sendmsg+0x67f/0xb50
? netlink_unicast+0x5d0/0x5d0
? copy_msghdr_from_user+0x206/0x340
? netlink_unicast+0x5d0/0x5d0
sock_sendmsg+0xb3/0xf0
___sys_sendmsg+0x60a/0x8b0
? copy_msghdr_from_user+0x340/0x340
? lock_downgrade+0x5e0/0x5e0
? tty_write_lock+0x18/0x50
? kvm_sched_clock_read+0x1a/0x30
? sched_clock+0x5/0x10
? sched_clock_cpu+0x18/0x170
? find_held_lock+0x39/0x1d0
? lock_downgrade+0x5e0/0x5e0
? lock_acquire+0x10b/0x330
? __audit_syscall_entry+0x316/0x690
? current_kernel_time64+0x6b/0xd0
? __fget_light+0x55/0x1f0
? __sys_sendmsg+0xd2/0x170
__sys_sendmsg+0xd2/0x170
? __ia32_sys_shutdown+0x70/0x70
? syscall_trace_enter+0x57a/0xd60
? rcu_read_lock_sched_held+0xdc/0x110
? __bpf_trace_sys_enter+0x10/0x10
? do_syscall_64+0x22/0x480
do_syscall_64+0xa5/0x480
entry_SYSCALL_64_after_hwframe+0x49/0xbe
RIP: 0033:0x7fd646988ba0
RSP: 002b:00007fffc9fab3c8 EFLAGS: 00000246 ORIG_RAX: 000000000000002e
RAX: ffffffffffffffda RBX: 00007fffc9fab4f0 RCX: 00007fd646988ba0
RDX: 0000000000000000 RSI: 00007fffc9fab440 RDI: 0000000000000003
RBP: 000000005b28c8b3 R08: 0000000000000002 R09: 0000000000000000
R10: 00007fffc9faae20 R11: 0000000000000246 R12: 0000000000000000
R13: 00007fffc9fab504 R14: 0000000000000001 R15: 000000000066c100
Fixes: 4e8c861550 ("net sched: net sched: ife action fix late binding")
Fixes: ef6980b6be ("introduce IFE action")
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Acked-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When blackhole is used on top of classful qdisc like hfsc it breaks
qlen and backlog counters because packets are disappear without notice.
In HFSC non-zero qlen while all classes are inactive triggers warning:
WARNING: ... at net/sched/sch_hfsc.c:1393 hfsc_dequeue+0xba4/0xe90 [sch_hfsc]
and schedules watchdog work endlessly.
This patch return __NET_XMIT_BYPASS in addition to NET_XMIT_SUCCESS,
this flag tells upper layer: this packet is gone and isn't queued.
Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
