These groups will contain socket-destruction events for
AF_INET/AF_INET6, IPPROTO_TCP/IPPROTO_UDP.
Near the end of socket destruction, a check for listeners is
performed. In the presence of a listener, rather than completely
cleanup the socket, a unit of work will be added to a private
work queue which will first broadcast information about the socket
and then finish the cleanup operation.
Signed-off-by: Craig Gallek <kraig@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Or Gerlitz says:
====================
mlx4 driver update (+ new VF ndo)
This series from Eran and Hadar is further dealing with traffic
counters in the mlx4 driver, this time mostly around SRIOV.
We added a new ndo to read the VF counters through the PF netdev
netlink infrastructure plus mlx4 implementation for that ndo.
changes from V0:
- applied feedback from John to use nested netlink encoding
for the VF counters so we can extend it later
- add handling of single ported VFs in the mlx4_en driver new ndo
- avoid chopping the FW counters from 64 to 32 bits in mlx4_en PF flow
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Implement the ndo to gather VF statistics through the PF.
All counters related to this VF are stored in a per slave
list, run over the slave's list and collect all statistics.
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add ndo_get_vf_stats where the PF retrieves and fills the VFs traffic
statistics. We encode the VF stats in a nested manner to allow for
future extensions.
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Allow the user to observe the PF own statistics using ethtool with pf_
prefixed counter names.
Those counters are the PF statistics out of the overall port statistics.
Every PF QP is attached to a counter and the summary of those counters
is the PF statistics.
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is an infrastructure step for querying VF and PF counters.
This code was in the IB driver, move it to the mlx4 core driver
so it will be accessible for more use cases.
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As IB VFs are not capable to read the port counters through MADs,
move there to read their own QP counters to gather statistics.
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is an infrastructure step to attach all the QPs opened from the
IB driver to a counter in order to collect VF stats from the PF using
those counters.
If the port's type is Ethernet, the counter policy demands two counters
per port (one for RoCE and one for Ethernet). The port default counter
(allocated in mlx4_core) is used for the Ethernet netdev QPs and we
allocate another counter for RoCE.
If the port's traffic is Infiniband, the counter policy demands
one counter per port, so it can use the port's default counter.
Also, Add 'allocated' flag for each counter in order to clean it at
unload.
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Default counter per port will be allocated at the mlx4 core driver load.
Every QP opened by the Ethernet driver will be attached to the port's default
counter. This is an infrastructure step to collect VF statistics from the PF.
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Counter will get its port attribute within the resource tracker when
the first QP attached to it is modified to RTR. If a QP is counter-less,
an attempt to create a new counter with assigned port will be made.
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Each physical function has a guarantee of two counters per port, one
for a default counter and one for the IB driver.
Each virtual function has a guarantee of one counter per port.
All other counters are free and can be obtained on demand.
This is a preparation step for supporting a get_vf_stats ndo call,
so we can promise a counter for every VF in order to collect their
statistics from the PF context.
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since virtual functions get their counters indices allocation from the PF,
allocate counters indices bitmap only in case the function isn't virtual.
Also, check that the device has counters to allocate before creating the
indices bitmap table.
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Reserve the last valid counter index for "sink" counter, when a
new counter cannot be allocated, the driver will use this counter.
In order to avoid allocating this counter on any other flow, fix the
indices bitmap allocation range, and reserve the sink counter index.
Add macro for the sink counter index and replace all appearences of the
index with the macro.
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add resetting the counter data to the free counter flow, so the counter's
data won't be accessible anymore if querying the counter. Also, on next
counter allocation (to another VM for example), it will be fresh and clear.
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If counters are not supported by the device. The indices bitmap table is not
allocated during initialization. Add the symmetrical check before cleaning
the counters bitmap table or freeing a counter.
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We need to delete from offload the device externally learnded fdbs when any
one of these events happen:
1) Bridge ages out fdb. (When bridge is doing ageing vs. device doing
ageing. If device is doing ageing, it would send SWITCHDEV_FDB_DEL
directly).
2) STP state change flushes fdbs on port.
3) User uses sysfs interface to flush fdbs from bridge or bridge port:
echo 1 >/sys/class/net/BR_DEV/bridge/flush
echo 1 >/sys/class/net/BR_PORT/brport/flush
4) Offload driver send event SWITCHDEV_FDB_DEL to delete fdb entry.
For rocker, we can now get called to delete fdb entry in wait and nowait
contexts, so set NOWAIT flag when deleting fdb entry.
Signed-off-by: Scott Feldman <sfeldma@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is the NFC pull request for 4.2.
- NCI drivers can now define their own handlers for processing
proprietary NCI responses and notifications.
- NFC vendors can use a dedicated netlink API to send their own
proprietary commands, like e.g. all commands needed to implement
vendor specific manufacturing tools.
- A new generic NCI over UART driver against which any NCI chipset
running on top of a serial interface can register.
- The st21nfcb driver is renamed to st-nci as it can and will support
most of ST Microelectronics NCI chipsets.
- The st21nfcb driver can put its CLF in hibernate mode and save
significant amount of power.
- A few st21nfcb minor fixes.
- The NXP NCI driver now supports ACPI enumeration.
- The Marvell NCI driver now supports both USB and serial
physical interfaces.
- The Marvell NCI drivers also supports NCI frames being muxed
over HCI. This is a setting that can be defined by a DT property.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=+IG2
-----END PGP SIGNATURE-----
Merge tag 'nfc-next-4.2-1' of git://git.kernel.org/pub/scm/linux/kernel/git/sameo/nfc-next
Samuel Ortiz says:
====================
NFC 4.2 pull request
This is the NFC pull request for 4.2.
- NCI drivers can now define their own handlers for processing
proprietary NCI responses and notifications.
- NFC vendors can use a dedicated netlink API to send their own
proprietary commands, like e.g. all commands needed to implement
vendor specific manufacturing tools.
- A new generic NCI over UART driver against which any NCI chipset
running on top of a serial interface can register.
- The st21nfcb driver is renamed to st-nci as it can and will support
most of ST Microelectronics NCI chipsets.
- The st21nfcb driver can put its CLF in hibernate mode and save
significant amount of power.
- A few st21nfcb minor fixes.
- The NXP NCI driver now supports ACPI enumeration.
- The Marvell NCI driver now supports both USB and serial
physical interfaces.
- The Marvell NCI drivers also supports NCI frames being muxed
over HCI. This is a setting that can be defined by a DT property.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Nikolay Aleksandrov says:
====================
bonding: extend the 3ad exported attributes
These are two small patches that export actor_oper_port_state and
partner_oper_port_state via netlink and sysfs, until now they were only
exported via bond's proc entry. If this set gets accepted I have an iproute2
patch prepared that will export them with which I tested these changes.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Export the partner_oper_port_state of each port via sysfs and netlink.
In 802.3ad mode it is valuable for the user to be able to check the
partner_oper state, it is already exported via bond's proc entry.
Signed-off-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: Andy Gospodarek <gospo@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Export the actor_oper_port_state of each port via sysfs and netlink.
In 802.3ad mode it is valuable for the user to be able to check the
actor_oper state, it is already exported via bond's proc entry.
Signed-off-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: Andy Gospodarek <gospo@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Scott Feldman says:
====================
rocker: revert back to support for nowait processes
One of the items removed from the rocker driver in the Spring Cleanup patch
series was the ability to mark processing in the driver as "no wait" for
those contexts where we cannot sleep. Turns out, we have "no wait"
contexts where we want to program the device and we don't want to defer the
processing to a process context. So re-add the ROCKER_OP_FLAG_NOWAIT flag
to mark such processes, and propagate flags to mem allocator and to the
device cmd executor. With NOWAIT, mem allocs are GFP_ATOMIC and device
cmds are queued to the device, but the driver will not wait (sleep) for the
response back from the device.
My bad for removing NOWAIT support in the first place; I thought we could
swing non-sleep contexts to process context using a work queue, for
example, but there is push-back to keep processing in original context.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
rocker_port_stop can be called from atomic and non-atomic contexts. Since
we can't test what context we're getting called in, do the processing as
'no wait', which will cover all cases.
Signed-off-by: Scott Feldman <sfeldma@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We can get STP updates from the bridge driver in atomic and non-atomic
contexts. Since we can't test what context we're getting called in,
do the STP processing as 'no wait', which will cover all cases.
Signed-off-by: Scott Feldman <sfeldma@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Neigh update event handler runs in a context where we can't sleep, so mark
processing in driver with ROCKER_OP_FLAG_NOWAIT. NOWAIT will use
GFP_ATOMIC for allocations and will queue cmds to the device's cmd ring but
will not wait (sleep) for cmd response back from device.
Signed-off-by: Scott Feldman <sfeldma@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
One of the items removed from the rocker driver in the Spring Cleanup patch
series was the ability to mark processing in the driver as "no wait" for
those contexts where we cannot sleep. Turns out, we have "no wait"
contexts where we want to program the device. So re-add the
ROCKER_OP_FLAG_NOWAIT flag to mark such processes, and propagate flags to
mem allocator and to the device cmd executor. With NOWAIT, mem allocs are
GFP_ATOMIC and device cmds are queued to the device, but the driver will
not wait (sleep) for the response back from the device.
My bad for removing NOWAIT support in the first place; I thought we could
swing non-sleep contexts to process context using a work queue, for
example, but there is push-back to keep processing in original context.
Signed-off-by: Scott Feldman <sfeldma@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
rocker->neigh_tbl_next_index is used to generate unique indices for neigh
entries programmed into the device. The way new indices were generated was
racy with the new prepare-commit transaction model. A simple fix here
removes the race. The race was with two processes getting the same index,
one process using prepare-commit, the other not:
Proc A Proc B
PREPARE phase
get neigh_tbl_next_index
NONE phase
get neigh_tbl_next_index
neigh_tbl_next_index++
COMMIT phase
neigh_tbl_next_index++
Both A and B got the same index. The fix is to store and increment
neigh_tbl_next_index in the PREPARE (or NONE) phase and use value in COMMIT
phase:
Proc A Proc B
PREPARE phase
get neigh_tbl_next_index
neigh_tbl_next_index++
NONE phase
get neigh_tbl_next_index
neigh_tbl_next_index++
COMMIT phase
// use value stashed in PREPARE phase
Reported-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Scott Feldman <sfeldma@gmail.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The ports array is filled in as ports are probed, but if probing doesn't
finish, we need to stop only those ports that where probed successfully.
Check the ports array for NULL to skip un-probed ports when stopping.
Signed-off-by: Scott Feldman <sfeldma@gmail.com>
Acked-by: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: David S. Miller <davem@davemloft.net>
Using a function instead of a macro is cleaner and remove
following W=1 warnings (extract)
In file included from net/ipv6/ip6_vti.c:29:0:
net/ipv6/ip6_vti.c: In function ‘vti6_dev_init_gen’:
include/linux/netdevice.h:2029:18: warning: variable ‘stat’ set but not
used [-Wunused-but-set-variable]
typeof(type) *stat; \
^
net/ipv6/ip6_vti.c:862:16: note: in expansion of macro
‘netdev_alloc_pcpu_stats’
dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats);
^
CC [M] net/ipv6/sit.o
In file included from net/ipv6/sit.c:30:0:
net/ipv6/sit.c: In function ‘ipip6_tunnel_init’:
include/linux/netdevice.h:2029:18: warning: variable ‘stat’ set but not
used [-Wunused-but-set-variable]
typeof(type) *stat; \
^
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
v2:
Move struct switchdev_obj automatics to inner scope where there used.
v1:
To maintain backward compatibility with the existing iproute2 "bridge vlan"
command, let bridge's setlink/dellink handler call into either the port
driver's 8021q ndo ops or the port driver's bridge_setlink/dellink ops.
This allows port driver to choose 8021q ops or the newer
bridge_setlink/dellink ops when implementing VLAN add/del filtering on the
device. The iproute "bridge vlan" command does not need to be modified.
To summarize using the "bridge vlan" command examples, we have:
1) bridge vlan add|del vid VID dev DEV
Here iproute2 sets MASTER flag. Bridge's bridge_setlink/dellink is called.
Vlan is set on bridge for port. If port driver implements ndo 8021q ops,
call those to port driver can install vlan filter on device. Otherwise, if
port driver implements bridge_setlink/dellink ops, call those to install
vlan filter to device. This option only works if port is bridged.
2) bridge vlan add|del vid VID dev DEV master
Same as 1)
3) bridge vlan add|del vid VID dev DEV self
Bridge's bridge_setlink/dellink isn't called. Port driver's
bridge_setlink/dellink is called, if implemented. This option works if
port is bridged or not. If port is not bridged, a VLAN can still be
added/deleted to device filter using this variant.
4) bridge vlan add|del vid VID dev DEV master self
This is a combination of 1) and 3), but will only work if port is bridged.
Signed-off-by: Scott Feldman <sfeldma@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Alexei Starovoitov says:
====================
v1->v2: switched to init_user_ns from current_user_ns as suggested by Andy
Introduce new helpers to access 'struct task_struct'->pid, tgid, uid, gid, comm
fields in tracing and networking.
Share bpf_trace_printk() and bpf_get_smp_processor_id() helpers between
tracing and networking.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
It's useful to do per-cpu histograms.
Suggested-by: Daniel Wagner <daniel.wagner@bmw-carit.de>
Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
bpf_trace_printk() is a helper function used to debug eBPF programs.
Let socket and TC programs use it as well.
Note, it's DEBUG ONLY helper. If it's used in the program,
the kernel will print warning banner to make sure users don't use
it in production.
Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
eBPF programs attached to kprobes need to filter based on
current->pid, uid and other fields, so introduce helper functions:
u64 bpf_get_current_pid_tgid(void)
Return: current->tgid << 32 | current->pid
u64 bpf_get_current_uid_gid(void)
Return: current_gid << 32 | current_uid
bpf_get_current_comm(char *buf, int size_of_buf)
stores current->comm into buf
They can be used from the programs attached to TC as well to classify packets
based on current task fields.
Update tracex2 example to print histogram of write syscalls for each process
instead of aggregated for all.
Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pablo Neira Ayuso says:
====================
Netfilter updates for net-next
This a bit large (and late) patchset that contains Netfilter updates for
net-next. Most relevantly br_netfilter fixes, ipset RCU support, removal of
x_tables percpu ruleset copy and rework of the nf_tables netdev support. More
specifically, they are:
1) Warn the user when there is a better protocol conntracker available, from
Marcelo Ricardo Leitner.
2) Fix forwarding of IPv6 fragmented traffic in br_netfilter, from Bernhard
Thaler. This comes with several patches to prepare the change in first place.
3) Get rid of special mtu handling of PPPoE/VLAN frames for br_netfilter. This
is not needed anymore since now we use the largest fragment size to
refragment, from Florian Westphal.
4) Restore vlan tag when refragmenting in br_netfilter, also from Florian.
5) Get rid of the percpu ruleset copy in x_tables, from Florian. Plus another
follow up patch to refine it from Eric Dumazet.
6) Several ipset cleanups, fixes and finally RCU support, from Jozsef Kadlecsik.
7) Get rid of parens in Netfilter Kconfig files.
8) Attach the net_device to the basechain as opposed to the initial per table
approach in the nf_tables netdev family.
9) Subscribe to netdev events to detect the removal and registration of a
device that is referenced by a basechain.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
In case the net_device is gone, we have to unregister the hooks and put back
the reference on the net_device object. Once it comes back, register them
again. This also covers the device rename case.
This patch also adds a new flag to indicate that the basechain is disabled, so
their hooks are not registered. This flag is used by the netdev family to
handle the case where the net_device object is gone. Currently this flag is not
exposed to userspace.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
The device is part of the hook configuration, so instead of a global
configuration per table, set it to each of the basechain that we create.
This patch reworks ebddf1a8d7 ("netfilter: nf_tables: allow to bind table to
net_device").
Note that this adds a dev_name field in the nft_base_chain structure which is
required the netdev notification subscription that follows up in a patch to
handle gone net_devices.
Suggested-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
After Florian patches, there is no need for XT_TABLE_INFO_SZ anymore :
Only one copy of table is kept, instead of one copy per cpu.
We also can avoid a dereference if we put table data right after
xt_table_info. It reduces register pressure and helps compiler.
Then, we attempt a kmalloc() if total size is under order-3 allocation,
to reduce TLB pressure, as in many cases, rules fit in 32 KB.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Jozsef Kadlecsik says:
====================
ipset patches for nf-next
Please consider to apply the next bunch of patches for ipset. First
comes the small changes, then the bugfixes and at the end the RCU
related patches.
* Use MSEC_PER_SEC consistently instead of the number.
* Use SET_WITH_*() helpers to test set extensions from Sergey Popovich.
* Check extensions attributes before getting extensions from Sergey Popovich.
* Permit CIDR equal to the host address CIDR in IPv6 from Sergey Popovich.
* Make sure we always return line number on batch in the case of error
from Sergey Popovich.
* Check CIDR value only when attribute is given from Sergey Popovich.
* Fix cidr handling for hash:*net* types, reported by Jonathan Johnson.
* Fix parallel resizing and listing of the same set so that the original
set is kept for the whole dumping.
* Make sure listing doesn't grab a set which is just being destroyed.
* Remove rbtree from ip_set_hash_netiface.c in order to introduce RCU.
* Replace rwlock_t with spinlock_t in "struct ip_set", change the locking
in the core and simplifications in the timeout routines.
* Introduce RCU locking in bitmap:* types with a slight modification in the
logic on how an element is added.
* Introduce RCU locking in hash:* types. This is the most complex part of
the changes.
* Introduce RCU locking in list type where standard rculist is used.
* Fix coding styles reported by checkpatch.pl.
====================
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
According to the reporter, they are not needed.
Reported-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Three types of data need to be protected in the case of the hash types:
a. The hash buckets: standard rcu pointer operations are used.
b. The element blobs in the hash buckets are stored in an array and
a bitmap is used for book-keeping to tell which elements in the array
are used or free.
c. Networks per cidr values and the cidr values themselves are stored
in fix sized arrays and need no protection. The values are modified
in such an order that in the worst case an element testing is repeated
once with the same cidr value.
The ipset hash approach uses arrays instead of lists and therefore is
incompatible with rhashtable.
Performance is tested by Jesper Dangaard Brouer:
Simple drop in FORWARD
~~~~~~~~~~~~~~~~~~~~~~
Dropping via simple iptables net-mask match::
iptables -t raw -N simple || iptables -t raw -F simple
iptables -t raw -I simple -s 198.18.0.0/15 -j DROP
iptables -t raw -D PREROUTING -j simple
iptables -t raw -I PREROUTING -j simple
Drop performance in "raw": 11.3Mpps
Generator: sending 12.2Mpps (tx:12264083 pps)
Drop via original ipset in RAW table
~~~~~~~~~~~~~~~~~~~~~~~~~~~
Create a set with lots of elements::
sudo ./ipset destroy test
echo "create test hash:ip hashsize 65536" > test.set
for x in `seq 0 255`; do
for y in `seq 0 255`; do
echo "add test 198.18.$x.$y" >> test.set
done
done
sudo ./ipset restore < test.set
Dropping via ipset::
iptables -t raw -F
iptables -t raw -N net198 || iptables -t raw -F net198
iptables -t raw -I net198 -m set --match-set test src -j DROP
iptables -t raw -I PREROUTING -j net198
Drop performance in "raw" with ipset: 8Mpps
Perf report numbers ipset drop in "raw"::
+ 24.65% ksoftirqd/1 [ip_set] [k] ip_set_test
- 21.42% ksoftirqd/1 [kernel.kallsyms] [k] _raw_read_lock_bh
- _raw_read_lock_bh
+ 99.88% ip_set_test
- 19.42% ksoftirqd/1 [kernel.kallsyms] [k] _raw_read_unlock_bh
- _raw_read_unlock_bh
+ 99.72% ip_set_test
+ 4.31% ksoftirqd/1 [ip_set_hash_ip] [k] hash_ip4_kadt
+ 2.27% ksoftirqd/1 [ixgbe] [k] ixgbe_fetch_rx_buffer
+ 2.18% ksoftirqd/1 [ip_tables] [k] ipt_do_table
+ 1.81% ksoftirqd/1 [ip_set_hash_ip] [k] hash_ip4_test
+ 1.61% ksoftirqd/1 [kernel.kallsyms] [k] __netif_receive_skb_core
+ 1.44% ksoftirqd/1 [kernel.kallsyms] [k] build_skb
+ 1.42% ksoftirqd/1 [kernel.kallsyms] [k] ip_rcv
+ 1.36% ksoftirqd/1 [kernel.kallsyms] [k] __local_bh_enable_ip
+ 1.16% ksoftirqd/1 [kernel.kallsyms] [k] dev_gro_receive
+ 1.09% ksoftirqd/1 [kernel.kallsyms] [k] __rcu_read_unlock
+ 0.96% ksoftirqd/1 [ixgbe] [k] ixgbe_clean_rx_irq
+ 0.95% ksoftirqd/1 [kernel.kallsyms] [k] __netdev_alloc_frag
+ 0.88% ksoftirqd/1 [kernel.kallsyms] [k] kmem_cache_alloc
+ 0.87% ksoftirqd/1 [xt_set] [k] set_match_v3
+ 0.85% ksoftirqd/1 [kernel.kallsyms] [k] inet_gro_receive
+ 0.83% ksoftirqd/1 [kernel.kallsyms] [k] nf_iterate
+ 0.76% ksoftirqd/1 [kernel.kallsyms] [k] put_compound_page
+ 0.75% ksoftirqd/1 [kernel.kallsyms] [k] __rcu_read_lock
Drop via ipset in RAW table with RCU-locking
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
With RCU locking, the RW-lock is gone.
Drop performance in "raw" with ipset with RCU-locking: 11.3Mpps
Performance-tested-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
There's nothing much required because the bitmap types use atomic
bit operations. However the logic of adding elements slightly changed:
first the MAC address updated (which is not atomic), then the element
activated (added). The extensions may call kfree_rcu() therefore we
call rcu_barrier() at module removal.
Signed-off-by: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
Replace rwlock_t with spinlock_t in "struct ip_set" and change the locking
accordingly. Convert the comment extension into an rcu-avare object. Also,
simplify the timeout routines.
Signed-off-by: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
There was a small window when all sets are destroyed and a concurrent
listing of all sets could grab a set which is just being destroyed.
Signed-off-by: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
When elements added to a hash:* type of set and resizing triggered,
parallel listing could start to list the original set (before resizing)
and "continue" with listing the new set. Fix it by references and
using the original hash table for listing. Therefore the destroying of
the original hash table may happen from the resizing or listing functions.
Signed-off-by: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>