Commit 91a178258a ("netfilter: rpfilter: Convert
rpfilter_lookup_reverse to new dev helper") removed the need for the
'ret' variable. This went unnoticed because of the __maybe_unused
annotation.
Signed-off-by: Guillaume Nault <gnault@redhat.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
When the SPDX-License-Identifier tag has been added, the corresponding
license text has not been removed.
Remove it now.
Also, in xt_connmark.h, move the copyright text at the top of the file
which is a much more common pattern.
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Florian Westphal <fw@strlen.de>
The previous commit changed the way the rescheduling delay is computed
which has a side effect: the bias is now represented as much as the
other entries in the rescheduling delay which makes the logic to kick in
only with very large sets, as the initial interval is very large
(INT_MAX).
Revisit the GC initial bias to allow more frequent GC for smaller sets
while still avoiding wakeups when a machine is mostly idle. We're moving
from a large initial value to pretending we have 100 entries expiring at
the upper bound. This way only a few entries having a small timeout
won't impact much the rescheduling delay and non-idle machines will have
enough entries to lower the delay when needed. This also improves
readability as the initial bias is now linked to what is computed
instead of being an arbitrary large value.
Fixes: 2cfadb761d ("netfilter: conntrack: revisit gc autotuning")
Suggested-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Antoine Tenart <atenart@kernel.org>
Signed-off-by: Florian Westphal <fw@strlen.de>
Commit 2cfadb761d ("netfilter: conntrack: revisit gc autotuning")
changed the eviction rescheduling to the use average expiry of scanned
entries (within 1-60s) by doing:
for (...) {
expires = clamp(nf_ct_expires(tmp), ...);
next_run += expires;
next_run /= 2;
}
The issue is the above will make the average ('next_run' here) more
dependent on the last expiration values than the firsts (for sets > 2).
Depending on the expiration values used to compute the average, the
result can be quite different than what's expected. To fix this we can
do the following:
for (...) {
expires = clamp(nf_ct_expires(tmp), ...);
next_run += (expires - next_run) / ++count;
}
Fixes: 2cfadb761d ("netfilter: conntrack: revisit gc autotuning")
Cc: Florian Westphal <fw@strlen.de>
Signed-off-by: Antoine Tenart <atenart@kernel.org>
Signed-off-by: Florian Westphal <fw@strlen.de>
- Delete the repeated word 'to' in the comment.
- Add the missing 'use' word within the sentence.
- Correct spelling on 'malformation', 'needs'.
Signed-off-by: Ruffalo Lavoisier <RuffaloLavoisier@gmail.com>
Reviewed-by: Randy Dunlap <rdunlap@infradead.org>
Link: https://lore.kernel.org/r/20220919053447.5702-1-RuffaloLavoisier@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
If CONFIG_DCB is not set,
make ARCH=x86_64 CROSS_COMPILE=x86_64-linux-gnu-,
will be failed, like this:
drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c: In function ‘otx2_select_queue’:
drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c:1886:19: error: unused variable ‘pf’ [-Werror=unused-variable]
struct otx2_nic *pf = netdev_priv(netdev);
^~
cc1: all warnings being treated as errors
To fix this build error, put the definition of *pf under the CONFIG_DCB.
Fixes: 99c969a83d ("octeontx2-pf: Add egress PFC support")
Signed-off-by: Ren Zhijie <renzhijie2@huawei.com>
Link: https://lore.kernel.org/r/20220919025840.256411-1-renzhijie2@huawei.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
E810 products can support low latency Tx timestamp register read.
This requires usage of threaded IRQ instead of kthread to reduce the
kthread start latency (spikes up to 20 ms).
Add a check for the device capability and use the new method if
supported.
Signed-off-by: Karol Kolacinski <karol.kolacinski@intel.com>
Tested-by: Gurucharan <gurucharanx.g@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Link: https://lore.kernel.org/r/20220916201728.241510-1-anthony.l.nguyen@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Fabio Porcedda says:
====================
Add a secondary AT port to the Telit FN990
In order to add a secondary AT port to the Telit FN990 first add "DUN2"
to mhi_wwan_ctrl.c, after that add a seconday AT port to the
Telit FN990 in pci_generic.c
====================
Link: https://lore.kernel.org/r/20220916144329.243368-1-fabio.porcedda@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add a secondary AT port using one of OEM reserved channel.
Signed-off-by: Fabio Porcedda <fabio.porcedda@gmail.com>
Reviewed-by: Loic Poulain <loic.poulain@linaro.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
In order to have a secondary AT port add "DUN2".
Signed-off-by: Fabio Porcedda <fabio.porcedda@gmail.com>
Reviewed-by: Loic Poulain <loic.poulain@linaro.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Zhengchao Shao says:
====================
refactor duplicate codes in the tc cls walk function
The walk implementation of most tc cls modules is basically the same.
That is, the values of count and skip are checked first. If count is
greater than or equal to skip, the registered fn function is executed.
Otherwise, increase the value of count. So the code can be refactored.
Then use helper function to replace the code of each cls module in
alphabetical order.
The walk function is invoked during dump. Therefore, test cases related
to the tdc filter need to be added.
====================
Link: https://lore.kernel.org/r/20220916020251.190097-1-shaozhengchao@huawei.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Test 0811: Add multiple basic filter with cmp ematch u8/link layer and
default action and dump them
Test 5129: List basic filters
Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Reviewed-by: Victor Nogueira <victor@mojatatu.com>
Tested-by: Victor Nogueira <victor@mojatatu.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Test 8293: Add tcindex filter with default action
Test 7281: Add tcindex filter with hash size and pass action
Test b294: Add tcindex filter with mask shift and reclassify action
Test 0532: Add tcindex filter with pass_on and continue actions
Test d473: Add tcindex filter with pipe action
Test 2940: Add tcindex filter with miltiple actions
Test 1893: List tcindex filters
Test 2041: Change tcindex filter with pass action
Test 9203: Replace tcindex filter with pass action
Test 7957: Delete tcindex filter with drop action
Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Reviewed-by: Victor Nogueira <victor@mojatatu.com>
Tested-by: Victor Nogueira <victor@mojatatu.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Test 2141: Add rsvp filter with tcp proto and specific IP address
Test 5267: Add rsvp filter with udp proto and specific IP address
Test 2819: Add rsvp filter with src ip and src port
Test c967: Add rsvp filter with tunnelid and continue action
Test 5463: Add rsvp filter with tunnel and pipe action
Test 2332: Add rsvp filter with miltiple actions
Test 8879: Add rsvp filter with tunnel and skp flag
Test 8261: List rsvp filters
Test 8989: Delete rsvp filter
Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Reviewed-by: Victor Nogueira <victor@mojatatu.com>
Tested-by: Victor Nogueira <victor@mojatatu.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Test e122: Add route filter with from and to tag
Test 6573: Add route filter with fromif and to tag
Test 1362: Add route filter with to flag and reclassify action
Test 4720: Add route filter with from flag and continue actions
Test 2812: Add route filter with form tag and pipe action
Test 7994: Add route filter with miltiple actions
Test 4312: List route filters
Test 2634: Delete route filter with pipe action
Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Reviewed-by: Victor Nogueira <victor@mojatatu.com>
Tested-by: Victor Nogueira <victor@mojatatu.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Test 5294: Add flow filter with map key and ops
Test 3514: Add flow filter with map key or ops
Test 7534: Add flow filter with map key xor ops
Test 4524: Add flow filter with map key rshift ops
Test 0230: Add flow filter with map key addend ops
Test 2344: Add flow filter with src map key
Test 9304: Add flow filter with proto map key
Test 9038: Add flow filter with proto-src map key
Test 2a03: Add flow filter with proto-dst map key
Test a073: Add flow filter with iif map key
Test 3b20: Add flow filter with priority map key
Test 8945: Add flow filter with mark map key
Test c034: Add flow filter with nfct map key
Test 0205: Add flow filter with nfct-src map key
Test 5315: Add flow filter with nfct-src map key
Test 7849: Add flow filter with nfct-proto-src map key
Test 9902: Add flow filter with nfct-proto-dst map key
Test 6742: Add flow filter with rt-classid map key
Test 5432: Add flow filter with sk-uid map key
Test 4134: Add flow filter with sk-gid map key
Test 4522: Add flow filter with vlan-tag map key
Test 4253: Add flow filter with rxhash map key
Test 4452: Add flow filter with hash key list
Test 4341: Add flow filter with muliple ops
Test 4392: List flow filters
Test 4322: Change flow filter with map key num
Test 2320: Replace flow filter with map key num
Test 3213: Delete flow filter with map key num
Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Reviewed-by: Victor Nogueira <victor@mojatatu.com>
Tested-by: Victor Nogueira <victor@mojatatu.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Test 6273: Add cgroup filter with cmp ematch u8/link layer and drop action
Test 4721: Add cgroup filter with cmp ematch u8/link layer with trans
flag and pass action
Test d392: Add cgroup filter with cmp ematch u16/link layer and pipe action
Test 0234: Add cgroup filter with cmp ematch u32/link layer and miltiple
actions
Test 8499: Add cgroup filter with cmp ematch u8/network layer and pass
action
Test b273: Add cgroup filter with cmp ematch u8/network layer with trans
flag and drop action
Test 1934: Add cgroup filter with cmp ematch u16/network layer and pipe
action
Test 2733: Add cgroup filter with cmp ematch u32/network layer and
miltiple actions
Test 3271: Add cgroup filter with NOT cmp ematch rule and pass action
Test 2362: Add cgroup filter with two ANDed cmp ematch rules and single
action
Test 9993: Add cgroup filter with two ORed cmp ematch rules and single
action
Test 2331: Add cgroup filter with two ANDed cmp ematch rules and one ORed
ematch rule and single action
Test 3645: Add cgroup filter with two ANDed cmp ematch rules and one NOT
ORed ematch rule and single action
Test b124: Add cgroup filter with u32 ematch u8/zero offset and drop
action
Test 7381: Add cgroup filter with u32 ematch u8/zero offset and invalid
value >0xFF
Test 2231: Add cgroup filter with u32 ematch u8/positive offset and drop
action
Test 1882: Add cgroup filter with u32 ematch u8/invalid mask >0xFF
Test 1237: Add cgroup filter with u32 ematch u8/missing offset
Test 3812: Add cgroup filter with u32 ematch u8/missing AT keyword
Test 1112: Add cgroup filter with u32 ematch u8/missing value
Test 3241: Add cgroup filter with u32 ematch u8/non-numeric value
Test e231: Add cgroup filter with u32 ematch u8/non-numeric mask
Test 4652: Add cgroup filter with u32 ematch u8/negative offset and pass
Test 1331: Add cgroup filter with u32 ematch u16/zero offset and pipe
action
Test e354: Add cgroup filter with u32 ematch u16/zero offset and invalid
value >0xFFFF
Test 3538: Add cgroup filter with u32 ematch u16/positive offset and drop
action
Test 4576: Add cgroup filter with u32 ematch u16/invalid mask >0xFFFF
Test b842: Add cgroup filter with u32 ematch u16/missing offset
Test c924: Add cgroup filter with u32 ematch u16/missing AT keyword
Test cc93: Add cgroup filter with u32 ematch u16/missing value
Test 123c: Add cgroup filter with u32 ematch u16/non-numeric value
Test 3675: Add cgroup filter with u32 ematch u16/non-numeric mask
Test 1123: Add cgroup filter with u32 ematch u16/negative offset and drop
action
Test 4234: Add cgroup filter with u32 ematch u16/nexthdr+ offset and pass
action
Test e912: Add cgroup filter with u32 ematch u32/zero offset and pipe
action
Test 1435: Add cgroup filter with u32 ematch u32/positive offset and drop
action
Test 1282: Add cgroup filter with u32 ematch u32/missing offset
Test 6456: Add cgroup filter with u32 ematch u32/missing AT keyword
Test 4231: Add cgroup filter with u32 ematch u32/missing value
Test 2131: Add cgroup filter with u32 ematch u32/non-numeric value
Test f125: Add cgroup filter with u32 ematch u32/non-numeric mask
Test 4316: Add cgroup filter with u32 ematch u32/negative offset and drop
action
Test 23ae: Add cgroup filter with u32 ematch u32/nexthdr+ offset and pipe
action
Test 23a1: Add cgroup filter with canid ematch and single SFF
Test 324f: Add cgroup filter with canid ematch and single SFF with mask
Test 2576: Add cgroup filter with canid ematch and multiple SFF
Test 4839: Add cgroup filter with canid ematch and multiple SFF with masks
Test 6713: Add cgroup filter with canid ematch and single EFF
Test ab9d: Add cgroup filter with canid ematch and multiple EFF with masks
Test 5349: Add cgroup filter with canid ematch and a combination of
SFF/EFF
Test c934: Add cgroup filter with canid ematch and a combination of
SFF/EFF with masks
Test 4319: Replace cgroup filter with diffferent match
Test 4636: Delete cgroup filter
Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Reviewed-by: Victor Nogueira <victor@mojatatu.com>
Tested-by: Victor Nogueira <victor@mojatatu.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Test 23c3: Add cBPF filter with valid bytecode
Test 1563: Add cBPF filter with invalid bytecode
Test 2334: Add eBPF filter with valid object-file
Test 2373: Add eBPF filter with invalid object-file
Test 4423: Replace cBPF bytecode
Test 5122: Delete cBPF filter
Test e0a9: List cBPF filters
Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Reviewed-by: Victor Nogueira <victor@mojatatu.com>
Tested-by: Victor Nogueira <victor@mojatatu.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
use tc_cls_stats_dump() in filter.
Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Reviewed-by: Victor Nogueira <victor@mojatatu.com>
Tested-by: Victor Nogueira <victor@mojatatu.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The walk implementation of most tc cls modules is basically the same.
That is, the values of count and skip are checked first. If count is
greater than or equal to skip, the registered fn function is executed.
Otherwise, increase the value of count. So we can reconstruct them.
Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Reviewed-by: Victor Nogueira <victor@mojatatu.com>
Tested-by: Victor Nogueira <victor@mojatatu.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reading MAC from OF may return -EPROBE_DEFER if underlaying NVMEM device
isn't ready yet. In such case pass that error code up and "wait" to be
probed later.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Link: https://lore.kernel.org/r/20220915133013.2243-1-zajec5@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
It makes little sense to ask if networking namespace or net device refcount
tracking shall be enabled for debug kernel builds without network support.
This is similar to the commit eb0b39efb7 ("net: CONFIG_DEBUG_NET depends
on CONFIG_NET").
Signed-off-by: Lukas Bulwahn <lukas.bulwahn@gmail.com>
Link: https://lore.kernel.org/r/20220915124256.32512-1-lukas.bulwahn@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Vladimir Oltean says:
====================
Small tc-taprio improvements
This series contains:
- the proper protected variant of rcu_dereference() of admin and oper
schedules for accesses from the slow path
- a removal of an extra function pointer indirection for
qdisc->dequeue() and qdisc->peek()
- a removal of WARN_ON_ONCE() checks that can never trigger
- the addition of netlink extack messages to some qdisc->init() failures
These were split from an earlier patch set, hence the v2.
====================
Link: https://lore.kernel.org/r/20220915105046.2404072-1-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The WARN_ON_ONCE() checks introduced in commit 13511704f8 ("net:
taprio offload: enforce qdisc to netdev queue mapping") take a small
toll on performance, but otherwise, the conditions are never expected to
happen. Replace them with comments, such that the information is still
conveyed to developers.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Stop contributing to the proverbial user unfriendliness of tc, and tell
the user what is wrong wherever possible.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Since commit 13511704f8 ("net: taprio offload: enforce qdisc to netdev
queue mapping"), taprio_dequeue_soft() and taprio_peek_soft() are de
facto the only implementations for Qdisc_ops :: dequeue and Qdisc_ops ::
peek that taprio provides.
This is because in full offload mode, __dev_queue_xmit() will select a
txq->qdisc which is never root taprio qdisc. So if nothing is enqueued
in the root qdisc, it will never be run and nothing will get dequeued
from it.
Therefore, we can remove the private indirection from taprio, and always
point Qdisc_ops :: dequeue to taprio_dequeue_soft (now simply named
taprio_dequeue) and Qdisc_ops :: peek to taprio_peek_soft (now simply
named taprio_peek).
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Since commit 13511704f8 ("net: taprio offload: enforce qdisc to netdev
queue mapping"), __dev_queue_xmit() will select a txq->qdisc for the
full offload case of taprio which isn't the root taprio qdisc, so
qdisc enqueues will never pass through taprio_enqueue().
That commit already introduced one safety precaution check for
FULL_OFFLOAD_IS_ENABLED(); a second one is really not needed, so
simplify the conditional for entering into the GSO segmentation logic.
Also reword the comment a little, to appear more natural after the code
change.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Sparse complains that taprio_destroy() dereferences q->oper_sched and
q->admin_sched without rcu_dereference(), since they are marked as __rcu
in the taprio private structure.
1671:28: warning: incorrect type in argument 1 (different address spaces)
1671:28: expected struct callback_head *head
1671:28: got struct callback_head [noderef] __rcu *
1674:28: warning: incorrect type in argument 1 (different address spaces)
1674:28: expected struct callback_head *head
1674:28: got struct callback_head [noderef] __rcu *
To silence that build warning, do actually use rtnl_dereference(), since
we know the rtnl_mutex is held at the time of q->destroy().
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Since the writer-side lock is taken here, we do not need to open an RCU
read-side critical section, instead we can use rtnl_dereference() to
tell lockdep we are serialized with concurrent writes.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The locking in taprio_offload_config_changed() is wrong (but also
inconsequentially so). The current_entry_lock does not serialize changes
to the admin and oper schedules, only to the current entry. In fact, the
rtnl_mutex does that, and that is taken at the time when taprio_change()
is called.
Replace the rcu_dereference_protected() method with the proper RCU
annotation, and drop the unnecessary spin lock.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Simon Horman says:
====================
nfp: flower: police validation and ct enhancements
this series enhances the flower hardware offload
facility provided by the nfp driver.
1. Add validation of police actions created independently of flows
2. Add support offload of ct NAT action
3. Support offload of rule which has both vlan push/pop/mangle
and ct action
====================
Link: https://lore.kernel.org/r/20220914160604.1740282-1-simon.horman@corigine.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Support hardware offload of rule which has both vlan push/pop/mangle
and ct action.
Signed-off-by: Hui Zhou <hui.zhou@corigine.com>
Reviewed-by: Louis Peens <louis.peens@corigine.com>
Signed-off-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
support ct nat action when pre_ct merge with post_ct
and nft. at the same time, add the extra checksum action
and hardware stats for nft to meet the action check when
do nat.
Signed-off-by: Hui Zhou <hui.zhou@corigine.com>
Reviewed-by: Louis Peens <louis.peens@corigine.com>
Signed-off-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Validation of police actions was added to offload drivers in
commit d97b4b105c ("flow_offload: reject offload for all drivers with
invalid police parameters")
This patch extends that validation in the nfp driver to include
police actions which are created independently of flows.
Signed-off-by: Ziyang Chen <ziyang.chen@corigine.com>
Reviewed-by: Baowen Zheng <baowen.zheng@corigine.com>
Reviewed-by: Louis Peens <louis.peens@corigine.com>
Signed-off-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Kuniyuki Iwashima says:
====================
tcp: Introduce optional per-netns ehash.
The more sockets we have in the hash table, the longer we spend looking
up the socket. While running a number of small workloads on the same
host, they penalise each other and cause performance degradation.
The root cause might be a single workload that consumes much more
resources than the others. It often happens on a cloud service where
different workloads share the same computing resource.
On EC2 c5.24xlarge instance (196 GiB memory and 524288 (1Mi / 2) ehash
entries), after running iperf3 in different netns, creating 24Mi sockets
without data transfer in the root netns causes about 10% performance
regression for the iperf3's connection.
thash_entries sockets length Gbps
524288 1 1 50.7
24Mi 48 45.1
It is basically related to the length of the list of each hash bucket.
For testing purposes to see how performance drops along the length,
I set 131072 (1Mi / 8) to thash_entries, and here's the result.
thash_entries sockets length Gbps
131072 1 1 50.7
1Mi 8 49.9
2Mi 16 48.9
4Mi 32 47.3
8Mi 64 44.6
16Mi 128 40.6
24Mi 192 36.3
32Mi 256 32.5
40Mi 320 27.0
48Mi 384 25.0
To resolve the socket lookup degradation, we introduce an optional
per-netns hash table for TCP, but it's just ehash, and we still share
the global bhash, bhash2 and lhash2.
With a smaller ehash, we can look up non-listener sockets faster and
isolate such noisy neighbours. Also, we can reduce lock contention.
For details, please see the last patch.
patch 1 - 4: prep for per-netns ehash
patch 5: small optimisation for netns dismantle without TIME_WAIT sockets
patch 6: add per-netns ehash
Many thanks to Eric Dumazet for reviewing and advising.
====================
Link: https://lore.kernel.org/r/20220908011022.45342-1-kuniyu@amazon.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The more sockets we have in the hash table, the longer we spend looking
up the socket. While running a number of small workloads on the same
host, they penalise each other and cause performance degradation.
The root cause might be a single workload that consumes much more
resources than the others. It often happens on a cloud service where
different workloads share the same computing resource.
On EC2 c5.24xlarge instance (196 GiB memory and 524288 (1Mi / 2) ehash
entries), after running iperf3 in different netns, creating 24Mi sockets
without data transfer in the root netns causes about 10% performance
regression for the iperf3's connection.
thash_entries sockets length Gbps
524288 1 1 50.7
24Mi 48 45.1
It is basically related to the length of the list of each hash bucket.
For testing purposes to see how performance drops along the length,
I set 131072 (1Mi / 8) to thash_entries, and here's the result.
thash_entries sockets length Gbps
131072 1 1 50.7
1Mi 8 49.9
2Mi 16 48.9
4Mi 32 47.3
8Mi 64 44.6
16Mi 128 40.6
24Mi 192 36.3
32Mi 256 32.5
40Mi 320 27.0
48Mi 384 25.0
To resolve the socket lookup degradation, we introduce an optional
per-netns hash table for TCP, but it's just ehash, and we still share
the global bhash, bhash2 and lhash2.
With a smaller ehash, we can look up non-listener sockets faster and
isolate such noisy neighbours. In addition, we can reduce lock contention.
We can control the ehash size by a new sysctl knob. However, depending
on workloads, it will require very sensitive tuning, so we disable the
feature by default (net.ipv4.tcp_child_ehash_entries == 0). Moreover,
we can fall back to using the global ehash in case we fail to allocate
enough memory for a new ehash. The maximum size is 16Mi, which is large
enough that even if we have 48Mi sockets, the average list length is 3,
and regression would be less than 1%.
We can check the current ehash size by another read-only sysctl knob,
net.ipv4.tcp_ehash_entries. A negative value means the netns shares
the global ehash (per-netns ehash is disabled or failed to allocate
memory).
# dmesg | cut -d ' ' -f 5- | grep "established hash"
TCP established hash table entries: 524288 (order: 10, 4194304 bytes, vmalloc hugepage)
# sysctl net.ipv4.tcp_ehash_entries
net.ipv4.tcp_ehash_entries = 524288 # can be changed by thash_entries
# sysctl net.ipv4.tcp_child_ehash_entries
net.ipv4.tcp_child_ehash_entries = 0 # disabled by default
# ip netns add test1
# ip netns exec test1 sysctl net.ipv4.tcp_ehash_entries
net.ipv4.tcp_ehash_entries = -524288 # share the global ehash
# sysctl -w net.ipv4.tcp_child_ehash_entries=100
net.ipv4.tcp_child_ehash_entries = 100
# ip netns add test2
# ip netns exec test2 sysctl net.ipv4.tcp_ehash_entries
net.ipv4.tcp_ehash_entries = 128 # own a per-netns ehash with 2^n buckets
When more than two processes in the same netns create per-netns ehash
concurrently with different sizes, we need to guarantee the size in
one of the following ways:
1) Share the global ehash and create per-netns ehash
First, unshare() with tcp_child_ehash_entries==0. It creates dedicated
netns sysctl knobs where we can safely change tcp_child_ehash_entries
and clone()/unshare() to create a per-netns ehash.
2) Control write on sysctl by BPF
We can use BPF_PROG_TYPE_CGROUP_SYSCTL to allow/deny read/write on
sysctl knobs.
Note that the global ehash allocated at the boot time is spread over
available NUMA nodes, but inet_pernet_hashinfo_alloc() will allocate
pages for each per-netns ehash depending on the current process's NUMA
policy. By default, the allocation is done in the local node only, so
the per-netns hash table could fully reside on a random node. Thus,
depending on the NUMA policy the netns is created with and the CPU the
current thread is running on, we could see some performance differences
for highly optimised networking applications.
Note also that the default values of two sysctl knobs depend on the ehash
size and should be tuned carefully:
tcp_max_tw_buckets : tcp_child_ehash_entries / 2
tcp_max_syn_backlog : max(128, tcp_child_ehash_entries / 128)
As a bonus, we can dismantle netns faster. Currently, while destroying
netns, we call inet_twsk_purge(), which walks through the global ehash.
It can be potentially big because it can have many sockets other than
TIME_WAIT in all netns. Splitting ehash changes that situation, where
it's only necessary for inet_twsk_purge() to clean up TIME_WAIT sockets
in each netns.
With regard to this, we do not free the per-netns ehash in inet_twsk_kill()
to avoid UAF while iterating the per-netns ehash in inet_twsk_purge().
Instead, we do it in tcp_sk_exit_batch() after calling tcp_twsk_purge() to
keep it protocol-family-independent.
In the future, we could optimise ehash lookup/iteration further by removing
netns comparison for the per-netns ehash.
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
While destroying netns, we call inet_twsk_purge() in tcp_sk_exit_batch()
and tcpv6_net_exit_batch() for AF_INET and AF_INET6. These commands
trigger the kernel to walk through the potentially big ehash twice even
though the netns has no TIME_WAIT sockets.
# ip netns add test
# ip netns del test
or
# unshare -n /bin/true >/dev/null
When tw_refcount is 1, we need not call inet_twsk_purge() at least
for the net. We can save such unneeded iterations if all netns in
net_exit_list have no TIME_WAIT sockets. This change eliminates
the tax by the additional unshare() described in the next patch to
guarantee the per-netns ehash size.
Tested:
# mount -t debugfs none /sys/kernel/debug/
# echo cleanup_net > /sys/kernel/debug/tracing/set_ftrace_filter
# echo inet_twsk_purge >> /sys/kernel/debug/tracing/set_ftrace_filter
# echo function > /sys/kernel/debug/tracing/current_tracer
# cat ./add_del_unshare.sh
for i in `seq 1 40`
do
(for j in `seq 1 100` ; do unshare -n /bin/true >/dev/null ; done) &
done
wait;
# ./add_del_unshare.sh
Before the patch:
# cat /sys/kernel/debug/tracing/trace_pipe
kworker/u128:0-8 [031] ...1. 174.162765: cleanup_net <-process_one_work
kworker/u128:0-8 [031] ...1. 174.240796: inet_twsk_purge <-cleanup_net
kworker/u128:0-8 [032] ...1. 174.244759: inet_twsk_purge <-tcp_sk_exit_batch
kworker/u128:0-8 [034] ...1. 174.290861: cleanup_net <-process_one_work
kworker/u128:0-8 [039] ...1. 175.245027: inet_twsk_purge <-cleanup_net
kworker/u128:0-8 [046] ...1. 175.290541: inet_twsk_purge <-tcp_sk_exit_batch
kworker/u128:0-8 [037] ...1. 175.321046: cleanup_net <-process_one_work
kworker/u128:0-8 [024] ...1. 175.941633: inet_twsk_purge <-cleanup_net
kworker/u128:0-8 [025] ...1. 176.242539: inet_twsk_purge <-tcp_sk_exit_batch
After:
# cat /sys/kernel/debug/tracing/trace_pipe
kworker/u128:0-8 [038] ...1. 428.116174: cleanup_net <-process_one_work
kworker/u128:0-8 [038] ...1. 428.262532: cleanup_net <-process_one_work
kworker/u128:0-8 [030] ...1. 429.292645: cleanup_net <-process_one_work
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
We will soon introduce an optional per-netns ehash.
This means we cannot use tcp_hashinfo directly in most places.
Instead, access it via net->ipv4.tcp_death_row.hashinfo.
The access will be valid only while initialising tcp_hashinfo
itself and creating/destroying each netns.
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
We will soon introduce an optional per-netns ehash.
This means we cannot use the global sk->sk_prot->h.hashinfo
to fetch a TCP hashinfo.
Instead, set NULL to sk->sk_prot->h.hashinfo for TCP and get
a proper hashinfo from net->ipv4.tcp_death_row.hashinfo.
Note that we need not use sk->sk_prot->h.hashinfo if DCCP is
disabled.
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
We will soon introduce an optional per-netns ehash and access hash
tables via net->ipv4.tcp_death_row->hashinfo instead of &tcp_hashinfo
in most places.
It could harm the fast path because dereferences of two fields in net
and tcp_death_row might incur two extra cache line misses. To save one
dereference, let's place tcp_death_row back in netns_ipv4 and fetch
hashinfo via net->ipv4.tcp_death_row"."hashinfo.
Note tcp_death_row was initially placed in netns_ipv4, and commit
fbb8295248 ("tcp: allocate tcp_death_row outside of struct netns_ipv4")
changed it to a pointer so that we can fire TIME_WAIT timers after freeing
net. However, we don't do so after commit 04c494e68a ("Revert "tcp/dccp:
get rid of inet_twsk_purge()""), so we need not define tcp_death_row as a
pointer.
Also, we move refcount_dec_and_test(&tw_refcount) from tcp_sk_exit() to
tcp_sk_exit_batch() as a debug check.
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This patch adds no functional change and cleans up some functions
that the following patches touch around so that we make them tidy
and easy to review/revert. The changes are
- Keep reverse christmas tree order
- Remove unnecessary init of port in inet_csk_find_open_port()
- Use req_to_sk() once in reqsk_queue_unlink()
- Use sock_net(sk) once in tcp_time_wait() and tcp_v[46]_connect()
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This is a regression test for commit 592335a416 ("bonding: accept
unsolicited NA message") and commit b7f14132bf ("bonding: use unspecified
address if no available link local address"). When the bond interface
up and no available link local address, unspecified address(::) is used to
send the NS message. The unsolicited NA message should also be accepted
for validation.
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Acked-by: Jonathan Toppins <jtoppins@redhat.com>
Link: https://lore.kernel.org/r/20220920033047.173244-1-liuhangbin@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
dev_err() can be replace with dev_err_probe() which will check if error
code is -EPROBE_DEFER.
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20220915065043.665138-3-yangyingliang@huawei.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
dev_err() can be replace with dev_err_probe() which will check if error
code is -EPROBE_DEFER.
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20220915065043.665138-2-yangyingliang@huawei.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
dev_err() can be replace with dev_err_probe() which will check if error
code is -EPROBE_DEFER.
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20220915065043.665138-1-yangyingliang@huawei.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
EMAC IP found on RZ/G2L Gb ethernet supports MII interface.
This patch adds support for selecting MII interface mode.
Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Link: https://lore.kernel.org/r/20220914192604.265859-1-biju.das.jz@bp.renesas.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Unlike with bridges, one can't add an interface to a bond and set it up
at the same time:
| # ip link set dummy0 down
| # ip link set dummy0 master bond0 up
| Error: Device can not be enslaved while up.
Of all drivers with ndo_add_slave callback, bond and team decline if
IFF_UP flag is set, vrf cycles the interface (i.e., sets it down and
immediately up again) and the others just don't care.
Support the common notion of setting the interface up after enslaving it
by sorting the operations accordingly.
Signed-off-by: Phil Sutter <phil@nwl.cc>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20220914150623.24152-1-phil@nwl.cc
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Radhey Shyam Pandey says:
====================
macb: add zynqmp SGMII dynamic configuration support
This patchset add firmware and driver support to do SD/GEM dynamic
configuration. In traditional flow GEM secure space configuration
is done by FSBL. However in specific usescases like dynamic designs
where GEM is not enabled in base vivado design, FSBL skips GEM
initialization and we need a mechanism to configure GEM secure space
in linux space at runtime.
====================
Link: https://lore.kernel.org/r/1663158796-14869-1-git-send-email-radhey.shyam.pandey@amd.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add support for the dynamic configuration which takes care of
configuring the GEM secure space configuration registers
using EEMI APIs.
High level sequence is to:
- Check for the PM dynamic configuration support, if no error proceed with
GEM dynamic configurations(next steps) otherwise skip the dynamic
configuration.
- Configure GEM Fixed configurations.
- Configure GEM_CLK_CTRL (gemX_sgmii_mode).
- Trigger GEM reset.
Signed-off-by: Radhey Shyam Pandey <radhey.shyam.pandey@amd.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Tested-by: Conor Dooley <conor.dooley@microchip.com> (for MPFS)
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
