- Untangle the header spaghetti which causes build failures in various
situations caused by the lockdep additions to seqcount to validate that
the write side critical sections are non-preemptible.
- The seqcount associated lock debug addons which were blocked by the
above fallout.
seqcount writers contrary to seqlock writers must be externally
serialized, which usually happens via locking - except for strict per
CPU seqcounts. As the lock is not part of the seqcount, lockdep cannot
validate that the lock is held.
This new debug mechanism adds the concept of associated locks.
sequence count has now lock type variants and corresponding
initializers which take a pointer to the associated lock used for
writer serialization. If lockdep is enabled the pointer is stored and
write_seqcount_begin() has a lockdep assertion to validate that the
lock is held.
Aside of the type and the initializer no other code changes are
required at the seqcount usage sites. The rest of the seqcount API is
unchanged and determines the type at compile time with the help of
_Generic which is possible now that the minimal GCC version has been
moved up.
Adding this lockdep coverage unearthed a handful of seqcount bugs which
have been addressed already independent of this.
While generaly useful this comes with a Trojan Horse twist: On RT
kernels the write side critical section can become preemtible if the
writers are serialized by an associated lock, which leads to the well
known reader preempts writer livelock. RT prevents this by storing the
associated lock pointer independent of lockdep in the seqcount and
changing the reader side to block on the lock when a reader detects
that a writer is in the write side critical section.
- Conversion of seqcount usage sites to associated types and initializers.
-----BEGIN PGP SIGNATURE-----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=7Gj+
-----END PGP SIGNATURE-----
Merge tag 'locking-urgent-2020-08-10' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull locking updates from Thomas Gleixner:
"A set of locking fixes and updates:
- Untangle the header spaghetti which causes build failures in
various situations caused by the lockdep additions to seqcount to
validate that the write side critical sections are non-preemptible.
- The seqcount associated lock debug addons which were blocked by the
above fallout.
seqcount writers contrary to seqlock writers must be externally
serialized, which usually happens via locking - except for strict
per CPU seqcounts. As the lock is not part of the seqcount, lockdep
cannot validate that the lock is held.
This new debug mechanism adds the concept of associated locks.
sequence count has now lock type variants and corresponding
initializers which take a pointer to the associated lock used for
writer serialization. If lockdep is enabled the pointer is stored
and write_seqcount_begin() has a lockdep assertion to validate that
the lock is held.
Aside of the type and the initializer no other code changes are
required at the seqcount usage sites. The rest of the seqcount API
is unchanged and determines the type at compile time with the help
of _Generic which is possible now that the minimal GCC version has
been moved up.
Adding this lockdep coverage unearthed a handful of seqcount bugs
which have been addressed already independent of this.
While generally useful this comes with a Trojan Horse twist: On RT
kernels the write side critical section can become preemtible if
the writers are serialized by an associated lock, which leads to
the well known reader preempts writer livelock. RT prevents this by
storing the associated lock pointer independent of lockdep in the
seqcount and changing the reader side to block on the lock when a
reader detects that a writer is in the write side critical section.
- Conversion of seqcount usage sites to associated types and
initializers"
* tag 'locking-urgent-2020-08-10' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (25 commits)
locking/seqlock, headers: Untangle the spaghetti monster
locking, arch/ia64: Reduce <asm/smp.h> header dependencies by moving XTP bits into the new <asm/xtp.h> header
x86/headers: Remove APIC headers from <asm/smp.h>
seqcount: More consistent seqprop names
seqcount: Compress SEQCNT_LOCKNAME_ZERO()
seqlock: Fold seqcount_LOCKNAME_init() definition
seqlock: Fold seqcount_LOCKNAME_t definition
seqlock: s/__SEQ_LOCKDEP/__SEQ_LOCK/g
hrtimer: Use sequence counter with associated raw spinlock
kvm/eventfd: Use sequence counter with associated spinlock
userfaultfd: Use sequence counter with associated spinlock
NFSv4: Use sequence counter with associated spinlock
iocost: Use sequence counter with associated spinlock
raid5: Use sequence counter with associated spinlock
vfs: Use sequence counter with associated spinlock
timekeeping: Use sequence counter with associated raw spinlock
xfrm: policy: Use sequence counters with associated lock
netfilter: nft_set_rbtree: Use sequence counter with associated rwlock
netfilter: conntrack: Use sequence counter with associated spinlock
sched: tasks: Use sequence counter with associated spinlock
...
Pablo Neira Ayuso says:
====================
Netfilter fixes for net
The following patchset contains Netfilter fixes for net:
1) Flush the cleanup xtables worker to make sure destructors
have completed, from Florian Westphal.
2) iifgroup is matching erroneously, also from Florian.
3) Add selftest for meta interface matching, from Florian Westphal.
4) Move nf_ct_offload_timeout() to header, from Roi Dayan.
5) Call nf_ct_offload_timeout() from flow_offload_add() to
make sure garbage collection does not evict offloaded flow,
from Roi Dayan.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
To be used by callers from other modules.
[ Rename DAY to NF_CT_DAY to avoid possible symbol name pollution
issue --Pablo ]
Signed-off-by: Roi Dayan <roid@mellanox.com>
Reviewed-by: Oz Shlomo <ozsh@mellanox.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Pablo Neira found that after recent update of xt_IDLETIMER the
iptables-nft tests sometimes show an error.
He tracked this down to the delayed cleanup used by nf_tables core:
del rule (transaction A)
add rule (transaction B)
Its possible that by time transaction B (both in same netns) runs,
the xt target destructor has not been invoked yet.
For native nft expressions this is no problem because all expressions
that have such side effects make sure these are handled from the commit
phase, rather than async cleanup.
For nft_compat however this isn't true.
Instead of forcing synchronous behaviour for nft_compat, keep track
of the number of outstanding destructor calls.
When we attempt to create a new expression, flush the cleanup worker
to make sure destructors have completed.
With lots of help from Pablo Neira.
Reported-by: Pablo Neira Ayso <pablo@netfilter.org>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
A sequence counter write side critical section must be protected by some
form of locking to serialize writers. A plain seqcount_t does not
contain the information of which lock must be held when entering a write
side critical section.
Use the new seqcount_spinlock_t data type, which allows to associate a
spinlock with the sequence counter. This enables lockdep to verify that
the spinlock used for writer serialization is held when the write side
critical section is entered.
If lockdep is disabled this lock association is compiled out and has
neither storage size nor runtime overhead.
Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20200720155530.1173732-15-a.darwish@linutronix.de
This new chain flag specifies that:
* the kernel dynamically allocates the chain name, if no chain name
is specified.
* If the immediate expression that refers to this chain is removed,
then this bound chain (and its content) is destroyed.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
This enum definition was never exposed through UAPI. Rename
NFT_BASE_CHAIN to NFT_CHAIN_BASE for consistency.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
This netlink attribute allows you to refer to chains inside a
transaction as an alternative to the name and the handle. The chain
binding support requires this new chain ID approach.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Currently, nf_flow_table_offload_add/del_cb are exported by nf_flow_table
module, therefore modules using them will have hard-dependency
on nf_flow_table and will require loading it all the time.
This can lead to an unnecessary overhead on systems that do not
use this API.
To relax the hard-dependency between the modules, we unexport these
functions and make them static inline.
Fixes: 978703f425 ("netfilter: flowtable: Add API for registering to flow table events")
Signed-off-by: Alaa Hleihel <alaa@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Reviewed-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pablo Neira Ayuso says:
====================
Netfilter updates for net-next
The following patchset contains Netfilter updates for net-next
to extend ctnetlink and the flowtable infrastructure:
1) Extend ctnetlink kernel side netlink dump filtering capabilities,
from Romain Bellan.
2) Generalise the flowtable hook parser to take a hook list.
3) Pass a hook list to the flowtable hook registration/unregistration.
4) Add a helper function to release the flowtable hook list.
5) Update the flowtable event notifier to pass a flowtable hook list.
6) Allow users to add new devices to an existing flowtables.
7) Allow users to remove devices to an existing flowtables.
8) Allow for registering a flowtable with no initial devices.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
This function schedules the flow teardown state and it forces a gc run.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Conntrack dump does not support kernel side filtering (only get exists,
but it returns only one entry. And user has to give a full valid tuple)
It means that userspace has to implement filtering after receiving many
irrelevant entries, consuming resources (conntrack table is sometimes
very huge, much more than a routing table for example).
This patch adds filtering in kernel side. To achieve this goal, we:
* Add a new CTA_FILTER netlink attributes, actually a flag list to
parametize filtering
* Convert some *nlattr_to_tuple() functions, to allow a partial parsing
of CTA_TUPLE_ORIG and CTA_TUPLE_REPLY (so nf_conntrack_tuple it not
fully set)
Filtering is now possible on:
* IP SRC/DST values
* Ports for TCP and UDP flows
* IMCP(v6) codes types and IDs
Filtering is done as an "AND" operator. For example, when flags
PROTO_SRC_PORT, PROTO_NUM and IP_SRC are sets, only entries matching all
values are dumped.
Changes since v1:
Set NLM_F_DUMP_FILTERED in nlm flags if entries are filtered
Changes since v2:
Move several constants to nf_internals.h
Move a fix on netlink values check in a separate patch
Add a check on not-supported flags
Return EOPNOTSUPP if CDA_FILTER is set in ctnetlink_flush_conntrack
(not yet implemented)
Code style issues
Changes since v3:
Fix compilation warning reported by kbuild test robot
Changes since v4:
Fix a regression introduced in v3 (returned EINVAL for valid netlink
messages without CTA_MARK)
Changes since v5:
Change definition of CTA_FILTER_F_ALL
Fix a regression when CTA_TUPLE_ZONE is not set
Signed-off-by: Romain Bellan <romain.bellan@wifirst.fr>
Signed-off-by: Florent Fourcot <florent.fourcot@wifirst.fr>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Move the bpf verifier trace check into the new switch statement in
HEAD.
Resolve the overlapping changes in hinic, where bug fixes overlap
the addition of VF support.
Signed-off-by: David S. Miller <davem@davemloft.net>
Gc step can queue offloaded flow del work or stats work.
Those work items can race each other and a flow could be freed
before the stats work is executed and querying it.
To avoid that, add a pending bit that if a work exists for a flow
don't queue another work for it.
This will also avoid adding multiple stats works in case stats work
didn't complete but gc step started again.
Signed-off-by: Paul Blakey <paulb@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
gcc-10 warns around a suspicious access to an empty struct member:
net/netfilter/nf_conntrack_core.c: In function '__nf_conntrack_alloc':
net/netfilter/nf_conntrack_core.c:1522:9: warning: array subscript 0 is outside the bounds of an interior zero-length array 'u8[0]' {aka 'unsigned char[0]'} [-Wzero-length-bounds]
1522 | memset(&ct->__nfct_init_offset[0], 0,
| ^~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from net/netfilter/nf_conntrack_core.c:37:
include/net/netfilter/nf_conntrack.h:90:5: note: while referencing '__nfct_init_offset'
90 | u8 __nfct_init_offset[0];
| ^~~~~~~~~~~~~~~~~~
The code is correct but a bit unusual. Rework it slightly in a way that
does not trigger the warning, using an empty struct instead of an empty
array. There are probably more elegant ways to do this, but this is the
smallest change.
Fixes: c41884ce05 ("netfilter: conntrack: avoid zeroing timer")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Initialize set lookup matching element to NULL. Otherwise, the
NFT_LOOKUP_F_INV flag reverses the matching logic and it leads to
deference an uninitialized pointer to the matching element. Make sure
element data area and stateful expression are accessed if there is a
matching set element.
This patch undoes 24791b9aa1 ("netfilter: nft_set_bitmap: initialize set
element extension in lookups") which is not required anymore.
Fixes: 339706bc21 ("netfilter: nft_lookup: update element stateful expression")
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Add nf_ct_acct_add function to update the conntrack counter
with packets and bytes.
Signed-off-by: wenxu <wenxu@ucloud.cn>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
The refcount is done via entry->skb, which does work fine.
Major problem: When putting the refcount of the bridge ports, we
must always put the references while the skb is still around.
However, we will need to put the references after okfn() to avoid
a possible 1 -> 0 -> 1 refcount transition, so we cannot use the
skb pointer anymore.
Place the physports in the queue entry structure instead to allow
for refcounting changes in the next patch.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
This is a preparation patch, no logical changes.
Move free_entry into core and rename it to something more sensible.
Will ease followup patches which will complicate the refcount handling.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Currently flow offload threads are synchronized by the flow block mutex.
Use rw lock instead to increase flow insertion (read) concurrency.
Signed-off-by: Paul Blakey <paulb@mellanox.com>
Reviewed-by: Oz Shlomo <ozsh@mellanox.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
This patch allows users to specify the stateful expression for the
elements in this set via NFTA_SET_EXPR. This new feature allows you to
turn on counters for all of the elements in this set.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
If the AVX2 set is available, we can exploit the repetitive
characteristic of this algorithm to provide a fast, vectorised
version by using 256-bit wide AVX2 operations for bucket loads and
bitwise intersections.
In most cases, this implementation consistently outperforms rbtree
set instances despite the fact they are configured to use a given,
single, ranged data type out of the ones used for performance
measurements by the nft_concat_range.sh kselftest.
That script, injecting packets directly on the ingoing device path
with pktgen, reports, averaged over five runs on a single AMD Epyc
7402 thread (3.35GHz, 768 KiB L1D$, 12 MiB L2$), the figures below.
CONFIG_RETPOLINE was not set here.
Note that this is not a fair comparison over hash and rbtree set
types: non-ranged entries (used to have a reference for hash types)
would be matched faster than this, and matching on a single field
only (which is the case for rbtree) is also significantly faster.
However, it's not possible at the moment to choose this set type
for non-ranged entries, and the current implementation also needs
a few minor adjustments in order to match on less than two fields.
---------------.-----------------------------------.------------.
AMD Epyc 7402 | baselines, Mpps | this patch |
1 thread |___________________________________|____________|
3.35GHz | | | | | |
768KiB L1D$ | netdev | hash | rbtree | | |
---------------| hook | no | single | | pipapo |
type entries | drop | ranges | field | pipapo | AVX2 |
---------------|--------|--------|--------|--------|------------|
net,port | | | | | |
1000 | 19.0 | 10.4 | 3.8 | 4.0 | 7.5 +87% |
---------------|--------|--------|--------|--------|------------|
port,net | | | | | |
100 | 18.8 | 10.3 | 5.8 | 6.3 | 8.1 +29% |
---------------|--------|--------|--------|--------|------------|
net6,port | | | | | |
1000 | 16.4 | 7.6 | 1.8 | 2.1 | 4.8 +128% |
---------------|--------|--------|--------|--------|------------|
port,proto | | | | | |
30000 | 19.6 | 11.6 | 3.9 | 0.5 | 2.6 +420% |
---------------|--------|--------|--------|--------|------------|
net6,port,mac | | | | | |
10 | 16.5 | 5.4 | 4.3 | 3.4 | 4.7 +38% |
---------------|--------|--------|--------|--------|------------|
net6,port,mac, | | | | | |
proto 1000 | 16.5 | 5.7 | 1.9 | 1.4 | 3.6 +26% |
---------------|--------|--------|--------|--------|------------|
net,mac | | | | | |
1000 | 19.0 | 8.4 | 3.9 | 2.5 | 6.4 +156% |
---------------'--------'--------'--------'--------'------------'
A similar strategy could be easily reused to implement specialised
versions for other SIMD sets, and I plan to post at least a NEON
version at a later time.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
This patch support both ipv4 and ipv6 tunnel_id, tunnel_src and
tunnel_dst match for flowtable offload
Signed-off-by: wenxu <wenxu@ucloud.cn>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
The current codebase makes use of the zero-length array language
extension to the C90 standard, but the preferred mechanism to declare
variable-length types such as these ones is a flexible array member[1][2],
introduced in C99:
struct foo {
int stuff;
struct boo array[];
};
By making use of the mechanism above, we will get a compiler warning
in case the flexible array does not occur last in the structure, which
will help us prevent some kind of undefined behavior bugs from being
inadvertently introduced[3] to the codebase from now on.
Also, notice that, dynamic memory allocations won't be affected by
this change:
"Flexible array members have incomplete type, and so the sizeof operator
may not be applied. As a quirk of the original implementation of
zero-length arrays, sizeof evaluates to zero."[1]
Lastly, fix checkpatch.pl warning
WARNING: __aligned(size) is preferred over __attribute__((aligned(size)))
in net/bridge/netfilter/ebtables.c
This issue was found with the help of Coccinelle.
[1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html
[2] https://github.com/KSPP/linux/issues/21
[3] commit 7649773293 ("cxgb3/l2t: Fix undefined behaviour")
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
They do not need to be writeable anymore.
v2: remove left-over __read_mostly annotation in set_pipapo.c (Stefano)
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Placing nftables set support in an extra module is pointless:
1. nf_tables needs dynamic registeration interface for sake of one module
2. nft heavily relies on sets, e.g. even simple rule like
"nft ... tcp dport { 80, 443 }" will not work with _SETS=n.
IOW, either nftables isn't used or both nf_tables and nf_tables_set
modules are needed anyway.
With extra module:
307K net/netfilter/nf_tables.ko
79K net/netfilter/nf_tables_set.ko
text data bss dec filename
146416 3072 545 150033 nf_tables.ko
35496 1817 0 37313 nf_tables_set.ko
This patch:
373K net/netfilter/nf_tables.ko
178563 4049 545 183157 nf_tables.ko
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
If driver deleted an FT entry, a FT failed to offload, or registered to the
flow table after flows were already added, we still get packets in
software.
For those packets, while restoring the ct state from the flow table
entry, refresh it's hardware offload.
Signed-off-by: Paul Blakey <paulb@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
NF flow table API associate 5-tuple rule with an action list by calling
the flow table type action() CB to fill the rule's actions.
In action CB of act_ct, populate the ct offload entry actions with a new
ct_metadata action. Initialize the ct_metadata with the ct mark, label and
zone information. If ct nat was performed, then also append the relevant
packet mangle actions (e.g. ipv4/ipv6/tcp/udp header rewrites).
Drivers that offload the ft entries may match on the 5-tuple and perform
the action list.
Signed-off-by: Paul Blakey <paulb@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Let drivers to add their cb allowing them to receive flow offload events
of type TC_SETUP_CLSFLOWER (REPLACE/DEL/STATS) for flows managed by the
flow table.
Signed-off-by: Paul Blakey <paulb@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This new set type allows for intervals in concatenated fields,
which are expressed in the usual way, that is, simple byte
concatenation with padding to 32 bits for single fields, and
given as ranges by specifying start and end elements containing,
each, the full concatenation of start and end values for the
single fields.
Ranges are expanded to composing netmasks, for each field: these
are inserted as rules in per-field lookup tables. Bits to be
classified are divided in 4-bit groups, and for each group, the
lookup table contains 4^2 buckets, representing all the possible
values of a bit group. This approach was inspired by the Grouper
algorithm:
http://www.cse.usf.edu/~ligatti/projects/grouper/
Matching is performed by a sequence of AND operations between
bucket values, with buckets selected according to the value of
packet bits, for each group. The result of this sequence tells
us which rules matched for a given field.
In order to concatenate several ranged fields, per-field rules
are mapped using mapping arrays, one per field, that specify
which rules should be considered while matching the next field.
The mapping array for the last field contains a reference to
the element originally inserted.
The notes in nft_set_pipapo.c cover the algorithm in deeper
detail.
A pure hash-based approach is of no use here, as ranges need
to be classified. An implementation based on "proxying" the
existing red-black tree set type, creating a tree for each
field, was considered, but deemed impractical due to the fact
that elements would need to be shared between trees, at least
as long as we want to keep UAPI changes to a minimum.
A stand-alone implementation of this algorithm is available at:
https://pipapo.lameexcu.se
together with notes about possible future optimisations
(in pipapo.c).
This algorithm was designed with data locality in mind, and can
be highly optimised for SIMD instruction sets, as the bulk of
the matching work is done with repetitive, simple bitwise
operations.
At this point, without further optimisations, nft_concat_range.sh
reports, for one AMD Epyc 7351 thread (2.9GHz, 512 KiB L1D$, 8 MiB
L2$):
TEST: performance
net,port [ OK ]
baseline (drop from netdev hook): 10190076pps
baseline hash (non-ranged entries): 6179564pps
baseline rbtree (match on first field only): 2950341pps
set with 1000 full, ranged entries: 2304165pps
port,net [ OK ]
baseline (drop from netdev hook): 10143615pps
baseline hash (non-ranged entries): 6135776pps
baseline rbtree (match on first field only): 4311934pps
set with 100 full, ranged entries: 4131471pps
net6,port [ OK ]
baseline (drop from netdev hook): 9730404pps
baseline hash (non-ranged entries): 4809557pps
baseline rbtree (match on first field only): 1501699pps
set with 1000 full, ranged entries: 1092557pps
port,proto [ OK ]
baseline (drop from netdev hook): 10812426pps
baseline hash (non-ranged entries): 6929353pps
baseline rbtree (match on first field only): 3027105pps
set with 30000 full, ranged entries: 284147pps
net6,port,mac [ OK ]
baseline (drop from netdev hook): 9660114pps
baseline hash (non-ranged entries): 3778877pps
baseline rbtree (match on first field only): 3179379pps
set with 10 full, ranged entries: 2082880pps
net6,port,mac,proto [ OK ]
baseline (drop from netdev hook): 9718324pps
baseline hash (non-ranged entries): 3799021pps
baseline rbtree (match on first field only): 1506689pps
set with 1000 full, ranged entries: 783810pps
net,mac [ OK ]
baseline (drop from netdev hook): 10190029pps
baseline hash (non-ranged entries): 5172218pps
baseline rbtree (match on first field only): 2946863pps
set with 1000 full, ranged entries: 1279122pps
v4:
- fix build for 32-bit architectures: 64-bit division needs
div_u64() (kbuild test robot <lkp@intel.com>)
v3:
- rework interface for field length specification,
NFT_SET_SUBKEY disappears and information is stored in
description
- remove scratch area to store closing element of ranges,
as elements now come with an actual attribute to specify
the upper range limit (Pablo Neira Ayuso)
- also remove pointer to 'start' element from mapping table,
closing key is now accessible via extension data
- use bytes right away instead of bits for field lengths,
this way we can also double the inner loop of the lookup
function to take care of upper and lower bits in a single
iteration (minor performance improvement)
- make it clearer that set operations are actually atomic
API-wise, but we can't e.g. implement flush() as one-shot
action
- fix type for 'dup' in nft_pipapo_insert(), check for
duplicates only in the next generation, and in general take
care of differentiating generation mask cases depending on
the operation (Pablo Neira Ayuso)
- report C implementation matching rate in commit message, so
that AVX2 implementation can be compared (Pablo Neira Ayuso)
v2:
- protect access to scratch maps in nft_pipapo_lookup() with
local_bh_disable/enable() (Florian Westphal)
- drop rcu_read_lock/unlock() from nft_pipapo_lookup(), it's
already implied (Florian Westphal)
- explain why partial allocation failures don't need handling
in pipapo_realloc_scratch(), rename 'm' to clone and update
related kerneldoc to make it clear we're not operating on
the live copy (Florian Westphal)
- add expicit check for priv->start_elem in
nft_pipapo_insert() to avoid ending up in nft_pipapo_walk()
with a NULL start element, and also zero it out in every
operation that might make it invalid, so that insertion
doesn't proceed with an invalid element (Florian Westphal)
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Introduce a new nested netlink attribute, NFTA_SET_DESC_CONCAT, used
to specify the length of each field in a set concatenation.
This allows set implementations to support concatenation of multiple
ranged items, as they can divide the input key into matching data for
every single field. Such set implementations would be selected as
they specify support for NFT_SET_INTERVAL and allow desc->field_count
to be greater than one. Explicitly disallow this for nft_set_rbtree.
In order to specify the interval for a set entry, userspace would
include in NFTA_SET_DESC_CONCAT attributes field lengths, and pass
range endpoints as two separate keys, represented by attributes
NFTA_SET_ELEM_KEY and NFTA_SET_ELEM_KEY_END.
While at it, export the number of 32-bit registers available for
packet matching, as nftables will need this to know the maximum
number of field lengths that can be specified.
For example, "packets with an IPv4 address between 192.0.2.0 and
192.0.2.42, with destination port between 22 and 25", can be
expressed as two concatenated elements:
NFTA_SET_ELEM_KEY: 192.0.2.0 . 22
NFTA_SET_ELEM_KEY_END: 192.0.2.42 . 25
and NFTA_SET_DESC_CONCAT attribute would contain:
NFTA_LIST_ELEM
NFTA_SET_FIELD_LEN: 4
NFTA_LIST_ELEM
NFTA_SET_FIELD_LEN: 2
v4: No changes
v3: Complete rework, NFTA_SET_DESC_CONCAT instead of NFTA_SET_SUBKEY
v2: No changes
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Add NFTA_SET_ELEM_KEY_END attribute to convey the closing element of the
interval between kernel and userspace.
This patch also adds the NFT_SET_EXT_KEY_END extension to store the
closing element value in this interval.
v4: No changes
v3: New patch
[sbrivio: refactor error paths and labels; add corresponding
nft_set_ext_type for new key; rebase]
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
If nf_flow_offload_add() fails to add the flow to hardware, then the
NF_FLOW_HW_REFRESH flag bit is set and the flow remains in the flowtable
software path.
If flowtable hardware offload is enabled, this patch enqueues a new
request to offload this flow to hardware.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
This function checks for the NF_FLOWTABLE_HW_OFFLOAD flag, meaning that
the flowtable hardware offload is enabled.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Originally, all flow flag bits were set on only from the workqueue. With
the introduction of the flow teardown state and hardware offload this is
no longer true. Let's be safe and use atomic bitwise operation to
operation with flow flags.
Fixes: 59c466dd68 ("netfilter: nf_flow_table: add a new flow state for tearing down offloading")
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
The dying bit removes the conntrack entry if the netdev that owns this
flow is going down. Instead, use the teardown mechanism to push back the
flow to conntrack to let the classic software path decide what to do
with it.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
This patch adds nf_flowtable_time_stamp and updates the existing code to
use it.
This patch is also implicitly fixing up hardware statistic fetching via
nf_flow_offload_stats() where casting to u32 is missing. Use
nf_flow_timeout_delta() to fix this.
Fixes: c29f74e0df ("netfilter: nf_flow_table: hardware offload support")
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Acked-by: wenxu <wenxu@ucloud.cn>
Replace all the occurrences of FIELD_SIZEOF() with sizeof_field() except
at places where these are defined. Later patches will remove the unused
definition of FIELD_SIZEOF().
This patch is generated using following script:
EXCLUDE_FILES="include/linux/stddef.h|include/linux/kernel.h"
git grep -l -e "\bFIELD_SIZEOF\b" | while read file;
do
if [[ "$file" =~ $EXCLUDE_FILES ]]; then
continue
fi
sed -i -e 's/\bFIELD_SIZEOF\b/sizeof_field/g' $file;
done
Signed-off-by: Pankaj Bharadiya <pankaj.laxminarayan.bharadiya@intel.com>
Link: https://lore.kernel.org/r/20190924105839.110713-3-pankaj.laxminarayan.bharadiya@intel.com
Co-developed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Kees Cook <keescook@chromium.org>
Acked-by: David Miller <davem@davemloft.net> # for net
This patch constifies the pointer to source register data that is passed
as an input parameter.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Linus Torvalds
David S. Miller
Roi Dayan
Florian Westphal
Ahmed S. Darwish
Pablo Neira Ayuso
Alaa Hleihel
Romain Bellan
Paul Blakey
Arnd Bergmann
wenxu
Stefano Brivio
Gustavo A. R. Silva
Pankaj Bharadiya