04 Apr, 2014
2 commits
-
Eric points out that the locks can be global.
Moreover, both Jesper and Eric note that using only 32 locks increases
false sharing as only two cache lines are used.This increases locks to 256 (16 cache lines assuming 64byte cacheline and
4 bytes per spinlock).Suggested-by: Jesper Dangaard Brouer
Suggested-by: Eric Dumazet
Signed-off-by: Florian Westphal
Signed-off-by: Pablo Neira Ayuso -
cannot use ARRAY_SIZE() if spinlock_t is empty struct.
Fixes: 1442e7507dd597 ("netfilter: connlimit: use keyed locks")
Reported-by: kbuild test robot
Signed-off-by: Florian Westphal
Signed-off-by: Pablo Neira Ayuso
17 Mar, 2014
3 commits
-
With current match design every invocation of the connlimit_match
function means we have to perform (number_of_conntracks % 256) lookups
in the conntrack table [ to perform GC/delete stale entries ].
This is also the reason why ____nf_conntrack_find() in perf top has
> 20% cpu time per core.This patch changes the storage to rbtree which cuts down the number of
ct objects that need testing.When looking up a new tuple, we only test the connections of the host
objects we visit while searching for the wanted host/network (or
the leaf we need to insert at).The slot count is reduced to 32. Increasing slot count doesn't
speed up things much because of rbtree nature.before patch (50kpps rx, 10kpps tx):
+ 20.95% ksoftirqd/0 [nf_conntrack] [k] ____nf_conntrack_find
+ 20.50% ksoftirqd/1 [nf_conntrack] [k] ____nf_conntrack_find
+ 20.27% ksoftirqd/2 [nf_conntrack] [k] ____nf_conntrack_find
+ 5.76% ksoftirqd/1 [nf_conntrack] [k] hash_conntrack_raw
+ 5.39% ksoftirqd/2 [nf_conntrack] [k] hash_conntrack_raw
+ 5.35% ksoftirqd/0 [nf_conntrack] [k] hash_conntrack_rawafter (90kpps, 51kpps tx):
+ 17.24% swapper [nf_conntrack] [k] ____nf_conntrack_find
+ 6.60% ksoftirqd/2 [nf_conntrack] [k] ____nf_conntrack_find
+ 2.73% swapper [nf_conntrack] [k] hash_conntrack_raw
+ 2.36% swapper [xt_connlimit] [k] count_treeObvious disadvantages to previous version are the increase in code
complexity and the increased memory cost.Partially based on Eric Dumazets fq scheduler.
Reviewed-by: Jesper Dangaard Brouer
Signed-off-by: Florian Westphal
Signed-off-by: Pablo Neira Ayuso -
currently returns 1 if they're the same. Make it work like mem/strcmp
so it can be used as rbtree search function.Reviewed-by: Jesper Dangaard Brouer
Signed-off-by: Florian Westphal
Signed-off-by: Pablo Neira Ayuso -
connlimit currently suffers from spinlock contention, example for
4-core system with rps enabled:+ 20.84% ksoftirqd/2 [kernel.kallsyms] [k] _raw_spin_lock_bh
+ 20.76% ksoftirqd/1 [kernel.kallsyms] [k] _raw_spin_lock_bh
+ 20.42% ksoftirqd/0 [kernel.kallsyms] [k] _raw_spin_lock_bh
+ 6.07% ksoftirqd/2 [nf_conntrack] [k] ____nf_conntrack_find
+ 6.07% ksoftirqd/1 [nf_conntrack] [k] ____nf_conntrack_find
+ 5.97% ksoftirqd/0 [nf_conntrack] [k] ____nf_conntrack_find
+ 2.47% ksoftirqd/2 [nf_conntrack] [k] hash_conntrack_raw
+ 2.45% ksoftirqd/0 [nf_conntrack] [k] hash_conntrack_raw
+ 2.44% ksoftirqd/1 [nf_conntrack] [k] hash_conntrack_rawMay allow parallel lookup/insert/delete if the entry is hashed to
another slot. With patch:+ 20.95% ksoftirqd/0 [nf_conntrack] [k] ____nf_conntrack_find
+ 20.50% ksoftirqd/1 [nf_conntrack] [k] ____nf_conntrack_find
+ 20.27% ksoftirqd/2 [nf_conntrack] [k] ____nf_conntrack_find
+ 5.76% ksoftirqd/1 [nf_conntrack] [k] hash_conntrack_raw
+ 5.39% ksoftirqd/2 [nf_conntrack] [k] hash_conntrack_raw
+ 5.35% ksoftirqd/0 [nf_conntrack] [k] hash_conntrack_raw
+ 2.00% ksoftirqd/1 [kernel.kallsyms] [k] __rcu_read_unlockImproved rx processing rate from ~35kpps to ~50 kpps.
Reviewed-by: Jesper Dangaard Brouer
Signed-off-by: Florian Westphal
Signed-off-by: Pablo Neira Ayuso
12 Mar, 2014
4 commits
-
We might allocate thousands of these (one object per connection).
Use distinct kmem cache to permit simplte tracking on how many
objects are currently used by the connlimit match via the sysfs.Reviewed-by: Jesper Dangaard Brouer
Signed-off-by: Florian Westphal
Signed-off-by: Pablo Neira Ayuso -
Allows easier code-reuse in followup patches.
Reviewed-by: Jesper Dangaard Brouer
Signed-off-by: Florian Westphal
Signed-off-by: Pablo Neira Ayuso -
Instead of freeing the entry from our list and then adding
it back again in the 'packet to closing connection' case just keep the
matching entry around. Also drop the found_ct != NULL test as
nf_ct_tuplehash_to_ctrack is just container_of().Reviewed-by: Jesper Dangaard Brouer
Signed-off-by: Florian Westphal
Signed-off-by: Pablo Neira Ayuso -
Simplifies followup patch that introduces separate locks for each of
the hash slots.Reviewed-by: Jesper Dangaard Brouer
Signed-off-by: Florian Westphal
Signed-off-by: Pablo Neira Ayuso
28 Feb, 2013
1 commit
-
I'm not sure why, but the hlist for each entry iterators were conceived
list_for_each_entry(pos, head, member)
The hlist ones were greedy and wanted an extra parameter:
hlist_for_each_entry(tpos, pos, head, member)
Why did they need an extra pos parameter? I'm not quite sure. Not only
they don't really need it, it also prevents the iterator from looking
exactly like the list iterator, which is unfortunate.Besides the semantic patch, there was some manual work required:
- Fix up the actual hlist iterators in linux/list.h
- Fix up the declaration of other iterators based on the hlist ones.
- A very small amount of places were using the 'node' parameter, this
was modified to use 'obj->member' instead.
- Coccinelle didn't handle the hlist_for_each_entry_safe iterator
properly, so those had to be fixed up manually.The semantic patch which is mostly the work of Peter Senna Tschudin is here:
@@
iterator name hlist_for_each_entry, hlist_for_each_entry_continue, hlist_for_each_entry_from, hlist_for_each_entry_rcu, hlist_for_each_entry_rcu_bh, hlist_for_each_entry_continue_rcu_bh, for_each_busy_worker, ax25_uid_for_each, ax25_for_each, inet_bind_bucket_for_each, sctp_for_each_hentry, sk_for_each, sk_for_each_rcu, sk_for_each_from, sk_for_each_safe, sk_for_each_bound, hlist_for_each_entry_safe, hlist_for_each_entry_continue_rcu, nr_neigh_for_each, nr_neigh_for_each_safe, nr_node_for_each, nr_node_for_each_safe, for_each_gfn_indirect_valid_sp, for_each_gfn_sp, for_each_host;type T;
expression a,c,d,e;
identifier b;
statement S;
@@-T b;
[akpm@linux-foundation.org: drop bogus change from net/ipv4/raw.c]
[akpm@linux-foundation.org: drop bogus hunk from net/ipv6/raw.c]
[akpm@linux-foundation.org: checkpatch fixes]
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foudnation.org: redo intrusive kvm changes]
Tested-by: Peter Senna Tschudin
Acked-by: Paul E. McKenney
Signed-off-by: Sasha Levin
Cc: Wu Fengguang
Cc: Marcelo Tosatti
Cc: Gleb Natapov
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
07 Jun, 2012
1 commit
-
It was scheduled to be removed.
Cc: Jan Engelhardt
Signed-off-by: Cong Wang
Signed-off-by: Pablo Neira Ayuso
15 Mar, 2011
4 commits
-
A potential race condition when generating connlimit_rnd is also fixed.
Signed-off-by: Changli Gao
Signed-off-by: Patrick McHardy -
The header of hlist is smaller than list.
Signed-off-by: Changli Gao
Signed-off-by: Patrick McHardy -
All the members are initialized after kzalloc().
Signed-off-by: Changli Gao
Signed-off-by: Patrick McHardy -
We use the reply tuples when limiting the connections by the destination
addresses, however, in SNAT scenario, the final reply tuples won't be
ready until SNAT is done in POSTROUING or INPUT chain, and the following
nf_conntrack_find_get() in count_tem() will get nothing, so connlimit
can't work as expected.In this patch, the original tuples are always used, and an additional
member addr is appended to save the address in either end.Signed-off-by: Changli Gao
Signed-off-by: Patrick McHardy
14 Feb, 2011
1 commit
-
This reverts commit 44bd4de9c2270b22c3c898310102bc6be9ed2978.
I have to revert the early loop termination in connlimit since it generates
problems when an iptables statement does not use -m state --state NEW before
the connlimit match extension.Signed-off-by: Stefan Berger
Signed-off-by: Patrick McHardy
12 Feb, 2011
1 commit
-
The patch below introduces an early termination of the loop that is
counting matches. It terminates once the counter has exceeded the
threshold provided by the user. There's no point in continuing the loop
afterwards and looking at other entries.It plays together with the following code further below:
return (connections > info->limit) ^ info->inverse;
where connections is the result of the counted connection, which in turn
is the matches variable in the loop. So once-> matches = info->limit + 1
alias -> matches > info->limit
alias -> matches > thresholdwe can terminate the loop.
Signed-off-by: Stefan Berger
Signed-off-by: Patrick McHardy
26 Jan, 2011
1 commit
-
xt_connlimit normally records the "original" tuples in a hashlist
(such as "1.2.3.4 -> 5.6.7.8"), and looks in this list for iph->daddr
when counting.When the user however uses DNAT in PREROUTING, looking for
iph->daddr -- which is now 192.168.9.10 -- will not match. Thus in
daddr mode, we need to record the reverse direction tuple
("192.168.9.10 -> 1.2.3.4") instead. In the reverse tuple, the dst
addr is on the src side, which is convenient, as count_them still uses
&conn->tuple.src.u3.Signed-off-by: Jan Engelhardt
20 Jan, 2011
1 commit
-
This adds destination address-based selection. The old "inverse"
member is overloaded (memory-wise) with a new "flags" variable,
similar to how J.Park did it with xt_string rev 1. Since revision 0
userspace only sets flag 0x1, no great changes are made to explicitly
test for different revisions.Signed-off-by: Jan Engelhardt
18 Jan, 2011
1 commit
-
Signed-off-by: Richard Weinberger
Signed-off-by: Jan Engelhardt
12 May, 2010
3 commits
-
Since xt_action_param is writable, let's use it. The pointer to
'bool hotdrop' always worried (8 bytes (64-bit) to write 1 byte!).
Surprisingly results in a reduction in size:text data bss filename
5457066 692730 357892 vmlinux.o-prev
5456554 692730 357892 vmlinux.oSigned-off-by: Jan Engelhardt
-
In future, layer-3 matches will be an xt module of their own, and
need to set the fragoff and thoff fields. Adding more pointers would
needlessy increase memory requirements (esp. so for 64-bit, where
pointers are wider).Signed-off-by: Jan Engelhardt
-
Signed-off-by: Jan Engelhardt
20 Apr, 2010
1 commit
-
Conflicts:
Documentation/feature-removal-schedule.txt
net/ipv6/netfilter/ip6t_REJECT.c
net/netfilter/xt_limit.cSigned-off-by: Patrick McHardy
30 Mar, 2010
1 commit
-
…it slab.h inclusion from percpu.h
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
25 Mar, 2010
3 commits
-
When extended status codes are available, such as ENOMEM on failed
allocations, or subsequent functions (e.g. nf_ct_get_l3proto), passing
them up to userspace seems like a good idea compared to just always
EINVAL.Signed-off-by: Jan Engelhardt
-
The following semantic patch does part of the transformation:
//
@ rule1 @
struct xt_match ops;
identifier check;
@@
ops.checkentry = check;@@
identifier rule1.check;
@@
check(...) { }@@
identifier rule1.check;
@@
check(...) { }
//Signed-off-by: Jan Engelhardt
-
Restore function signatures from bool to int so that we can report
memory allocation failures or similar using -ENOMEM rather than
always having to pass -EINVAL back.This semantic patch may not be too precise (checking for functions
that use xt_mtchk_param rather than functions referenced by
xt_match.checkentry), but reviewed, it produced the intended result.//
@@
type bool;
identifier check, par;
@@
-bool check
+int check
(struct xt_mtchk_param *par) { ... }
//Signed-off-by: Jan Engelhardt
18 Mar, 2010
1 commit
-
Signed-off-by: Jan Engelhardt
17 Mar, 2010
1 commit
-
Signed-off-by: Jan Engelhardt
16 Feb, 2010
1 commit
-
Normally, each connection needs a unique identity. Conntrack zones allow
to specify a numerical zone using the CT target, connections in different
zones can use the same identity.Example:
iptables -t raw -A PREROUTING -i veth0 -j CT --zone 1
iptables -t raw -A OUTPUT -o veth1 -j CT --zone 1Signed-off-by: Patrick McHardy
18 Jan, 2010
1 commit
-
Signed-off-by: Alexey Dobriyan
Signed-off-by: Patrick McHardy
04 Jan, 2010
1 commit
-
We can initialize the random hash bytes on checkentry. This is
preferable since it is outside the hot path.Reference: http://bugzilla.netfilter.org/show_bug.cgi?id=621
Signed-off-by: Jan Engelhardt
Signed-off-by: Patrick McHardy
07 Nov, 2009
1 commit
-
Commit v2.6.28-rc1~717^2~109^2~2 was slightly incomplete; not all
instances of par->match->family were changed to par->family.References: http://bugzilla.netfilter.org/show_bug.cgi?id=610
Signed-off-by: Jan Engelhardt
Signed-off-by: Patrick McHardy
Signed-off-by: David S. Miller
26 Mar, 2009
1 commit
-
Use "hlist_nulls" infrastructure we added in 2.6.29 for RCUification of UDP & TCP.
This permits an easy conversion from call_rcu() based hash lists to a
SLAB_DESTROY_BY_RCU one.Avoiding call_rcu() delay at nf_conn freeing time has numerous gains.
First, it doesnt fill RCU queues (up to 10000 elements per cpu).
This reduces OOM possibility, if queued elements are not taken into account
This reduces latency problems when RCU queue size hits hilimit and triggers
emergency mode.- It allows fast reuse of just freed elements, permitting better use of
CPU cache.- We delete rcu_head from "struct nf_conn", shrinking size of this structure
by 8 or 16 bytes.This patch only takes care of "struct nf_conn".
call_rcu() is still used for less critical conntrack parts, that may
be converted later if necessary.Signed-off-by: Eric Dumazet
Signed-off-by: Patrick McHardy
08 Oct, 2008
5 commits
-
Using ->family in struct xt_*_param, multiple struct xt_{match,target}
can be squashed together.Signed-off-by: Jan Engelhardt
Signed-off-by: Patrick McHardy -
This patch does this for match extensions' destroy functions.
Signed-off-by: Jan Engelhardt
Signed-off-by: Patrick McHardy -
This patch does this for match extensions' checkentry functions.
Signed-off-by: Jan Engelhardt
Signed-off-by: Patrick McHardy -
The function signatures for Xtables extensions have grown over time.
It involves a lot of typing/replication, and also a bit of stack space
even if they are not used. Realize an NFWS2008 idea and pack them into
structs. The skb remains outside of the struct so gcc can continue to
apply its optimizations.This patch does this for match extensions' match functions.
A few ambiguities have also been addressed. The "offset" parameter for
example has been renamed to "fragoff" (there are so many different
offsets already) and "protoff" to "thoff" (there is more than just one
protocol here, so clarify).Signed-off-by: Jan Engelhardt
Signed-off-by: Patrick McHardy -
* make per-netns conntrack hash
Other solution is to add ->ct_net pointer to tuplehashes and still has one
hash, I tried that it's ugly and requires more code deep down in protocol
modules et al.* propagate netns pointer to where needed, e. g. to conntrack iterators.
Signed-off-by: Alexey Dobriyan
Signed-off-by: Patrick McHardy