With simple extension to the binding mechanism, which allows to bind more
than 64k sockets (or smaller amount, depending on sysctl parameters),
we have to traverse the whole bind hash table to find out empty bucket.
And while it is not a problem for example for 32k connections, bind()
completion time grows exponentially (since after each successful binding
we have to traverse one bucket more to find empty one) even if we start
each time from random offset inside the hash table.
So, when hash table is full, and we want to add another socket, we have
to traverse the whole table no matter what, so effectivelly this will be
the worst case performance and it will be constant.
Attached picture shows bind() time depending on number of already bound
sockets.
Green area corresponds to the usual binding to zero port process, which
turns on kernel port selection as described above. Red area is the bind
process, when number of reuse-bound sockets is not limited by 64k (or
sysctl parameters). The same exponential growth (hidden by the green
area) before number of ports reaches sysctl limit.
At this time bind hash table has exactly one reuse-enbaled socket in a
bucket, but it is possible that they have different addresses. Actually
kernel selects the first port to try randomly, so at the beginning bind
will take roughly constant time, but with time number of port to check
after random start will increase. And that will have exponential growth,
but because of above random selection, not every next port selection
will necessary take longer time than previous. So we have to consider
the area below in the graph (if you could zoom it, you could find, that
there are many different times placed there), so area can hide another.
Blue area corresponds to the port selection optimization.
This is rather simple design approach: hashtable now maintains (unprecise
and racely updated) number of currently bound sockets, and when number
of such sockets becomes greater than predefined value (I use maximum
port range defined by sysctls), we stop traversing the whole bind hash
table and just stop at first matching bucket after random start. Above
limit roughly corresponds to the case, when bind hash table is full and
we turned on mechanism of allowing to bind more reuse-enabled sockets,
so it does not change behaviour of other sockets.
Signed-off-by: Evgeniy Polyakov <zbr@ioremap.net>
Tested-by: Denys Fedoryschenko <denys@visp.net.lb>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since all feature-negotiation processing now takes place in feat.c,
functions for producing verbose debugging output are concentrated
there.
New functions to print out values, entry records, and options are
provided, and also a macro is defined to not always have the function
name in the output line.
Thanks a lot to Wei Yongjun and Giuseppe Galeota for help and
discussion with an earlier revision of this patch.
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Acked-by: Ian McDonald <ian.mcdonald@jandi.co.nz>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch takes care of initialising and type-checking sysctls
related to feature negotiation. Type checking is important since some
of the sysctls now directly impact the feature-negotiation process.
The sysctls are initialised with the known default values for each
feature. For the type-checking the value constraints from RFC 4340
are used:
* Sequence Window uses the specified Wmin=32, the maximum is ulong (4 bytes),
tested and confirmed that it works up to 4294967295 - for Gbps speed;
* Ack Ratio is between 0 .. 0xffff (2-byte unsigned integer);
* CCIDs are between 0 .. 255;
* request_retries, retries1, retries2 also between 0..255 for good measure;
* tx_qlen is checked to be non-negative;
* sync_ratelimit remains as before.
Notes:
------
1. Die s@sysctl_dccp_feat@sysctl_dccp@g since the sysctls are now in feat.c.
2. As pointed out by Arnaldo, the pattern of type-checking repeats itself in
other places, sometimes with exactly the same kind of definitions (e.g.
"static int zero;"). It may be a good idea (kernel janitors?) to consolidate
type checking. For the sake of keeping the changeset small and in order not
to affect other subsystems, I have not strived to generalise here.
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Acked-by: Ian McDonald <ian.mcdonald@jandi.co.nz>
Signed-off-by: David S. Miller <davem@davemloft.net>
This adds full support for local/remote Sequence Window feature, from which the
* sequence-number-validity (W) and
* acknowledgment-number-validity (W') windows
derive as specified in RFC 4340, 7.5.3.
Specifically, the following is contained in this patch:
* integrated new socket fields into dccp_sk;
* updated the update_gsr/gss routines with regard to these fields;
* updated handler code: the Sequence Window feature is located at the TX side,
so the local feature is meant if the handler-rx flag is false;
* the initialisation of `rcv_wnd' in reqsk is removed, since
- rcv_wnd is not used by the code anywhere;
- sequence number checks are not done in the LISTEN state (cf. 7.5.3);
- dccp_check_req checks the Ack number validity more rigorously;
* the `struct dccp_minisock' became empty and is now removed.
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Acked-by: Ian McDonald <ian.mcdonald@jandi.co.nz>
Signed-off-by: David S. Miller <davem@davemloft.net>
This initialises feature negotiation from two tables, which are in
turn are initialised from sysctls.
As a novel feature, specifics of the implementation (e.g. that short
seqnos and ECN are not yet available) are advertised for robustness.
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Acked-by: Ian McDonald <ian.mcdonald@jandi.co.nz>
Signed-off-by: David S. Miller <davem@davemloft.net>
With net_device_ops if set_mac_address is null, then error
is -EOPNOTSUPPORTED.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Acked-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Acked-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Acked-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Acked-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Now that stats are in net_device, use them.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Caused by call to request_module() while holding nf_conntrack_lock.
Reported-and-tested-by: Kövesdi György <kgy@teledigit.hu>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The previous fix to paged packets broke the merging because it
reset the skb->len before we added it to the merged packet. This
wasn't detected because it simply resulted in the truncation of
the packet while the missing bit is subsequently retransmitted.
The fix is to store skb->len before we clobber it.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
When a frag is shorter than an Ethernet header, we'd return a
zeroed packet instead of aborting. This patch fixes that.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
We need to perform skb_postpull_rcsum after pulling the IPv6
header in order to maintain the correctness of the complete
checksum.
This patch also adds a missing iph reload after pulling.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
register_pernet_gen_subsys omits mutex_unlock in one fail path.
Fix it.
Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This reverts commit fc8c7dc1b2.
As indicated by Jiri Klimes, this won't work. These numbers are
not only used the size validation, they are also used to locate
attributes sitting after the message.
Signed-off-by: David S. Miller <davem@davemloft.net>
The trick in socket splicing where we try to convert the skb->data
into a page based reference using virt_to_page() does not work so
well.
The idea is to pass the virt_to_page() reference via the pipe
buffer, and refcount the buffer using a SKB reference.
But if we are splicing from a socket to a socket (via sendpage)
this doesn't work.
The from side processing will grab the page (and SKB) references.
The sendpage() calls will grab page references only, return, and
then the from side processing completes and drops the SKB ref.
The page based reference to skb->data is not enough to keep the
kmalloc() buffer backing it from being reused. Yet, that is
all that the socket send side has at this point.
This leads to data corruption if the skb->data buffer is reused
by SLAB before the send side socket actually gets the TX packet
out to the device.
The fix employed here is to simply allocate a page and copy the
skb->data bytes into that page.
This will hurt performance, but there is no clear way to fix this
properly without a copy at the present time, and it is important
to get rid of the data corruption.
With fixes from Herbert Xu.
Tested-by: Willy Tarreau <w@1wt.eu>
Foreseen-by: Changli Gao <xiaosuo@gmail.com>
Diagnosed-by: Willy Tarreau <w@1wt.eu>
Reported-by: Willy Tarreau <w@1wt.eu>
Fixed-by: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
I'm trying to track down why people're hitting the checksum warning
in skb_gso_segment. As the problem seems to be hitting lots of
people and I can't reproduce it or locate the bug, here is a patch
to print out more details which hopefully should help us to track
this down.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
The country IE number of channels on 5 GHz specifies the number
of 5 GHz channels, not the number of sequential channel numbers.
For example, if in a country IEs if the first channel given is 36
and the number of channels passed is 4 then the individual channel
numbers defined for the 5 GHz PHY by these parameters
are: 36, 40, 44, 48
not: 36, 37, 38, 39
See: http://tinyurl.com/11d-clarification
Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
This fixes a regression on disallowing bands introduced with the new
802.11d support. The issue is that IEEE-802.11 allows APs to send
a subset of what a country regulatory domain defines. This was clarified
in this document:
http://tinyurl.com/11d-clarification
As such it is possible, and this is what is done in practice, that a
single band 2.4 GHz AP will only send 2.4 GHz band regulatory information
through the 802.11 country information element and then the current
intersection with what CRDA provided yields a regulatory domain with
no 5 GHz information -- even though that country may actually allow
5 GHz operation. We correct this by only applying the intersection rules
on a channel if the the intersection yields a regulatory rule on the
same band the channel is on.
Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
This allows us to make more wiphy specific judgements when
handling the channels later on.
Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Fix (delete) more mac80211 kernel-doc:
Warning(linux-2.6.28-git13//include/net/mac80211.h:375): Excess struct/union/enum/typedef member 'retry_count' description in 'ieee80211_tx_info'
Warning(linux-2.6.28-git13//net/mac80211/sta_info.h:308): Excess struct/union/enum/typedef member 'last_txrate' description in 'sta_info'
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (95 commits)
b44: GFP_DMA skb should not escape from driver
korina: do not use IRQF_SHARED with IRQF_DISABLED
korina: do not stop queue here
korina: fix handling tx_chain_tail
korina: do tx at the right position
korina: do schedule napi after testing for it
korina: rework korina_rx() for use with napi
korina: disable napi on close and restart
korina: reset resource buffer size to 1536
korina: fix usage of driver_data
bnx2x: First slow path interrupt race
bnx2x: MTU Filter
bnx2x: Indirection table initialization index
bnx2x: Missing brackets
bnx2x: Fixing the doorbell size
bnx2x: Endianness issues
bnx2x: VLAN tagged packets without VLAN offload
bnx2x: Protecting the link change indication
bnx2x: Flow control updated before reporting the link
bnx2x: Missing mask when calculating flow control
...
If INET=y and INFINIBAND=y, but IPV6=m then INFINIBAND_ADDR_TRANS is set
to n and the RDMA CM functions rdma_connect() et al are not built.
However, the current config dependencies allow NET_9P_RDMA to be selected
in this, which leads to a build failure. Fix this by adding a dependency
on INFINIBAND_ADDR_TRANS to disallow NET_9P_RDMA in this case.
Reported-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Acked-by: Randy Dunlap <randy.dunlap@oracle.com>
Tested-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Due to the loopback functionality in can_send() we can not invoke it
from hardirq context which was done inside the
bcm_tx_timeout_handler() hrtimer callback:
[ 700.361154] [<c012228c>] warn_slowpath+0x80/0xb6
[ 700.361163] [<c013d559>] valid_state+0x125/0x136
[ 700.361171] [<c013d858>] mark_lock+0x18e/0x332
[ 700.361180] [<c013e300>] __lock_acquire+0x12e/0xb1e
[ 700.361189] [<f8ab5915>] bcm_tx_timeout_handler+0x0/0xbc [can_bcm]
[ 700.361198] [<c031e20a>] dev_queue_xmit+0x191/0x479
[ 700.361206] [<c01262a7>] __local_bh_disable+0x2b/0x64
[ 700.361213] [<c031e20a>] dev_queue_xmit+0x191/0x479
[ 700.361225] [<f8aa69a1>] can_send+0xd7/0x11a [can]
[ 700.361235] [<f8ab522b>] bcm_can_tx+0x9d/0xd9 [can_bcm]
[ 700.361245] [<f8ab597f>] bcm_tx_timeout_handler+0x6a/0xbc [can_bcm]
[ 700.361255] [<f8ab5915>] bcm_tx_timeout_handler+0x0/0xbc [can_bcm]
[ 700.361263] [<c0134143>] __run_hrtimer+0x5a/0x86
[ 700.361273] [<f8ab5915>] bcm_tx_timeout_handler+0x0/0xbc [can_bcm]
[ 700.361282] [<c0134a50>] hrtimer_interrupt+0xb9/0x110
This patch moves the rest of the functionality from the hrtimer
callback to the already existing tasklet to fix this slowpath problem.
Signed-off-by: Oliver Hartkopp <oliver@hartkopp.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This adds an init_dummy_netdev() function that gets a network device
structure (allocation and lifetime entirely under caller's control) and
initialize the minimum amount of fields so it can be used to schedule
NAPI polls without registering a full blown interface. This is to be
used by drivers that need to tie several hardware interfaces to a single
NAPI poll scheduler due to HW limitations.
It also updates the ibm_newemac driver to use that, this fixing the
oops on 2.6.29 due to passing NULL as "dev" to netif_napi_add()
Symbol is exported GPL only a I don't think we want binary drivers doing
that sort of acrobatics (if we want them at all).
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Tested-by: Geert Uytterhoeven <Geert.Uytterhoeven@sonycom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When we get a GSO packet from an untrusted source, we need to
ensure that it is sufficiently long so that we don't end up
crashing.
Based on discovery and patch by Ian Campbell.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Tested-by: Ian Campbell <ian.campbell@citrix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When an skb with page frags is merged into an existing one, we
cannibalise its reference count. This is OK when the skb is
reused because we set nr_frags to zero in that case. However,
for the case where the skb is freed through kfree_skb, we didn't
clear nr_frags which causes the page to be freed prematurely.
This is fixed by moving the skb resetting into skb_gro_receive.
Reported-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
As GRO cannot be applied to packets with frag_list we need to
make sure that we reject such packets if they are fed to us,
e.g., through a tunnel device.
Also there is no point in applying GRO on GSO packets so they
too should be rejected. This allows GRO to be used in virtio-net
which may produce GSO packets directly but may still benefit
from GRO if the other end of it doesn't support GSO.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
When a fib6 table dump is prematurely ended, we won't unlink
its walker from the list. This causes all sorts of grief for
other users of the list later.
Reported-by: Chris Caputo <ccaputo@alt.net>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
As spotted by Willy Tarreau, current splice() from tcp socket to pipe is not
optimal. It processes at most one segment per call.
This results in low performance and very high overhead due to syscall rate
when splicing from interfaces which do not support LRO.
Willy provided a patch inside tcp_splice_read(), but a better fix
is to let tcp_read_sock() process as many segments as possible, so
that tcp_rcv_space_adjust() and tcp_cleanup_rbuf() are called less
often.
With this change, splice() behaves like tcp_recvmsg(), being able
to consume many skbs in one system call. With typical 1460 bytes
of payload per frame, that means splice(SPLICE_F_NONBLOCK) can return
16*1460 = 23360 bytes.
Signed-off-by: Willy Tarreau <w@1wt.eu>
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
net/mac80211/ht.c: In function ‘ieee80211_start_tx_ba_session’:
net/mac80211/ht.c:472: warning: ‘ret’ may be used uninitialized in this function
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Currently htb_do_events() breaks events recounting for a level after 2
jiffies, but there is no reason to repeat this for next levels and
increase delays even more (with softirqs disabled). htb_dequeue_tree()
can add to this too, btw. In such a case q->now time is invalid anyway.
Thanks to Patrick McHardy for spotting an error around earlier version
of this patch.
Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Next event time should consider jiffies used for recounting. Otherwise
qdisc_watchdog_schedule() triggers hrtimer immediately with the event
in the past, and may cause very high ksoftirqd cpu usage (if highres
is on).
There is also removed checking "event" for zero in htb_dequeue(): it's
always true in this place.
Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
netfilter: xt_time: print timezone for user information
Let users have a way to figure out if their distro set the kernel
timezone at all.
Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
nf_conntrack_alloc cannot return NULL, so there is no need to check for
NULL before using the value. I have also removed the initialization of ct
to NULL in nf_conntrack_alloc, since the value is never used, and since
perhaps it might lead one to think that return ct at the end might return
NULL.
The semantic patch that finds this problem is as follows:
(http://www.emn.fr/x-info/coccinelle/)
// <smpl>
@match exists@
expression x, E;
position p1,p2;
statement S1, S2;
@@
x@p1 = nf_conntrack_alloc(...)
... when != x = E
(
if (x@p2 == NULL || ...) S1 else S2
|
if (x@p2 == NULL && ...) S1 else S2
)
@other_match exists@
expression match.x, E1, E2;
position p1!=match.p1,match.p2;
@@
x@p1 = E1
... when != x = E2
x@p2
@ script:python depends on !other_match@
p1 << match.p1;
p2 << match.p2;
@@
print "%s: call to nf_conntrack_alloc %s bad test %s" % (p1[0].file,p1[0].line,p2[0].line)
// </smpl>
Signed-off-by: Julia Lawall <julia@diku.dk>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
An old bug crept back into the ICMP/ICMPv6 conntrack protocols: the timeout
values are defined as unsigned longs, the sysctl's maxsize is set to
sizeof(unsigned int). Use unsigned int for the timeout values as in the
other conntrack protocols.
Reported-by: Jean-Mickael Guerin <jean-mickael.guerin@6wind.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit 8cc784ee (netfilter: change return types of match functions
for ebtables extensions) broke ebtables matches by inverting the
sense of match/nomatch.
Reported-by: Matt Cross <matthltc@us.ibm.com>
Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Evgeniy Polyakov