Main difference between napi_frags_skb() and napi_gro_receive() is that
the later is called while ethernet header was already pulled by the NIC
driver (eth_type_trans() was called before napi_gro_receive())
Jerry Chu in commit 299603e837 ("net-gro: Prepare GRO stack for the
upcoming tunneling support") tried to remove this difference by calling
eth_type_trans() from napi_frags_skb() instead of doing this later from
napi_frags_finish()
Goal was that napi_gro_complete() could call
ptype->callbacks.gro_complete(skb, 0) (offset of first network header =
0)
Also, xxx_gro_receive() handlers all use off = skb_gro_offset(skb) to
point to their own header, for the current skb and ones held in gro_list
Problem is this cleanup work defeated the frag0 optimization:
It turns out the consecutive pskb_may_pull() calls are too expensive.
This patch brings back the frag0 stuff in napi_frags_skb().
As all skb have their mac header in skb head, we no longer need
skb_gro_mac_header()
Reported-by: Michal Schmidt <mschmidt@redhat.com>
Fixes: 299603e837 ("net-gro: Prepare GRO stack for the upcoming tunneling support")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Jerry Chu <hkchu@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Binding might result in a NULL device which is later dereferenced
without checking.
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This allows to monitor carrier on/off transitions and detect link
flapping issues:
- new /sys/class/net/X/carrier_changes
- new rtnetlink IFLA_CARRIER_CHANGES (getlink)
Tested:
- grep . /sys/class/net/*/carrier_changes
+ ip link set dev X down/up
+ plug/unplug cable
- updated iproute2: prints IFLA_CARRIER_CHANGES
- iproute2 20121211-2 (debian): unchanged behavior
Signed-off-by: David Decotigny <decot@googlers.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The issue raises when adding policy route, specify a particular
NIC as oif, the policy route did not take effect. The reason is
that fl6.oif is not set and route map failed. From the
tcp_v6_send_response function, if the binding address is linklocal,
fl6.oif is set, but not for global address.
Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: Wang Yufen <wangyufen@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Move the whole rt6_need_strict as static inline into ip6_route.h,
so that it can be reused
Signed-off-by: Wang Yufen <wangyufen@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use existing function instead of trying to use our own.
Signed-off-by: Vlad Yasevich <vyasevic@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch replaces/reworks the kernel-internal BPF interpreter with
an optimized BPF instruction set format that is modelled closer to
mimic native instruction sets and is designed to be JITed with one to
one mapping. Thus, the new interpreter is noticeably faster than the
current implementation of sk_run_filter(); mainly for two reasons:
1. Fall-through jumps:
BPF jump instructions are forced to go either 'true' or 'false'
branch which causes branch-miss penalty. The new BPF jump
instructions have only one branch and fall-through otherwise,
which fits the CPU branch predictor logic better. `perf stat`
shows drastic difference for branch-misses between the old and
new code.
2. Jump-threaded implementation of interpreter vs switch
statement:
Instead of single table-jump at the top of 'switch' statement,
gcc will now generate multiple table-jump instructions, which
helps CPU branch predictor logic.
Note that the verification of filters is still being done through
sk_chk_filter() in classical BPF format, so filters from user- or
kernel space are verified in the same way as we do now, and same
restrictions/constraints hold as well.
We reuse current BPF JIT compilers in a way that this upgrade would
even be fine as is, but nevertheless allows for a successive upgrade
of BPF JIT compilers to the new format.
The internal instruction set migration is being done after the
probing for JIT compilation, so in case JIT compilers are able to
create a native opcode image, we're going to use that, and in all
other cases we're doing a follow-up migration of the BPF program's
instruction set, so that it can be transparently run in the new
interpreter.
In short, the *internal* format extends BPF in the following way (more
details can be taken from the appended documentation):
- Number of registers increase from 2 to 10
- Register width increases from 32-bit to 64-bit
- Conditional jt/jf targets replaced with jt/fall-through
- Adds signed > and >= insns
- 16 4-byte stack slots for register spill-fill replaced
with up to 512 bytes of multi-use stack space
- Introduction of bpf_call insn and register passing convention
for zero overhead calls from/to other kernel functions
- Adds arithmetic right shift and endianness conversion insns
- Adds atomic_add insn
- Old tax/txa insns are replaced with 'mov dst,src' insn
Performance of two BPF filters generated by libpcap resp. bpf_asm
was measured on x86_64, i386 and arm32 (other libpcap programs
have similar performance differences):
fprog #1 is taken from Documentation/networking/filter.txt:
tcpdump -i eth0 port 22 -dd
fprog #2 is taken from 'man tcpdump':
tcpdump -i eth0 'tcp port 22 and (((ip[2:2] - ((ip[0]&0xf)<<2)) -
((tcp[12]&0xf0)>>2)) != 0)' -dd
Raw performance data from BPF micro-benchmark: SK_RUN_FILTER on the
same SKB (cache-hit) or 10k SKBs (cache-miss); time in ns per call,
smaller is better:
--x86_64--
fprog #1 fprog #1 fprog #2 fprog #2
cache-hit cache-miss cache-hit cache-miss
old BPF 90 101 192 202
new BPF 31 71 47 97
old BPF jit 12 34 17 44
new BPF jit TBD
--i386--
fprog #1 fprog #1 fprog #2 fprog #2
cache-hit cache-miss cache-hit cache-miss
old BPF 107 136 227 252
new BPF 40 119 69 172
--arm32--
fprog #1 fprog #1 fprog #2 fprog #2
cache-hit cache-miss cache-hit cache-miss
old BPF 202 300 475 540
new BPF 180 270 330 470
old BPF jit 26 182 37 202
new BPF jit TBD
Thus, without changing any userland BPF filters, applications on
top of AF_PACKET (or other families) such as libpcap/tcpdump, cls_bpf
classifier, netfilter's xt_bpf, team driver's load-balancing mode,
and many more will have better interpreter filtering performance.
While we are replacing the internal BPF interpreter, we also need
to convert seccomp BPF in the same step to make use of the new
internal structure since it makes use of lower-level API details
without being further decoupled through higher-level calls like
sk_unattached_filter_{create,destroy}(), for example.
Just as for normal socket filtering, also seccomp BPF experiences
a time-to-verdict speedup:
05-sim-long_jumps.c of libseccomp was used as micro-benchmark:
seccomp_rule_add_exact(ctx,...
seccomp_rule_add_exact(ctx,...
rc = seccomp_load(ctx);
for (i = 0; i < 10000000; i++)
syscall(199, 100);
'short filter' has 2 rules
'large filter' has 200 rules
'short filter' performance is slightly better on x86_64/i386/arm32
'large filter' is much faster on x86_64 and i386 and shows no
difference on arm32
--x86_64-- short filter
old BPF: 2.7 sec
39.12% bench libc-2.15.so [.] syscall
8.10% bench [kernel.kallsyms] [k] sk_run_filter
6.31% bench [kernel.kallsyms] [k] system_call
5.59% bench [kernel.kallsyms] [k] trace_hardirqs_on_caller
4.37% bench [kernel.kallsyms] [k] trace_hardirqs_off_caller
3.70% bench [kernel.kallsyms] [k] __secure_computing
3.67% bench [kernel.kallsyms] [k] lock_is_held
3.03% bench [kernel.kallsyms] [k] seccomp_bpf_load
new BPF: 2.58 sec
42.05% bench libc-2.15.so [.] syscall
6.91% bench [kernel.kallsyms] [k] system_call
6.25% bench [kernel.kallsyms] [k] trace_hardirqs_on_caller
6.07% bench [kernel.kallsyms] [k] __secure_computing
5.08% bench [kernel.kallsyms] [k] sk_run_filter_int_seccomp
--arm32-- short filter
old BPF: 4.0 sec
39.92% bench [kernel.kallsyms] [k] vector_swi
16.60% bench [kernel.kallsyms] [k] sk_run_filter
14.66% bench libc-2.17.so [.] syscall
5.42% bench [kernel.kallsyms] [k] seccomp_bpf_load
5.10% bench [kernel.kallsyms] [k] __secure_computing
new BPF: 3.7 sec
35.93% bench [kernel.kallsyms] [k] vector_swi
21.89% bench libc-2.17.so [.] syscall
13.45% bench [kernel.kallsyms] [k] sk_run_filter_int_seccomp
6.25% bench [kernel.kallsyms] [k] __secure_computing
3.96% bench [kernel.kallsyms] [k] syscall_trace_exit
--x86_64-- large filter
old BPF: 8.6 seconds
73.38% bench [kernel.kallsyms] [k] sk_run_filter
10.70% bench libc-2.15.so [.] syscall
5.09% bench [kernel.kallsyms] [k] seccomp_bpf_load
1.97% bench [kernel.kallsyms] [k] system_call
new BPF: 5.7 seconds
66.20% bench [kernel.kallsyms] [k] sk_run_filter_int_seccomp
16.75% bench libc-2.15.so [.] syscall
3.31% bench [kernel.kallsyms] [k] system_call
2.88% bench [kernel.kallsyms] [k] __secure_computing
--i386-- large filter
old BPF: 5.4 sec
new BPF: 3.8 sec
--arm32-- large filter
old BPF: 13.5 sec
73.88% bench [kernel.kallsyms] [k] sk_run_filter
10.29% bench [kernel.kallsyms] [k] vector_swi
6.46% bench libc-2.17.so [.] syscall
2.94% bench [kernel.kallsyms] [k] seccomp_bpf_load
1.19% bench [kernel.kallsyms] [k] __secure_computing
0.87% bench [kernel.kallsyms] [k] sys_getuid
new BPF: 13.5 sec
76.08% bench [kernel.kallsyms] [k] sk_run_filter_int_seccomp
10.98% bench [kernel.kallsyms] [k] vector_swi
5.87% bench libc-2.17.so [.] syscall
1.77% bench [kernel.kallsyms] [k] __secure_computing
0.93% bench [kernel.kallsyms] [k] sys_getuid
BPF filters generated by seccomp are very branchy, so the new
internal BPF performance is better than the old one. Performance
gains will be even higher when BPF JIT is committed for the
new structure, which is planned in future work (as successive
JIT migrations).
BPF has also been stress-tested with trinity's BPF fuzzer.
Joint work with Daniel Borkmann.
Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Cc: Hagen Paul Pfeifer <hagen@jauu.net>
Cc: Kees Cook <keescook@chromium.org>
Cc: Paul Moore <pmoore@redhat.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: H. Peter Anvin <hpa@linux.intel.com>
Cc: linux-kernel@vger.kernel.org
Acked-by: Kees Cook <keescook@chromium.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
There are currently pch_gbe, cpts, and ixp4xx_eth drivers that open-code
and reimplement a BPF classifier for the PTP protocol. Since all of them
effectively do the very same thing and load the very same PTP/BPF filter,
we can just consolidate that code by introducing ptp_classify_raw() in
the time-stamping core framework which can be used in drivers.
As drivers get initialized after bootstrapping the core networking
subsystem, they can make use of ptp_insns wrapped through
ptp_classify_raw(), which allows to simplify and remove PTP classifier
setup code in drivers.
Joint work with Alexei Starovoitov.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Cc: Richard Cochran <richard.cochran@omicron.at>
Cc: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch migrates an open-coded sk_run_filter() implementation with
proper use of the BPF API, that is, sk_unattached_filter_create(). This
migration is needed, as we will be internally transforming the filter
to a different representation, and therefore needs to be decoupled.
It is okay to do so as skb_timestamping_init() is called during
initialization of the network stack in core initcall via sock_init().
This would effectively also allow for PTP filters to be jit compiled if
bpf_jit_enable is set.
For better readability, there are also some newlines introduced, also
ptp_classify.h is only in kernel space.
Joint work with Alexei Starovoitov.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Cc: Richard Cochran <richard.cochran@omicron.at>
Cc: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch basically does two things, i) removes the extern keyword
from the include/linux/filter.h file to be more consistent with the
rest of Joe's changes, and ii) moves filter accounting into the filter
core framework.
Filter accounting mainly done through sk_filter_{un,}charge() take
care of the case when sockets are being cloned through sk_clone_lock()
so that removal of the filter on one socket won't result in eviction
as it's still referenced by the other.
These functions actually belong to net/core/filter.c and not
include/net/sock.h as we want to keep all that in a central place.
It's also not in fast-path so uninlining them is fine and even allows
us to get rd of sk_filter_release_rcu()'s EXPORT_SYMBOL and a forward
declaration.
Joint work with Alexei Starovoitov.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Cc: Pavel Emelyanov <xemul@parallels.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In order to open up the possibility to internally transform a BPF program
into an alternative and possibly non-trivial reversible representation, we
need to keep the original BPF program around, so that it can be passed back
to user space w/o the need of a complex decoder.
The reason for that use case resides in commit a8fc927780 ("sk-filter:
Add ability to get socket filter program (v2)"), that is, the ability
to retrieve the currently attached BPF filter from a given socket used
mainly by the checkpoint-restore project, for example.
Therefore, we add two helpers sk_{store,release}_orig_filter for taking
care of that. In the sk_unattached_filter_create() case, there's no such
possibility/requirement to retrieve a loaded BPF program. Therefore, we
can spare us the work in that case.
This approach will simplify and slightly speed up both, sk_get_filter()
and sock_diag_put_filterinfo() handlers as we won't need to successively
decode filters anymore through sk_decode_filter(). As we still need
sk_decode_filter() later on, we're keeping it around.
Joint work with Alexei Starovoitov.
Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Cc: Pavel Emelyanov <xemul@parallels.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds a jited flag into sk_filter struct in order to indicate
whether a filter is currently jited or not. The size of sk_filter is
not being expanded as the 32 bit 'len' member allows upper bits to be
reused since a filter can currently only grow as large as BPF_MAXINSNS.
Therefore, there's enough room also for other in future needed flags to
reuse 'len' field if necessary. The jited flag also allows for having
alternative interpreter functions running as currently, we can only
detect jit compiled filters by testing fp->bpf_func to not equal the
address of sk_run_filter().
Joint work with Alexei Starovoitov.
Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Cc: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Conflicts:
drivers/net/ethernet/marvell/mvneta.c
The mvneta.c conflict is a case of overlapping changes,
a conversion to devm_ioremap_resource() vs. a conversion
to netdev_alloc_pcpu_stats.
Signed-off-by: David S. Miller <davem@davemloft.net>
ERROR: "(foo*)" should be "(foo *)"
ERROR: "foo * bar" should be "foo *bar"
Suggested-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: Wang Yufen <wangyufen@huawei.com>
Acked-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
ERROR: open brace '{' following enum go on the same line
ERROR: open brace '{' following struct go on the same line
ERROR: trailing statements should be on next line
Signed-off-by: Wang Yufen <wangyufen@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
WARNING: please, no space before tabs
WARNING: please, no spaces at the start of a line
ERROR: spaces required around that ':' (ctx:VxW)
ERROR: spaces required around that '>' (ctx:VxV)
ERROR: spaces required around that '>=' (ctx:VxV)
Signed-off-by: Wang Yufen <wangyufen@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Stop taking the transmit lock when a network device has specified
NETIF_F_LLTX.
If no locks needed to trasnmit a packet this is the ideal scenario for
netpoll as all packets can be trasnmitted immediately.
Even if some locks are needed in ndo_start_xmit skipping any unnecessary
serialization is desirable for netpoll as it makes it more likely a
debugging packet may be trasnmitted immediately instead of being
deferred until later.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove the assumption that the skbs that make it to
netpoll_send_skb_on_dev are allocated with find_skb, such that
skb->users == 1 and nothing is attached that would prevent the skbs from
being freed from hard irq context.
Remove this assumption by replacing __kfree_skb on error paths with
dev_kfree_skb_irq (in hard irq context) and kfree_skb (in process
context).
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The netpoll_rx_enable and netpoll_rx_disable functions have always
controlled polling the network drivers transmit and receive queues.
Rename them to netpoll_poll_enable and netpoll_poll_disable to make
their functionality clear.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Today netpoll_rx_enable and netpoll_rx_disable are called from
dev_close and and __dev_close, and not from dev_close_many.
Move the calls into __dev_close_many so that we have a single call
site to maintain, and so that dev_close_many gains this protection as
well. Which importantly makes batched network device deletes safe.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Factor out the code that needs to surround ndo_start_xmit
from netpoll_send_skb_on_dev into netpoll_start_xmit.
It is an unfortunate fact that as the netpoll code has been maintained
the primary call site ndo_start_xmit learned how to handle vlans
and timestamps but the second call of ndo_start_xmit in queue_process
did not.
With the introduction of netpoll_start_xmit this associated logic now
happens at both call sites of ndo_start_xmit and should make it easy
for that to continue into the future.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The gfp parameter was added in:
commit 47be03a28c
Author: Amerigo Wang <amwang@redhat.com>
Date: Fri Aug 10 01:24:37 2012 +0000
netpoll: use GFP_ATOMIC in slave_enable_netpoll() and __netpoll_setup()
slave_enable_netpoll() and __netpoll_setup() may be called
with read_lock() held, so should use GFP_ATOMIC to allocate
memory. Eric suggested to pass gfp flags to __netpoll_setup().
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: "David S. Miller" <davem@davemloft.net>
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Cong Wang <amwang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The reason for the gfp parameter was removed in:
commit c4cdef9b71
Author: dingtianhong <dingtianhong@huawei.com>
Date: Tue Jul 23 15:25:27 2013 +0800
bonding: don't call slave_xxx_netpoll under spinlocks
The slave_xxx_netpoll will call synchronize_rcu_bh(),
so the function may schedule and sleep, it should't be
called under spinlocks.
bond_netpoll_setup() and bond_netpoll_cleanup() are always
protected by rtnl lock, it is no need to take the read lock,
as the slave list couldn't be changed outside rtnl lock.
Signed-off-by: Ding Tianhong <dingtianhong@huawei.com>
Cc: Jay Vosburgh <fubar@us.ibm.com>
Cc: Andy Gospodarek <andy@greyhouse.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Nothing else that calls __netpoll_setup or ndo_netpoll_setup
requires a gfp paramter, so remove the gfp parameter from both
of these functions making the code clearer.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Some drivers incorrectly assign vlan acceleration features to
vlan_features thus causing issues for Q-in-Q vlan configurations.
Warn the user of such cases.
Signed-off-by: Vlad Yasevich <vyasevic@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When the vlan filtering is enabled on the bridge, but
the filter is not configured on the bridge device itself,
running tcpdump on the bridge device will result in a
an Oops with NULL pointer dereference. The reason
is that br_pass_frame_up() will bypass the vlan
check because promisc flag is set. It will then try
to get the table pointer and process the packet based
on the table. Since the table pointer is NULL, we oops.
Catch this special condition in br_handle_vlan().
Reported-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp>
CC: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp>
Signed-off-by: Vlad Yasevich <vyasevic@redhat.com>
Acked-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp>
Signed-off-by: David S. Miller <davem@davemloft.net>
skb_network_protocol() already accounts for multiple vlan
headers that may be present in the skb. However, skb_mac_gso_segment()
doesn't know anything about it and assumes that skb->mac_len
is set correctly to skip all mac headers. That may not
always be the case. If we are simply forwarding the packet (via
bridge or macvtap), all vlan headers may not be accounted for.
A simple solution is to allow skb_network_protocol to return
the vlan depth it has calculated. This way skb_mac_gso_segment
will correctly skip all mac headers.
Signed-off-by: Vlad Yasevich <vyasevic@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
addrconf_join_solict and addrconf_join_anycast may cause actions which
need rtnl locked, especially on first address creation.
A new DAD state is introduced which defers processing of the initial
DAD processing into a workqueue.
To get rtnl lock we need to push the code paths which depend on those
calls up to workqueues, specifically addrconf_verify and the DAD
processing.
(v2)
addrconf_dad_failure needs to be queued up to the workqueue, too. This
patch introduces a new DAD state and stop the DAD processing in the
workqueue (this is because of the possible ipv6_del_addr processing
which removes the solicited multicast address from the device).
addrconf_verify_lock is removed, too. After the transition it is not
needed any more.
As we are not processing in bottom half anymore we need to be a bit more
careful about disabling bottom half out when we lock spin_locks which are also
used in bh.
Relevant backtrace:
[ 541.030090] RTNL: assertion failed at net/core/dev.c (4496)
[ 541.031143] CPU: 0 PID: 0 Comm: swapper/0 Tainted: G O 3.10.33-1-amd64-vyatta #1
[ 541.031145] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2007
[ 541.031146] ffffffff8148a9f0 000000000000002f ffffffff813c98c1 ffff88007c4451f8
[ 541.031148] 0000000000000000 0000000000000000 ffffffff813d3540 ffff88007fc03d18
[ 541.031150] 0000880000000006 ffff88007c445000 ffffffffa0194160 0000000000000000
[ 541.031152] Call Trace:
[ 541.031153] <IRQ> [<ffffffff8148a9f0>] ? dump_stack+0xd/0x17
[ 541.031180] [<ffffffff813c98c1>] ? __dev_set_promiscuity+0x101/0x180
[ 541.031183] [<ffffffff813d3540>] ? __hw_addr_create_ex+0x60/0xc0
[ 541.031185] [<ffffffff813cfe1a>] ? __dev_set_rx_mode+0xaa/0xc0
[ 541.031189] [<ffffffff813d3a81>] ? __dev_mc_add+0x61/0x90
[ 541.031198] [<ffffffffa01dcf9c>] ? igmp6_group_added+0xfc/0x1a0 [ipv6]
[ 541.031208] [<ffffffff8111237b>] ? kmem_cache_alloc+0xcb/0xd0
[ 541.031212] [<ffffffffa01ddcd7>] ? ipv6_dev_mc_inc+0x267/0x300 [ipv6]
[ 541.031216] [<ffffffffa01c2fae>] ? addrconf_join_solict+0x2e/0x40 [ipv6]
[ 541.031219] [<ffffffffa01ba2e9>] ? ipv6_dev_ac_inc+0x159/0x1f0 [ipv6]
[ 541.031223] [<ffffffffa01c0772>] ? addrconf_join_anycast+0x92/0xa0 [ipv6]
[ 541.031226] [<ffffffffa01c311e>] ? __ipv6_ifa_notify+0x11e/0x1e0 [ipv6]
[ 541.031229] [<ffffffffa01c3213>] ? ipv6_ifa_notify+0x33/0x50 [ipv6]
[ 541.031233] [<ffffffffa01c36c8>] ? addrconf_dad_completed+0x28/0x100 [ipv6]
[ 541.031241] [<ffffffff81075c1d>] ? task_cputime+0x2d/0x50
[ 541.031244] [<ffffffffa01c38d6>] ? addrconf_dad_timer+0x136/0x150 [ipv6]
[ 541.031247] [<ffffffffa01c37a0>] ? addrconf_dad_completed+0x100/0x100 [ipv6]
[ 541.031255] [<ffffffff8105313a>] ? call_timer_fn.isra.22+0x2a/0x90
[ 541.031258] [<ffffffffa01c37a0>] ? addrconf_dad_completed+0x100/0x100 [ipv6]
Hunks and backtrace stolen from a patch by Stephen Hemminger.
Reported-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Dropping packets in __dev_queue_xmit() when transmit queue
is stopped (NIC TX ring buffer full or BQL limit reached) currently
outputs a syslog message.
It would be better to get a precise count of such events available in
netdevice stats so that monitoring tools can have a clue.
This extends the work done in caf586e5f2
("net: add a core netdev->rx_dropped counter")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Quite often it can be useful to test with dummy or similar
devices as a blackhole sink for skbs. Such devices are only
equipped with a single txq, but marked as NETIF_F_LLTX as
they do not require locking their internal queues on xmit
(or implement locking themselves). Therefore, rather use
HARD_TX_{UN,}LOCK API, so that NETIF_F_LLTX will be respected.
trafgen mmap/TX_RING example against dummy device with config
foo: { fill(0xff, 64) } results in the following performance
improvements for such scenarios on an ordinary Core i7/2.80GHz:
Before:
Performance counter stats for 'trafgen -i foo -o du0 -n100000000' (10 runs):
160,975,944,159 instructions:k # 0.55 insns per cycle ( +- 0.09% )
293,319,390,278 cycles:k # 0.000 GHz ( +- 0.35% )
192,501,104 branch-misses:k ( +- 1.63% )
831 context-switches:k ( +- 9.18% )
7 cpu-migrations:k ( +- 7.40% )
69,382 cache-misses:k # 0.010 % of all cache refs ( +- 2.18% )
671,552,021 cache-references:k ( +- 1.29% )
22.856401569 seconds time elapsed ( +- 0.33% )
After:
Performance counter stats for 'trafgen -i foo -o du0 -n100000000' (10 runs):
133,788,739,692 instructions:k # 0.92 insns per cycle ( +- 0.06% )
145,853,213,256 cycles:k # 0.000 GHz ( +- 0.17% )
59,867,100 branch-misses:k ( +- 4.72% )
384 context-switches:k ( +- 3.76% )
6 cpu-migrations:k ( +- 6.28% )
70,304 cache-misses:k # 0.077 % of all cache refs ( +- 1.73% )
90,879,408 cache-references:k ( +- 1.35% )
11.719372413 seconds time elapsed ( +- 0.24% )
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Cc: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
It seems I missed one change in get_timewait4_sock() to compute
the remaining time before deletion of IPV4 timewait socket.
This could result in wrong output in /proc/net/tcp for tm->when field.
Fixes: 96f817fede ("tcp: shrink tcp6_timewait_sock by one cache line")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If a bridge with vlan_filtering enabled receives frames with stacked
vlan tags, i.e., they have two vlan tags, br_vlan_untag() strips not
only the outer tag but also the inner tag.
br_vlan_untag() is called only from br_handle_vlan(), and in this case,
it is enough to set skb->vlan_tci to 0 here, because vlan_tci has already
been set before calling br_handle_vlan().
Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp>
Acked-by: Vlad Yasevich <vyasevic@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Bridge vlan code (br_vlan_get_tag()) assumes that all frames have vlan_tci
if they are tagged, but if vlan tx offload is manually disabled on bridge
device and frames are sent from vlan device on the bridge device, the tags
are embedded in skb->data and they break this assumption.
Extract embedded vlan tags and move them to vlan_tci at ingress.
Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp>
Acked-by: Vlad Yasevich <vyasevic@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The node discovery domain is assigned when a bearer is enabled.
In the previous commit we reflect this attribute directly in the
bearer structure since it's needed to reinitialize the node
discovery mechanism after a hardware address change.
There's no need to replicate this attribute anywhere else, so we
remove it from the tipc_link_req structure.
Signed-off-by: Erik Hugne <erik.hugne@ericsson.com>
Reviewed-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If the hardware address of a underlying netdevice is changed, it is
not enough to simply reset the bearer/links over this device. We
also need to reflect this change in the TIPC bearer and node
discovery structures aswell.
This patch adds the necessary reinitialization of the node disovery
mechanism following a hardware address change so that the correct
originating media address is advertised in the discovery messages.
Signed-off-by: Erik Hugne <erik.hugne@ericsson.com>
Reported-by: Dong Liu <dliu.cn@gmail.com>
Reviewed-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
skb_zerocopy can copy elements of the frags array between skbs, but it doesn't
orphan them. Also, it doesn't handle errors, so this patch takes care of that
as well, and modify the callers accordingly. skb_tx_error() is also added to
the callers so they will signal the failed delivery towards the creator of the
skb.
Signed-off-by: Zoltan Kiss <zoltan.kiss@citrix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use del_timer_sync to ensure that the timer is stopped on all CPUs before
the driver exists.
This change was suggested by Thomas Gleixner.
The semantic patch that makes this change is as follows:
(http://coccinelle.lip6.fr/)
// <smpl>
@r@
declarer name module_exit;
identifier ex;
@@
module_exit(ex);
@@
identifier r.ex;
@@
ex(...) {
<...
- del_timer
+ del_timer_sync
(...)
...>
}
// </smpl>
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use del_timer_sync to ensure that the timer is stopped on all CPUs before
the driver exists.
This change was suggested by Thomas Gleixner.
The semantic patch that makes this change is as follows:
(http://coccinelle.lip6.fr/)
// <smpl>
@r@
declarer name module_exit;
identifier ex;
@@
module_exit(ex);
@@
identifier r.ex;
@@
ex(...) {
<...
- del_timer
+ del_timer_sync
(...)
...>
}
// </smpl>
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
There is no need to allocate 15 bytes in excess for a SYNACK packet,
as it contains no data, only headers.
SYNACK are always generated in softirq context, and contain a single
segment, we can use TCP_INC_STATS_BH()
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If an IPv6 host route with metrics exists, an attempt to add a
new route for the same target with different metrics fails but
rewrites the metrics anyway:
12sp0:~ # ip route add fec0::1 dev eth0 rto_min 1000
12sp0:~ # ip -6 route show
fe80::/64 dev eth0 proto kernel metric 256
fec0::1 dev eth0 metric 1024 rto_min lock 1s
12sp0:~ # ip route add fec0::1 dev eth0 rto_min 1500
RTNETLINK answers: File exists
12sp0:~ # ip -6 route show
fe80::/64 dev eth0 proto kernel metric 256
fec0::1 dev eth0 metric 1024 rto_min lock 1.5s
This is caused by all IPv6 host routes using the metrics in
their inetpeer (or the shared default). This also holds for the
new route created in ip6_route_add() which shares the metrics
with the already existing route and thus ip6_route_add()
rewrites the metrics even if the new route ends up not being
used at all.
Another problem is that old metrics in inetpeer can reappear
unexpectedly for a new route, e.g.
12sp0:~ # ip route add fec0::1 dev eth0 rto_min 1000
12sp0:~ # ip route del fec0::1
12sp0:~ # ip route add fec0::1 dev eth0
12sp0:~ # ip route change fec0::1 dev eth0 hoplimit 10
12sp0:~ # ip -6 route show
fe80::/64 dev eth0 proto kernel metric 256
fec0::1 dev eth0 metric 1024 hoplimit 10 rto_min lock 1s
Resolve the first problem by moving the setting of metrics down
into fib6_add_rt2node() to the point we are sure we are
inserting the new route into the tree. Second problem is
addressed by introducing new flag DST_METRICS_FORCE_OVERWRITE
which is set for a new host route in ip6_route_add() and makes
ipv6_cow_metrics() always overwrite the metrics in inetpeer
(even if they are not "new"); it is reset after that.
v5: use a flag in _metrics member rather than one in flags
v4: fix a typo making a condition always true (thanks to Hannes
Frederic Sowa)
v3: rewritten based on David Miller's idea to move setting the
metrics (and allocation in non-host case) down to the point we
already know the route is to be inserted. Also rebased to
net-next as it is quite late in the cycle.
Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, if the card supports CTAG acceleration we do not
account for the vlan header even if we are configuring an
8021AD vlan. This may not be best since we'll do software
tagging for 8021AD which will cause data copy on skb head expansion
Configure the length based on available hw offload capabilities and
vlan protocol.
CC: Patrick McHardy <kaber@trash.net>
Signed-off-by: Vlad Yasevich <vyasevic@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Without properly implicit or explicit read memory barrier, it's
unsafe to read an atomic variable with atomic_read() from another
thread which is different with the thread of changing the atomic
variable with atomic_inc() or atomic_dec(). So a stale tipc_num_links
may be got with atomic_read() in tipc_node_get_links(). If the
tipc_num_links variable type is converted from atomic to unsigned
integer and node list lock is used to protect it, the issue would
be avoided.
Signed-off-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Erik Hugne <erik.hugne@ericsson.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As tipc_node_list is protected by rcu read lock on read side, it's
unnecessary to hold node_list_lock to protect tipc_node_list in
tipc_node_get_links(). Instead, node_list_lock should just protects
tipc_num_nodes in the function.
Signed-off-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Erik Hugne <erik.hugne@ericsson.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Convert tipc_node_list list and node_htable hash list to RCU lists.
On read side, the two lists are protected with RCU read lock, and
on update side, node_list_lock is applied to them.
Signed-off-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Erik Hugne <erik.hugne@ericsson.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When a node is created, tipc_net_lock read lock is first held and
then node_create_lock is grabbed in order to prevent the same node
from being created and inserted into both node list and hlist twice.
But when we query node from the two node lists, we only hold
tipc_net_lock read lock without grabbing node_create_lock. Obviously
this locking policy is unable to guarantee that the two node lists
are always synchronized especially when the operation of changing
and accessing them occurs in different contexts like currently doing.
Therefore, rename node_create_lock to node_list_lock to protect the
two node lists, that is, whenever node is inserted into them or node
is queried from them, the node_list_lock should be always held. As a
result, tipc_net_lock read lock becomes redundant and then can be
removed from the node query functions.
Signed-off-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Erik Hugne <erik.hugne@ericsson.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Now unicast bearer is dynamically allocated and placed into its
identity specified slot of bearer_list array. When we search
bearer_list array with a bearer identity, the corresponding bearer
instance can be found. But broadcast bearer is statically allocated
and it is not located in the bearer_list array yet. So we decide to
enlarge bearer_list array into MAX_BEARERS + 1 slots, and its last
slot stores the broadcast bearer so that the broadcast bearer can
be found from bearer_list array with MAX_BEARERS as index. The
change will help us reduce the complex relationship between bearer
and link in the future.
Signed-off-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Erik Hugne <erik.hugne@ericsson.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
After the allocation of tipc_bearer structure instance is converted
from statical way to dynamical way, we identify whether a certain
tipc_bearer structure pointer is valid by checking whether the pointer
is NULL or not. So the active flag in tipc_bearer structure becomes
redundant.
Signed-off-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Erik Hugne <erik.hugne@ericsson.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As part of the effort to introduce RCU protection for the bearer
list, we first need to change it to a list of pointers.
Signed-off-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Erik Hugne <erik.hugne@ericsson.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The 'tipc_node_list' is guarded by tipc_net_lock and 'links' array
defined in 'tipc_node' structure is protected by node lock as well.
Without acquiring the two locks in named_cluster_distribute() a fatal
oops may happen in case that a destroyed link might be got and then
accessed. Therefore, above mentioned two locks must be held in
named_cluster_distribute() to prevent the issue from happening
accidentally.
As 'links' array in node struct must be protected by node lock,
we have to move the code of selecting an active link from
tipc_link_xmit() to named_cluster_distribute() and then call
__tipc_link_xmit() with the selected link to deliver name messages.
Signed-off-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Erik Hugne <erik.hugne@ericsson.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
