The bpf syscall and selftests conflicts were trivial
overlapping changes.
The r8169 change involved moving the added mdelay from 'net' into a
different function.
A TLS close bug fix overlapped with the splitting of the TLS state
into separate TX and RX parts. I just expanded the tests in the bug
fix from "ctx->conf == X" into "ctx->tx_conf == X && ctx->rx_conf
== X".
Signed-off-by: David S. Miller <davem@davemloft.net>
Damir reported a breakage of SO_BINDTODEVICE for UDP sockets.
In absence of VRF devices, after commit fb74c27735 ("net:
ipv4: add second dif to udp socket lookups") the dif mismatch
isn't fatal anymore for UDP socket lookup with non null
sk_bound_dev_if, breaking SO_BINDTODEVICE semantics.
This changeset addresses the issue making the dif match mandatory
again in the above scenario.
Reported-by: Damir Mansurov <dnman@oktetlabs.ru>
Fixes: fb74c27735 ("net: ipv4: add second dif to udp socket lookups")
Fixes: 1801b570dd ("net: ipv6: add second dif to udp socket lookups")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Acked-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
No changes in refcount semantics -- key init is false; replace
static_key_enable with static_branch_enable
static_key_slow_inc|dec with static_branch_inc|dec
static_key_false with static_branch_unlikely
Added a '_key' suffix to udp and udpv6 encap_needed, for better
self documentation.
Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Syzbot managed to send a udp gso packet without checksum offload into
the gso stack by disabling tx checksum (UDP_NO_CHECK6_TX). This
triggered the skb_warn_bad_offload.
RIP: 0010:skb_warn_bad_offload+0x2bc/0x600 net/core/dev.c:2658
skb_gso_segment include/linux/netdevice.h:4038 [inline]
validate_xmit_skb+0x54d/0xd90 net/core/dev.c:3120
__dev_queue_xmit+0xbf8/0x34c0 net/core/dev.c:3577
dev_queue_xmit+0x17/0x20 net/core/dev.c:3618
UDP_NO_CHECK6_TX sets skb->ip_summed to CHECKSUM_NONE just after the
udp gso integrity checks in udp_(v6_)send_skb. Extend those checks to
catch and fail in this case.
After the integrity checks jump directly to the CHECKSUM_PARTIAL case
to avoid reading the no_check_tx flags again (a TOCTTOU race).
Fixes: bec1f6f697 ("udp: generate gso with UDP_SEGMENT")
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Allow specifying segment size in the send call.
The new control message performs the same function as socket option
UDP_SEGMENT while avoiding the extra system call.
[ Export udp_cmsg_send for ipv6. -DaveM ]
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Support generic segmentation offload for udp datagrams. Callers can
concatenate and send at once the payload of multiple datagrams with
the same destination.
To set segment size, the caller sets socket option UDP_SEGMENT to the
length of each discrete payload. This value must be smaller than or
equal to the relevant MTU.
A follow-up patch adds cmsg UDP_SEGMENT to specify segment size on a
per send call basis.
Total byte length may then exceed MTU. If not an exact multiple of
segment size, the last segment will be shorter.
The implementation adds a gso_size field to the udp socket, ip(v6)
cmsg cookie and inet_cork structure to be able to set the value at
setsockopt or cmsg time and to work with both lockless and corked
paths.
Initial benchmark numbers show UDP GSO about as expensive as TCP GSO.
tcp tso
3197 MB/s 54232 msg/s 54232 calls/s
6,457,754,262 cycles
tcp gso
1765 MB/s 29939 msg/s 29939 calls/s
11,203,021,806 cycles
tcp without tso/gso *
739 MB/s 12548 msg/s 12548 calls/s
11,205,483,630 cycles
udp
876 MB/s 14873 msg/s 624666 calls/s
11,205,777,429 cycles
udp gso
2139 MB/s 36282 msg/s 36282 calls/s
11,204,374,561 cycles
[*] after reverting commit 0a6b2a1dc2
("tcp: switch to GSO being always on")
Measured total system cycles ('-a') for one core while pinning both
the network receive path and benchmark process to that core:
perf stat -a -C 12 -e cycles \
./udpgso_bench_tx -C 12 -4 -D "$DST" -l 4
Note the reduction in calls/s with GSO. Bytes per syscall drops
increases from 1470 to 61818.
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
UDP segmentation offload needs access to inet_cork in the udp layer.
Pass the struct to ip(6)_make_skb instead of allocating it on the
stack in that function itself.
This patch is a noop otherwise.
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
A new RTF_CACHE route can be created between ip6_sk_dst_lookup_flow()
and ip6_dst_store() calls in udpv6_sendmsg(), when datagram sending
results to ICMPV6_PKT_TOOBIG error:
udp_v6_send_skb(), for example with vti6 tunnel:
vti6_xmit(), get ICMPV6_PKT_TOOBIG error
skb_dst_update_pmtu(), can create a RTF_CACHE clone
icmpv6_send()
...
udpv6_err()
ip6_sk_update_pmtu()
ip6_update_pmtu(), can create a RTF_CACHE clone
...
ip6_datagram_dst_update()
ip6_dst_store()
And after commit 33c162a980 ("ipv6: datagram: Update dst cache of
a connected datagram sk during pmtu update"), the UDPv6 error handler
can update socket's dst cache, but it can happen before the update in
the end of udpv6_sendmsg(), preventing getting the new dst cache on
the next udpv6_sendmsg() calls.
In order to fix it, save dst in a connected socket only if the current
socket's dst cache is invalid.
The previous patch prepared ip6_sk_dst_lookup_flow() to do that with
the new argument, and this patch enables it in udpv6_sendmsg().
Fixes: 33c162a980 ("ipv6: datagram: Update dst cache of a connected datagram sk during pmtu update")
Fixes: 45e4fd2668 ("ipv6: Only create RTF_CACHE routes after encountering pmtu exception")
Signed-off-by: Alexey Kodanev <alexey.kodanev@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This should make it consistent with ip6_sk_dst_lookup_flow()
that is accepting the new 'connected' parameter of type bool.
Signed-off-by: Alexey Kodanev <alexey.kodanev@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add 'connected' parameter to ip6_sk_dst_lookup_flow() and update
the cache only if ip6_sk_dst_check() returns NULL and a socket
is connected.
The function is used as before, the new behavior for UDP sockets
in udpv6_sendmsg() will be enabled in the next patch.
Signed-off-by: Alexey Kodanev <alexey.kodanev@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
== The problem ==
See description of the problem in the initial patch of this patch set.
== The solution ==
The patch provides much more reliable in-kernel solution for the 2nd
part of the problem: making outgoing connecttion from desired IP.
It adds new attach types `BPF_CGROUP_INET4_CONNECT` and
`BPF_CGROUP_INET6_CONNECT` for program type
`BPF_PROG_TYPE_CGROUP_SOCK_ADDR` that can be used to override both
source and destination of a connection at connect(2) time.
Local end of connection can be bound to desired IP using newly
introduced BPF-helper `bpf_bind()`. It allows to bind to only IP though,
and doesn't support binding to port, i.e. leverages
`IP_BIND_ADDRESS_NO_PORT` socket option. There are two reasons for this:
* looking for a free port is expensive and can affect performance
significantly;
* there is no use-case for port.
As for remote end (`struct sockaddr *` passed by user), both parts of it
can be overridden, remote IP and remote port. It's useful if an
application inside cgroup wants to connect to another application inside
same cgroup or to itself, but knows nothing about IP assigned to the
cgroup.
Support is added for IPv4 and IPv6, for TCP and UDP.
IPv4 and IPv6 have separate attach types for same reason as sys_bind
hooks, i.e. to prevent reading from / writing to e.g. user_ip6 fields
when user passes sockaddr_in since it'd be out-of-bound.
== Implementation notes ==
The patch introduces new field in `struct proto`: `pre_connect` that is
a pointer to a function with same signature as `connect` but is called
before it. The reason is in some cases BPF hooks should be called way
before control is passed to `sk->sk_prot->connect`. Specifically
`inet_dgram_connect` autobinds socket before calling
`sk->sk_prot->connect` and there is no way to call `bpf_bind()` from
hooks from e.g. `ip4_datagram_connect` or `ip6_datagram_connect` since
it'd cause double-bind. On the other hand `proto.pre_connect` provides a
flexible way to add BPF hooks for connect only for necessary `proto` and
call them at desired time before `connect`. Since `bpf_bind()` is
allowed to bind only to IP and autobind in `inet_dgram_connect` binds
only port there is no chance of double-bind.
bpf_bind() sets `force_bind_address_no_port` to bind to only IP despite
of value of `bind_address_no_port` socket field.
bpf_bind() sets `with_lock` to `false` when calling to __inet_bind()
and __inet6_bind() since all call-sites, where bpf_bind() is called,
already hold socket lock.
Signed-off-by: Andrey Ignatov <rdna@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
This patch moves the udp_rmem_min, udp_wmem_min
to namespace and init the udp_l3mdev_accept explicitly.
The udp_rmem_min/udp_wmem_min affect udp rx/tx queue,
with this patch namespaces can set them differently.
Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
/proc has been ignoring struct file_operations::owner field for 10 years.
Specifically, it started with commit 786d7e1612
("Fix rmmod/read/write races in /proc entries"). Notice the chunk where
inode->i_fop is initialized with proxy struct file_operations for
regular files:
- if (de->proc_fops)
- inode->i_fop = de->proc_fops;
+ if (de->proc_fops) {
+ if (S_ISREG(inode->i_mode))
+ inode->i_fop = &proc_reg_file_ops;
+ else
+ inode->i_fop = de->proc_fops;
+ }
VFS stopped pinning module at this point.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch moves the udp[46]_portaddr_hash()
to net/ip[v6].h. The function name is renamed to
ipv[46]_portaddr_hash().
It will be used by a later patch which adds a second listener
hashtable hashed by the address and port.
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since commit e32ea7e747 ("soreuseport: fast reuseport UDP socket
selection") and commit c125e80b88 ("soreuseport: fast reuseport
TCP socket selection") the relevant reuseport socket matching the current
packet is selected by the reuseport_select_sock() call. The only
exceptions are invalid BPF filters/filters returning out-of-range
indices.
In the latter case the code implicitly falls back to using the hash
demultiplexing, but instead of selecting the socket inside the
reuseport_select_sock() function, it relies on the hash selection
logic introduced with the early soreuseport implementation.
With this patch, in case of a BPF filter returning a bad socket
index value, we fall back to hash-based selection inside the
reuseport_select_sock() body, so that we can drop some duplicate
code in the ipv4 and ipv6 stack.
This also allows faster lookup in the above scenario and will allow
us to avoid computing the hash value for successful, BPF based
demultiplexing - in a later patch.
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Acked-by: Craig Gallek <kraig@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Please do not apply this to mainline directly, instead please re-run the
coccinelle script shown below and apply its output.
For several reasons, it is desirable to use {READ,WRITE}_ONCE() in
preference to ACCESS_ONCE(), and new code is expected to use one of the
former. So far, there's been no reason to change most existing uses of
ACCESS_ONCE(), as these aren't harmful, and changing them results in
churn.
However, for some features, the read/write distinction is critical to
correct operation. To distinguish these cases, separate read/write
accessors must be used. This patch migrates (most) remaining
ACCESS_ONCE() instances to {READ,WRITE}_ONCE(), using the following
coccinelle script:
----
// Convert trivial ACCESS_ONCE() uses to equivalent READ_ONCE() and
// WRITE_ONCE()
// $ make coccicheck COCCI=/home/mark/once.cocci SPFLAGS="--include-headers" MODE=patch
virtual patch
@ depends on patch @
expression E1, E2;
@@
- ACCESS_ONCE(E1) = E2
+ WRITE_ONCE(E1, E2)
@ depends on patch @
expression E;
@@
- ACCESS_ONCE(E)
+ READ_ONCE(E)
----
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: davem@davemloft.net
Cc: linux-arch@vger.kernel.org
Cc: mpe@ellerman.id.au
Cc: shuah@kernel.org
Cc: snitzer@redhat.com
Cc: thor.thayer@linux.intel.com
Cc: tj@kernel.org
Cc: viro@zeniv.linux.org.uk
Cc: will.deacon@arm.com
Link: http://lkml.kernel.org/r/1508792849-3115-19-git-send-email-paulmck@linux.vnet.ibm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
While trying an ESP transport mode encryption for UDPv6 packets of
datagram size 1436 with MTU 1500, checksum error was observed in
the secondary fragment.
This error occurs due to the UDP payload checksum being missed out
when computing the full checksum for these packets in
udp6_hwcsum_outgoing().
Fixes: d39d938c82 ("ipv6: Introduce udpv6_send_skb()")
Signed-off-by: Subash Abhinov Kasiviswanathan <subashab@codeaurora.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pull networking updates from David Miller:
1) Support ipv6 checksum offload in sunvnet driver, from Shannon
Nelson.
2) Move to RB-tree instead of custom AVL code in inetpeer, from Eric
Dumazet.
3) Allow generic XDP to work on virtual devices, from John Fastabend.
4) Add bpf device maps and XDP_REDIRECT, which can be used to build
arbitrary switching frameworks using XDP. From John Fastabend.
5) Remove UFO offloads from the tree, gave us little other than bugs.
6) Remove the IPSEC flow cache, from Florian Westphal.
7) Support ipv6 route offload in mlxsw driver.
8) Support VF representors in bnxt_en, from Sathya Perla.
9) Add support for forward error correction modes to ethtool, from
Vidya Sagar Ravipati.
10) Add time filter for packet scheduler action dumping, from Jamal Hadi
Salim.
11) Extend the zerocopy sendmsg() used by virtio and tap to regular
sockets via MSG_ZEROCOPY. From Willem de Bruijn.
12) Significantly rework value tracking in the BPF verifier, from Edward
Cree.
13) Add new jump instructions to eBPF, from Daniel Borkmann.
14) Rework rtnetlink plumbing so that operations can be run without
taking the RTNL semaphore. From Florian Westphal.
15) Support XDP in tap driver, from Jason Wang.
16) Add 32-bit eBPF JIT for ARM, from Shubham Bansal.
17) Add Huawei hinic ethernet driver.
18) Allow to report MD5 keys in TCP inet_diag dumps, from Ivan
Delalande.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1780 commits)
i40e: point wb_desc at the nvm_wb_desc during i40e_read_nvm_aq
i40e: avoid NVM acquire deadlock during NVM update
drivers: net: xgene: Remove return statement from void function
drivers: net: xgene: Configure tx/rx delay for ACPI
drivers: net: xgene: Read tx/rx delay for ACPI
rocker: fix kcalloc parameter order
rds: Fix non-atomic operation on shared flag variable
net: sched: don't use GFP_KERNEL under spin lock
vhost_net: correctly check tx avail during rx busy polling
net: mdio-mux: add mdio_mux parameter to mdio_mux_init()
rxrpc: Make service connection lookup always check for retry
net: stmmac: Delete dead code for MDIO registration
gianfar: Fix Tx flow control deactivation
cxgb4: Ignore MPS_TX_INT_CAUSE[Bubble] for T6
cxgb4: Fix pause frame count in t4_get_port_stats
cxgb4: fix memory leak
tun: rename generic_xdp to skb_xdp
tun: reserve extra headroom only when XDP is set
net: dsa: bcm_sf2: Configure IMP port TC2QOS mapping
net: dsa: bcm_sf2: Advertise number of egress queues
...
Twice patches trying to constify inet{6}_protocol have been reverted:
39294c3df2 ("Revert "ipv6: constify inet6_protocol structures"") to
revert 3a3a4e3054 and then 03157937fe ("Revert "ipv4: make
net_protocol const"") to revert aa8db499ea.
Add a comment that the structures can not be const because the
early_demux field can change based on a sysctl.
Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, in the udp6 code, the dst cookie is not initialized/updated
concurrently with the RX dst used by early demux.
As a result, the dst_check() in the early_demux path always fails,
the rx dst cache is always invalidated, and we can't really
leverage significant gain from the demux lookup.
Fix it adding udp6 specific variant of sk_rx_dst_set() and use it
to set the dst cookie when the dst entry is really changed.
The issue is there since the introduction of early demux for ipv6.
Fixes: 5425077d73 ("net: ipv6: Add early demux handler for UDP unicast")
Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Due to commit e6afc8ace6 ("udp: remove
headers from UDP packets before queueing"), when udp packets are being
peeked the requested extra offset is always 0 as there is no need to skip
the udp header. However, when the offset is 0 and the next skb is
of length 0, it is only returned once. The behaviour can be seen with
the following python script:
from socket import *;
f=socket(AF_INET6, SOCK_DGRAM | SOCK_NONBLOCK, 0);
g=socket(AF_INET6, SOCK_DGRAM | SOCK_NONBLOCK, 0);
f.bind(('::', 0));
addr=('::1', f.getsockname()[1]);
g.sendto(b'', addr)
g.sendto(b'b', addr)
print(f.recvfrom(10, MSG_PEEK));
print(f.recvfrom(10, MSG_PEEK));
Where the expected output should be the empty string twice.
Instead, make sk_peek_offset return negative values, and pass those values
to __skb_try_recv_datagram/__skb_try_recv_from_queue. If the passed offset
to __skb_try_recv_from_queue is negative, the checked skb is never skipped.
__skb_try_recv_from_queue will then ensure the offset is reset back to 0
if a peek is requested without an offset, unless no packets are found.
Also simplify the if condition in __skb_try_recv_from_queue. If _off is
greater then 0, and off is greater then or equal to skb->len, then
(_off || skb->len) must always be true assuming skb->len >= 0 is always
true.
Also remove a redundant check around a call to sk_peek_offset in af_unix.c,
as it double checked if MSG_PEEK was set in the flags.
V2:
- Moved the negative fixup into __skb_try_recv_from_queue, and remove now
redundant checks
- Fix peeking in udp{,v6}_recvmsg to report the right value when the
offset is 0
V3:
- Marked new branch in __skb_try_recv_from_queue as unlikely.
Signed-off-by: Matthew Dawson <matthew@mjdsystems.ca>
Acked-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Any use of key->enabled (that is static_key_enabled and static_key_count)
outside jump_label_lock should handle its own serialization. The only
two that are not doing so are the UDP encapsulation static keys. Change
them to use static_key_enable, which now correctly tests key->enabled under
the jump label lock.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Jason Baron <jbaron@akamai.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1501601046-35683-3-git-send-email-pbonzini@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Add a second device index, sdif, to inet6 socket lookups. sdif is the
index for ingress devices enslaved to an l3mdev. It allows the lookups
to consider the enslaved device as well as the L3 domain when searching
for a socket.
TCP moves the data in the cb. Prior to tcp_v4_rcv (e.g., early demux) the
ingress index is obtained from IPCB using inet_sdif and after tcp_v4_rcv
tcp_v4_sdif is used.
Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a second device index, sdif, to udp socket lookups. sdif is the
index for ingress devices enslaved to an l3mdev. It allows the lookups
to consider the enslaved device as well as the L3 domain when searching
for a socket.
Early demux lookups are handled in the next patch as part of INET_MATCH
changes.
Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since commit 67a51780ae ("ipv6: udp: leverage scratch area
helpers") udp6_recvmsg() read the skb len from the scratch area,
to avoid a cache miss.
But the UDP6 rx path support RFC 2675 UDPv6 jumbograms, and their
length exceeds the 16 bits available in the scratch area. As a side
effect the length returned by recvmsg() is:
<ingress datagram len> % (1<<16)
This commit addresses the issue allocating one more bit in the
IP6CB flags field and setting it for incoming jumbograms.
Such field is still in the first cacheline, so at recvmsg()
time we can check it and fallback to access skb->len if
required, without a measurable overhead.
Fixes: 67a51780ae ("ipv6: udp: leverage scratch area helpers")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When an early demuxed packet reaches __udp6_lib_lookup_skb(), the
sk reference is retrieved and used, but the relevant reference
count is leaked and the socket destructor is never called.
Beyond leaking the sk memory, if there are pending UDP packets
in the receive queue, even the related accounted memory is leaked.
In the long run, this will cause persistent forward allocation errors
and no UDP skbs (both ipv4 and ipv6) will be able to reach the
user-space.
Fix this by explicitly accessing the early demux reference before
the lookup, and properly decreasing the socket reference count
after usage.
Also drop the skb_steal_sock() in __udp6_lib_lookup_skb(), and
the now obsoleted comment about "socket cache".
The newly added code is derived from the current ipv4 code for the
similar path.
v1 -> v2:
fixed the __udp6_lib_rcv() return code for resubmission,
as suggested by Eric
Reported-by: Sam Edwards <CFSworks@gmail.com>
Reported-by: Marc Haber <mh+netdev@zugschlus.de>
Fixes: 5425077d73 ("net: ipv6: Add early demux handler for UDP unicast")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
refcount_t type and corresponding API should be
used instead of atomic_t when the variable is used as
a reference counter. This allows to avoid accidental
refcounter overflows that might lead to use-after-free
situations.
This patch uses refcount_inc_not_zero() instead of
atomic_inc_not_zero_hint() due to absense of a _hint()
version of refcount API. If the hint() version must
be used, we might need to revisit API.
Signed-off-by: Elena Reshetova <elena.reshetova@intel.com>
Signed-off-by: Hans Liljestrand <ishkamiel@gmail.com>
Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: David Windsor <dwindsor@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The commit b65ac44674 ("udp: try to avoid 2 cache miss on dequeue")
leveraged the scratched area helpers for UDP v4 but I forgot to
update accordingly the IPv6 code path.
This change extends the scratch area usage to the IPv6 code, synching
the two implementations and giving some performance benefit.
IPv6 is again almost on the same level of IPv4, performance-wide.
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In __ip6_datagram_connect(), reset sk->sk_v6_daddr and inet->dport if
error occurs.
In udp_v6_early_demux(), check for sk_state to make sure it is in
TCP_ESTABLISHED state.
Together, it makes sure unconnected UDP socket won't be considered as a
valid candidate for early demux.
v3: add TCP_ESTABLISHED state check in udp_v6_early_demux()
v2: fix compilation error
Fixes: 5425077d73 ("net: ipv6: Add early demux handler for UDP unicast")
Signed-off-by: Wei Wang <weiwan@google.com>
Acked-by: Maciej Żenczykowski <maze@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
very similar to commit dd99e425be ("udp: prefetch
rmem_alloc in udp_queue_rcv_skb()"), this allows saving a cache
miss when the BH is bottle-neck for UDP over ipv6 packet
processing, e.g. for small packets when a single RX NIC ingress
queue is in use.
Performances under flood when multiple NIC RX queues used are
unaffected, but when a single NIC rx queue is in use, this
gives ~8% performance improvement.
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In udp_v4/6_early_demux() code, we try to hold dst->__refcnt for
dst with DST_NOCACHE flag. This is because later in udp_sk_rx_dst_set()
function, we will try to cache this dst in sk for connected case.
However, a better way to achieve this is to not try to hold dst in
early_demux(), but in udp_sk_rx_dst_set(), call dst_hold_safe(). This
approach is also more consistant with how tcp is handling it. And it
will make later changes simpler.
Signed-off-by: Wei Wang <weiwan@google.com>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since the udp memory accounting refactor, we don't need any more
to export the *udp*_queue_rcv_skb(). Make them static and fix
a couple of sparse warnings:
net/ipv4/udp.c:1615:5: warning: symbol 'udp_queue_rcv_skb' was not
declared. Should it be static?
net/ipv6/udp.c:572:5: warning: symbol 'udpv6_queue_rcv_skb' was not
declared. Should it be static?
Fixes: 850cbaddb5 ("udp: use it's own memory accounting schema")
Fixes: c915fe13cb ("udplite: fix NULL pointer dereference")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
under udp flood the sk_receive_queue spinlock is heavily contended.
This patch try to reduce the contention on such lock adding a
second receive queue to the udp sockets; recvmsg() looks first
in such queue and, only if empty, tries to fetch the data from
sk_receive_queue. The latter is spliced into the newly added
queue every time the receive path has to acquire the
sk_receive_queue lock.
The accounting of forward allocated memory is still protected with
the sk_receive_queue lock, so udp_rmem_release() needs to acquire
both locks when the forward deficit is flushed.
On specific scenarios we can end up acquiring and releasing the
sk_receive_queue lock multiple times; that will be covered by
the next patch
Suggested-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern reported that 5425077d73 ("net: ipv6: Add early demux
handler for UDP unicast") breaks udp_l3mdev_accept=0 since early
demux for IPv6 UDP was doing a generic socket lookup which does not
require an exact match. Fix this by making UDPv6 early demux match
connected sockets only.
v1->v2: Take reference to socket after match as suggested by Eric
v2->v3: Add comment before break
Fixes: 5425077d73 ("net: ipv6: Add early demux handler for UDP unicast")
Reported-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: Subash Abhinov Kasiviswanathan <subashab@codeaurora.org>
Cc: Eric Dumazet <edumazet@google.com>
Acked-by: David Ahern <dsa@cumulusnetworks.com>
Tested-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Certain system process significant unconnected UDP workload.
It would be preferrable to disable UDP early demux for those systems
and enable it for TCP only.
By disabling UDP demux, we see these slight gains on an ARM64 system-
782 -> 788Mbps unconnected single stream UDPv4
633 -> 654Mbps unconnected UDPv4 different sources
The performance impact can change based on CPU architecure and cache
sizes. There will not much difference seen if entire UDP hash table
is in cache.
Both sysctls are enabled by default to preserve existing behavior.
v1->v2: Change function pointer instead of adding conditional as
suggested by Stephen.
v2->v3: Read once in callers to avoid issues due to compiler
optimizations. Also update commit message with the tests.
v3->v4: Store and use read once result instead of querying pointer
again incorrectly.
v4->v5: Refactor to avoid errors due to compilation with IPV6={m,n}
Signed-off-by: Subash Abhinov Kasiviswanathan <subashab@codeaurora.org>
Suggested-by: Eric Dumazet <edumazet@google.com>
Cc: Stephen Hemminger <stephen@networkplumber.org>
Cc: Tom Herbert <tom@herbertland.com>
Cc: David Miller <davem@davemloft.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Conflicts:
drivers/net/ethernet/broadcom/genet/bcmmii.c
drivers/net/hyperv/netvsc.c
kernel/bpf/hashtab.c
Almost entirely overlapping changes.
Signed-off-by: David S. Miller <davem@davemloft.net>
In the case udp_sk(sk)->pending is AF_INET6, udpv6_sendmsg() would
jump to do_append_data, skipping the initialization of sockc.tsflags.
Fix the problem by moving sockc.tsflags initialization earlier.
The bug was detected with KMSAN.
Fixes: c14ac9451c ("sock: enable timestamping using control messages")
Signed-off-by: Alexander Potapenko <glider@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
While running a single stream UDPv6 test, we observed that amount
of CPU spent in NET_RX softirq was much greater than UDPv4 for an
equivalent receive rate. The test here was run on an ARM64 based
Android system. On further analysis with perf, we found that UDPv6
was spending significant time in the statistics netfilter targets
which did socket lookup per packet. These statistics rules perform
a lookup when there is no socket associated with the skb. Since
there are multiple instances of these rules based on UID, there
will be equal number of lookups per skb.
By introducing early demux for UDPv6, we avoid the redundant lookups.
This also helped to improve the performance (800Mbps -> 870Mbps) on a
CPU limited system in a single stream UDPv6 receive test with 1450
byte sized datagrams using iperf.
v1->v2: Use IPv6 cookie to validate dst instead of 0 as suggested
by Eric
Signed-off-by: Subash Abhinov Kasiviswanathan <subashab@codeaurora.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds a check on the type of the source address for the case
where the destination address is in6addr_any. If the source is an
IPv4-mapped IPv6 source address, the destination is changed to
::ffff:127.0.0.1, and otherwise the destination is changed to ::1. This
is done in three locations to handle UDP calls to either connect() or
sendmsg() and TCP calls to connect(). Note that udpv6_sendmsg() delays
handling an in6addr_any destination until very late, so the patch only
needs to handle the case where the source is an IPv4-mapped IPv6
address.
Signed-off-by: Jonathan T. Leighton <jtleight@udel.edu>
Signed-off-by: David S. Miller <davem@davemloft.net>
The conflict was an interaction between a bug fix in the
netvsc driver in 'net' and an optimization of the RX path
in 'net-next'.
Signed-off-by: David S. Miller <davem@davemloft.net>
When same struct dst_entry can be used for many different
neighbours we can not use it for pending confirmations.
The datagram protocols can use MSG_CONFIRM to confirm the
neighbour. When used with MSG_PROBE we do not reach the
code where neighbour is confirmed, so we have to do the
same slow lookup by using the dst_confirm_neigh() helper.
When MSG_PROBE is not used, ip_append_data/ip6_append_data
will set the skb flag dst_pending_confirm.
Reported-by: YueHaibing <yuehaibing@huawei.com>
Fixes: 5110effee8 ("net: Do delayed neigh confirmation.")
Fixes: f2bb4bedf3 ("ipv4: Cache output routes in fib_info nexthops.")
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Dmitry reported that UDP sockets being destroyed would trigger the
WARN_ON(atomic_read(&sk->sk_rmem_alloc)); in inet_sock_destruct()
It turns out we do not properly destroy skb(s) that have wrong UDP
checksum.
Thanks again to syzkaller team.
Fixes : 7c13f97ffd ("udp: do fwd memory scheduling on dequeue")
Reported-by: Dmitry Vyukov <dvyukov@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Paolo Abeni <pabeni@redhat.com>
Cc: Hannes Frederic Sowa <hannes@stressinduktion.org>
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Packets arriving in a VRF currently are delivered to UDP sockets that
aren't bound to any interface. TCP defaults to not delivering packets
arriving in a VRF to unbound sockets. IP route lookup and socket
transmit both assume that unbound means using the default table and
UDP applications that haven't been changed to be aware of VRFs may not
function correctly in this case since they may not be able to handle
overlapping IP address ranges, or be able to send packets back to the
original sender if required.
So add a sysctl, udp_l3mdev_accept, to control this behaviour with it
being analgous to the existing tcp_l3mdev_accept, namely to allow a
process to have a VRF-global listen socket. Have this default to off
as this is the behaviour that users will expect, given that there is
no explicit mechanism to set unmodified VRF-unaware application into a
default VRF.
Signed-off-by: Robert Shearman <rshearma@brocade.com>
Acked-by: David Ahern <dsa@cumulusnetworks.com>
Tested-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We pass these per-protocol equal functions around in various places, but
we can just have one function that checks the sk->sk_family and then do
the right comparison function. I've also changed the ipv4 version to
not cast to inet_sock since it is unneeded.
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This was entirely automated, using the script by Al:
PATT='^[[:blank:]]*#[[:blank:]]*include[[:blank:]]*<asm/uaccess.h>'
sed -i -e "s!$PATT!#include <linux/uaccess.h>!" \
$(git grep -l "$PATT"|grep -v ^include/linux/uaccess.h)
to do the replacement at the end of the merge window.
Requested-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
udplite conflict is resolved by taking what 'net-next' did
which removed the backlog receive method assignment, since
it is no longer necessary.
Two entries were added to the non-priv ethtool operations
switch statement, one in 'net' and one in 'net-next, so
simple overlapping changes.
Signed-off-by: David S. Miller <davem@davemloft.net>
In commits 93821778de ("udp: Fix rcv socket locking") and
f7ad74fef3 ("net/ipv6/udp: UDP encapsulation: break backlog_rcv into
__udpv6_queue_rcv_skb") UDP backlog handlers were renamed, but UDPlite
was forgotten.
This leads to crashes if UDPlite header is pulled twice, which happens
starting from commit e6afc8ace6 ("udp: remove headers from UDP packets
before queueing")
Bug found by syzkaller team, thanks a lot guys !
Note that backlog use in UDP/UDPlite is scheduled to be removed starting
from linux-4.10, so this patch is only needed up to linux-4.9
Fixes: 93821778de ("udp: Fix rcv socket locking")
Fixes: f7ad74fef3 ("net/ipv6/udp: UDP encapsulation: break backlog_rcv into __udpv6_queue_rcv_skb")
Fixes: e6afc8ace6 ("udp: remove headers from UDP packets before queueing")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Andrey Konovalov <andreyknvl@google.com>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
All conflicts were simple overlapping changes except perhaps
for the Thunder driver.
That driver has a change_mtu method explicitly for sending
a message to the hardware. If that fails it returns an
error.
Normally a driver doesn't need an ndo_change_mtu method becuase those
are usually just range changes, which are now handled generically.
But since this extra operation is needed in the Thunder driver, it has
to stay.
However, if the message send fails we have to restore the original
MTU before the change because the entire call chain expects that if
an error is thrown by ndo_change_mtu then the MTU did not change.
Therefore code is added to nicvf_change_mtu to remember the original
MTU, and to restore it upon nicvf_update_hw_max_frs() failue.
Signed-off-by: David S. Miller <davem@davemloft.net>
UDP_SKB_CB(skb)->partial_cov is located at offset 66 in skb,
requesting a cold cache line being read in cpu cache.
We can avoid this cache line miss for UDP sockets,
as partial_cov has a meaning only for UDPLite.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
UDP busy polling is restricted to connected UDP sockets.
This is because sk_busy_loop() only takes care of one NAPI context.
There are cases where it could be extended.
1) Some hosts receive traffic on a single NIC, with one RX queue.
2) Some applications use SO_REUSEPORT and associated BPF filter
to split the incoming traffic on one UDP socket per RX
queue/thread/cpu
3) Some UDP sockets are used to send/receive traffic for one flow, but
they do not bother with connect()
This patch records the napi_id of first received skb, giving more
reach to busy polling.
Tested:
lpaa23:~# echo 70 >/proc/sys/net/core/busy_read
lpaa24:~# echo 70 >/proc/sys/net/core/busy_read
lpaa23:~# for f in `seq 1 10`; do ./super_netperf 1 -H lpaa24 -t UDP_RR -l 5; done
Before patch :
27867 28870 37324 41060 41215
36764 36838 44455 41282 43843
After patch :
73920 73213 70147 74845 71697
68315 68028 75219 70082 73707
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Honor udptable parameter that is passed to __udp*_lib_mcast_deliver(),
otherwise udplite broadcast/multicast use the wrong table and it breaks.
Fixes: 2dc41cff75 ("udp: Use hash2 for long hash1 chains in __udp*_lib_mcast_deliver.")
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pablo Neira Ayuso says:
====================
Netfilter updates for net-next
The following patchset contains a second batch of Netfilter updates for
your net-next tree. This includes a rework of the core hook
infrastructure that improves Netfilter performance by ~15% according to
synthetic benchmarks. Then, a large batch with ipset updates, including
a new hash:ipmac set type, via Jozsef Kadlecsik. This also includes a
couple of assorted updates.
Regarding the core hook infrastructure rework to improve performance,
using this simple drop-all packets ruleset from ingress:
nft add table netdev x
nft add chain netdev x y { type filter hook ingress device eth0 priority 0\; }
nft add rule netdev x y drop
And generating traffic through Jesper Brouer's
samples/pktgen/pktgen_bench_xmit_mode_netif_receive.sh script using -i
option. perf report shows nf_tables calls in its top 10:
17.30% kpktgend_0 [nf_tables] [k] nft_do_chain
15.75% kpktgend_0 [kernel.vmlinux] [k] __netif_receive_skb_core
10.39% kpktgend_0 [nf_tables_netdev] [k] nft_do_chain_netdev
I'm measuring here an improvement of ~15% in performance with this
patchset, so we got +2.5Mpps more. I have used my old laptop Intel(R)
Core(TM) i5-3320M CPU @ 2.60GHz 4-cores.
This rework contains more specifically, in strict order, these patches:
1) Remove compile-time debugging from core.
2) Remove obsolete comments that predate the rcu era. These days it is
well known that a Netfilter hook always runs under rcu_read_lock().
3) Remove threshold handling, this is only used by br_netfilter too.
We already have specific code to handle this from br_netfilter,
so remove this code from the core path.
4) Deprecate NF_STOP, as this is only used by br_netfilter.
5) Place nf_state_hook pointer into xt_action_param structure, so
this structure fits into one single cacheline according to pahole.
This also implicit affects nftables since it also relies on the
xt_action_param structure.
6) Move state->hook_entries into nf_queue entry. The hook_entries
pointer is only required by nf_queue(), so we can store this in the
queue entry instead.
7) use switch() statement to handle verdict cases.
8) Remove hook_entries field from nf_hook_state structure, this is only
required by nf_queue, so store it in nf_queue_entry structure.
9) Merge nf_iterate() into nf_hook_slow() that results in a much more
simple and readable function.
10) Handle NF_REPEAT away from the core, so far the only client is
nf_conntrack_in() and we can restart the packet processing using a
simple goto to jump back there when the TCP requires it.
This update required a second pass to fix fallout, fix from
Arnd Bergmann.
11) Set random seed from nft_hash when no seed is specified from
userspace.
12) Simplify nf_tables expression registration, in a much smarter way
to save lots of boiler plate code, by Liping Zhang.
13) Simplify layer 4 protocol conntrack tracker registration, from
Davide Caratti.
14) Missing CONFIG_NF_SOCKET_IPV4 dependency for udp4_lib_lookup, due
to recent generalization of the socket infrastructure, from Arnd
Bergmann.
15) Then, the ipset batch from Jozsef, he describes it as it follows:
* Cleanup: Remove extra whitespaces in ip_set.h
* Cleanup: Mark some of the helpers arguments as const in ip_set.h
* Cleanup: Group counter helper functions together in ip_set.h
* struct ip_set_skbinfo is introduced instead of open coded fields
in skbinfo get/init helper funcions.
* Use kmalloc() in comment extension helper instead of kzalloc()
because it is unnecessary to zero out the area just before
explicit initialization.
* Cleanup: Split extensions into separate files.
* Cleanup: Separate memsize calculation code into dedicated function.
* Cleanup: group ip_set_put_extensions() and ip_set_get_extensions()
together.
* Add element count to hash headers by Eric B Munson.
* Add element count to all set types header for uniform output
across all set types.
* Count non-static extension memory into memsize calculation for
userspace.
* Cleanup: Remove redundant mtype_expire() arguments, because
they can be get from other parameters.
* Cleanup: Simplify mtype_expire() for hash types by removing
one level of intendation.
* Make NLEN compile time constant for hash types.
* Make sure element data size is a multiple of u32 for the hash set
types.
* Optimize hash creation routine, exit as early as possible.
* Make struct htype per ipset family so nets array becomes fixed size
and thus simplifies the struct htype allocation.
* Collapse same condition body into a single one.
* Fix reported memory size for hash:* types, base hash bucket structure
was not taken into account.
* hash:ipmac type support added to ipset by Tomasz Chilinski.
* Use setup_timer() and mod_timer() instead of init_timer()
by Muhammad Falak R Wani, individually for the set type families.
16) Remove useless connlabel field in struct netns_ct, patch from
Florian Westphal.
17) xt_find_table_lock() doesn't return ERR_PTR() anymore, so simplify
{ip,ip6,arp}tables code that uses this.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Since commit ca065d0cf8 ("udp: no longer use SLAB_DESTROY_BY_RCU")
the udp6_lib_lookup and udp4_lib_lookup functions are only
provided when it is actually possible to call them.
However, moving the callers now caused a link error:
net/built-in.o: In function `nf_sk_lookup_slow_v6':
(.text+0x131a39): undefined reference to `udp6_lib_lookup'
net/ipv4/netfilter/nf_socket_ipv4.o: In function `nf_sk_lookup_slow_v4':
nf_socket_ipv4.c:(.text.nf_sk_lookup_slow_v4+0x114): undefined reference to `udp4_lib_lookup'
This extends the #ifdef so we also provide the functions when
CONFIG_NF_SOCKET_IPV4 or CONFIG_NF_SOCKET_IPV6, respectively
are set.
Fixes: 8db4c5be88 ("netfilter: move socket lookup infrastructure to nf_socket_ipv{4,6}.c")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
A new argument is added to __skb_recv_datagram to provide
an explicit skb destructor, invoked under the receive queue
lock.
The UDP protocol uses such argument to perform memory
reclaiming on dequeue, so that the UDP protocol does not
set anymore skb->desctructor.
Instead explicit memory reclaiming is performed at close() time and
when skbs are removed from the receive queue.
The in kernel UDP protocol users now need to call a
skb_recv_udp() variant instead of skb_recv_datagram() to
properly perform memory accounting on dequeue.
Overall, this allows acquiring only once the receive queue
lock on dequeue.
Tested using pktgen with random src port, 64 bytes packet,
wire-speed on a 10G link as sender and udp_sink as the receiver,
using an l4 tuple rxhash to stress the contention, and one or more
udp_sink instances with reuseport.
nr sinks vanilla patched
1 440 560
3 2150 2300
6 3650 3800
9 4450 4600
12 6250 6450
v1 -> v2:
- do rmem and allocated memory scheduling under the receive lock
- do bulk scheduling in first_packet_length() and in udp_destruct_sock()
- avoid the typdef for the dequeue callback
Suggested-by: Eric Dumazet <edumazet@google.com>
Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
So that we can use it even after orphaining the skbuff.
Suggested-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
- Use the UID in routing lookups made by protocol connect() and
sendmsg() functions.
- Make sure that routing lookups triggered by incoming packets
(e.g., Path MTU discovery) take the UID of the socket into
account.
- For packets not associated with a userspace socket, (e.g., ping
replies) use UID 0 inside the user namespace corresponding to
the network namespace the socket belongs to. This allows
all namespaces to apply routing and iptables rules to
kernel-originated traffic in that namespaces by matching UID 0.
This is better than using the UID of the kernel socket that is
sending the traffic, because the UID of kernel sockets created
at namespace creation time (e.g., the per-processor ICMP and
TCP sockets) is the UID of the user that created the socket,
which might not be mapped in the namespace.
Tested: compiles allnoconfig, allyesconfig, allmodconfig
Tested: https://android-review.googlesource.com/253302
Signed-off-by: Lorenzo Colitti <lorenzo@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Mostly simple overlapping changes.
For example, David Ahern's adjacency list revamp in 'net-next'
conflicted with an adjacency list traversal bug fix in 'net'.
Signed-off-by: David S. Miller <davem@davemloft.net>
First bug was added in commit ad6f939ab1 ("ip: Add offset parameter to
ip_cmsg_recv") : Tom missed that ipv4 udp messages could be received on
AF_INET6 socket. ip_cmsg_recv(msg, skb) should have been replaced by
ip_cmsg_recv_offset(msg, skb, sizeof(struct udphdr));
Then commit e6afc8ace6 ("udp: remove headers from UDP packets before
queueing") forgot to adjust the offsets now UDP headers are pulled
before skb are put in receive queue.
Fixes: ad6f939ab1 ("ip: Add offset parameter to ip_cmsg_recv")
Fixes: e6afc8ace6 ("udp: remove headers from UDP packets before queueing")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Sam Kumar <samanthakumar@google.com>
Cc: Willem de Bruijn <willemb@google.com>
Tested-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Completely avoid default sock memory accounting and replace it
with udp-specific accounting.
Since the new memory accounting model encapsulates completely
the required locking, remove the socket lock on both enqueue and
dequeue, and avoid using the backlog on enqueue.
Be sure to clean-up rx queue memory on socket destruction, using
udp its own sk_destruct.
Tested using pktgen with random src port, 64 bytes packet,
wire-speed on a 10G link as sender and udp_sink as the receiver,
using an l4 tuple rxhash to stress the contention, and one or more
udp_sink instances with reuseport.
nr readers Kpps (vanilla) Kpps (patched)
1 170 440
3 1250 2150
6 3000 3650
9 4200 4450
12 5700 6250
v4 -> v5:
- avoid unneeded test in first_packet_length
v3 -> v4:
- remove useless sk_rcvqueues_full() call
v2 -> v3:
- do not set the now unsed backlog_rcv callback
v1 -> v2:
- add memory pressure support
- fixed dropwatch accounting for ipv6
Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Now RCU lookups of ipv6 udp sockets no longer dereference
pinet6 field, we can get rid of udp_v6_clear_sk() helper.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This implements SOCK_DESTROY for UDP sockets similar to what was done
for TCP with commit c1e64e298b ("net: diag: Support destroying TCP
sockets.") A process with a UDP socket targeted for destroy is awakened
and recvmsg fails with ECONNABORTED.
Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
After commit ca065d0cf8 ("udp: no longer use SLAB_DESTROY_BY_RCU")
we do not need this special allocation mode anymore, even if it is
harmless.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
After a612769774 ("udp: prevent bugcheck if filter truncates packet
too much"), there followed various other fixes for similar cases such
as f4979fcea7 ("rose: limit sk_filter trim to payload").
Latter introduced a new helper sk_filter_trim_cap(), where we can pass
the trim limit directly to the socket filter handling. Make use of it
here as well with sizeof(struct udphdr) as lower cap limit and drop the
extra skb->len test in UDP's input path.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Cc: Willem de Bruijn <willemb@google.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If socket filter truncates an udp packet below the length of UDP header
in udpv6_queue_rcv_skb() or udp_queue_rcv_skb(), it will trigger a
BUG_ON in skb_pull_rcsum(). This BUG_ON (and therefore a system crash if
kernel is configured that way) can be easily enforced by an unprivileged
user which was reported as CVE-2016-6162. For a reproducer, see
http://seclists.org/oss-sec/2016/q3/8
Fixes: e6afc8ace6 ("udp: remove headers from UDP packets before queueing")
Reported-by: Marco Grassi <marco.gra@gmail.com>
Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Acked-by: Eric Dumazet <edumazet@google.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Several cases of overlapping changes, except the packet scheduler
conflicts which deal with the addition of the free list parameter
to qdisc_enqueue().
Signed-off-by: David S. Miller <davem@davemloft.net>
There is a corner case in which udp packets belonging to a same
flow are hashed to different socket when hslot->count changes from 10
to 11:
1) When hslot->count <= 10, __udp_lib_lookup() searches udp_table->hash,
and always passes 'daddr' to udp_ehashfn().
2) When hslot->count > 10, __udp_lib_lookup() searches udp_table->hash2,
but may pass 'INADDR_ANY' to udp_ehashfn() if the sockets are bound to
INADDR_ANY instead of some specific addr.
That means when hslot->count changes from 10 to 11, the hash calculated by
udp_ehashfn() is also changed, and the udp packets belonging to a same
flow will be hashed to different socket.
This is easily reproduced:
1) Create 10 udp sockets and bind all of them to 0.0.0.0:40000.
2) From the same host send udp packets to 127.0.0.1:40000, record the
socket index which receives the packets.
3) Create 1 more udp socket and bind it to 0.0.0.0:44096. The number 44096
is 40000 + UDP_HASH_SIZE(4096), this makes the new socket put into the
same hslot as the aformentioned 10 sockets, and makes the hslot->count
change from 10 to 11.
4) From the same host send udp packets to 127.0.0.1:40000, and the socket
index which receives the packets will be different from the one received
in step 2.
This should not happen as the socket bound to 0.0.0.0:44096 should not
change the behavior of the sockets bound to 0.0.0.0:40000.
It's the same case for IPv6, and this patch also fixes that.
Signed-off-by: Su, Xuemin <suxm@chinanetcenter.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In IPv6 the ToS values are part of the flowlabel in flowi6 and get
extracted during fib rule lookup, but we forgot to correctly initialize
the flowlabel before the routing lookup.
Reported-by: <liam.mcbirnie@boeing.com>
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Paul Moore tracked a regression caused by a recent commit, which
mistakenly assumed that sk_filter() could be avoided if socket
had no current BPF filter.
The intent was to avoid udp_lib_checksum_complete() overhead.
But sk_filter() also checks skb_pfmemalloc() and
security_sock_rcv_skb(), so better call it.
Fixes: e6afc8ace6 ("udp: remove headers from UDP packets before queueing")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Paul Moore <paul@paul-moore.com>
Tested-by: Paul Moore <paul@paul-moore.com>
Tested-by: Stephen Smalley <sds@tycho.nsa.gov>
Cc: samanthakumar <samanthakumar@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In case we find a socket with encapsulation enabled we should call
the encap_recv function even if just a udp header without payload is
available. The callbacks are responsible for correctly verifying and
dropping the packets.
Also, in case the header validation fails for geneve and vxlan we
shouldn't put the skb back into the socket queue, no one will pick
them up there. Instead we can simply discard them in the respective
encap_recv functions.
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
While testing an OpenStack configuration using VXLANs I saw the following
call trace:
RIP: 0010:[<ffffffff815fad49>] udp4_lib_lookup_skb+0x49/0x80
RSP: 0018:ffff88103867bc50 EFLAGS: 00010286
RAX: ffff88103269bf00 RBX: ffff88103269bf00 RCX: 00000000ffffffff
RDX: 0000000000004300 RSI: 0000000000000000 RDI: ffff880f2932e780
RBP: ffff88103867bc60 R08: 0000000000000000 R09: 000000009001a8c0
R10: 0000000000004400 R11: ffffffff81333a58 R12: ffff880f2932e794
R13: 0000000000000014 R14: 0000000000000014 R15: ffffe8efbfd89ca0
FS: 0000000000000000(0000) GS:ffff88103fd80000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000488 CR3: 0000000001c06000 CR4: 00000000001426e0
Stack:
ffffffff81576515 ffffffff815733c0 ffff88103867bc98 ffffffff815fcc17
ffff88103269bf00 ffffe8efbfd89ca0 0000000000000014 0000000000000080
ffffe8efbfd89ca0 ffff88103867bcc8 ffffffff815fcf8b ffff880f2932e794
Call Trace:
[<ffffffff81576515>] ? skb_checksum+0x35/0x50
[<ffffffff815733c0>] ? skb_push+0x40/0x40
[<ffffffff815fcc17>] udp_gro_receive+0x57/0x130
[<ffffffff815fcf8b>] udp4_gro_receive+0x10b/0x2c0
[<ffffffff81605863>] inet_gro_receive+0x1d3/0x270
[<ffffffff81589e59>] dev_gro_receive+0x269/0x3b0
[<ffffffff8158a1b8>] napi_gro_receive+0x38/0x120
[<ffffffffa0871297>] gro_cell_poll+0x57/0x80 [vxlan]
[<ffffffff815899d0>] net_rx_action+0x160/0x380
[<ffffffff816965c7>] __do_softirq+0xd7/0x2c5
[<ffffffff8107d969>] run_ksoftirqd+0x29/0x50
[<ffffffff8109a50f>] smpboot_thread_fn+0x10f/0x160
[<ffffffff8109a400>] ? sort_range+0x30/0x30
[<ffffffff81096da8>] kthread+0xd8/0xf0
[<ffffffff81693c82>] ret_from_fork+0x22/0x40
[<ffffffff81096cd0>] ? kthread_park+0x60/0x60
The following trace is seen when receiving a DHCP request over a flow-based
VXLAN tunnel. I believe this is caused by the metadata dst having a NULL
dev value and as a result dev_net(dev) is causing a NULL pointer dereference.
To resolve this I am replacing the check for skb_dst(skb)->dev with just
skb->dev. This makes sense as the callers of this function are usually in
the receive path and as such skb->dev should always be populated. In
addition other functions in the area where these are called are already
using dev_net(skb->dev) to determine the namespace the UDP packet belongs
in.
Fixes: 63058308cd ("udp: Add udp6_lib_lookup_skb and udp4_lib_lookup_skb")
Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In the sendmsg function of UDP, raw, ICMP and l2tp sockets, we use local
variables like hlimits, tclass, opt and dontfrag and pass them to corresponding
functions like ip6_make_skb, ip6_append_data and xxx_push_pending_frames.
This is not a good practice and makes it hard to add new parameters.
This fix introduces a new struct ipcm6_cookie similar to ipcm_cookie in
ipv4 and include the above mentioned variables. And we only pass the
pointer to this structure to corresponding functions. This makes it easier
to add new parameters in the future and makes the function cleaner.
Signed-off-by: Wei Wang <weiwan@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
UDP uses the generic socket backlog code, and this will soon
be changed to not disable BH when protocol is called back.
We need to use appropriate SNMP accessors.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Rename ICMP6_INC_STATS_BH() to __ICMP6_INC_STATS()
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Rename UDP_INC_STATS_BH() to __UDP_INC_STATS(),
and UDP6_INC_STATS_BH() to __UDP6_INC_STATS()
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In the old days (before linux-3.0), SNMP counters were duplicated,
one for user context, and one for BH context.
After commit 8f0ea0fe3a ("snmp: reduce percpu needs by 50%")
we have a single copy, and what really matters is preemption being
enabled or disabled, since we use this_cpu_inc() or __this_cpu_inc()
respectively.
We therefore kill SNMP_INC_STATS_USER(), SNMP_ADD_STATS_USER(),
NET_INC_STATS_USER(), NET_ADD_STATS_USER(), SCTP_INC_STATS_USER(),
SNMP_INC_STATS64_USER(), SNMP_ADD_STATS64_USER(), TCP_ADD_STATS_USER(),
UDP_INC_STATS_USER(), UDP6_INC_STATS_USER(), and XFRM_INC_STATS_USER()
Following patches will rename __BH helpers to make clear their
usage is not tied to BH being disabled.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Conflicts were two cases of simple overlapping changes,
nothing serious.
In the UDP case, we need to add a hlist_add_tail_rcu()
to linux/rculist.h, because we've moved UDP socket handling
away from using nulls lists.
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds a release_cb for UDPv6. It does a route lookup
and updates sk->sk_dst_cache if it is needed. It picks up the
left-over job from ip6_sk_update_pmtu() if the sk was owned
by user during the pmtu update.
It takes a rcu_read_lock to protect the __sk_dst_get() operations
because another thread may do ip6_dst_store() without taking the
sk lock (e.g. sendmsg).
Fixes: 45e4fd2668 ("ipv6: Only create RTF_CACHE routes after encountering pmtu exception")
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Reported-by: Wei Wang <weiwan@google.com>
Cc: Cong Wang <xiyou.wangcong@gmail.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Wei Wang <weiwan@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add externally visible functions to lookup a UDP socket by skb. This
will be used for GRO in UDP sockets. These functions also check
if skb->dst is set, and if it is not skb->dev is used to get dev_net.
This allows calling lookup functions before dst has been set on the
skbuff.
Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Enable peeking at UDP datagrams at the offset specified with socket
option SOL_SOCKET/SO_PEEK_OFF. Peek at any datagram in the queue, up
to the end of the given datagram.
Implement the SO_PEEK_OFF semantics introduced in commit ef64a54f6e
("sock: Introduce the SO_PEEK_OFF sock option"). Increase the offset
on peek, decrease it on regular reads.
When peeking, always checksum the packet immediately, to avoid
recomputation on subsequent peeks and final read.
The socket lock is not held for the duration of udp_recvmsg, so
peek and read operations can run concurrently. Only the last store
to sk_peek_off is preserved.
Signed-off-by: Sam Kumar <samanthakumar@google.com>
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove UDP transport headers before queueing packets for reception.
This change simplifies a follow-up patch to add MSG_PEEK support.
Signed-off-by: Sam Kumar <samanthakumar@google.com>
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Tom Herbert would like not touching UDP socket refcnt for encapsulated
traffic. For this to happen, we need to use normal RCU rules, with a grace
period before freeing a socket. UDP sockets are not short lived in the
high usage case, so the added cost of call_rcu() should not be a concern.
This actually removes a lot of complexity in UDP stack.
Multicast receives no longer need to hold a bucket spinlock.
Note that ip early demux still needs to take a reference on the socket.
Same remark for functions used by xt_socket and xt_PROXY netfilter modules,
but this might be changed later.
Performance for a single UDP socket receiving flood traffic from
many RX queues/cpus.
Simple udp_rx using simple recvfrom() loop :
438 kpps instead of 374 kpps : 17 % increase of the peak rate.
v2: Addressed Willem de Bruijn feedback in multicast handling
- keep early demux break in __udp4_lib_demux_lookup()
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Tom Herbert <tom@herbertland.com>
Cc: Willem de Bruijn <willemb@google.com>
Tested-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, SOL_TIMESTAMPING can only be enabled using setsockopt.
This is very costly when users want to sample writes to gather
tx timestamps.
Add support for enabling SO_TIMESTAMPING via control messages by
using tsflags added in `struct sockcm_cookie` (added in the previous
patches in this series) to set the tx_flags of the last skb created in
a sendmsg. With this patch, the timestamp recording bits in tx_flags
of the skbuff is overridden if SO_TIMESTAMPING is passed in a cmsg.
Please note that this is only effective for overriding the recording
timestamps flags. Users should enable timestamp reporting (e.g.,
SOF_TIMESTAMPING_SOFTWARE | SOF_TIMESTAMPING_OPT_ID) using
socket options and then should ask for SOF_TIMESTAMPING_TX_*
using control messages per sendmsg to sample timestamps for each
write.
Signed-off-by: Soheil Hassas Yeganeh <soheil@google.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Process socket-level control messages by invoking
__sock_cmsg_send in ip6_datagram_send_ctl for control messages on
the SOL_SOCKET layer.
This makes sure whenever ip6_datagram_send_ctl is called for
udp and raw, we also process socket-level control messages.
This is a bit uglier than IPv4, since IPv6 does not have
something like ipcm_cookie. Perhaps we can later create
a control message cookie for IPv6?
Note that this commit interprets new control messages that
were ignored before. As such, this commit does not change
the behavior of IPv6 control messages.
Signed-off-by: Soheil Hassas Yeganeh <soheil@google.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
IPv6 counters updates use a different macro than IPv4.
Fixes: 36cbb2452c ("udp: Increment UDP_MIB_IGNOREDMULTI for arriving unmatched multicasts")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Rick Jones <rick.jones2@hp.com>
Cc: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Several cases of overlapping changes, as well as one instance
(vxlan) of a bug fix in 'net' overlapping with code movement
in 'net-next'.
Signed-off-by: David S. Miller <davem@davemloft.net>
IPv4 interprets a negative return value from a protocol handler as a
request to redispatch to a new protocol. In contrast, IPv6 interprets a
negative value as an error, and interprets a positive value as a request
for redispatch.
UDP for IPv6 was unaware of this difference. Change __udp6_lib_rcv() to
return a positive value for redispatch. Note that the socket's
encap_rcv hook still needs to return a negative value to request
dispatch, and in the case of IPv6 packets, adjust IP6CB(skb)->nhoff to
identify the byte containing the next protocol.
Signed-off-by: Bill Sommerfeld <wsommerfeld@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In ipv4, when the machine receives a ICMP_FRAG_NEEDED message, the
connected UDP socket will get EMSGSIZE message on its next read from the
socket.
However, this is not the case for ipv6.
This fix modifies the udp err handler in Ipv6 for ICMP6_PKT_TOOBIG to
make it similar to ipv4 behavior. That is when the machine gets an
ICMP6_PKT_TOOBIG message, the connected UDP socket will get EMSGSIZE
message on its next read from the socket.
Signed-off-by: Wei Wang <weiwan@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In order to support fast lookups for TCP sockets with SO_REUSEPORT,
the function that adds sockets to the listening hash set needs
to be able to check receive address equality. Since this equality
check is different for IPv4 and IPv6, we will need two different
socket hashing functions.
This patch adds inet6_hash identical to the existing inet_hash function
and updates the appropriate references. A following patch will
differentiate the two by passing different comparison functions to
__inet_hash.
Additionally, in order to use the IPv6 address equality function from
inet6_hashtables (which is compiled as a built-in object when IPv6 is
enabled) it also needs to be in a built-in object file as well. This
moves ipv6_rcv_saddr_equal into inet_hashtables to accomplish this.
Signed-off-by: Craig Gallek <kraig@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Using a combination of connected and un-connected sockets, Dmitry
was able to trigger soft lockups with his fuzzer.
The problem is that sockets in the SO_REUSEPORT array might have
different scores.
Right after sk2=socket(), setsockopt(sk2,...,SO_REUSEPORT, on) and
bind(sk2, ...), but _before_ the connect(sk2) is done, sk2 is added into
the soreuseport array, with a score which is smaller than the score of
first socket sk1 found in hash table (I am speaking of the regular UDP
hash table), if sk1 had the connect() done, giving a +8 to its score.
hash bucket [X] -> sk1 -> sk2 -> NULL
sk1 score = 14 (because it did a connect())
sk2 score = 6
SO_REUSEPORT fast selection is an optimization. If it turns out the
score of the selected socket does not match score of first socket, just
fallback to old SO_REUSEPORT logic instead of trying to be too smart.
Normal SO_REUSEPORT users do not mix different kind of sockets, as this
mechanism is used for load balance traffic.
Fixes: e32ea7e747 ("soreuseport: fast reuseport UDP socket selection")
Reported-by: Dmitry Vyukov <dvyukov@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Craig Gallek <kraigatgoog@gmail.com>
Acked-by: Craig Gallek <kraig@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This socket-lookup path did not pass along the skb in question
in my original BPF-based socket selection patch. The skb in the
udpN_lib_lookup2 path can be used for BPF-based socket selection just
like it is in the 'traditional' udpN_lib_lookup path.
udpN_lib_lookup2 kicks in when there are greater than 10 sockets in
the same hlist slot. Coincidentally, I chose 10 sockets per
reuseport group in my functional test, so the lookup2 path was not
excersised. This adds an additional set of tests with 20 sockets.
Fixes: 538950a1b7 ("soreuseport: setsockopt SO_ATTACH_REUSEPORT_[CE]BPF")
Fixes: 3ca8e40299 ("soreuseport: BPF selection functional test")
Suggested-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Craig Gallek <kraig@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Expose socket options for setting a classic or extended BPF program
for use when selecting sockets in an SO_REUSEPORT group. These options
can be used on the first socket to belong to a group before bind or
on any socket in the group after bind.
This change includes refactoring of the existing sk_filter code to
allow reuse of the existing BPF filter validation checks.
Signed-off-by: Craig Gallek <kraig@google.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
