13 Dec, 2016
14 commits
-
Could be useful for debugging memory consumption issues,
and perhaps power-save as well.Signed-off-by: Ben Greear
Signed-off-by: Johannes Berg -
Previously, kernel sends NEW_PEER_CANDIDATE event to user land even if
the found peer does not have any room to accept other peer. This causes
continuous connection trials.Signed-off-by: Masashi Honma
Signed-off-by: Johannes Berg -
Add the ability for an AP (and associated VLANs) to perform
multicast-to-unicast conversion for ARP, IPv4 and IPv6 frames
(possibly within 802.1Q). If enabled, such frames are to be sent
to each station separately, with the DA replaced by their own
MAC address rather than the group address.Note that this may break certain expectations of the receiver,
such as the ability to drop unicast IP packets received within
multicast L2 frames, or the ability to not send ICMP destination
unreachable messages for packets received in L2 multicast (which
is required, but the receiver can't tell the difference if this
new option is enabled.)This also doesn't implement the 802.11 DMS (directed multicast
service).Signed-off-by: Michael Braun
[use true/false, rename label to the correct "multicast",
use __be16 for ethertype and network order for constants]
Signed-off-by: Johannes Berg -
Commit 4a733ef1bea7 (mac80211: remove PM-QoS listener) removed all use
of 'beaconint_us' from ieee80211_recalc_ps() but left the variable
intact. Compiling with W=1 gives the following warning, fix it.
net/mac80211/mlme.c: In function ‘ieee80211_recalc_ps’:
net/mac80211/mlme.c:1481:7: warning: variable ‘beaconint_us’ set but not used [-Wunused-but-set-variable]iee80211_tu_to_usec has no side-effects and is safe to remove.
Fixes: 4a733ef1bea7 ("mac80211: remove PM-QoS listener")
Cc: Johannes Berg
Signed-off-by: Kirtika Ruchandani
Signed-off-by: Johannes Berg -
Commit b1bce14a7954 (mac80211: update opmode when adding new station)
refactored ieee80211_vht_handle_opmode into __ieee80211_vht_handle_opmode
and ieee80211_vht_handle_opmode leaving a set but unused variable
(sband) in the former. Compiling with W=1 gives the following warning,
fix it.net/mac80211/vht.c: In function ‘__ieee80211_vht_handle_opmode’:
net/mac80211/vht.c:424:35: warning: variable ‘sband’ set but not used [-Wunused-but-set-variable]Remove 'struct ieee80211_local* local' as well, it was only used to
set sband.This is a harmless warning, and is only being fixed to reduce the
noise with W=1 in the kernel.Fixes: b1bce14a7954 ("mac80211: update opmode when adding new station")
Cc: Marek Kwaczynski
Cc: Johannes Berg
Signed-off-by: Kirtika Ruchandani
Signed-off-by: Johannes Berg -
Commit 633e27132625 (mac80211: split sched scan IEs) introduced the
len variable to keep track of the return value of
ieee80211_build_preq_ies() but did not use it. Compiling with W=1
gives the following warning, fix it.net/mac80211/scan.c: In function ‘__ieee80211_request_sched_scan_start’:
net/mac80211/scan.c:1123:9: warning: variable ‘len’ set but not used [-Wunused-but-set-variable]This is a harmless warning and is only being fixed to reduce the noise
with W=1 in the kernel.Fixes: 633e27132625 ("mac80211: split sched scan IEs")
Cc: David Spinadel
Cc: Alexander Bondar
Cc: Johannes Berg
Signed-off-by: Kirtika Ruchandani
Signed-off-by: Johannes Berg -
Commit 5bcae31d9 (mac80211: implement multi-vif in-place reservations)
introduced ieee80211_vif_use_reserved_switch() with a counter variable
'i' that is set but not used. Compiling with W=1 gives the following
warning, fix it.
net/mac80211/chan.c: In function ‘ieee80211_vif_use_reserved_switch’:
net/mac80211/chan.c:1273:6: warning: variable ‘i’ set but not used [-Wunused-but-set-variable]This is a harmless warning, and is only being fixed to reduce the
noise obtained with W=1 in the kernel.Fixes: 5bcae31d9 ("mac80211: implement multi-vif in-place reservations")
Cc: Michal Kazior
Cc: Johannes Berg
Signed-off-by: Kirtika Ruchandani
Signed-off-by: Johannes Berg -
Commit 3b17fbf87d5d introduced sta_get_expected_throughput()
leaving variable 'struct rate_control_ref* ref' set but unused.
Compiling with W=1 gives the following warning, fix it.net/mac80211/sta_info.c: In function ‘sta_set_sinfo’:
net/mac80211/sta_info.c:2052:27: warning: variable ‘ref’ set but not used [-Wunused-but-set-variable]Fixes: 3b17fbf87d5d ("mac80211: mesh: Add support for HW RC implementation")
Cc: Johannes Berg
Cc: Maxim Altshul
Signed-off-by: Kirtika Ruchandani
Signed-off-by: Johannes Berg -
Commit f027c2aca0cf introduced 'rates_idx' in
ieee80211_tx_status_noskb but did not use it. Compiling with W=1
gives the following warning, fix it.mac80211/status.c: In function ‘ieee80211_tx_status_noskb’:
mac80211/status.c:636:6: warning: variable ‘rates_idx’ set but not used [-Wunused-but-set-variable]This is a harmless warning, and is only being fixed to reduce the
noise generated with W=1.Fixes: f027c2aca0cf ("mac80211: add ieee80211_tx_status_noskb")
Cc: Johannes Berg
Cc: Felix Fietkau
Signed-off-by: Kirtika Ruchandani
Signed-off-by: Johannes Berg -
Commit 554891e63a29 introduced 'struct ieee80211_rx_status' in
ieee80211_rx_h_defragment but did not use it. Compiling with W=1
gives the following warning, fix it.net/mac80211/rx.c: In function ‘ieee80211_rx_h_defragment’:
net/mac80211/rx.c:1911:30: warning: variable ‘status’ set but not used [-Wunused-but-set-variable]Fixes: 554891e63a29 ("mac80211: move packet flags into packet")
Cc: Johannes Berg
Cc: John W. Linville
Signed-off-by: Kirtika Ruchandani
Signed-off-by: Johannes Berg -
There is no need to prevent toggling multicast_to_unicast while
interface is already up. This change simplifies reconfiguration
from hostapd.Signed-off-by: Michael Braun
Signed-off-by: Johannes Berg -
The presence of the NL80211_ATTR_SCHED_SCAN_INTERVAL attribute was
checked in nl80211_parse_sched_scan() and
nl80211_parse_sched_scan_plans() which might be a bit redundant
so removing one.Signed-off-by: Arend van Spriel
Signed-off-by: Johannes Berg -
The comment on the name indirection suggested an issue but turned out
to be untrue. Digging in older kernel version showed issue with ipw2x00
but that is no longer true so get rid on the name indirection.Signed-off-by: Arend van Spriel
Signed-off-by: Johannes Berg -
Simplify the two conditions gating the schedule_work() into
a single one and get rid of the additional exit point from
the function in doing so.Signed-off-by: Johannes Berg
11 Dec, 2016
9 commits
-
Dump and reset doesn't work unless cmpxchg64() is used both from packet
and control plane paths. This approach is going to be slow though.
Instead, use a percpu seqcount to fetch counters consistently, then
subtract bytes and packets in case a reset was requested.The cpu that running over the reset code is guaranteed to own this stats
exclusively, we have to turn counters into signed 64bit though so stats
update on reset don't get wrong on underflow.This patch is based on original sketch from Eric Dumazet.
Fixes: 43da04a593d8 ("netfilter: nf_tables: atomic dump and reset for stateful objects")
Suggested-by: Eric Dumazet
Signed-off-by: Pablo Neira Ayuso
Signed-off-by: David S. Miller -
Signed-off-by: Asbjoern Sloth Toennesen
Signed-off-by: David S. Miller -
Move the L2TP_MSG_* definitions to UAPI, as it is part of
the netlink API.Signed-off-by: Asbjoern Sloth Toennesen
Signed-off-by: David S. Miller -
802.1D [1] specifies that the bridges must use a short value to age out
dynamic entries in the Filtering Database for a period, once a topology
change has been communicated by the root bridge.Add a bridge_ageing_time member in the net_bridge structure to store the
bridge ageing time value configured by the user (ioctl/netlink/sysfs).If we are using in-kernel STP, shorten the ageing time value to twice
the forward delay used by the topology when the topology change flag is
set. When the flag is cleared, restore the configured ageing time.[1] "8.3.5 Notifying topology changes ",
http://profesores.elo.utfsm.cl/~agv/elo309/doc/802.1D-1998.pdfSigned-off-by: Vivien Didelot
Signed-off-by: David S. Miller -
Add a __br_set_topology_change helper to set the topology change value.
This can be later extended to add actions when the topology change flag
is set or cleared.Signed-off-by: Vivien Didelot
Signed-off-by: David S. Miller -
The SWITCHDEV_ATTR_ID_BRIDGE_AGEING_TIME switchdev attr is actually set
when initializing a bridge port, and when configuring the bridge ageing
time from ioctl/netlink/sysfs.Add a __set_ageing_time helper to offload the ageing time to physical
switches, and add the SWITCHDEV_F_DEFER flag since it can be called
under bridge lock.Signed-off-by: Vivien Didelot
Signed-off-by: David S. Miller -
This patch removes a newline which was added
in socket.c file in net-nextSigned-off-by: Amit Kushwaha
Signed-off-by: David S. Miller -
netlink_chain is called in ->release(), which is apparently
a process context, so we don't have to use an atomic notifier
here.Signed-off-by: Cong Wang
Signed-off-by: David S. Miller
10 Dec, 2016
6 commits
-
It seems attackers can also send UDP packets with no payload at all.
skb_condense() can still be a win in this case.
It will be possible to replace the custom code in tcp_add_backlog()
to get full benefit from skb_condense()Signed-off-by: Eric Dumazet
Signed-off-by: David S. Miller -
…inux/kernel/git/jberg/mac80211-next
Johannes Berg says:
====================
Three fixes:
* fix a logic bug introduced by a previous cleanup
* fix nl80211 attribute confusing (trying to use
a single attribute for two purposes)
* fix a long-standing BSS leak that happens when an
association attempt is abandoned
====================Signed-off-by: David S. Miller <davem@davemloft.net>
-
In flood situations, keeping sk_rmem_alloc at a high value
prevents producers from touching the socket.It makes sense to lower sk_rmem_alloc only at the end
of udp_rmem_release() after the thread draining receive
queue in udp_recvmsg() finished the writes to sk_forward_alloc.Signed-off-by: Eric Dumazet
Signed-off-by: David S. Miller -
If udp_recvmsg() constantly releases sk_rmem_alloc
for every read packet, it gives opportunity for
producers to immediately grab spinlocks and desperatly
try adding another packet, causing false sharing.We can add a simple heuristic to give the signal
by batches of ~25 % of the queue capacity.This patch considerably increases performance under
flood by about 50 %, since the thread draining the queue
is no longer slowed by false sharing.Signed-off-by: Eric Dumazet
Signed-off-by: David S. Miller -
In UDP RX handler, we currently clear skb->dev before skb
is added to receive queue, because device pointer is no longer
available once we exit from RCU section.Since this first cache line is always hot, lets reuse this space
to store skb->truesize and thus avoid a cache line miss at
udp_recvmsg()/udp_skb_destructor time while receive queue
spinlock is held.Signed-off-by: Eric Dumazet
Signed-off-by: David S. Miller -
Idea of busylocks is to let producers grab an extra spinlock
to relieve pressure on the receive_queue spinlock shared by consumer.This behavior is requested only once socket receive queue is above
half occupancy.Under flood, this means that only one producer can be in line
trying to acquire the receive_queue spinlock.These busylock can be allocated on a per cpu manner, instead of a
per socket one (that would consume a cache line per socket)This patch considerably improves UDP behavior under stress,
depending on number of NIC RX queues and/or RPS spread.Signed-off-by: Eric Dumazet
Signed-off-by: David S. Miller
09 Dec, 2016
11 commits
-
When mac80211 abandons an association attempt, it may free
all the data structures, but inform cfg80211 and userspace
about it only by sending the deauth frame it received, in
which case cfg80211 has no link to the BSS struct that was
used and will not cfg80211_unhold_bss() it.Fix this by providing a way to inform cfg80211 of this with
the BSS entry passed, so that it can clean up properly, and
use this ability in the appropriate places in mac80211.This isn't ideal: some code is more or less duplicated and
tracing is missing. However, it's a fairly small change and
it's thus easier to backport - cleanups can come later.Cc: stable@vger.kernel.org
Signed-off-by: Johannes Berg -
NL80211_ATTR_MAC was used to set both the specific BSSID to be scanned
and the random MAC address to be used when privacy is enabled. When both
the features are enabled, both the BSSID and the local MAC address were
getting same value causing Probe Request frames to go with unintended
DA. Hence, this has been fixed by using a different NL80211_ATTR_BSSID
attribute to set the specific BSSID (which was the more recent addition
in cfg80211) for a scan.Backwards compatibility with old userspace software is maintained to
some extent by allowing NL80211_ATTR_MAC to be used to set the specific
BSSID when scanning without enabling random MAC address use.Scanning with random source MAC address was introduced by commit
ad2b26abc157 ("cfg80211: allow drivers to support random MAC addresses
for scan") and the issue was introduced with the addition of the second
user for the same attribute in commit 818965d39177 ("cfg80211: Allow a
scan request for a specific BSSID").Fixes: 818965d39177 ("cfg80211: Allow a scan request for a specific BSSID")
Signed-off-by: Vamsi Krishna
Signed-off-by: Jouni Malinen
Signed-off-by: Johannes Berg -
Arend inadvertently inverted the logic while converting to
wdev_running(), fix that.Fixes: 73c7da3dae1e ("cfg80211: add generic helper to check interface is running")
Signed-off-by: Johannes Berg -
This patch cleanup checkpatch.pl warning
WARNING: __aligned(size) is preferred over __attribute__((aligned(size)))Signed-off-by: Amit Kushwaha
Signed-off-by: David S. Miller -
…etooth/bluetooth-next
Johan Hedberg says:
====================
pull request: bluetooth-next 2016-12-08I didn't miss your "net-next is closed" email, but it did come as a bit
of a surprise, and due to time-zone differences I didn't have a chance
to react to it until now. We would have had a couple of patches in
bluetooth-next that we'd still have wanted to get to 4.10.Out of these the most critical one is the H7/CT2 patch for Bluetooth
Security Manager Protocol, something that couldn't be published before
the Bluetooth 5.0 specification went public (yesterday). If these really
can't go to net-next we'll likely be sending at least this patch through
bluetooth.git to net.git for rc1 inclusion.
====================Signed-off-by: David S. Miller <davem@davemloft.net>
-
This patch allows XDP prog to extend/remove the packet
data at the head (like adding or removing header). It is
done by adding a new XDP helper bpf_xdp_adjust_head().It also renames bpf_helper_changes_skb_data() to
bpf_helper_changes_pkt_data() to better reflect
that XDP prog does not work on skb.This patch adds one "xdp_adjust_head" bit to bpf_prog for the
XDP-capable driver to check if the XDP prog requires
bpf_xdp_adjust_head() support. The driver can then decide
to error out during XDP_SETUP_PROG.Signed-off-by: Martin KaFai Lau
Acked-by: Daniel Borkmann
Acked-by: Alexei Starovoitov
Acked-by: John Fastabend
Signed-off-by: David S. Miller -
Under UDP flood, many softirq producers try to add packets to
UDP receive queue, and one user thread is burning one cpu trying
to dequeue packets as fast as possible.Two parts of the per packet cost are :
- copying payload from kernel space to user space,
- freeing memory pieces associated with skb.If socket is under pressure, softirq handler(s) can try to pull in
skb->head the payload of the packet if it fits.Meaning the softirq handler(s) can free/reuse the page fragment
immediately, instead of letting udp_recvmsg() do this hundreds of usec
later, possibly from another node.Additional gains :
- We reduce skb->truesize and thus can store more packets per SO_RCVBUF
- We avoid cache line misses at copyout() time and consume_skb() time,
and avoid one put_page() with potential alien freeing on NUMA hosts.This comes at the cost of a copy, bounded to available tail room, which
is usually small. (We might have to fix GRO_MAX_HEAD which looks bigger
than necessary)This patch gave me about 5 % increase in throughput in my tests.
skb_condense() helper could probably used in other contexts.
Signed-off-by: Eric Dumazet
Cc: Paolo Abeni
Signed-off-by: David S. Miller -
RFS is not commonly used, so add a jump label to avoid some conditionals
in fast path.Signed-off-by: Eric Dumazet
Cc: Paolo Abeni
Signed-off-by: David S. Miller -
Support matching on ICMP type and code.
Example usage:
tc qdisc add dev eth0 ingress
tc filter add dev eth0 protocol ip parent ffff: flower \
indev eth0 ip_proto icmp type 8 code 0 action droptc filter add dev eth0 protocol ipv6 parent ffff: flower \
indev eth0 ip_proto icmpv6 type 128 code 0 action dropSigned-off-by: Simon Horman
Acked-by: Jiri Pirko
Signed-off-by: David S. Miller -
Allow dissection of ICMP(V6) type and code. This should only occur
if a packet is ICMP(V6) and the dissector has FLOW_DISSECTOR_KEY_ICMP set.There are currently no users of FLOW_DISSECTOR_KEY_ICMP.
A follow-up patch will allow FLOW_DISSECTOR_KEY_ICMP to be used by
the flower classifier.Signed-off-by: Simon Horman
Acked-by: Jiri Pirko
Signed-off-by: David S. Miller -
Add UAPI to provide set of flags for matching, where the flags
provided from user-space are mapped to flow-dissector flags.The 1st flag allows to match on whether the packet is an
IP fragment and corresponds to the FLOW_DIS_IS_FRAGMENT flag.Signed-off-by: Or Gerlitz
Reviewed-by: Paul Blakey
Acked-by: Jiri Pirko
Signed-off-by: David S. Miller
