| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
The csa_active flag was added in sdata a while ago and made
IEEE80211_STA_CSA_RECEIVED redundant. The new flag is also used to
mark when CSA is ongoing on other iftypes and took over the old one as
the preferred method for checking whether we're in the middle of a
channel switch. Remove the old, redundant flag.
Signed-off-by: Luciano Coelho <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
|
|
Since the teardown packet is created while the queues are
stopped, it isn't sent immediately, but rather is pending.
To be sure that when we flush the queues prior to destroying
the station we also send this packet - the tasklet handling
pending packets is invoked to flush the packets.
Signed-off-by: Liad Kaufman <[email protected]>
Reviewed-by: ArikX Nemtsov <[email protected]>
Reviewed-by: Johannes Berg <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
|
|
If the AP receives actions frames destined for other peers, it may
mistakenly toggle BA-sessions from itself to a peer.
Ignore TDLS data packets as well - the AP should not handle them.
Signed-off-by: Arik Nemtsov <[email protected]>
Reviewed-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
|
|
Some VHT TDLS peers (Google Nexus 5) include the VHT-AID IE in their
TDLS setup request/response. Usermode passes this aid as the station
aid, causing it to fail verifiction, since this happens in the
"set_station" stage. Make an exception for the TDLS use-case.
Signed-off-by: Arik Nemtsov <[email protected]>
Reviewed-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
|
|
TDLS VHT support requires some more information elements during setup.
While these are not there, mask out the peer's VHT capabilities so that
VHT rates are not mistakenly used.
Signed-off-by: Arik Nemtsov <[email protected]>
Reviewed-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
|
|
Set for completeness mostly, currently unused in the code.
Signed-off-by: Arik Nemtsov <[email protected]>
Reviewed-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
|
|
Add the HT capabilities and HT operation information elements to TDLS
setup packets where appropriate.
Signed-off-by: Arik Nemtsov <[email protected]>
Reviewed-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
|
|
We can only be a station for TDLS connections. Also fix a bug where
a delayed work could be left scheduled if the station interface was
brought down during TDLS setup.
Signed-off-by: Arik Nemtsov <[email protected]>
Reviewed-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
|
|
When TDLS QoS is supported by the the peer and the local card, add
the WMM parameter IE to the setup-confirm frame. Take the QoS settings
from the current AP, or if unsupported, use the default values from
the specification. This behavior is mandated by IEEE802.11-2012 section
10.22.4.
Signed-off-by: Arik Nemtsov <[email protected]>
Reviewed-by: Liad Kaufman <[email protected]>
Reviewed-by: Johannes Berg <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
|
|
If QoS is supported by the card, add an appropriate IE to TDLS setup-
request and setup-response frames.
Consolidate the setting of the WMM info IE across mac80211.
Signed-off-by: Arik Nemtsov <[email protected]>
Reviewed-by: Liad Kaufman <[email protected]>
Reviewed-by: Johannes Berg <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
|
|
When sending setup-failure frames, set the capability field to zero, as
mandated by the specification (IEEE802.11-2012 8.5.13).
Signed-off-by: Arik Nemtsov <[email protected]>
Reviewed-by: Liad Kaufman <[email protected]>
Reviewed-by: Johannes Berg <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
|
|
Most setup-specific information elements are not to be added when a
setup frame is sent with an error status code.
Signed-off-by: Arik Nemtsov <[email protected]>
Reviewed-by: Liad Kaufman <[email protected]>
Reviewed-by: Johannes Berg <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
|
|
When building TDLS setup frames, use the IE order mandates in the
specification, splitting extra IEs coming from usermode.
Signed-off-by: Arik Nemtsov <[email protected]>
Reviewed-by: Johannes Berg <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
|
|
Add all information elements for TDLS discovery and setup in the same
function.
Signed-off-by: Arik Nemtsov <[email protected]>
Reviewed-by: Liad Kaufman <[email protected]>
Reviewed-by: Johannes Berg <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
|
|
The patch "8f02e6b mac80211: make sure TDLS peer STA exists during
setup" broke TDLS error paths where the STA doesn't exist when sending
the error.
Fix it by only testing for STA existence during a non-error flow.
Signed-off-by: Arik Nemtsov <[email protected]>
Reviewed-by: Johannes Berg <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
|
|
Infer the TDLS initiator and track it in mac80211 via a STA flag. This
avoids breaking old userspace that doesn't pass it via nl80211 APIs.
The only case where userspace will need to pass the initiator is when the
STA is removed due to unreachability before a teardown packet is sent.
Support for unreachability was only recently added to wpa_supplicant, so
it won't be a problem in practice.
Signed-off-by: Arik Nemtsov <[email protected]>
Reviewed-by: Johannes Berg <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
|
|
When a VLAN interface (on top of batX) is removed and
re-added within a short timeframe TT does not have enough
time to properly cleanup. This creates an internal TT state
mismatch as the newly created softif_vlan will be
initialized from scratch with a TT client count of zero
(even if TT entries for this VLAN still exist). The
resulting TT messages are bogus due to the counter / tt
client listing mismatch, thus creating inconsistencies on
every node in the network
To fix this issue destroy_vlan() has to not free the VLAN
object immediately but it has to be kept alive until all the
TT entries for this VLAN have been removed. destroy_vlan()
still removes the sysfs folder so that the user has the
feeling that everything went fine.
If the same VLAN is re-added before the old object is free'd,
then the latter is resurrected and re-used.
Implement such behaviour by increasing the reference counter
of a softif_vlan object every time a new local TT entry for
such VLAN is created and remove the object from the list
only when all the TT entries have been destroyed.
Signed-off-by: Antonio Quartulli <[email protected]>
Signed-off-by: Marek Lindner <[email protected]>
|
|
Since bridge loop avoidance only supports untagged or simple 802.1q
tagged VLAN claim frames, claim frames with stacked VLAN headers (QinQ)
should be detected and dropped. Transporting the over the mesh may cause
problems on the receivers, or create bogus entries in the local tt
tables.
Reported-by: Antonio Quartulli <[email protected]>
Signed-off-by: Simon Wunderlich <[email protected]>
Signed-off-by: Marek Lindner <[email protected]>
Signed-off-by: Antonio Quartulli <[email protected]>
|
|
*_result[len] is parsed as *(_result[len]) which is not at all what we
want to touch here.
Signed-off-by: Ben Hutchings <[email protected]>
Fixes: 84a7c0b1db1c ("dns_resolver: assure that dns_query() result is null-terminated")
Signed-off-by: David S. Miller <[email protected]>
|
|
Fixes the following sparse warnings:
net/tipc/socket.c:545:5: warning:
symbol 'tipc_sk_proto_rcv' was not declared. Should it be static?
net/tipc/socket.c:2015:5: warning:
symbol 'tipc_ioctl' was not declared. Should it be static?
Signed-off-by: Wei Yongjun <[email protected]>
Acked-by: Ying Xue <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=80611
Reported-by: David Binderman <[email protected]>
Signed-off-by: Andrey Utkin <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Pablo Neira Ayuso says:
====================
Netfilter updates for net-next
The following patchset contains updates for your net-next tree,
they are:
1) Use kvfree() helper function from x_tables, from Eric Dumazet.
2) Remove extra timer from the conntrack ecache extension, use a
workqueue instead to redeliver lost events to userspace instead,
from Florian Westphal.
3) Removal of the ulog targets for ebtables and iptables. The nflog
infrastructure superseded this almost 9 years ago, time to get rid
of this code.
4) Replace the list of loggers by an array now that we can only have
two possible non-overlapping logger flavours, ie. kernel ring buffer
and netlink logging.
5) Move Eric Dumazet's log buffer code to nf_log to reuse it from
all of the supported per-family loggers.
6) Consolidate nf_log_packet() as an unified interface for packet logging.
After this patch, if the struct nf_loginfo is available, it explicitly
selects the logger that is used.
7) Move ip and ip6 logging code from xt_LOG to the corresponding
per-family loggers. Thus, x_tables and nf_tables share the same code
for packet logging.
8) Add generic ARP packet logger, which is used by nf_tables. The
format aims to be consistent with the output of xt_LOG.
9) Add generic bridge packet logger. Again, this is used by nf_tables
and it routes the packets to the real family loggers. As a result,
we get consistent logging format for the bridge family. The ebt_log
logging code has been intentionally left in place not to break
backward compatibility since the logging output differs from xt_LOG.
10) Update nft_log to explicitly request the required family logger when
needed.
11) Finish nft_log so it supports arp, ip, ip6, bridge and inet families.
Allowing selection between netlink and kernel buffer ring logging.
12) Several fixes coming after the netfilter core logging changes spotted
by robots.
13) Use IS_ENABLED() macros whenever possible in the netfilter tree,
from Duan Jiong.
14) Removal of a couple of unnecessary branch before kfree, from Fabian
Frederick.
====================
Signed-off-by: David S. Miller <[email protected]>
|
|
When kernel generates a handle for a u32 filter, it tries to start
from the max in the bucket. So when we have a filter with the max (fff)
handle, it will cause kernel always generates the same handle for new
filters. This can be shown by the following command:
tc qdisc add dev eth0 ingress
tc filter add dev eth0 parent ffff: protocol ip pref 770 handle 800::fff u32 match ip protocol 1 0xff
tc filter add dev eth0 parent ffff: protocol ip pref 770 u32 match ip protocol 1 0xff
...
we will get some u32 filters with same handle:
# tc filter show dev eth0 parent ffff:
filter protocol ip pref 770 u32
filter protocol ip pref 770 u32 fh 800: ht divisor 1
filter protocol ip pref 770 u32 fh 800::fff order 4095 key ht 800 bkt 0
match 00010000/00ff0000 at 8
filter protocol ip pref 770 u32 fh 800::fff order 4095 key ht 800 bkt 0
match 00010000/00ff0000 at 8
filter protocol ip pref 770 u32 fh 800::fff order 4095 key ht 800 bkt 0
match 00010000/00ff0000 at 8
filter protocol ip pref 770 u32 fh 800::fff order 4095 key ht 800 bkt 0
match 00010000/00ff0000 at 8
handles should be unique. This patch fixes it by looking up a bitmap,
so that can guarantee the handle is as unique as possible. For compatibility,
we still start from 0x800.
Cc: "David S. Miller" <[email protected]>
Signed-off-by: Cong Wang <[email protected]>
Signed-off-by: Cong Wang <[email protected]>
Signed-off-by: Jamal Hadi Salim <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Currently it's done silently (from the kernel part), and thus it might be
hard to track the renames from logs.
Add a simple netdev_info() to notify the rename, but only in case the
previous name was valid.
CC: "David S. Miller" <[email protected]>
CC: Eric Dumazet <[email protected]>
CC: Vlad Yasevich <[email protected]>
CC: stephen hemminger <[email protected]>
CC: Jerry Chu <[email protected]>
CC: Ben Hutchings <[email protected]>
CC: David Laight <[email protected]>
Signed-off-by: Veaceslav Falico <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
This way we'll always know in what status the device is, unless it's
running normally (i.e. NETDEV_REGISTERED).
Also, emit a warning once in case of a bad reg_state.
CC: "David S. Miller" <[email protected]>
CC: Jason Baron <[email protected]>
CC: Eric Dumazet <[email protected]>
CC: Vlad Yasevich <[email protected]>
CC: stephen hemminger <[email protected]>
CC: Jerry Chu <[email protected]>
CC: Ben Hutchings <[email protected]>
CC: Joe Perches <[email protected]>
Signed-off-by: Veaceslav Falico <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
We modify mirred action (m->tcfm_dev) in netdev event, we need to
prevent on-going mirred actions from reading freed m->tcfm_dev.
So we need to acquire this spin lock.
Cc: Jamal Hadi Salim <[email protected]>
Cc: "David S. Miller" <[email protected]>
Signed-off-by: Cong Wang <[email protected]>
Signed-off-by: Cong Wang <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Add new "NFC_DIGITAL_FRAMING_*" calls to the digital
layer so the driver can make the necessary adjustments
when performing anticollision while in target mode.
The driver must ensure that the effect of these calls
happens after the following response has been sent but
before reception of the next request begins.
Acked-by: Thierry Escande <[email protected]>
Signed-off-by: Mark A. Greer <[email protected]>
Signed-off-by: Samuel Ortiz <[email protected]>
|
|
Currently, digital_target_found() has a race between
the events started by calling nfc_targets_found()
(which ultimately expect ddev->poll_tech_count to be
zero) and setting ddev->poll_tech_count to zero after
the call to nfc_targets_found(). When the race is
"lost" (i.e., ddev->poll_tech_count is found to not
be zero by the events started by nfc_targets_found()),
an error message is printed and the target is not found.
A similar race exists when digital_tg_recv_atr_req()
calls nfc_tm_activated().
Fix this by first saving the current value of
ddev->poll_tech_count and then clearing it before
calling nfc_targets_found()/nfc_tm_activated().
Clearing ddev->poll_tech_count before calling
nfc_targets_found()/nfc_tm_activated() eliminates
the race. Saving the value is required so it can be
restored when nfc_targets_found()/nfc_tm_activated()
fails and polling needs to continue.
Acked-by: Thierry Escande <[email protected]>
Signed-off-by: Mark A. Greer <[email protected]>
Signed-off-by: Samuel Ortiz <[email protected]>
|
|
In digital_in_recv_sel_res(), the code that determines
the tag type will interpret bits 7:6 (lsb being b1 as
per the Digital Specification) of a SEL RES set to 11b
as a Type 4 tag. This is okay except that the neard
will interpret the same value as an NFC-DEP device
(in src/tag.c:set_tag_type() in the neard source).
Make the digital layer's interpretation match neard's
interpretation by changing the order of the checks in
digital_in_recv_sel_res() so that a value of 11b in
bits 7:6 is interpreted as an NFC-DEP device instead
of a Type 4 tag.
Acked-by: Thierry Escande <[email protected]>
Signed-off-by: Mark A. Greer <[email protected]>
Signed-off-by: Samuel Ortiz <[email protected]>
|
|
The variable ret does not need to be assigned when declaring it. So
remove this initial assignment.
Signed-off-by: Marcel Holtmann <[email protected]>
Signed-off-by: Johan Hedberg <[email protected]>
|
|
When HCI_CONNECTABLE is set the code has been enabling passive scanning
in order to be consistent with BR/EDR and accept connections from any
device doing directed advertising to us. However, some hardware
(particularly CSR) can get very noisy even when doing duplicates
filtering, making this feature waste resources.
Considering that the feature is for fairly corner-case use (devices
who'd use directed advertising would likely be in the whitelist anyway)
it's better to disable it for now. It may still be brought back later,
possibly with a better implementation (e.g. through improved scan
parameters).
Signed-off-by: Johan Hedberg <[email protected]>
Signed-off-by: Marcel Holtmann <[email protected]>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth-next
|
|
|
|
Quell another sparse warning.
Signed-off-by: Trond Myklebust <[email protected]>
Signed-off-by: J. Bruce Fields <[email protected]>
|
|
The current code always selects XPRT_TRANSPORT_BC_TCP for the back
channel, even when the forward channel was not TCP (eg, RDMA). When
a 4.1 mount is attempted with RDMA, the server panics in the TCP BC
code when trying to send CB_NULL.
Instead, construct the transport protocol number from the forward
channel transport or'd with XPRT_TRANSPORT_BC. Transports that do
not support bi-directional RPC will not have registered a "BC"
transport, causing create_backchannel_client() to fail immediately.
Fixes: https://bugzilla.linux-nfs.org/show_bug.cgi?id=265
Signed-off-by: Chuck Lever <[email protected]>
Signed-off-by: J. Bruce Fields <[email protected]>
|
|
The EOPNOTSUPP and ENOTSUPP errors are very similar in meaning, but
ENOTSUPP is a fairly new addition to POSIX. Not all libc versions know
about the value the kernel uses for ENOTSUPP so it's better to use
EOPNOTSUPP to ensure understandable error messages.
Signed-off-by: Johan Hedberg <[email protected]>
Signed-off-by: Marcel Holtmann <[email protected]>
|
|
tsc can be NULL (mac80211 currently always passes NULL),
resulting in NULL-dereference. check before copying it.
Cc: [email protected]
Signed-off-by: Eliad Peller <[email protected]>
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
|
|
When lockdep isn't compiled, a local variable isn't used
(it's only in a macro argument), annotate it to suppress
the compiler warning.
Signed-off-by: Johannes Berg <[email protected]>
|
|
v2: fixed issue with checking return of dcbnl_rtnl_ops->getapp()
Signed-off-by: Anish Bhatt <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Like other places, we need to cancel the nest attribute after
we start. Fortunately the netlink message will not be sent on
failure, so it's not a big problem at all.
Cc: Jamal Hadi Salim <[email protected]>
Cc: "David S. Miller" <[email protected]>
Signed-off-by: Cong Wang <[email protected]>
Signed-off-by: Cong Wang <[email protected]>
Signed-off-by: Jamal Hadi Salim <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Special kernel keys, such as those used to hold DNS results for AFS, CIFS and
NFS and those used to hold idmapper results for NFS, used to be
'invalidateable' with key_revoke(). However, since the default permissions for
keys were reduced:
Commit: 96b5c8fea6c0861621051290d705ec2e971963f1
KEYS: Reduce initial permissions on keys
it has become impossible to do this.
Add a key flag (KEY_FLAG_ROOT_CAN_INVAL) that will permit a key to be
invalidated by root. This should not be used for system keyrings as the
garbage collector will try and remove any invalidate key. For system keyrings,
KEY_FLAG_ROOT_CAN_CLEAR can be used instead.
After this, from userspace, keyctl_invalidate() and "keyctl invalidate" can be
used by any possessor of CAP_SYS_ADMIN (typically root) to invalidate DNS and
idmapper keys. Invalidated keys are immediately garbage collected and will be
immediately rerequested if needed again.
Signed-off-by: David Howells <[email protected]>
Tested-by: Steve Dickson <[email protected]>
|
|
When we're not pairable we should still allow us to act as initiators
for pairing, i.e. the HCI_PAIRABLE flag should only be affecting
incoming pairing attempts. This patch fixes the relevant checks for the
hci_io_capa_request_evt() and hci_pin_code_request_evt() functions.
Signed-off-by: Johan Hedberg <[email protected]>
Signed-off-by: Marcel Holtmann <[email protected]>
|
|
Even though our side requests authentication, the original action that
caused it may be remotely triggered, such as an incoming L2CAP or RFCOMM
connect request. To track this information introduce a new hci_conn flag
called HCI_CONN_AUTH_INITIATOR.
Signed-off-by: Johan Hedberg <[email protected]>
Signed-off-by: Marcel Holtmann <[email protected]>
|
|
We're interested in whether an authentication request is because of a
remote or local action. So far hci_conn_security() has been used both
for incoming and outgoing actions (e.g. RFCOMM or L2CAP connect
requests) so without some modifications it cannot know which peer is
responsible for requesting authentication.
This patch adds a new "bool initiator" parameter to hci_conn_security()
to indicate which side is responsible for the request and updates the
current users to pass this information correspondingly.
Signed-off-by: Johan Hedberg <[email protected]>
Signed-off-by: Marcel Holtmann <[email protected]>
|
|
When a new hci_conn object is created the remote SSP authentication
requirement is set to the invalid value 0xff to indicate that it is
unknown. Once pairing completes however the code was leaving it as-is.
In case a new pairing happens over the same connection it is important
that we reset the value back to unknown so that the pairing code doesn't
make false assumptions about the requirements.
Signed-off-by: Johan Hedberg <[email protected]>
Signed-off-by: Marcel Holtmann <[email protected]>
|
|
If the current process is exiting, lingering on socket close will make
it unkillable, so we should avoid it.
Reproducer:
#include <sys/types.h>
#include <sys/socket.h>
#define BTPROTO_L2CAP 0
#define BTPROTO_SCO 2
#define BTPROTO_RFCOMM 3
int main()
{
int fd;
struct linger ling;
fd = socket(PF_BLUETOOTH, SOCK_STREAM, BTPROTO_RFCOMM);
//or: fd = socket(PF_BLUETOOTH, SOCK_DGRAM, BTPROTO_L2CAP);
//or: fd = socket(PF_BLUETOOTH, SOCK_SEQPACKET, BTPROTO_SCO);
ling.l_onoff = 1;
ling.l_linger = 1000000000;
setsockopt(fd, SOL_SOCKET, SO_LINGER, &ling, sizeof(ling));
return 0;
}
Signed-off-by: Vladimir Davydov <[email protected]>
Signed-off-by: Marcel Holtmann <[email protected]>
Cc: [email protected]
|
|
If user space has a NoInputNoOutput IO capability it makes no sense to
bother it with confirmation requests. This patch updates both SSP and
SMP to check for the local IO capability before sending a user
confirmation request to user space.
Signed-off-by: Johan Hedberg <[email protected]>
Signed-off-by: Marcel Holtmann <[email protected]>
|
|
Commit 6c53823ae0e10e723131055e1e65dd6a328a228e reshuffled the way the
authentication requirement gets set in the hci_io_capa_request_evt()
function, but at the same time it failed to update an if-statement where
cp.authentication is used before it has been initialized. The correct
value the code should be looking for in this if-statement is
conn->auth_type.
Signed-off-by: Johan Hedberg <[email protected]>
Signed-off-by: Marcel Holtmann <[email protected]>
Cc: [email protected] # 3.16
|
|
Generic netlink tables can be const.
Signed-off-by: Stephen Hemminger <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Many multicast sources can have the same port which can result in a very
large list when hashing by port only. Hash by address and port instead
if this is the case. This makes multicast more similar to unicast.
On a 24-core machine receiving from 500 multicast sockets on the same
port, before this patch 80% of system CPU was used up by spin locking
and only ~25% of packets were successfully delivered.
With this patch, all packets are delivered and kernel overhead is ~8%
system CPU on spinlocks.
Signed-off-by: David Held <[email protected]>
Signed-off-by: Eric Dumazet <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|