Draft P802.11ax_D2.5 defines the following capabilities that
can be negotiated using RSNXE capabilities:

- Secure LTF measurement exchange protocol.
- Secure RTT measurement exchange protocol.
- Management frame protection for all management frames exchanged
  during the negotiation and range measurement procedure.

Extend the nl80211 API to allow drivers to declare support for
these new capabilities as part of extended feature.
Signed-off-by: NIlan Peer <ilan.peer@intel.com>
Signed-off-by: NLuca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210409123755.8280e31d8091.Ifcb29f84f432290338f80c8378aa5c9e0a390c93@changeidSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

In case the frequency is invalid, ieee80211_parse_ch_switch_ie
will fail and we may not even reach the check in
ieee80211_sta_process_chanswitch. Drop the connection
in case ieee80211_parse_ch_switch_ie failed, but still
take into account the CSA mode to remember not to send
a deauth frame in case if it is forbidden to.
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: NLuca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210409123755.34712ef96a0a.I75d7ad7f1d654e8b0aa01cd7189ff00a510512b3@changeidSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

This will allow the low level driver to get the wdev during
the add_interface flow.
In order to do that, remove a few checks from there and do
not return NULL for vifs that were not yet added to the
driver. Note that all the current callers of this helper
function assume that the vif already exists:
 - The callers from the drivers already have a vif pointer.
 Before this change, ieee80211_vif_to_wdev would return NULL
 in some cases, but those callers don't even check they
 get a non-NULL pointer from ieee80211_vif_to_wdev.
 - The callers from net/mac80211/cfg.c assume the vif is
 already added to the driver as well.

So, this change has no impact on existing callers of this
helper function.
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: NLuca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210409123755.6078d3517095.I1907a45f267a62dab052bcc44428aa7a2005ffc9@changeidSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

Add a flag that indicates that the ISTA shall indicate support for
LMR feedback in NDP ranging negotiation.
Signed-off-by: NAvraham Stern <avraham.stern@intel.com>
Signed-off-by: NLuca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210409123755.eff546283504.I2606161e700ac24d94d0b50c8edcdedd4c0395c2@changeidSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

Add an enum with the values of the ranging parameters max LTF total
field, as defined in IEEE802.11az_D2.6, table Table 9-322h23fc.
Signed-off-by: NAvraham Stern <avraham.stern@intel.com>
Signed-off-by: NLuca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210409123755.d2588ebb1974.I9424c8ade13c4c938cb9999d8ce99d0d4c1cc198@changeidSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

As crypto_shash_setkey() can fail, we should check the return value.

Addresses-Coverity-ID: 1401813 ("Unchecked return value")
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NLuca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210409123755.533ff7acf1d2.I034bafa201c4a6823333f8410aeaa60cca5ee9e0@changeidSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

The increment of idx is indented one level too deeply, clean up the
code by removing the extraneous tab.
Signed-off-by: NColin Ian King <colin.king@canonical.com>
Link: https://lore.kernel.org/r/20210416095137.2033469-1-colin.king@canonical.comSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

My previous commit to not apply flow control to management frames
that are going over TXQs (which is currently only the case for
iwlwifi, I think) broke things, with iwlwifi firmware crashing on
certain frames. As it turns out, that was due to the frame being
too short: space for the MIC wasn't added at the end of encrypted
management frames.

Clearly, this is due to using the 'frags' queue - this is meant
only for frames that have already been processed for TX, and the
code in ieee80211_tx_dequeue() just returns them. This caused all
management frames to now not get any TX processing.

To fix this, use IEEE80211_TX_INTCFL_NEED_TXPROCESSING (which is
currently used only in other circumstances) to indicate that the
frames need processing, and clear it immediately after so that,
at least in theory, MMPDUs can be fragmented.

Fixes: 73bc9e0a ("mac80211: don't apply flow control on management frames")
Acked-by: NToke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/r/20210416134702.ef8486a64293.If0a9025b39c71bb91b11dd6ac45547aba682df34@changeidSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

It's not modified so make it const with the eventual goal of moving
data to text for various static struct ieee80211_rate arrays.
Signed-off-by: NJoe Perches <joe@perches.com>
Link: https://lore.kernel.org/r/8b210b5f5972e39eded269b35a1297cf824c4181.camel@perches.comSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

Fix multiple kernel-doc warnings in cfg80211.h.

cfg80211.h:363: warning: missing initial short description on line:
 * struct ieee80211_sband_iftype_data
cfg80211.h:6743: warning: missing initial short description on line:
 * cfg80211_vendor_cmd_get_sender
Signed-off-by: NRandy Dunlap <rdunlap@infradead.org>
Link: https://lore.kernel.org/r/20210417060142.1648-1-rdunlap@infradead.orgSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

The htmldoc produces this warning which was introduced
bu the commit below.

include/net/cfg80211.h:6643: warning: expecting prototype for wiphy_rfkill_set_hw_state().
Prototype was for wiphy_rfkill_set_hw_state_reason() instead

Fixes: 6f779a66 ("cfg80211: allow specifying a reason for hw_rfkill")
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>
Link: https://lore.kernel.org/r/20210413113850.59098-1-emmanuel.grumbach@intel.comSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

Remove a wrong length check for RNR information element as it
can have arbitrary length.
Signed-off-by: NIlan Peer <ilan.peer@intel.com>
Link: https://lore.kernel.org/r/20210408143224.c7eeaf1a5270.Iead7762982e941a1cbff93f68bf8b5139447ff0c@changeidSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

If any of the cipher schemes specified by the driver are invalid, bail
out and fail the registration rather than just warning.  Otherwise, we
might later crash when we try to use the invalid cipher scheme, e.g.
if the hdr_len is (significantly) less than the pn_offs + pn_len, we'd
have an out-of-bounds access in RX validation.

Fixes: 2475b1cc ("mac80211: add generic cipher scheme support")
Link: https://lore.kernel.org/r/20210408143149.38a3a13a1b19.I6b7f5790fa0958ed8049cf02ac2a535c61e9bc96@changeidSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

After channel switch, we should consider any beacon with a
CSA IE as a new switch. If the CSA IE is a leftover from
before the switch that the AP forgot to remove, we'll get
a CSA-to-Self.

This caused issues in iwlwifi where the firmware saw a beacon
with a CSA-to-Self with mode = 1 on the new channel after a
switch. The firmware considered this a new switch and closed
its queues. Since the beacon didn't change between before and
after the switch, we wouldn't handle it (the CRC is the same)
and we wouldn't let the firmware open its queues again or
disconnect if the CSA IE stays for too long.

Clear the CRC valid state after we switch to make sure that
we handle the beacon and handle the CSA IE as required.
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>
Link: https://lore.kernel.org/r/20210408143124.b9e68aa98304.I465afb55ca2c7d59f7bf610c6046a1fd732b4c28@changeidSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

Some drivers, for example mt76, use the skb priority field, and
expects that to be consistent with the skb queue mapping. On some
frame injection code paths that was not true, and it broke frame
injection. Now the skb queue mapping is set according to the skb
priority value when the frame is injected. The skb priority value
is also derived from the frame data for all frame types, as it
was done prior to commit dbd50a85 (only allocate one queue
when using iTXQs). Fixes frame injection with the mt76 driver on
MT7610E chipset.
Signed-off-by: NJohan Almbladh <johan.almbladh@anyfinetworks.com>
Link: https://lore.kernel.org/r/20210401164455.978245-1-johan.almbladh@anyfinetworks.comSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

Some HW/driver can support passing ethernet rx decap frames and
raw 802.11 frames for the monitor interface concurrently and
via separate RX calls to mac80211. Packets going to the monitor
interface(s) would be in 802.11 format and thus not have the
RX_FLAG_8023 set, and 802.11 format monitoring frames should have
RX_FLAG_ONLY_MONITOR set.

Drivers doing such can enable the SUPPORTS_CONC_MON_RX_DECAP to
allow using ethernet decap offload while a monitor interface is
active, currently RX decapsulation offload gets disabled when a
monitor interface is added.
Signed-off-by: NSriram R <srirrama@codeaurora.org>
Link: https://lore.kernel.org/r/1617068116-32253-1-git-send-email-srirrama@codeaurora.org
[add proper documentation, rewrite commit message]
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>

Enhance enum nl80211_tdls_peer_capability to configure TDLS peer's
support for HE mode. Userspace decodes the TDLS setup response frame
and confugures the HE mode support to driver if the peer has advertized
HE mode support in TDLS setup response frame. The driver uses this
information to decide whether to include HE operation IE in TDLS setup
confirmation frame.
Signed-off-by: NVamsi Krishna <vamsin@codeaurora.org>
Link: https://lore.kernel.org/r/1614696636-30144-1-git-send-email-vamsin@codeaurora.orgSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

rfkill now allows to report a reason for the hw_rfkill state.
Allow cfg80211 drivers to specify this reason.
Keep the current API to use the default reason
(RFKILL_HARD_BLOCK_SIGNAL).
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>
Link: https://lore.kernel.org/r/20210322204633.102581-4-emmanuel.grumbach@intel.comSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

The docs were very sparse with how exactly CMD_ROAM should be
used. Specifically related to BSS information normally obtained
through a user space scan.
Signed-off-by: NJames Prestwood <prestwoj@gmail.com>
Link: https://lore.kernel.org/r/20210311230333.103934-1-prestwoj@gmail.comSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

Similar to 802.11 txpath, set socket sk_pacing_shift for 802.3 tx path.
Signed-off-by: NLorenzo Bianconi <lorenzo@kernel.org>
Link: https://lore.kernel.org/r/7230abc48dcf940657838546cdaef7dce691ecdd.1615240733.git.lorenzo@kernel.orgSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

In some cases (depending on the driver, but it's true e.g. for
iwlwifi) we're using an internal TXQ for management packets,
mostly to simplify the code and to have a place to queue them.
However, it appears that in certain cases we can confuse the
code and management frames are dropped, which is certainly not
what we want.

Short-circuit the processing of management frames. To keep the
impact minimal, only put them on the frags queue and check the
tid == management only for doing that and to skip the airtime
fairness checks, if applicable.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Acked-by: NToke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/r/20210319232800.0e876c800866.Id2b66eb5a17f3869b776c39b5ca713272ea09d5d@changeidSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

Add NL80211_FILS_DISCOVERY_ATTR_TMPL explicitly in
nl80211_fils_discovery_policy definition.
Signed-off-by: NAloka Dixit <alokad@codeaurora.org>
Link: https://lore.kernel.org/r/20210222212059.22492-1-alokad@codeaurora.orgSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

The variable result is being assigned a value that is never
read and it is being updated later with a new value. The
initialization is redundant and can be removed.

Addresses-Coverity: ("Unused value")
Signed-off-by: NColin Ian King <colin.king@canonical.com>
Link: https://lore.kernel.org/r/20210328213729.65819-1-colin.king@canonical.comSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

GCC reports the following warning with W=1:

net/mac80211/rc80211_minstrel_ht.c:871:34: warning:
 variable 'mg' set but not used [-Wunused-but-set-variable]
  871 |  struct minstrel_mcs_group_data *mg;
      |                                  ^~

This variable is not used in function , this commit
remove it to fix the warning.
Reported-by: NHulk Robot <hulkci@huawei.com>
Signed-off-by: NWei Yongjun <weiyongjun1@huawei.com>
Link: https://lore.kernel.org/r/20210326024843.987941-1-weiyongjun1@huawei.comSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

spinlock can be initialized automatically with DEFINE_SPINLOCK()
rather than explicitly calling spin_lock_init().
Reported-by: NHulk Robot <hulkci@huawei.com>
Signed-off-by: NQiheng Lin <linqiheng@huawei.com>
Link: https://lore.kernel.org/r/20210325143854.13186-1-linqiheng@huawei.comSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

spinlock can be initialized automatically with DEFINE_SPINLOCK()
rather than explicitly calling spin_lock_init().
Reported-by: NHulk Robot <hulkci@huawei.com>
Signed-off-by: NGuobin Huang <huangguobin4@huawei.com>
Link: https://lore.kernel.org/r/1616839160-6654-1-git-send-email-huangguobin4@huawei.comSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

spinlock can be initialized automatically with DEFINE_SPINLOCK()
rather than explicitly calling spin_lock_init().
Reported-by: NHulk Robot <hulkci@huawei.com>
Signed-off-by: NGuobin Huang <huangguobin4@huawei.com>
Link: https://lore.kernel.org/r/1617711116-49370-1-git-send-email-huangguobin4@huawei.comSigned-off-by: NJohannes Berg <johannes.berg@intel.com>

Use the module_pci_driver() macro to make the code simpler
by eliminating module_init and module_exit calls.
Signed-off-by: NWei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Use the module_pci_driver() macro to make the code simpler
by eliminating module_init and module_exit calls.
Signed-off-by: NWei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Use the module_pci_driver() macro to make the code simpler
by eliminating module_init and module_exit calls.
Signed-off-by: NWei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Use the module_pci_driver() macro to make the code simpler
by eliminating module_init and module_exit calls.
Signed-off-by: NWei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Use the module_pci_driver() macro to make the code simpler
by eliminating module_init and module_exit calls.
Signed-off-by: NWei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Use the module_pci_driver() macro to make the code simpler
by eliminating module_init and module_exit calls.
Signed-off-by: NWei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

module_spi_driver() makes the code simpler by eliminating
boilerplate code.
Signed-off-by: NWei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

According to the changelog, asynchronous mode was dropped sometime
before v2.2. Let's get rid of the unused driver-specific async state as
well so that it doesn't show up when doing tree-wide tty work.
Signed-off-by: NJohan Hovold <johan@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The macro CN23XX_PEM_BAR1_INDEX_REG is being used to shift oct->pcie_port
(a u16) left 24 places. There are two subtle issues here, first the
shift gets promoted to an signed int and then sign extended to a u64.
If oct->pcie_port is 0x80 or more then the upper bits get sign extended
to 1. Secondly shfiting a u16 24 bits will lead to an overflow so it
needs to be cast to a u64 for all the bits to not overflow.

It is entirely possible that the u16 port value is never large enough
for this to fail, but it is useful to fix unintended overflows such
as this.

Fix this by casting the port parameter to the macro to a u64 before
the shift.

Addresses-Coverity: ("Unintended sign extension")
Fixes: 5bc67f58 ("liquidio: CN23XX register definitions")
Signed-off-by: NColin Ian King <colin.king@canonical.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The error check on err is always false as err is always 0 at the
port_found label. The code is redundant and can be removed.

Addresses-Coverity: ("Logically dead code")
Signed-off-by: NColin Ian King <colin.king@canonical.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Merge tag 'linux-can-next-for-5.13-20210407' of git://git.kernel.org/pub/scm/linux/kernel/git/mkl/linux-can-next

Marc Kleine-Budde says:

====================
pull-request: can-next 2021-04-07

this is a pull request of 6 patches for net-next/master.

The first patch targets the CAN driver infrastructure, it improves the
alloc_can{,fd}_skb() function to set the pointer to the CAN frame to
NULL if skb allocation fails.

The next patch adds missing error handling to the m_can driver's RX
path (the code was introduced in -next, no need to backport).

In the next patch an unused constant is removed from an enum in the
c_can driver.

The last 3 patches target the mcp251xfd driver. They add BQL support
and try to work around a sometimes broken CRC when reading the TBC
register.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Here is only one place where we want to specify new_ifindex. In all
other cases, callers pass 0 as new_ifindex. It looks reasonable to add a
low-level function with new_ifindex and to convert
dev_change_net_namespace to a static inline wrapper.

Fixes: eeb85a14 ("net: Allow to specify ifindex when device is moved to another namespace")
Suggested-by: NJakub Kicinski <kuba@kernel.org>
Signed-off-by: NAndrei Vagin <avagin@gmail.com>
Acked-by: NJakub Kicinski <kuba@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In this case, we don't need to check that new_ifindex is positive in
validate_linkmsg.

Fixes: eeb85a14 ("net: Allow to specify ifindex when device is moved to another namespace")
Suggested-by: NJakub Kicinski <kuba@kernel.org>
Signed-off-by: NAndrei Vagin <avagin@gmail.com>
Acked-by: NJakub Kicinski <kuba@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>