Emmanuel Grumbach [Sat, 5 Feb 2022 09:21:38 +0000 (11:21 +0200)]
iwlwifi: mvm: always remove the session protection after association
The firmware will soon stop removing the session protection for us after
association. While this was convenient, it was not symmetric.
Always remove the session protection after association, even for devices
that support the new API.
Miri Korenblit [Sat, 5 Feb 2022 09:21:36 +0000 (11:21 +0200)]
iwlwifi: mvm: refactor setting PPE thresholds in STA_HE_CTXT_CMD
We are setting the PPE Thresholds in STA_HE_CTXT_CMD according
to HE PHY Capabilities IE. As EHT is introduced, we will have to
set this thresholds according to EHT PHY Capabilities IE if we're
in an EHT connection. Some parts of the code can be used for both
HE and EHT. Put this parts in functions which will be used in the
patch which adds support for EHT PPE.
Luca Coelho [Sat, 5 Feb 2022 09:21:33 +0000 (11:21 +0200)]
iwlwifi: mvm: don't send BAID removal to the FW during hw_restart
With the new ML API, we can't send the BAID removal command to
firmware during hw_restart because it will cause an assertion failure
0x350D because the BAID doesn't exist at that point.
So avoid sending the command if we are performing a hw_restart.
Emmanuel Grumbach [Sat, 5 Feb 2022 09:21:32 +0000 (11:21 +0200)]
iwlwifi: don't dump_stack() when we get an unexpected interrupt
It is yet unclear if the WARNING really points to a real problem,
but for sure the stack dump doesn't help fixing it.
Just use a regular error print instead.
Sometimes RFIm can be deactivated in FW due to internal
errors. In this case, FW will send a notification to the
driver about that. Add a log message in this case since
FW logs are not always available.
Ilan Peer [Sat, 5 Feb 2022 09:21:30 +0000 (11:21 +0200)]
iwlwifi: mvm: Consider P2P GO operation during scan
A scan during active P2P GO operation, i.e., data traffic with
clients, can impact the throughput and latency of such traffic.
Thus, when scan is requested while there is an active P2P GO
and low latency is asserted:
- Ask the FW scan logic to respect the P2P GO activity during the
scheduling of the scan operation to minimize the impact on the
throughput and latency.
- Force scan to perform EBS before starting the scan to reduce the
number of scanned channels.
Johannes Berg [Fri, 4 Feb 2022 10:25:07 +0000 (12:25 +0200)]
iwlwifi: eeprom: clean up macros
There are two versions of the same definitions, with and without
IWL_ prefix. Use the ones with IWL_ prefix, keeping the correct
comment (microseconds).
Johannes Berg [Fri, 4 Feb 2022 10:25:06 +0000 (12:25 +0200)]
iwlwifi: drv: load tlv debug data earlier
There's no good reason to pick the opmode first and load this
under the mutex, so just load it before continuing. This will
let us load it asynchronously more easily later.
Johannes Berg [Fri, 4 Feb 2022 10:25:04 +0000 (12:25 +0200)]
iwlwifi: pcie: adjust to Bz completion descriptor
The Bz devices got a new completion descriptor again since
we only ever really used 4 out of 32 bytes anyway. Adjust
the code to deal with that. Note that the intention was to
reduce the size, but the hardware was implemented wrongly.
While at it, do some cleanups and remove the union to simplify
the code, clean up iwl_pcie_free_bd_size() to no longer need
an argument and add iwl_pcie_used_bd_size() with the logic to
selct completion descriptor size.
Ilan Peer [Fri, 4 Feb 2022 10:25:03 +0000 (12:25 +0200)]
iwlwifi: mvm: Passively scan non PSC channels only when requested so
Non PSC channels should generally be scanned based on information about
collocated APs obtained during scan on legacy bands, and otherwise
should not be scanned unless specifically requested so (as there are
relatively many non PSC channels, scanning them passively is time consuming
and interferes with regular data traffic).
Thus, modify the scan logic to avoid passively scanning PSC channels
if there is no information about collocated APs and the scan is not
a passive scan.
Mukesh Sisodiya [Fri, 4 Feb 2022 10:25:01 +0000 (12:25 +0200)]
iwlwifi: yoyo: Avoid using dram data if allocation failed
The config set TLV setting depend on dram allocation
and if allocation failed the data used in config set tlv
should not set this.
Adding the check if dram fragment is available or not.
Ilan Peer [Fri, 4 Feb 2022 10:25:00 +0000 (12:25 +0200)]
iwlwifi: mvm: Correctly set fragmented EBS
Currently, fragmented EBS was set for a channel only if the 'hb_type'
was set to fragmented or balanced scan. However, 'hb_type' is set only
in case of CDB, and thus fragmented EBS is never set for a channel for
non-CDB devices. Fix it.
Johannes Berg [Sun, 30 Jan 2022 09:53:04 +0000 (11:53 +0200)]
iwlwifi: make some functions friendly to sparse
Somehow, sparse cannot track the cond_lock() here properly,
but if we return directly from the inner basic block then
it doesn't complain. Refactor the code a bit to make it not
complain.
Johannes Berg [Sun, 30 Jan 2022 09:53:00 +0000 (11:53 +0200)]
iwlwifi: move symbols into a separate namespace
Nobody other than iwlwifi itself should be accessing the symbols
that it exports, so make that clear by moving them into a separate
new namespace ("IWLWIFI").
Andrei Otcheretianski [Sun, 30 Jan 2022 09:52:58 +0000 (11:52 +0200)]
iwlwifi: pcie: make sure iwl_rx_packet_payload_len() will not underflow
If the device is malfunctioning and reports too short rx descriptor
length, iwl_rx_packet_payload_len() will underflow, eventually resulting
in accessing memory out of bounds and other bad things. Prevent this.
Miri Korenblit [Sun, 30 Jan 2022 09:52:57 +0000 (11:52 +0200)]
iwlwifi: mvm: use debug print instead of WARN_ON()
In MCC_UPDATE_CMD we get from the FW the number of entries in the
channel info array. We used to WARN_ON if this parameter is greater
than we expected. Since this is not really a driver bug, and since it
might happen in some valid cases too, we shouldn't use a warning here.
Fix this by replacing the WARN_ON with a debug print.
Due to preg protection we cannot write to this register
while FW is running (when FW in Halt it is ok).
since we have some cases that we need to dump this
region while FW is running remove this writing from DRV.
FW will do this writing.
Johannes Berg [Sat, 29 Jan 2022 11:16:16 +0000 (13:16 +0200)]
iwlwifi: mvm: align locking in D3 test debugfs
Since commit a05829a7222e ("cfg80211: avoid holding the RTNL when
calling the driver") we're not only holding the RTNL when going
in and out of suspend, but also the wiphy->mtx. Add that to the
D3 test debugfs in iwlwifi since it's required for various calls
to mac80211.
We may not have all the interfaces added to the driver when we get the
THERMAL_DUAL_CHAIN_REQUEST notification from the FW, so instead of
iterating all vifs to update SMPS, iterate only the ones that are
already assigned. The interfaces that were not assigned yet, will be
updated accordingly when we start using them.
Mordechay Goodstein [Sat, 29 Jan 2022 11:16:13 +0000 (13:16 +0200)]
iwlwifi: mvm: only enable HE DCM if we also support TX
Currently we enable DCM if the peer supports RX without checking whether
we advertised that we support TX. Fix this by also checking that our TX
side is set.
Johannes Berg [Fri, 28 Jan 2022 13:34:32 +0000 (15:34 +0200)]
iwlwifi: mvm: support new BAID allocation command
Due to some rearchitecting inside the firmware, a new BAID
allocation command is being introduced. Support it. Note
that with it the firmware no longer returns "no space" but
will crash instead, so check for that before sending the
command to the firmware.
Johannes Berg [Fri, 28 Jan 2022 13:34:28 +0000 (15:34 +0200)]
iwlwifi: fix various more -Wcast-qual warnings
When enabling W=3 that gets us -Wcast-qual, fix those warnings
in the code and propagate constness properly, or cast it away
via (uintptr_t) where really needed.
Bjoern A. Zeeb [Fri, 28 Jan 2022 13:34:27 +0000 (15:34 +0200)]
iwlwifi: propagate (const) type qualifier
Most of this change is a continuation of commit 403ea939ea6a
("iwlwifi: dbg: Mark ucode tlv data as const") propagating the
(const) type qualifier for ucode based tlv data to avoid having
the impression that it is writeable.
The other part of the change preserves the (const) type qualifier
over casts and function calls where it was previously lost.
Both changes are needed to avoid compile time errors on system with
more strict error settings, in this case found with clang on FreeBSD.
Sponsored by: The FreeBSD Foundation Signed-off-by: Bjoern A. Zeeb <bz@FreeBSD.ORG> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
[fix double word in commit message] Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Link: https://lore.kernel.org/r/iwlwifi.20220128153014.3230c41312fc.I0032c597984834258d5a79b97052ed83dbe53b80@changeid Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Bjoern A. Zeeb [Fri, 28 Jan 2022 13:34:26 +0000 (15:34 +0200)]
iwlwifi: de-const properly where needed
In order to de-const variables simply casting through (void *) is
not enough: "cast from 'const .. *' to 'void *' drops const qualifier".
Cast through (uintptr_t) as well [1] to make this compile on systems
with more strict requirements.
In addition passing const void *data to dma_map_single() also
drops the (const) qualifier. De-constify on variable on assignment
which may be overwritten later. In either case the (void *) cast
to dma_map_single() is not needed (anymore) either.
Johannes Berg [Fri, 28 Jan 2022 13:34:25 +0000 (15:34 +0200)]
iwlwifi: make iwl_fw_lookup_cmd_ver() take a cmd_id
Instead of taking the group/command separately, make the function
take a combined command ID. In many cases, this allows us to pass
an existing command ID (e.g. cmd.id), or introduce a new variable
for it, so that we don't use the command ID twice.
This way, we can also use LONG_GROUP implicitly, so we don't need
to spell that out for many commands.
Apart from mvm.h, fw/img.{c,h} changes and some copyright and
indentation updates, this was done with spatch:
Johannes Berg [Fri, 28 Jan 2022 13:34:23 +0000 (15:34 +0200)]
iwlwifi: prefer WIDE_ID() over iwl_cmd_id()
The order of arguments for iwl_cmd_id() is confusing, and the
version is always 0 and thus a useless argument. Prefer the
WIDE_ID() macro (which needs to be a macro due to use in switch
cases etc.) over the iwl_cmd_id() function.
Obviously done with spatch:
@@
expression G, C;
@@
-iwl_cmd_id(C, G, 0)
+WIDE_ID(G, C)
Ayala Barazani [Fri, 28 Jan 2022 13:34:22 +0000 (15:34 +0200)]
iwlwifi: mvm: allow enabling UHB TAS in the USA via ACPI setting
Read a new bit defined in ACPI WTAS that allows OEMs to specify whether
TAS is allowed in UHB (6-7GHz) in the USA. This can be used by OEMs
that got certified to use this feature.
Nathan Errera [Fri, 28 Jan 2022 13:34:21 +0000 (15:34 +0200)]
iwlwifi: mvm: offload channel switch timing to FW
Since FW is now in charge of timing the channel switch, there is no need
to send the add/modify/remove time event command to fw with every (e)CSA
element.
However, the driver needs to cancel the channel switch if the CS start
notification arrives and it does not know about an ongoing channel switch.
Takashi Iwai [Fri, 21 Jan 2022 11:40:24 +0000 (12:40 +0100)]
iwlwifi: mvm: Don't call iwl_mvm_sta_from_mac80211() with NULL sta
The recent fix for NULL sta in iwl_mvm_get_tx_rate() still has a call
of iwl_mvm_sta_from_mac80211() that may be called with NULL sta.
Although this practically only points to the address and the actual
access doesn't happen due to the conditional evaluation at a later
point, it looks a bit flaky.
This patch drops the temporary variable above and evaluates
iwm_mvm_sta_from_mac80211() directly for avoiding confusions.
Mordechay Goodstein [Mon, 14 Feb 2022 16:30:05 +0000 (17:30 +0100)]
mac80211: parse AddBA request with extended AddBA element
In EHT requesting aggregation with 1K needs the use of extended
the AddBA element for the buffer size, so add the logic to parse
it and make sure it's in limits of the EHT aggregation size.
Johannes Berg [Mon, 14 Feb 2022 16:29:48 +0000 (17:29 +0100)]
mac80211_hwsim: check TX and STA bandwidth
Add checks to hwsim to validate that neither TX nor any
station's configured bandwidth can exceed the channel
(context) configuration previously requested.
Johannes Berg [Mon, 14 Feb 2022 16:29:22 +0000 (17:29 +0100)]
mac80211: parse only HE capability elements with valid size
The code validates the HE capability element size later,
but slightly wrong, so use the new helper to do it right
and only accept it if it has a good size.
Johannes Berg [Mon, 14 Feb 2022 16:29:21 +0000 (17:29 +0100)]
ieee80211: add helper to check HE capability element size
This element has a very dynamic structure, create a small helper
function to validate its size. We're currently checking it in
mac80211 in a conversion function, but that's actually slightly
buggy.
Ping-Ke Shih [Fri, 11 Feb 2022 07:59:53 +0000 (15:59 +0800)]
rtw89: handle TX/RX 160M bandwidth
Apply 160M bandwidth to RA (rate adaptive) mechanism, so it can transmit
packets with this bandwidth. On the other hand, convert 160M bandwidth
from RX desc to rx_info_bw.
Zong-Zhe Yang [Fri, 11 Feb 2022 07:59:50 +0000 (15:59 +0800)]
rtw89: extend subband for 6G band
Split 6G band into 8 sub-bands where indexes are from 0 to 7,
i.e. RTW89_CH_6G_BAND_IDX[0-7]. Then, decide subband by both
band and channel instead of just channel because conflicts
between 5G channels and 6G channels.
Moreover, add default case to the existing use of switch (subband).
Zong-Zhe Yang [Fri, 11 Feb 2022 07:59:49 +0000 (15:59 +0800)]
rtw89: refine naming of rfk helpers with prefix
Since these macro in rfk helpers are common now, a common naming
should be better. So, apply RTW89_ as prefix to them, and modify
the use correspondly. No logic is changed at all.
Zhao, Jiaqing [Wed, 9 Feb 2022 15:46:50 +0000 (15:46 +0000)]
brcmfmac: Add BCM43454/6 support
BCM43454/6 is a variant of BCM4345 which is exactly identical to
BCM4345/6, except the chip id is 0xa9be. This patch adds support
for BCM43454/6 by handing it in the same way as BCM4345.
Note: when loading some specific version of BCM4345 firmware, the
chip id may become 0x4345. This is an expected behavior, and it will
restore to 0xa9be after power cycle.
David S. Miller [Fri, 11 Feb 2022 14:19:23 +0000 (14:19 +0000)]
Merge tag 'wireless-next-2022-02-11' of git://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless-next
wireless-next patches for v5.18
First set of patches for v5.18, with both wireless and stack patches.
rtw89 now has AP mode support and wcn36xx has survey support. But
otherwise pretty normal.
Eric Dumazet [Thu, 10 Feb 2022 21:42:31 +0000 (13:42 -0800)]
ipv4: add (struct uncached_list)->quarantine list
This is an optimization to keep the per-cpu lists as short as possible:
Whenever rt_flush_dev() changes one rtable dst.dev
matching the disappearing device, it can can transfer the object
to a quarantine list, waiting for a final rt_del_uncached_list().
Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Thu, 10 Feb 2022 21:42:30 +0000 (13:42 -0800)]
ipv6: add (struct uncached_list)->quarantine list
This is an optimization to keep the per-cpu lists as short as possible:
Whenever rt6_uncached_list_flush_dev() changes one rt6_info
matching the disappearing device, it can can transfer the object
to a quarantine list, waiting for a final rt6_uncached_list_del().
Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Thu, 10 Feb 2022 21:42:29 +0000 (13:42 -0800)]
ipv6: give an IPv6 dev to blackhole_netdev
IPv6 addrconf notifiers wants the loopback device to
be the last device being dismantled at netns deletion.
This caused many limitations and work arounds.
Back in linux-5.3, Mahesh added a per host blackhole_netdev
that can be used whenever we need to make sure objects no longer
refer to a disappearing device.
If we attach to blackhole_netdev an ip6_ptr (allocate an idev),
then we can use this special device (which is never freed)
in place of the loopback_dev (which can be freed).
This will permit improvements in netdev_run_todo() and other parts
of the stack where had steps to make sure loopback_dev was
the last device to disappear.
Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Mahesh Bandewar <maheshb@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Holger Brunck [Thu, 10 Feb 2022 17:48:23 +0000 (18:48 +0100)]
dsa: mv88e6xxx: make serdes SGMII/Fiber tx amplitude configurable
The mv88e6352, mv88e6240 and mv88e6176 have a serdes interface. This patch
allows to configure the output swing to a desired value in the
phy-handle of the port. The value which is peak to peak has to be
specified in microvolts. As the chips only supports eight dedicated
values we return EINVAL if the value in the DTS does not match one of
these values.
Signed-off-by: Holger Brunck <holger.brunck@hitachienergy.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Reviewed-by: Marek Behún <kabel@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Common PHYs and network PCSes often have the possibility to specify
peak-to-peak voltage on the differential pair - the default voltage
sometimes needs to be changed for a particular board.
Add properties `tx-p2p-microvolt` and `tx-p2p-microvolt-names` for this
purpose. The second property is needed to specify the mode for the
corresponding voltage in the `tx-p2p-microvolt` property, if the voltage
is to be used only for speficic mode. More voltage-mode pairs can be
specified.
Example usage with only one voltage (it will be used for all supported
PHY modes, the `tx-p2p-microvolt-names` property is not needed in this
case):
projects / users / dwmw2 / linux.git / log