I hit a bug in which we started a CSA with an action frame,
but the AP changed its mind and didn't change the beacon.
The CSA wasn't cancelled and we lost the connection.
The beacons were ignored because they never changed: they
never contained any CSA IE. Because they never changed, the
CRC of the beacon didn't change either which made us ignore
the beacons instead of processing them.
Now what happens is:
1) beacon has CRC X and it is valid. No CSA IE in the beacon
2) as long as beacon's CRC X, don't process their IEs
3) rx action frame with CSA
4) invalidate the beacon's CRC
5) rx beacon, CRC is still X, but now it is invalid
6) process the beacon, detect there is no CSA IE
7) abort CSA
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Link: https://lore.kernel.org/r/iwlwifi.20201206145305.83470b8407e6.I739b907598001362744692744be15335436b8351@changeid
Signed-off-by: Johannes Berg <[email protected]>
The WLAN device may exist yet not be usable. This can happen
when the WLAN device is controllable by both the host and
some platform internal component.
We need some arbritration that is vendor specific, but when
the device is not available for the host, we need to reflect
this state towards the user space.
Add a reason field to the rfkill object (and event) so that
userspace can know why the device is in rfkill: because some
other platform component currently owns the device, or
because the actual hw rfkill signal is asserted.
Capable userspace can now determine the reason for the rfkill
and possibly do some negotiation on a side band channel using
a proprietary protocol to gain ownership on the device in case
the device is owned by some other component. When the host
gains ownership on the device, the kernel can remove the
RFKILL_HARD_BLOCK_NOT_OWNER reason and the hw rfkill state
will be off. Then, the userspace can bring the device up and
start normal operation.
The rfkill_event structure is enlarged to include the additional
byte, it is now 9 bytes long. Old user space will ask to read
only 8 bytes so that the kernel can know not to feed them with
more data. When the user space writes 8 bytes, new kernels will
just read what is present in the file descriptor. This new byte
is read only from the userspace standpoint anyway.
If a new user space uses an old kernel, it'll ask to read 9 bytes
but will get only 8, and it'll know that it didn't get the new
state. When it'll write 9 bytes, the kernel will again ignore
this new byte which is read only from the userspace standpoint.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Johannes Berg <[email protected]>
We no longer need code that was introduced to differentiate
between two early versions of 8260.
We can remove this convoluted way to get the hardware version
that was needed because of a bug in the register's
configuration.
Moreover, since we no longer need to access the PRPH
registers, we no longer need to wake up the device,
request ownership, etc...
Remove all that.
This allows us to get the rid of the obsolete comment
about the AUX bus MISC address space which should have
been moved when this code was moved away from here.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Link: https://lore.kernel.org/r/iwlwifi.20201209231352.4a5665ccd8a6.Iff3879405c15758ba661c430e77dc2160ddada1c@changeid
Signed-off-by: Luca Coelho <[email protected]>
The low level driver adds its own opaque information
in the BSS table in the cfg80211_bss structure.
The low level driver may need to signal that this information
is no longer relevant and needs to be recreated.
Add an API to allow the low level driver to do that.
iwlwifi needs this because it keeps there an information about
the firmware's internal clock. This is kept in mac80211's
struct ieee80211_bss::sync_device_ts.
This information is populated while we scan, we add the
internal firmware's clock to each beacon which allows us to
program the firmware correctly after association so that
it'll know when (in terms of its internal clock) the DTIM
and TBTT will happen.
When the firmware is reset this internal clock is reset as
well and ieee80211_bss::sync_device_ts is no longer accurate.
iwlwifi will call this new API any time the firmware is started.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Johannes Berg <[email protected]>
I changed the API for asking for a session protection but
I omitted the TDLS flows. Fix that now.
Note that for the TDLS flow, we need to block until the
session protection actually starts, so add this option
to iwl_mvm_schedule_session_protection.
This patch fixes a firmware assert in the TDLS flow since
the old TIME_EVENT_CMD is not supported anymore by newer
firwmare versions.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Fixes: fe959c7b20 ("iwlwifi: mvm: use the new session protection command")
Signed-off-by: Luca Coelho <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>
The purpose of this was to keep all the queues updated with
the Rx sequence numbers because unlikely yet possible
situations where queues can't understand if a specific
packet needs to be dropped or not.
Unfortunately, it was reported that this caused issues in
our DMA engine. We don't fully understand how this is related,
but this is being currently debugged. For now, just don't send
this notification to the Rx queues. This de-facto reverts my
commit 3c514bf831:
iwlwifi: mvm: add a loose synchronization of the NSSN across Rx queues
This issue was reported here:
https://bugzilla.kernel.org/show_bug.cgi?id=204873https://bugzilla.kernel.org/show_bug.cgi?id=205001
and others maybe.
Fixes: 3c514bf831 ("iwlwifi: mvm: add a loose synchronization of the NSSN across Rx queues")
CC: <[email protected]> # 5.3+
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>
The firmware API has changed a little bit but this change
has no impact on the flow and is backward compatible.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>
The firmware has now a new session protection command.
This new API allows the firmware to manage the protection
needed for association. It'll also remove the event when
the association is complete.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Sometimes the firmware won't be able to decrypt frames
because the keys were not installed yet or other scenarios.
The firmware will soon stop dropping multicast frames when
MAC_FILTER_ACCEPT_GRP is not set. The firmware will simply
always pass multicast frame in.
In order to avoid logging any such frame coming in when we
don't have the keys, drop the print.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Any firmware that supports the new channel switch flow is
able to close / re-open the queues when needed. It takes
into account the channel switch mode etc...
Don't open / close the queues or enable / disable beacon
abort before and after the channel switch in case the
firmware is able to do this by itself.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
This means:
1) stop calling pm_runtime_resume when starting the hardware
2) removing the unneeded low_power parameter to start / stop hw / fw
transport ops
3) squashing transport functions that are now the same
_iwl_trans_pcie_start_hw / iwl_trans_pcie_start_hw
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Remove the now unneeded functions that called those from the
transport layer.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
This flag should never be set unless integration work with the
platform is done. We don't support any platforms officially and don't
plan to do so in the near future, so we can remove this option
entirely in order to avoid having it enabled by mistake.
This has been marked with "depends on EXPERT", so there shouldn't be
many systems running with it set. And, if there are systems, they
should not be using this flag.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
This flag is never set on any host command. Remove it.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
We're now left with a status bit that is never set and a few
other leftovers.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Also change the signature of the power functions that won't
receive d0i3=true anymore.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Now that d0i3 is dead, this function can't be called from d0i3
flows. Change its signature and make it static.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
For runtime PM to work with d0i3 code, a lot of integration work needs
to be done with the platform (e.g. the out-of-band wake up interrupt)
and we currently don't have any platforms where this integration
happened. So, this code has been pretty much stale for a while and
when someone enables it, it just breaks things.
Therefore, to simplify the code base and make sure no one enables this
by mistake, we will remove the whole code.
This is only the very start, much more work is needed.
Remove the places where we check iwl_mvm_is_d0i3_supported
but leave all the refs, those will be removed in a different
patch.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
If mvm->fwrt.cur_fw_img != IWL_UCODE_INIT, then
rfkill_safe_init_done must be true since
rfkill_safe_init_done is set to true before we start to load
the runtime image.
Remove the redundant condition.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Starting from 22560, the byte count is expected to be in
bytes and we have now 14 bits. Ajust the code to this.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>
Accessing the hdr of an skb that was consumed already isn't
a good idea.
First ask if the skb is a QoS packet, then keep that data
on stack, and then consume the skb.
This was spotted by KASAN.
Cc: [email protected]
Fixes: 08f7d8b69a ("iwlwifi: mvm: bring back mvm GSO code")
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
The index for the elements of the ACPI object we dereference
was static. This means that if we called the function twice
we wouldn't start from 3 again, but rather from the latest
index we reached in the previous call.
This was dutifully reported by KASAN.
Fix this.
Cc: [email protected]
Fixes: 6996490501 ("iwlwifi: mvm: add support for EWRD (Dynamic SAR) ACPI table")
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
In order to remember how to unmap a memory (as single or
as page), we maintain a bit per Transmit Buffer (TBs) in
the meta data (structure iwl_cmd_meta).
We maintain a bitmap: 1 bit per TB.
If the TB is set, we will free the memory as a page.
This bitmap was never cleared. Fix this.
Cc: [email protected]
Fixes: 3cd1980b0c ("iwlwifi: pcie: introduce new tfd and tb formats")
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
An earlier patch made sure that the queues are not lagging
too far behind. This means that iwl_mvm_release_frames
should not be called with a head_sn too far behind NSSN.
Don't take the risk to change completely the entry
condition to iwl_mvm_release_frames, but don't update
the head_sn is the NSSN is more than 2048 packets ahead
of us. Since this just cannot be right. This means that
the scenario described here happened. We are queue 0.
Q:0 Q:1
head_sn: 0 -> 2047
head_sn: 2048
Lots of packets arrive:
head_sn: 2047 -> 2150
send NSSN_SYNC notification
Handle notification
from the firmware and
do NOT move the head_sn
back to 2048
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
In order to support MSI-X efficiently, we want to avoid
communication across Rx queues. Each Rx queue should have
all the data it needs to process a packet.
The reordering buffer is a challenge in the MSI-X world
since we can have a single BA session whose packets are
directed to different queues. This is why each queue has
its own reordering buffer. The hardware is able to hint
the driver whether we have a hole or not, which allows
the driver to know whether it can release a packet or not.
This indication is called NSSN. Roughly, if the packet's
SN is lower than the NSSN, we can release the packet to
the stack. The NSSN is the SN of the newest packet received
without any holes + 1.
This is working as long as we don't have packets that we
release because of a timeout. When that happens, we could
have taken the decision to release a packet after we have
been waiting for its predecessor for too long. If this
predecessor comes later, we have to drop it because we
can't release packets out of order. In that case, the
hardware will give us an indication that we can we release
the packet (SN < NSSN), but the packet still needs to be
dropped.
This is why we sometimes need to ignore the NSSN and we
track the head_sn in software.
Here is a specific example of this:
1) Rx queue 1 got packets: 480, 482, 483
2) We release 480 to to the stack and wait for 481
3) NSSN is now 481
4) The timeout expires
5) We release 482 and 483, NSSN is still 480
6) 481 arrives its NSSN is 484.
We need to drop 481 even if 481 < 484. This is why we'll
update the head_sn to 484 at step 2. The flow now is:
1) Rx queue 1 got packets: 480, 482, 483
2) We release 480 to to the stack and wait for 481
3) NSSN is now 481 / head_sn is 481
4) The timeout expires
5) We release 482 and 483, NSSN is still 480 but head_sn is 484.
6) 481 arrives its NSSN is 484, but head_sn is 484 and we drop it.
This code introduces another problem in case all the traffic
goes well (no hole, no timeout):
Rx queue 1: 0 -> 483 (head_sn = 484)
Rx queue 2: 501 -> 4095 (head_sn = 0)
Rx queue 2: 0 -> 480 (head_sn = 481)
Rx queue 1: 481 but head_sn = 484 and we drop it.
At this point, the SN of queue 1 is far behind: more than
4040 packets behind. Queue 1 will consider 481 "old"
because 481 is in [501-64:501] whereas it is a very new
packet.
In order to fix that, send an Rx notification from time to
time (twice across the full set of 4096 packets) to make
sure no Rx queue is lagging too far behind.
What will happen then is:
Rx queue 1: 0 -> 483 (head_sn = 484)
Rx queue 2: 501 -> 2047 (head_sn = 2048)
Rx queue 1: Sync nofication (head_sn = 2048)
Rx queue 2: 2048 -> 4095 (head_sn = 0)
Rx queue 1: Sync notification (head_sn = 0)
Rx queue 2: 1 -> 481 (head_sn = 482)
Rx queue 1: 481 and head_sn = 0.
In queue 1's data, head_sn is now 0, the packet coming in
is 481, it'll understand that the new packet is new and it
won't be dropped.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
Instead of allocating memory for which we have an upper
limit, use a small buffer on stack.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
We will soon be using a new notification that will be
initiated by the driver, sent to the firmware and sent
back to all the RSS queues by the firmware. This new
notification will be useful to synchronize the NSSN across
all the queues.
For now, don't send the notification, just add the code to
handle it. Later patch will add the code to actually send
it.
While at it, validate the baid coming from the firmware to
avoid accessing an array with a bad index in the driver.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
We will need a new type of synchronization message going
through all the RSS queues. Prepare the ground for this.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
iwl_mvm_send_cmd returns 0 when the command won't be sent
because RF-Kill is asserted. Do the same when we call
iwl_get_shared_mem_conf since it is not sent through
iwl_mvm_send_cmd but directly calls the transport layer.
Cc: [email protected]
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Sometimes the register status can include interrupts that
were masked. We can, for example, get the RF-Kill bit set
in the interrupt status register although this interrupt
was masked. Then if we get the ALIVE interrupt (for example)
that was not masked, we need to *not* service the RF-Kill
interrupt.
Fix this in the MSI-X interrupt handler.
Cc: [email protected]
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Newest devices have a new firmware load mechanism. This
mechanism is called the context info. It means that the
driver doesn't need to load the sections of the firmware.
The driver rather prepares a place in DRAM, with pointers
to the relevant sections of the firmware, and the firmware
loads itself.
At the end of the process, the firmware sends the ALIVE
interrupt. This is different from the previous scheme in
which the driver expected the FH_TX interrupt after each
section being transferred over the DMA.
In order to support this new flow, we enabled all the
interrupts. This broke the assumption that we have in the
code that the RF-Kill interrupt can't interrupt the firmware
load flow.
Change the context info flow to enable only the ALIVE
interrupt, and re-enable all the other interrupts only
after the firmware is alive. Then, we won't see the RF-Kill
interrupt until then. Getting the RF-Kill interrupt while
loading the firmware made us kill the firmware while it is
loading and we ended up dumping garbage instead of the firmware
state.
Re-enable the ALIVE | RX interrupts from the ISR when we
get the ALIVE interrupt to be able to get the RX interrupt
that comes immediately afterwards for the ALIVE
notification. This is needed for non MSI-X only.
Cc: [email protected]
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
We added code to restock the buffer upon ALIVE interrupt
when MSI-X is disabled. This was added as part of the context
info code. This code was added only if the ISR debug level
is set which is very unlikely to be related.
Move this code to run even when the ISR debug level is not
set.
Note that gen2 devices work with MSI-X in most cases so that
this path is seldom used.
Cc: [email protected]
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
TWT is still very new and we expect issues. Make its usage
configurable and disable it by default.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
FSEQ TLV should be parsed and read even when FMAC is not compiled.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
When we have a single image (same firmware image for INIT and
OPERATIONAL), we couldn't load the driver and register to the
stack if we had hardware RF-Kill asserted.
Fix this. This required a few changes:
1) Run the firmware as part of the INIT phase even if its
ucode_type is not IWL_UCODE_INIT.
2) Send the commands that are sent to the unified image in
INIT flow even in RF-Kill.
3) Don't ask the transport to stop the hardware upon RF-Kill
interrupt if the RF-Kill is asserted.
4) Allow the RF-Kill interrupt to take us out of L1A so that
the RF-Kill interrupt will be received by the host (to
enable the radio).
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>
We introduced a bug that prevented this old device from
working. The driver would simply not be able to complete
the INIT flow while spewing this warning:
CSR addresses aren't configured
WARNING: CPU: 0 PID: 819 at drivers/net/wireless/intel/iwlwifi/pcie/drv.c:917
iwl_pci_probe+0x160/0x1e0 [iwlwifi]
Cc: [email protected] # v4.18+
Fixes: a8cbb46f83 ("iwlwifi: allow different csr flags for different device families")
Signed-off-by: Emmanuel Grumbach <[email protected]>
Fixes: c8f1b51e50 ("iwlwifi: allow different csr flags for different device families")
Signed-off-by: Luca Coelho <[email protected]>
The spatial reuse 4 words fields are fetched from the HE-SIGA
by the firmware and propagated to the driver through the
Rx info. This is useful to populate the radiotap header.
We were looking at the wrong place in the firmware data and
got bogus values. Fix that.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Fixes: bdf180c8d3 ("iwlwifi: mvm: change PHY data RX for HE radiotap")
Signed-off-by: Luca Coelho <[email protected]>
Recently we started to send the WEP keys to the firmware so
that we could use hardware Tx encryption also on newer
devices that require the keys to be installed in the firmware
for encryption (as opposed to older devices that can get
the key in the Tx command for each Tx).
When we implemented that, we forgot to remove the key when
we remove a station leading to a situation where a station
that connects and disconnects a lot of times exhausts the
key database inside the firmware.
A fix was made for that, but we always removed the same
key: mvmvif->ap_wep_key which means that we removed the
same key entry in the firmware. This can make sense since
in WEP, the key is the same for all the stations, but the
internal implementation of iwl_mvm_set_sta_key and
iwl_mvm_remove_sta_key assumes that each station uses a
different key in the firmware's key database.
So now we got to the situation where we have a single
ieee80211_key_conf instance that means, a single
ieee80211_key_conf.hw_key_idx index for several stations
and hence for several keys.
ieee80211_key_conf.hw_key_idx is set to 0 when the first
station associates, and then it is overwritten to 1 when
the second station associates which is a buggy of course.
This led to the following message upon the removal of the
second station:
iwlwifi 0000:00:03.0: offset 1 not used in fw key table.
WARNING: CPU: 2 PID: 27883 at net/mac80211/sta_info.c:1122 __sta_info_destroy_part2+0x16b/0x180 [mac80211]
RIP: 0010:__sta_info_destroy_part2+0x16b/0x180 [mac80211]
Call Trace:
__sta_info_destroy+0x2a/0x40 [mac80211]
sta_info_destroy_addr_bss+0x38/0x60 [mac80211]
ieee80211_del_station+0x1d/0x30 [mac80211]
nl80211_del_station+0xe0/0x1f0 [cfg80211]
Fix this by copying the ieee80211_key_conf structure for
each and every station. This is the easiest way to properly
remove the keys with the right index. Another solution
would have been to allow several stations to use the same
key offset in the firmware. That would require to change
the way we track keys in iwlmvm and not really worth it.
Also, maintain correctly fw_key_table when we add a key
for the multicast station.
Remove the key when we remove the multicast station.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Fixes: 337bfc9881 ("iwlwifi: mvm: set wep key for all stations in soft ap mode")
Signed-off-by: Luca Coelho <[email protected]>
The firmware is changing the format of the beacon
notification to remove the dependency on the Tx response
format.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
We maintain a timer for each Tx queue to detect stalls and
be able to recover / debug.
When we work in AP mode, we can freeze the Tx queue timer if
a station goes to sleep, because we don't want to warn about
stalls that are caused by faulty clients that don't wake up
on time.
This mechanism was applied to the queues of the clients, but
the management queue was omitted. Fix that.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
The new (CDB) statistics API is used by non-CDB devices
as well. Look at the right TLV flag to know which version
of the statistics notification to use.
To avoid confusion, remove the _cdb suffix from the
structure name.
While at it, remove a structure that was never used.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Fixes: 678d9b6ddd ("iwlwifi: mvm: update rx statistics cmd api")
Signed-off-by: Luca Coelho <[email protected]>
Mac80211 will check both the HE Capability IE and the
Extended Capability IE, so set the TWT support bit when
mac80211 tells us to.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
We want to advertise support for TWT in the Extended
Capability IE. Since we don't want to set the bits for all
the interface types, define an interface specific
configuration.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
TWT is a feature that was added in 11ah and enhanced in
11ax. There are two bits that need to be set if we want
to use the feature in 11ax: one in the HE Capability IE
and one in the Extended Capability IE. This is because
of backward compatibility between 11ah and 11ax.
In order to simplify the flow for the low level driver
in managed mode, aggregate the two bits and add a boolean
that tells whether TWT is supported or not, but only if
11ax is supported.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
These bits are defined in ieee802.11ax to advertise support
for TWT in addition to the bits in the HE IE.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
We used to have many versions of statistics notification
coming from the firmware. In one of the cleanup patches,
we forgot to clean the code that checks if data->general
is set. Since it is always set, remove the check.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
The iTXQs stop/wake queue mechanism involves a whole bunch
of locks and this is probably why the call to
ieee80211_wake_txqs is deferred to a tasklet when called from
__ieee80211_wake_queue.
Another advantage of that is that ieee80211_wake_txqs might
call the wake_tx_queue() callback and then the driver may
call mac80211 which will call it back in the same context.
The bug I saw is that when we send a deauth frame as a
station we do:
flush(drop=1)
tx deauth
flush(drop=0)
While we flush we stop the queues and wake them up
immediately after we finished flushing. The problem here is
that the tasklet that de-facto enables the queue may not have
run until we send the deauth. Then the deauth frame is sent
to the driver (which is surprising by itself), but the driver
won't get anything useful from ieee80211_tx_dequeue because
the queue is stopped (or more precisely because
vif->txqs_stopped[0] is true).
Then the deauth is not sent. Later on, the tasklet will run,
but that'll be too late. We'll already have removed all the
vif etc...
Fix this by calling ieee80211_wake_txqs synchronously if we
are not waking up the queues from the driver (we check the
reason to determine that). This makes the code really
convoluted because we may call ieee80211_wake_txqs from
__ieee80211_wake_queue. The latter assumes that
queue_stop_reason_lock has been taken by the caller and
ieee80211_wake_txqs may release the lock to send the frames.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
NullFunc packets should never be duplicate just like
QoS-NullFunc packets.
We saw a client that enters / exits power save with
NullFunc frames (and not with QoS-NullFunc) despite the
fact that the association supports HT.
This specific client also re-uses a non-zero sequence number
for different NullFunc frames.
At some point, the client had to send a retransmission of
the NullFunc frame and we dropped it, leading to a
misalignment in the power save state.
Fix this by never consider a NullFunc frame as duplicate,
just like we do for QoS NullFunc frames.
This fixes https://bugzilla.kernel.org/show_bug.cgi?id=201449
CC: <[email protected]>
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
We need to check the TWT support of the peer and to
propagte the capability to the firmware.
The current implementation will enable TWT only if the TWT
support is advertised in the HE CAP IE and in the Extended
Capability IE.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
The firmware stopped looking at this field long ago.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
When the firmware starts, it doesn't have any regulatory
information, hence it uses the world wide limitations. The
driver can feed the firmware with previous knowledge that
was kept in the driver, but the firmware may still not
update its internal tables.
This happens when we start a BSS interface, and then the
firmware can change the regulatory tables based on our
location and it'll use more lenient, location specific
rules. Then, if the firmware is shut down (when the
interface is brought down), and then an AP interface is
created, the firmware will forget the country specific
rules.
The host will think that we are in a certain country that
may allow channels and will try to teach the firmware about
our location, but the firmware may still not allow to drop
the world wide limitations and apply country specific rules
because it was just re-started.
In this case, the firmware will reply with MCC_RESP_ILLEGAL
to the MCC_UPDATE_CMD. In that case, iwlwifi needs to let
the upper layers (cfg80211 / hostapd) know that the channel
list they know about has been updated.
This fixes https://bugzilla.kernel.org/show_bug.cgi?id=201105
Cc: [email protected]
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
The oldest firmware supported by iwlmvm do support getting
the average beacon RSSI. Enable the sta_statistics() call
from mac80211 even on older firmware versions.
Fixes: 33cef92563 ("iwlwifi: mvm: support beacon statistics for BSS client")
Cc: [email protected] # 4.2+
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
chanwidths isn't used now in debugfs-vif.c. Remove it.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Fix a bug that happens in the following scenario:
1) suspend without WoWLAN
2) mac80211 calls drv_stop because of the suspend
3) __iwl_mvm_mac_stop deallocates the aux station
4) during drv_stop the firmware crashes
5) iwlmvm:
* sets IWL_MVM_STATUS_HW_RESTART_REQUESTED
* asks mac80211 to kick the restart flow
6) mac80211 puts the restart worker into a freezable
queue which means that the worker will not run for now
since the workqueue is already frozen
7) ...
8) resume
9) mac80211 runs ieee80211_reconfig as part of the resume
10) mac80211 detects that a restart flow has been requested
and that we are now resuming from suspend and cancels
the restart worker
11) mac80211 calls drv_start()
12) __iwl_mvm_mac_start checks that IWL_MVM_STATUS_HW_RESTART_REQUESTED
clears it, sets IWL_MVM_STATUS_IN_HW_RESTART and calls
iwl_mvm_restart_cleanup()
13) iwl_fw_error_dump gets called and accesses the device
to get debug data
14) iwl_mvm_up adds the aux station
15) iwl_mvm_add_aux_sta() allocates an internal station for
the aux station
16) iwl_mvm_allocate_int_sta() tests IWL_MVM_STATUS_IN_HW_RESTART
and doesn't really allocate a station ID for the aux
station
17) a new queue is added for the aux station
Note that steps from 5 to 9 aren't really part of the
problem but were described for the sake of completeness.
Once the iwl_mvm_mac_stop() is called, the device is not
accessible, meaning that step 12) can't succeed and we'll
see the following:
drivers/net/wireless/intel/iwlwifi/pcie/trans.c:2122 iwl_trans_pcie_grab_nic_access+0xc0/0x1d6 [iwlwifi]()
Timeout waiting for hardware access (CSR_GP_CNTRL 0x080403d8)
Call Trace:
[<ffffffffc03e6ad3>] iwl_trans_pcie_grab_nic_access+0xc0/0x1d6 [iwlwifi]
[<ffffffffc03e6a13>] iwl_trans_pcie_dump_regs+0x3fd/0x3fd [iwlwifi]
[<ffffffffc03dad42>] iwl_fw_error_dump+0x4f5/0xe8b [iwlwifi]
[<ffffffffc04bd43e>] __iwl_mvm_mac_start+0x5a/0x21a [iwlmvm]
[<ffffffffc04bd6d2>] iwl_mvm_mac_start+0xd4/0x103 [iwlmvm]
[<ffffffffc042d378>] drv_start+0xa1/0xc5 [iwl7000_mac80211]
[<ffffffffc045a339>] ieee80211_reconfig+0x145/0xf50 [mac80211]
[<ffffffffc044788b>] ieee80211_resume+0x62/0x66 [mac80211]
[<ffffffffc0366c5b>] wiphy_resume+0xa9/0xc6 [cfg80211]
The station id of the aux station is set to 0xff in step 3
and because we don't really allocate a new station id for
the auxliary station (as explained in 16), we end up sending
a command to the firmware asking to connect the queue
to station id 0xff. This makes the firmware crash with the
following information:
0x00002093 | ADVANCED_SYSASSERT
0x000002F0 | trm_hw_status0
0x00000000 | trm_hw_status1
0x00000B38 | branchlink2
0x0001978C | interruptlink1
0x00000000 | interruptlink2
0xFF080501 | data1
0xDEADBEEF | data2
0xDEADBEEF | data3
Firmware error during reconfiguration - reprobe!
FW error in SYNC CMD SCD_QUEUE_CFG
Fix this by clearing IWL_MVM_STATUS_HW_RESTART_REQUESTED
in iwl_mvm_mac_stop(). We won't be able to collect debug
data anyway and when we will brought up again, we will
have a clean state from the firmware perspective.
Since we won't have IWL_MVM_STATUS_IN_HW_RESTART set in
step 12) we won't get to the 2093 ASSERT either.
Fixes: bf8b286f86 ("iwlwifi: mvm: defer setting IWL_MVM_STATUS_IN_HW_RESTART")
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
We no longer want to consider the RSSI if the beacon / probe
has been heard on an adjacent channel.
This was based on a firmware capability that is now
unavailable.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
We can dump data from the firmware either when it crashes,
or when the firmware is alive.
Not all the data is available if the firmware is running
(like the Tx / Rx FIFOs which are available only when the
firmware is halted), so we first check that the firmware
is alive to compute the required size for the dump and then
fill the buffer with the data.
When we allocate the buffer, we test the STATUS_FW_ERROR
bit to check if the firmware is alive or not. This bit
can be changed during the course of the dump since it is
modified in the interrupt handler.
We hit a case where we allocate the buffer while the
firmware is sill working, and while we start to fill the
buffer, the firmware crashes. Then we test STATUS_FW_ERROR
again and decide to fill the buffer with data like the
FIFOs even if no room was allocated for this data in the
buffer. This means that we overflow the buffer that was
allocated leading to memory corruption.
To fix this, test the STATUS_FW_ERROR bit only once and
rely on local variables to check if we should dump fifos
or other firmware components.
Fixes: 04fd2c2822 ("iwlwifi: mvm: add rxf and txf to dump data")
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
The SP length in the ADD_STA command is an actual number of
frames, and not the SP len as it appears in the WME IE.
Fix that comment. The actual code is fine.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
When tracing is enabled, all the debug messages are recorded and must
not exceed MAX_MSG_LEN (100) columns. Longer debug messages grant the
user with:
WARNING: CPU: 3 PID: 32642 at /tmp/wifi-core-20180806094828/src/iwlwifi-stack-dev/net/mac80211/./trace_msg.h:32 trace_event_raw_event_mac80211_msg_event+0xab/0xc0 [mac80211]
Workqueue: phy1 ieee80211_iface_work [mac80211]
RIP: 0010:trace_event_raw_event_mac80211_msg_event+0xab/0xc0 [mac80211]
Call Trace:
__sdata_dbg+0xbd/0x120 [mac80211]
ieee80211_ibss_rx_queued_mgmt+0x15f/0x510 [mac80211]
ieee80211_iface_work+0x21d/0x320 [mac80211]
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
If the driver fails to properly prepare for the channel
switch, mac80211 will disconnect. If the CSA IE had mode
set to 1, it means that the clients are not allowed to send
any Tx on the current channel, and that includes the
deauthentication frame.
Make sure that we don't send the deauthentication frame in
this case.
In iwlwifi, this caused a failure to flush queues since the
firmware already closed the queues after having parsed the
CSA IE. Then mac80211 would wait until the deauthentication
frame would go out (drv_flush(drop=false)) and that would
never happen.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
We hit a problem with iwlwifi that was caused by a bug in
mac80211. A bug in iwlwifi caused the firwmare to crash in
certain cases in channel switch. Because of that bug,
drv_pre_channel_switch would fail and trigger the restart
flow.
Now we had the hw restart worker which runs on the system's
workqueue and the csa_connection_drop_work worker that runs
on mac80211's workqueue that can run together. This is
obviously problematic since the restart work wants to
reconfigure the connection, while the csa_connection_drop_work
worker does the exact opposite: it tries to disconnect.
Fix this by cancelling the csa_connection_drop_work worker
in the restart worker.
Note that this can sound racy: we could have:
driver iface_work CSA_work restart_work
+++++++++++++++++++++++++++++++++++++++++++++
|
<--drv_cs ---|
<FW CRASH!>
-CS FAILED-->
| |
| cancel_work(CSA)
schedule |
CSA work |
| |
Race between those 2
But this is not possible because we flush the workqueue
in the restart worker before we cancel the CSA worker.
That would be bullet proof if we could guarantee that
we schedule the CSA worker only from the iface_work
which runs on the workqueue (and not on the system's
workqueue), but unfortunately we do have an instance
in which we schedule the CSA work outside the context
of the workqueue (ieee80211_chswitch_done).
Note also that we should probably cancel other workers
like beacon_connection_loss_work and possibly others
for different types of interfaces, at the very least,
IBSS should suffer from the exact same problem, but for
now, do the minimum to fix the actual bug that was actually
experienced and reproduced.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
Probe responses were sent to the multicast station while
they should be routed to the broadcast station.
This has no negative effect since the frame was still
routed to the right queue, but it looked very fishy
to send a frame to a (queue, station) tuple where
'queue' is not mapped to 'station'.
Fixes: 7c305de2b9 ("iwlwifi: mvm: Direct multicast frames to the correct station")
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
When the NIC is disconnected, we just can't do anything
besides seeking for help from the bus driver. Dumping the
device's memory is not necessary and just bloats the logs
with unusable data. Moreover, asking mac80211 to restart
the hardware is also useless. Bypass all this.
Also, use the STATUS_TRANS_DEAD status bit instead of a
bool inside the transport layer. The advantage of this is
that now, the transport and the op_mode can know what is the
situation and bypass the useless recovery steps mentioned
above.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
I changed the way mac80211 updates the PM state of the peer.
I forgot that we could also have multicast frames from the
peer and that those frame should of course not change the
PM state of the peer: A peer goes to power save when it
needs to scan, but it won't send the broadcast Probe Request
with the PM bit set.
This made us mark the peer as awake when it wasn't and then
Intel's firmware would fail to transmit because the peer is
asleep according to its database. The driver warned about
this and it looked like this:
WARNING: CPU: 0 PID: 184 at /usr/src/linux-4.16.14/drivers/net/wireless/intel/iwlwifi/mvm/tx.c:1369 iwl_mvm_rx_tx_cmd+0x53b/0x860
CPU: 0 PID: 184 Comm: irq/124-iwlwifi Not tainted 4.16.14 #1
RIP: 0010:iwl_mvm_rx_tx_cmd+0x53b/0x860
Call Trace:
iwl_pcie_rx_handle+0x220/0x880
iwl_pcie_irq_handler+0x6c9/0xa20
? irq_forced_thread_fn+0x60/0x60
? irq_thread_dtor+0x90/0x90
The relevant code that spits the WARNING is:
case TX_STATUS_FAIL_DEST_PS:
/* the FW should have stopped the queue and not
* return this status
*/
WARN_ON(1);
info->flags |= IEEE80211_TX_STAT_TX_FILTERED;
This fixes https://bugzilla.kernel.org/show_bug.cgi?id=199967.
Fixes: 9fef654433 ("mac80211: always update the PM state of a peer on MGMT / DATA frames")
Cc: <[email protected]> #4.16+
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
The periphery can't be accessed before we set the
INIT_DONE bit which initializes the device.
A previous patch added a reconfiguration of the MSI-X
tables upon resume, but at that point in the flow,
INIT_DONE wasn't set. Since the reconfiguration of the
MSI-X tables require periphery access, it failed.
The difference between WoWLAN and without WoWLAN is that
in WoWLAN, iwl_trans_pcie_d3_suspend clears the INIT_DONE
without clearing the STATUS_DEVICE_ENABLED bit in the
software status. Because of that, the resume code thinks
that the device is enabled, but the INIT_DONE bit has been
cleared.
To fix this, don't reconfigure the MSI-X tables in case
WoWLAN is enabled. It will be done in
iwl_trans_pcie_d3_resume anyway.
Fixes: 52848a79b9 ("iwlwifi: pcie: reconfigure MSI-X HW on resume")
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
A peer can limit the number of subframes it can handle in a
single A-MSDU. Honor this limit.
Note that the smallest limit is 8, and we are very unlikely to reach
that limit. So this isn't really a big deal.
Fixes: a6d5e32f24 ("iwlwifi: mvm: send large SKBs to the transport")
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Our hardware has a limited amount of buffer descriptors
for each Tx packet. Because of that, there is a short
piece of code that makes sure that that we don't push too
many subframes in an A-MSDU because of subframes needs 2
buffer descriptors. This code also takes into account the
number of fragment of the skb since we also need a buffer
descriptor for each fragment in the skb.
This piece of code though didn't check that the resulting
number of subframes wasn't 0.
A user reported that using NFS client, he could get skbs
that are so fragmented that the code mentioned above
returned 0 for the number of subframes making
skb_gso_segment fail and subconsequently iwlwifi would WARN.
Fix this by make sure that num_subframes is at least 1.
This fixes:
https://bugzilla.kernel.org/show_bug.cgi?id=199209
Fixes: a6d5e32f24 ("iwlwifi: mvm: send large SKBs to the transport")
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
22000 devices rely on this flag to install the key to the right
queues. For earlier devices we didn't have a key / queue mapping and
the key was sent along with the Tx command for each Tx hence the
problem didn't arise.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
When we enable TSO, we can have a lot of packets in the
operation mode that will be pushed to the transport
no matter what is the queue's fullness state.
To cope with that the transport can buffer those packets
and add them to the ring later when there is more room.
This implementation was missing in the Gen2 devices'
code.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
The primary channel is the channel that will be untouched by BT. The
secondary channel might be touched by BT. Hence, we want the primary
to be the most active channel. To do so, use the TCM infrastructure.
Since the BT keeps sending notifications, we can rely on them to
trigger the check. Every 10 seconds, we will check what is the most
active context and chose the right primary.
We need to wait 10 seconds before we modify the settings because
frequent changes in these settings can confuse BT.
Signed-off-by: Emmanuel Grumbach <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Emmanuel Grumbach