When trying to reach high Rx throughput of more than 500Mbps on
a device with a relatively weak CPU (Atom x5-Z8500), CPU utilization
may become a bottleneck. Analysis showed that we are looping in
iwl_pcie_rx_handle for very long periods which led to starvation
of other threads (iwl_pcie_rx_handle runs with _bh disabled).
We were handling Rx and allocating new buffers and the new buffers
were ready quickly enough to be available before we had finished
handling all the buffers available in the hardware. As a
consequence, we called iwl_pcie_rxq_restock to refill the hardware
with the new buffers, and start again handling new buffers without
exiting the function. Since we read the hardware pointer again when
we goto restart, new buffers were handled immediately instead of
exiting the function.

This patch avoids refilling RBs inside rx handling loop, unless an
emergency situation is reached. It also doesn't read the hardware
pointer again unless we are in an emergency (unlikely) case.
This significantly reduce the maximal time we spend in
iwl_pcie_rx_handle with _bh disabled.
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

iwl_mvm_tcool_get_cur_state is the function that returns the
cooling state index to the sysfs handler. This function returns
mvm->cooling_dev.cur_state but that variable was set to the
budget and not the cooling state index. Fix that.
Add a missing blank line while at it.
Signed-off-by: Chaya Rachel Ivgi <chaya.rachel.ivgi@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

mac80211 advertises this feature for all its drivers.
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

We need to track the next packet that we will reclaim in
order to know when the Tx queues are empty. This is useful
when we open or tear down an A-MPDU session which requires
to switch queue.
The next packet being reclaimed is identified by its WiFi
sequence number and this is relevant only when we use QoS.
QoS NDPs do have a TID but have a meaningless sequence
number. The spec mandates the receiver to ignore the
sequence number in this case, allowing the transmitter to
put any sequence number. Our implementation leaves it 0.
When we reclaim a QoS NDP, we can't update the next_relcaim
counter since the sequence number of the QoS NDP itself is
invalid.
We used to update the next_reclaim based on the sequence
number of the QoS NDP which reset it to 1 (0 + 1) and
because of this, we never knew when the queue got empty.
This had to sad consequence to stuck the A-MPDU state
machine in a transient state.
To fix this, don't update next_reclaim when we reclaim
a QoS NDP.

Alesya saw this bug when testing u-APSD. Because the
A-MPDU state machine was stuck in EMPTYING_DELBA, we
updated mac80211 that we still have frames for that
station when it got back to sleep. mac80211 then wrongly
set the TIM bit in the beacon and requested to release
non-existent frames from the A-MPDU queue. This led to
a situation where the client was trying to poll frames
but we had no frames to send.
Reported-by: Alesya Shapira <alesya.shapira@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

From 9000 family on, we need to get HW address from host
CSR registers.
OEM can override it by fusing the override registers - read
those first, and if those are 0 - read the OTP registers instead.

In addition - bail out if no valid mac address is present. Make
it shared for all NICs.
Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

We kick the allocator when we have 2 RBDs that don't have
attached RBs, and the allocator allocates 8 RBs meaning
that it needs another 6 RBDs to attach the RBs to.
The design is that allocator should always have enough RBDs
to fulfill requests, so we give in advance 6 RBDs to the
allocator so that when it is kicked, it gets additional 2 RBDs
and has enough RBDs.
These RBDs were taken from the Rx queue itself, meaning
that each Rx queue didn't have the maximal number of
RBDs, but MAX - 6.
Change initial number of RBDs in the system to include both
queue size and allocator reserves.
Note the multi-queue is always 511 instead of 512 to avoid a
full queue since we cannot detect this state easily enough in
the 9000 arch.
Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

When running async rx handler the framework holds the mvm->mutex
before starting the async handler, that might cause a deadlock in case
the handler calls to ops that lock the mutex as well.
Add support for running async rx handler without hold the mutex before
activating the handler.
Signed-off-by: Chaya Rachel Ivgi <chaya.rachel.ivgi@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

Currently when the FW sends start/stop aux roc time event
notification with an error status, the driver returns an
error value, but does not remove the time event, and does
not notify the stack above that the time event is over.

This causes problems that the stack above assumes we are still
in the middle of a time event, and therefore can block different
events, such as scanning.

On FW failure notification, cleanup the time event parameters and
notify the stack above that the time event is over.
Signed-off-by: Matti Gottlieb <matti.gottlieb@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

Our hardware de-aggregates AMSDUs but copies the mac header
as it to the de-aggregated MPDUs. We need to turn off the AMSDU
bit in the QoS control ourselves.
Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

Add debugfs entries to get the ctdp budget average
and to stop ctdp.
Signed-off-by: Chaya Rachel Ivgi <chaya.rachel.ivgi@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

Before authentication, we start a time event during
which we wait for a beacon in order to sync our timers.
If we didn't hear the beacon during this time - we abandon
the connection. However, in congested environment, it was
observed we might not hear beacons in that time slot.
Extend the time event to give the connection a better chance.
Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

The amsdu enum values are off by 1 bit. Fix it.
Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

The call to iwl_mvm_thermal_initialize() was too early in the
function.
Unregister will be performed when goto out_unregister is called,
but as the code was - out_free may be called and leave without
unregistering from thermal.
Signed-off-by: Chaya Rachel Ivgi <chaya.rachel.ivgi@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

It makes it slightly easier to follow. Pass the pointer to
the transport which allows to read WFMP_MAC_ADDR_X register
only when needed and to use IWL_ERR instead of the less
commonly used IWL_ERR_DEV logger macro.
Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

Thermal zone device registration can fail, and in this case
we don't want to remove WiFi functionality. This is why the
thermal zone registration function is void, and the flows
continue even if the thermal zone device registration failed.
Same applies for the cooling device.

This means that we at least need to remember that the thermal
zone device didn't register properly and take the minimal
precautions to avoid panic'ing when we access it.

This was missing.
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

128 byte chunk size is supported only on PCIe and not
on IOSF. For now, change it back to 64 byte.
Reported-by: Oren Givon <oren.givon@intel.com>
Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

Change the code to move rxbs directly from the allocator's
list to the queue's free list. This makes the code more
readable, saves the interim array and the double loop over
the free RBs.
Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

The pci driver keeps any unbound device in active state and forbids
runtime PM.  When our driver gets probed, we take control of the
state.  When the device is released (i.e. during unbind or module
removal), we should return the state to what it was before.  To do so,
we need to forbid RTPM in the driver remove op.

Additionally, remove an unnecessary pm_runtime_disable() call, move
the initial ref_count setting to a better place and add some comments
explaining what is going on.
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

Add a wrapper function to allow stopping SW queues from MVM
as well.
Signed-off-by: Liad Kaufman <liad.kaufman@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

If d0i3 is supported, we have released the initial transport reference
in iwl_op_mode_mvm_start(), so we should take it back in
iwl_op_mode_mvm_stop() to keep it balanced.
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

Add module parameter that disables VHT capabilities.
Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

If d0i3 is not supported by the firmware (or if it's disabled via
module parameters) we shouldn't release the initial transport
reference, so that we won't enter runtime suspend.
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

Do not allow entrance into DQA flows until feature is
completely ready and merged.
Signed-off-by: Liad Kaufman <liad.kaufman@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

In 9000 series A0 step the closed_rb_num is not wrapping around
properly. The queue is wrapping around as it should, so we can
W/A it by wrapping the closed_rb_num in the driver.
While at it, extend RX logging and add error handling of other
cases HW values may cause us to access invalid memory locations.
Add also a proper masking of vid value read from HW - this should
not have actual affect, but better to be on the safe side.
Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

Leaving ucode_loaded to true after stop_device() has been called
is a recipe for problems. Flows that are not sync'ed with the
driver life cycle (like debugfs hooks and thermal hooks) must
check that the firmware is loaded before they interact with it.
Therefore we need to keep this variable updated with the real
status of the firmware.
Signed-off-by: Chaya Rachel Ivgi <chaya.rachel.ivgi@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

Till today, the ucode consisted of two d0 images - regular,
in which the usniffer wasn't enabled, and usniffer, in which the
usniffer logs were enabled.
Lately, the two images were unified, so there is only one d0 image,
in which the usniffer logs are enabled.

Add new TLV capability for supporting the consolidated images
(set 2, bit 13).
Signed-off-by: Golan Ben-Ami <golan.ben.ami@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

Currently when entering D3 with WOWLAN configured, we enable in the
configuration flags beacon storing, and do not disable beacon
filtering, and do not wake up from a magic packet.

Having both enabled is wrong (should not have both enabled),
and causes problems in the RX queues in the FW, causing
the FW not to recognize the magic packet when it comes.

Disable beacon storing in wowlan configuration.
Signed-off-by: Matti Gottlieb <matti.gottlieb@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

iwl_mvm_enable_agg_txq has only one user. Kill it and
adapt the call site.
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

To ensure that the SNAP/TCP/IP headers are DW aligned, the firmware
may add 2-byte pad at the end of the mac header - after the IV, before
the SNAP.
In that case the mpdu descriptor pad bit will be turned on.
Driver should take it into consideration, and remove the padding before
passing the packet to mac80211. Do that.
Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

Beacon abort (ba) is set while sending power command, but only
after at least one beacon_filter command was successfully sent.

If we heard a beacon before starting association, this order
is maintained and ba is properly set.

However, if the first beacon is received after association,
we send the power command upon association, configure the
beacon filtering when the first beacon arrives, and in that case,
beacon abort is not set.

So identify this, and send a power command post the beacon_filter
command if needed.
Signed-off-by: Avri Altman <avri.altman@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

Older versions of the firmware don't support U-APSD for
P2P Client. Forbid U-APSD for P2P Client when an old
firmware is being used.
Signed-off-by: Avri Altman <avri.altman@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

Allow to publish RRM capabilities without the need to support a minimal
capability set. Since some RRM features(e.g. neighbor report) are fw
independent, set this capability unconditionally.
Signed-off-by: Beni Lev <beni.lev@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

* Remove uneeded includes:
	iwl-csr.h and devcoredump aren't used in mac80211.c.
* Remove uneeded empty line
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

When the device is in d0i3/d3 we will not receive the VHT
MU-MIMO group id management frame. Instead, firmware will
notify us upon exit on the current status and we can in turn
update mac80211. Support this notification.
While at it, also check as a precaution that the vif is indeed
the VHT MU-MIMO owner before updating the firmware.
Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

Commit 69c7fda4 removed the
users of iwl_mvm_tid_data.reduced_tpc. Due to a conflict,
I forgot to commit the hunk that removed the field itself.
Do this know.
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

In multi rx queue HW, without execessive locking, there is no sync
between the ctrl path (default queue) and the rest of the rx queues.
This might cause issues on certain situations. For example, in case
a delBA was processed on a default queue but out of order packets
still wait for processing on the other queue.

The solution is to introduce internal messaging between the CTRL path
and the other rx queues.
The driver will send a message to the firmware, which will echo it to
all the requested queues. The message will be in order inside the queue.
This way we can avoid CTRL path and RSS queues races.

Add support for this messaging mechanism. As the firmware is agnostic to
the data sent, add internal representation of the data as well.
Although currently only delBA flow will use it, the internal representation
will enable generic use of this infrastructure for future uses.
Next patch will utilize this messaging mechanism for the reorder buffer
delBA flow.
Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

Next hardware will direct TCP/UDP streams to different cores.
Packets belonging to the same stream will be directed to the same
core.
The result is that duplicates will be always directed to the same
rx queue were the first packet was received.
This enabled parallelizing the duplicate packet detection across
the different cores, without sharing data between the rx queues.
Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

The patch below introduced a variable shadowing. Fix that.

Fixes: 3955525d ("iwlwifi: pcie: buffer packets to avoid overflowing Tx queues")
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

With these ops, we can know when we are about to enter system suspend.
This allows us to exit D0i3 state before entering suspend.
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>