Enable runtime power management (RTPM) for PCIe devices and implement
the corresponding functions to enable D0i3 mode when the device is
idle.

Additionally, remove some unnecessary #ifdef's because the RTPM code
will not be called if runtime PM is not configured.
Signed-off-by: NLuca Coelho <luciano.coelho@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

When the Tx queues are full above a threshold, we
immediately stop the mac80211's queue to stop getting new
packets. This worked until TSO was enabled.
With TSO, one single packet from mac80211 can use many
descriptors since a large send needs to be split into
several segments.
This means that stopping mac80211's queues is not enough
and we also need to ensure that we don't overflow the Tx
queues with one single packet from mac80211.
Add code to transport layer to do just that. Stop
mac80211's queue as soon as the queue is full above the
same threshold as before, and keep pushing the current
packet along with its segments on the queue, but check
that we don't overflow. If that would happen, buffer the
segments, and send them when there is room in the Tx queue
again. Of course, we first need to send the buffered
segments and only then, wake up mac80211's queues.
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

All the callers used silent = false.
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

When the op_mode sends an skb whose payload is bigger than
MSS, PCIe will create an A-MSDU out of it. PCIe assumes
that the skb that is coming from the op_mode can fit in one
A-MSDU. It is the op_mode's responsibility to make sure
that this guarantee holds.

Additional headers need to be built for the subframes.
The TSO core code takes care of the IP / TCP headers and
the driver takes care of the 802.11 subframe headers.

These headers are stored on a per-cpu page that is re-used
for all the packets handled on that same CPU. Each skb
holds a reference to that page and releases the page when
it is reclaimed. When the page gets full, it is released
and a new one is allocated.

Since any SKB that doesn't go through the fast-xmit path
of mac80211 will be segmented, we can assume here that the
packet is not WEP / TKIP and has a proper SNAP header.
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

The code that handles the TBs that contain the WiFi payload
will be changed for TSO. Move the current code into a
separate function.
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

Allow to configure the driver to pretend to have TX CSUM
offload support. This will be useful to test the TSO flows
that will come in further patches.
This configuration is disabled by default.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

ilw@linux.intel.com is not available anymore.
linuxwifi@intel.com should be used instead.
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

Host commands now have a group id, express this in printed messages.
Signed-off-by: NSharon Dvir <sharon.dvir@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

This allows the op_mode to request from the transport to
call a callback when an ASYNC commands is completed by
the firmware. The same callback will be called for all the
commands. Pass the command whose response triggers the
callback as a parameter to the callback itself.
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

In certain flows (see next patches), the op_mode may need to
block the Tx queues for a short period. Provide an API for
that. The transport is in charge of counting the number of
times the queues are blocked since the op_mode may block the
queues several times in a row before unblocking them.
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

Part of reorganising wireless drivers directory and Kconfig.
Signed-off-by: NKalle Valo <kvalo@codeaurora.org>

Allow frag SKBs in PCIe and advertise the maximum number of frags
to the opmode. As a fallback. linearize the SKB if it exceeds the
maximum number of fragments. This allows using the hardware better
(filling more TBs) and should improve performance when used by the
opmode.

Also adjust tracing to be able to deal with this.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

Existing UMAC commands already use the long header, but are sent
with group 0 and the long header inserted manually. Move them to
the group 1 to take advantage of the header building in the low-
level transport.

Existing firmware ignores the group_id field (it's reserved) and
the first firmware that really supports long command headers can
parse all commands in both group 0 (with short header) and group
1 (with long header.)
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

As the firmware is slowly running out of command IDs and grouping of
commands is desirable anyway, the firmware is extending the command
header from 4 bytes to 8 bytes to introduce a group (in place of the
former flags field, since that's always 0 on commands and thus can
be easily used to distinguish between the two.

In order to support this most easily in the driver widen the command
command ID used in the command sending functions and encode the new
values (group and version) in the ID. That way existing code doesn't
have to be changed (since the higher bits are 0 automatically) and
newer code can easily use the new ID generation function to create a
value to use in place of just the command ID.
Signed-off-by: NAviya Erenfeld <aviya.erenfeld@intel.com>
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

The 'flags' field really has been reserved in the firmware API for a
very long time, probably since 4965. As a consequence, the field is
always 0 and checking for a IWL_CMD_FAILED_MSK flag makes no sense.

Rename the field to 'reserved', get rid of IWL_CMD_FAILED_MSK and
all the code for it.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

In iwlmvm firmwares, the Byte count written in the scheduler
byte count table is in DWORDs and not in bytes.
We should check that this value fits in the 12 bits and
the value can be either in bits of in DWORD or bytes
depending on the firmware. Check the value after the
translation to DWORDs is done (if needed).
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

With the previous patch series, no opmode continues using the
command or handler_status (i.e. the return value from the RX)
so it can be removed now.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

During NIC initialization shared HW is reset and this disables the
scheduler. Some HW platforms do not activate the scheduler after it.
Consequently all HCMD sent by the driver stay at the queues which cause
to queue stuck.
Set the scheduler to work on auto active mode so it would be activated upon
change over one of the queues' write pointer.
Signed-off-by: NHaim Dreyfuss <haim.dreyfuss@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

The stuck queue detection mechanism allows to detect queues
that are stuck. For sleeping clients, a queue may rightfully
be stuck: if a poor client implementation stays asleep for
more than 10s, then we don't want to trigger recovery flows
because of that client.
In order to cope with this, I added a mechanism that
monitors the state of the client: when a client goes to
sleep, the timer of his queues is frozen. When he wakes up,
the timer is reset to the right value so that if a client
was awake for more than 10s and the queues are stuck, only
then, the recovery flow will kick in.
This is valid only on non-shared queues: A-MPDU queues.

There was a bug in case we Tx to a sleeping client that has
an empty A-MPDU queue: the timer was armed to now + 10s.
This is bad, but pretty harmless.
The problem is that when the client wakes up, the timer is
modified to be now + remainder. But remainder is 0 since the
queue was empty when that client went to sleep...

Fix this by checking the state of the client before playing
with the timer when we add a packet to an empty queue.
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

This check for family type is redundant as the if clause above
checks a family-dependent Boolean (which is not set for family 8000 anyway).
Signed-off-by: NAvri Altman <avri.altman@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

The cmd_in_flight tracking was introduced to workaround faulty
power management hardware, by having the driver keep the NIC
awake as long as there are commands in flight. However, some of
the code handling this workaround was unconditionally executed,
which resulted with an inconsistent state where the driver assumed
that the NIC was awake although it wasn't.

Fix this by renaming 'cmd_in_flight' to 'cmd_hold_nic_awake' and
handling the NIC requested awake state only for hardwares for
which the workaround is needed.
Signed-off-by: NIlan Peer <ilan.peer@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

This allows the op_mode to let the transport know that a
queue is currently frozen and that its timer should be
stopped.
When the queue is unfrozen, its timer should be set to
expire after the remainder of the timeout has elapsed.
This can be used when stations go to sleep. When a station
goes to sleep, the op_mode can freeze the timer so that the
queue will never be considered as stuck. When the station
wakes up, the queue will be unfrozen.
This is meant to avoid false positives that would happen if
a buggy station goes to sleep for a very long time. In case
we have a dedicated queue for this station (BA agreement)
and it goes to sleep for a very long time, the queue would
rightfully be stopped during all that time. In this case,
the stuck queue timer could fire and that would be a false
positive.
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

We don't need to acquire MAC access for each access, it
makes much more sense to keep the MAC access. This speeds
up the Tx DMA stop flow significantly.
Moreover, if one channel can't be stopped, stop the others
but don't poll for them to avoid being stuck there for a
long time.

This solves a situation in which we were stuck in that flow
for way too long with a spinlock held which led to a kernel
panic.
Reviewed-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

Different queue can have different behavior. While it can be
unacceptable for a certain queue to be stuck for 2 seconds
(e.g. the command queue), it can happen that another queue
will stay stuck for even longer (a queue servicing a power
saving client in GO).
The op_mode can even make the timeout be a function of the
listen interval.
Reviewed-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

This watchdog allows to monitor the transmit queues. When a
queue doesn't progress for a too long time, a timer fires
and then, debug data can be collected.
This watchdog has never been enabled on dvm controlled
devices, so don't enable it there.
In order to have it running on mvm controlled devices, we
need to fix a small issue in the transport layer: mvm
controlled devices use the shadow registers optimization.
In this case, the watchdog wasn't running at all, even if
enabled by the module parameter. Fix that on the way.
Reviewed-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

The base address of the scheduler in the device's memory
(SRAM) comes from two different sources. The periphery
register and the alive notification from the firmware.
We have a check in iwl_pcie_tx_start that ensures that
they are the same.
When we resume from WoWLAN, the firmware may have crashed
for whatever reason. In that case, the whole device may be
reset which means that the periphery register will hold a
meaningless value. When we come to compare
trans_pcie->scd_base_addr (which really holds the value we
had when we loaded the WoWLAN firmware upon suspend) and
the current value of the register, we don't see a match
unsurprisingly.
Trick the check to avoid a loud yet harmless WARN.
Note that when the WoWLAN has crashed, we will see that
in iwl_trans_pcie_d3_resume which will let the op_mode
know. Once the op_mode is informed that the WowLAN firmware
has crashed, it can't do much besides resetting the whole
device.

CC: <stable@vger.kernel.org>
Reviewed-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

Some devices have 31 TFD queues. Don't enable it yet since
there are still issues with it, but at least prepare the
code for it. There was a bug in the read pointer assignment,
fix that. Also, move the inline functions to iwl-scd.h which
is the right place.
Reviewed-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

Printing all the scratch data of the TFDs of that queue is
useless and stuffed the kernel log with data. Remove that.
Reviewed-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

A new host command can be used to configure the scheduler
instead of accessing the scheduler's registers from the
driver. This is easier and less error prone since accessing
the hardware at certain moments can lead to races with the
firmware.
Prefer to use the host command whenever it is available.
Reviewed-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

Implement the ref/unref trans ops and track both tx and
host command queues (and hold references while they
are not empty).
Signed-off-by: NEliad Peller <eliadx.peller@intel.com>
Reviewed-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

A following patche will use trans_pcie->cmd_in_flight
for reference accounting as well. get ready for it.
Signed-off-by: NEliad Peller <eliadx.peller@intel.com>
Reviewed-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

In some rare cases, the firmware can put the device to
sleep after the driver requested the access. This is
because the access request can take a short time to be
propagated to the firmware.

If that happens, the driver may think that it has access
since the firmware hasn't put the device to sleep yet, but
right after the driver's check, the firmware might put the
device to sleep.

Warn when this happens by allowing the firmware to finish
the "put the device sleep" flow so that the driver will
not get access to the device. This will make the issue
visible.

This still doesn't fix the race, but at least it makes
it more visible.
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

When freeing the structures used for command data, clear their
memory as they may have contained key material at some point.
Also clear the duplicated buffer when freeing it to be safe;
currently key material is never put there but that may change.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

This configuration is not needed for dvm, and it actually
broke it.
Reported-by: NOliver Hartkopp <socketcan@hartkopp.net>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

Our legal structure changed at some point (see wikipedia), but
we forgot to immediately switch over to the new copyright
notice.

For files that we have modified in the time since the change,
add the proper copyright notice now.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

Currently the firmware is handling this, but that is wrong as it then
needs to assume a certain command queue, therefore this should be in
the driver; add it here so it can be removed from the firmware in the
future.
Signed-off-by: NAvri Altman <avri.altman@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

Currently a valid sta_id is assumed to mean that the queue is
meant to also be aggregated, but that assumption will not be
true in the future, so don't make it in the lower level but
only in the inline wrapper.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

In a later patch, the hardware configuration will be moved to
firmware. Prepare for this by allowing hardware configuration
in the transport to be skipped by not passing a configuration
on enable and passing configure_scd=false on disable.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

Instead of having all arguments passed to the function,
add a struct to hold them and only pass some directly.

This will make future work in this area cleaner.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>

Configuring the hw scheduler during queue enablement is done by
writing the appropriate values to the scheduler peripherals, and
it is essentially the same for all buses.

Whenever writing is done via the standard iwl_write_prph, we can
avoid duplicating the code for each bus. Those operations are
queue deactivation, RA/TID mapping, chain-building settings,
enabling/disabling aggregations and activating/deactivating the
TX FIFOs.

Consolidate this code using static inlines in a new header file.
Signed-off-by: NAvri Altman <avri.altman@intel.com>
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NEmmanuel Grumbach <emmanuel.grumbach@intel.com>