In order to verify d0i3 flow, add debug print to indicate
d0i3 exit was completed (right after tx was re-enabled),
along with the wakeup reasons.

Signed-off-by: Eliad Peller <eliadx.peller@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>

Recent fw version added assert to make sure wowlan configuration
is configured only when a station is connected.

Change the driver behavior to pass this configuration only
if we indeed have ap station id (i.e. connected).

Signed-off-by: Eliad Peller <eliadx.peller@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>

The RX API is currently mixed up into the general fw-api.h
file, but we're going to need to extend it significantly in
the future, so move it to its own file.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>

Remove some unused values from the RX API; these were used
with older firmware API that didn't have the RX energy API,
support for which was removed a long time ago.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>

Since the driver can never configure the data here, this field
will always be reported as 0 by the firmware. Even if this was
not the case, however, it wouldn't matter since the extra data
would be added beyond the end of the phy_info structure we use
in the driver, so wouldn't harm anything in this code either.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>

This message is useless - it's in the good case that always
happens so enabling it doesn't really help. Just remove it.
There are other ways to debug this (e.g. tracing) so there's
no need to add a message in the bad case.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>

It's possible for an AP interface to be UP but not actually
operating (i.e. not beaconing etc.) - in this case it can't
actually do ToF, so check for it.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>

This flag is set in all supported firmwares.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Luciano Coelho <luciano.coelho@intel.com>

This flag is set in all supported firmwares.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Luciano Coelho <luciano.coelho@intel.com>

All the supported firmwares have this flag set.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Luciano Coelho <luciano.coelho@intel.com>

All the supported firmwares use the new API.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Luciano Coelho <luciano.coelho@intel.com>

All the supported firwmares support the new API.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Luciano Coelho <luciano.coelho@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>

We don't support firmwares that don't use the new API.
This also allows to use all the SCD queues, so increase
the reported number of queues to 31.

Signed-off-by: Liad Kaufman <liad.kaufman@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Luciano Coelho <luciano.coelho@intel.com>

-13.ucode has been published long ago. Deprecate old
versions of the firmware.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Luciano Coelho <luciano.coelho@intel.com>

There's obviously no reason to write a NUL-terminating byte into
the debugfs file. Apparently nobody tried to use this from a tool
that doesn't strip NUL bytes though, so we never noticed. Fix it.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luciano Coelho <luciano.coelho@intel.com>

If we are in d0i3 when entering suspend, we leave d0i3 so that
mac80211 can call us to remove connections or whatever before going to
suspend.  We do this by calling pm_runtime_resume() early in the slave
transport flow and reactivating it later, when the wiphy suspend flow
runs.

The problem is that we queue a work in order to leave d0i3.  If this
work hasn't run yet when the wiphy suspend flow is called, we have a
race and entering d0i3 fails (because we're still holding the
IWL_MVM_REF_EXIT_WORK reference).

To solve this, simply flush the d0i3_exit_work at the beginning of the
iwl_mvm_suspend() function.

Signed-off-by: Luciano Coelho <luciano.coelho@intel.com>

When in AP mode, we need beacons from other APs for HT protection.
However, when there's no any associated station we will not do
any Tx and thus don't really need beacons. On the other hand, these
beacons will cause a lot of unnecessary wakeups which increase our
power consumption. Handle this by asking FW to pass beacons only when
there's at least one associated station.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Signed-off-by: Luciano Coelho <luciano.coelho@intel.com>

With CONFIG_IWLWIFI_DEVICE_TRACING=y, these functions are rather large,
too big for inlining.

With this .config: http://busybox.net/~vda/kernel_config

,
after uninlining these functions have sizes and callsite counts
as follows:

iwl_read32  475 bytes, 51 callsites
iwl_write32 477 bytes, 90 callsites
iwl_write8  493 bytes, 3 callsites

Reduction in size is about 74,000 bytes:

    text     data      bss       dec     hex filename
90758147 17226024 36659200 144643371 89f152b vmlinux0
90687995 17221928 36659200 144569123 89df323 vmlinux.after

Reported-by: Denys Vlasenko <dvlasenk@redhat.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Luciano Coelho <luciano.coelho@intel.com>

ECHO_CMD is a simple command that can be sent to the
firmware just to check that it is alive.
This command is useful for debug purpose, provide a debugfs
hook to send it.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Luciano Coelho <luciano.coelho@intel.com>

When receiving statistics notification there is a field of average energy.
This is defines as signed 8 bit, while FW refers to it as unsigned.
when the energy is higher than 127 this causes in iwl_mvm_stat_iterator
a wrong computation of the signal
     int sig = -stats->general.beacon_filter_average_energy;
resulting in incorrect CQM event (change from low to high).

While at it - update the rest of the fields to the correct type.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Luciano Coelho <luciano.coelho@intel.com>

This debugfs entry was implemented just as a proof-of-concept before
the full support for netdetect via cfg80211/mac80211 was implemented.
Now that we have a proper way to enable netdetect from userspace, this
entry is redundant and should be removed to avoid unnecessary
maintenance work.

Signed-off-by: Luciano Coelho <luciano.coelho@intel.com>

The driver is now able to handle -17.ucode.

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

Add support for the new (bigger) TX power command. This doesn't
actually take advantage of the new capabilities (to set per-chain
TX power limits) but makes the code compatible with newer firmware
images expecting the larger command.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

split long debug messages that may result warning in tracing.

Fixes: 1a84e771

 ("iwlwifi: mvm: add debug info to schedule scan complete message.")
Signed-off-by: Ayala Beker <ayala.beker@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

The command string array is assumed to be indexable by a full u8, so it
must have 256 entries (0-255), not just 255. A recent firmware change
(apparently) started using the command 0xff for a notification to the
host, causing the driver to crash in debug message/tracing code.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

when waking from d0i3 there is a wakeup reason.
Enum in driver is not up to date with FW api - fix it.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

My static checker complains that we don't check for underflows in
iwl_dbgfs_fw_dbg_conf_write().  This is harmless because we have a
sanity check in iwl_mvm_start_fw_dbg_conf(), but we may as well make
this unsigned and silence the underflow warning.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

When we enter D0i3, we must stop TXing otherwise the
sequence number we use might conflict with the firmware's
internal TX. In order to do so, we have
IWL_MVM_STATUS_IN_D0I3 which should prevent any Tx while we
enter D0i3. There is a bug in this code since we may Tx even
if IWL_MVM_STATUS_IN_D0I3 is set. This can happen as long as
mvm->d0i3_ap_sta_id is not set.

To make sure that we don't have any packet in the Tx path
while we set mvm->d0i3_ap_sta_id, call synchronize_net only
after we already set mvm->d0i3_ap_sta_id.

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

Currently if we wake up during D0I3 due to beacon loss we disconnect
immediately. This behaviour causes redundant disconnection, which could
be prevented by polling as it is usually done in mac80211.
Instead, we prefer reporting beacon loss and let mac80211 try polling
before disconnection.

Signed-off-by: David Spinadel <david.spinadel@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

KASan error report:
==================================================================
BUG: KASan: out of bounds access in iwl_init_sband_channels+0x207/0x260 [iwlwifi] at addr ffff8800c2d0aac8
Read of size 4 by task modprobe/329
==================================================================

Both loops of this function compare data from the 'chan' array and then
check if the index is valid.

The 2 conditions should be inverted to avoid an out-of-bounds access.

Signed-off-by: Adrien Schildknecht <adrien+dev@schischi.me>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

The driver is now able to handle -16.ucode.

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

Fix bug where MIMO is disabled for low latency TX on P2P VIF
regardless of configuration. Make it dependent on
IWL_MVM_RS_DISABLE_P2P_MIMO compilation option. Change configuration
so that MIMO will be disabled only in SDIO platforms.

Signed-off-by: Alexander Bondar <alexander.bondar@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

Currently the driver disable power management on all low-latency
interfaces, while it should disable it on WiDi interfaces only.
Non-P2P interfaces that runs voice and video traffic should enable
power management.

Signed-off-by: Avri Altman <avri.altman@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

When TDLS support is declared by the FW, set the bit indicating wider-BW
support as well.

Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

The &mvm->tof_data.range_req.ap[] array has IWL_MVM_TOF_MAX_APS elements
so the check should be >= instead of >.  Also the test can underflow so
I have changed "i" to unsigned.

Fixes: ce792918

 ('wlwifi: mvm: add basic Time of Flight (802.11mc FTM) support')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

There is no need to use hex_dump_to_buffer() in the cases like this:

	hexdump_to_buffer(buf, len, 16, 1, outbuf, outlen, false);	/* len <= 16 */
	sprintf("%s\n", outbuf);

since it maybe easily converted to simple:

	sprintf("%*ph\n", len, buf);

Note: it seems in one case the output is groupped by 2 bytes and looks like a
typo. Thus, patch changes that to plain byte stream.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

The mac80211 patch to pass the NAPI struct only changed iwlwifi to
store the NAPI struct, but we can do better: pass it directly from
the lower transport layer to the opmode during RX, and then on to
mac80211 from there.

When we add multiple RX queues, we can then pass the appropriate
NAPI struct properly.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

Another pull request for the next cycle, this time with quite
a bit of content:
 * mesh fixes/improvements from Alexis, Bob, Chun-Yeow and Jesse
 * TDLS higher bandwidth support (Arik)
 * OCB fixes from Bertold Van den Bergh
 * suspend/resume fixes from Eliad
 * dynamic SMPS support for minstrel-HT (Krishna Chaitanya)
 * VHT bitrate mask support (Lorenzo Bianconi)
 * better regulatory support for 5/10 MHz channels (Matthias May)
 * basic support for MU-MIMO to avoid the multi-vif issue (Sara Sharon)
along with a number of other cleanups.

Add more information to schedule scan complete message.

Signed-off-by: Ayala Beker <ayala.beker@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

Lmac sched scan supports partial scans, so we can set some channels to be
scanned on every scan iteration and others to be scanned only on some
iterations. Currently we set all channels to be scanned every iteration,
but still have some configuration of which iterations should be partial
and which should be full.

Remove all partial/full scan configuration to reduce confusions.

Signed-off-by: David Spinadel <david.spinadel@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>