It doesn't make much sense to overwrite send_cb
and recv_cb callbacks over and over again whenever
transport starts. Just make sure to unmask copy
engine interrupts when starting.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

It doesn't make sense to re-init irqs completely
whenever transport is started/stopped. Do it just
once upon probing/removing.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

Wrong register was being set up. This could
prevent firmware from booting in some rare cases
when using legacy interrupts.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

Firmware probing is done only once when driver is
registered and firmware version is guaranteed to
remain the same until driver is unregistered.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

Sometimes users forget to include important info like firmware version,
so better to print all the info.
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

Better to have a clear name for the function. While at it, clear up the title
for the register dump.
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

Store the firmware registers and other relevant data to a firmware crash dump
file and provide it to user-space via debugfs. Should help with figuring out
why the firmware crashed.

kvalo: remove dbglog support, rework and refactor the code to avoid ifdefs and
otherwise simplify it as well
Signed-off-by: NBen Greear <greearb@candelatech.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

ath10k_pci_diag_read32() is for reading u32 from a device and ath10k_pci_diag_read_hi()
is a helper for reading data using "host interest" table.
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

Service mapping for main firmware branch was incorrectly used for 10.x firmware
and vice-versa.  This caused wmi_services in debugfs to print wrong values.

This fixes commit cff990ce ("ath10k: fix wmi service bitmap debug") where
for some reason there was either a conflict that wasn't resolved properly or
git had a bad day.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

If spectral is unregistered after mac80211, the relayfs file has already
been removed recursively by mac/cfg80211, and spectral tries to remove
the file once more, thus leading to double free problems. Better clean
up spectral before to avoid that problem.
Reported-by: NKalle Valo <kvalo@qca.qualcomm.com>
Signed-off-by: NSimon Wunderlich <sw@simonwunderlich.de>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

tranmist -> transmit
Signed-off-by: NBen Greear <greearb@candelatech.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

Check vdev map has space before calling ffs,
fix invalid cleanup in failure to create vdev
case.

Open-code the BIT() logic since BIT does not properly
handle 64-bit bitfields and future patches will make
use of larger bitfields.
Signed-off-by: NBen Greear <greearb@candelatech.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

This is not necessary anymore. There are no more
uncontrolled htc tx entry points.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

Make probe/remove functions shorter and easier to
understand.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

The ATH10K_PCI_FEATURE_MSI_X was originally
introduced to support both chips QCA988Xv1 and
QCA988Xv2. Since v1 isn't supported anymore it
doesn't make sense to keep the feature flag
around. Since this is the last one remove the
whole thing.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

The soc powersave was disabled by default. It
never was fully tested. Some hw apparently had
problems with it and the implementation itself had
a possible race.

Just remove the refcounting and simply wake up the
device when probing and put to sleep when
removing.

kvalo: make ath10k_pci_wake() and _sleep() static
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

Use the common convention of embedding private
structures inside parent structures. This
reduces allocations and simplifies pci probing
code.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

This aims at fixing some rare scan bugs related to
firmware reporting unexpected scan event
sequences.

One such bug was if spectral scan phyerr reporting
prevented firmware from properly propagating scan
events to host. This led to scan timeout. After
that next scan would trigger scan completed event
first (before scan started event) leading to
ar->scan.in_progress and timeout timer states to
be overwritten incorrectly and making the very
next scan to hang forever.
Reported-by: NJanusz Dziedzic <janusz.dziedzic@tieto.com>
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

This also reduces the cruft of printing scan event
names in capitals.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

The 10.x and main firmware branches have
conflicting WMI service bitmap definitions.

This also fixes WMI services parsing on big-endian
hosts and changes debugfs output to be more human
friendly.

kvalo: remove braces and the last semicolon from SVCSTR()
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

Adds the spectral scan feature for ath10k. The spectral scan is triggered by
configuring a mode through a debugfs control file. Samples can be gathered via
another relay debugfs file.

Essentially, to try it out:

ip link set dev wlan0 up
echo background > /sys/kernel/debug/ieee80211/phy0/ath10k/spectral_scan_ctl
echo trigger > /sys/kernel/debug/ieee80211/phy0/ath10k/spectral_scan_ctl
iw dev wlan0 scan
echo disable > /sys/kernel/debug/ieee80211/phy0/ath10k/spectral_scan_ctl
cat /sys/kernel/debug/ieee80211/phy0/ath10k/spectral_scan0 > samples

This feature is still experimental. Based on the original RFC patch of
Sven Eckelmann.
Signed-off-by: NSimon Wunderlich <sw@simonwunderlich.de>
Signed-off-by: NMathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

In some cases during heavy tx vdev stop-start
would timeout on vdev synchronization causing
traffic to stall for a few seconds.

Instead of stop-starting use a dedicated vdev
restart command and down vdevs explicitly before
doing so.

This gets rid of the synchronization
warnings/timeouts and makes channel switching
smoother during traffic.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

Print sequence number, AMSDU_MORE flag and AC when additional
debug enabled in RX path. This is usefull for debugging purpose.
Signed-off-by: NJanusz Dziedzic <janusz.dziedzic@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

A-MSDU 4addr frames weren't reconstructed properly
and in some cases this resulted in a warning:

 br0: received packet on wlan0.sta1 with own address as source address

Since this was only related to A-MSDU it would
trigger when more intense traffic was generated.
Reported-by: NVu Hai NGUYEN <vh.nguyen@actiasodielec.fr>
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

The 10.2 firmware is a successor of 10.1 firmware
(formerly identified as 10.x). Both share a lot
but have some slight ABI differences that need to
be taken care of.

The 10.2 firmware introduces some new features but
those can be added in subsequent patches. This
patch makes ath10k boot and work with 10.2 with
comparable functionality to 10.1.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

Different firmware may support different numbers of
vdevs.  Use value that is always out of range for all
firmware.
Signed-off-by: NBen Greear <greearb@candelatech.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

In case of A-MSDU RX we should check attention flags
correctly to be sure we report correct FCS status for
A-MSDU subframes. Without a patch we could report A-MSDU
subframes with wrong FCS as a correct to the stack, next
get a lot of DUP ACK TCP packets. Finally TP drop is seen
and this drop depends on FCS errors ratio for A-MSDU frame.

Example test case when TP drop is seen:
- ath10k configured as an AP
- used ath10k station
- forced A-MSDU (7 frames) on STA
- other traffic on channel (often FCS errors)
- monitor iface added on AP
- TCP STA -> AP traffic (iperf)

a) Iperf logs for case without the patch:

echo "1 64" > htt_max_amsdu_ampdu // disable A-MSDU
[ ID] Interval       Transfer     Bandwidth
[  3]  0.0- 5.0 sec  56.6 MBytes  95.0 Mbits/sec
[  3]  5.0-10.0 sec  60.4 MBytes   101 Mbits/sec
[  3] 10.0-15.0 sec  60.2 MBytes   101 Mbits/sec
[  3] 15.0-20.0 sec  60.2 MBytes   101 Mbits/sec
[  3] 20.0-25.0 sec  63.8 MBytes   107 Mbits/sec
[  3] 25.0-30.0 sec  64.9 MBytes   109 Mbits/sec

echo "7 64" > htt_max_amsdu_ampdu  // set 7 A-MSDU subframes
[  3] 30.0-35.0 sec  40.0 MBytes  67.1 Mbits/sec
[  3] 35.0-40.0 sec  35.9 MBytes  60.2 Mbits/sec
[  3] 40.0-45.0 sec  36.9 MBytes  61.9 Mbits/sec
[  3] 45.0-50.0 sec  37.9 MBytes  63.5 Mbits/sec
[  3] 50.0-55.0 sec  34.5 MBytes  57.9 Mbits/sec
[  3] 55.0-60.0 sec  25.4 MBytes  42.6 Mbits/sec
[  3] 60.0-65.0 sec  48.2 MBytes  81.0 Mbits/sec
[  3] 65.0-70.0 sec  28.8 MBytes  48.2 Mbits/sec
[  3] 70.0-75.0 sec  29.2 MBytes  49.1 Mbits/sec
[  3] 75.0-80.0 sec  22.9 MBytes  38.4 Mbits/sec
[  3] 80.0-85.0 sec  26.4 MBytes  44.2 Mbits/sec
[  3] 85.0-90.0 sec  31.5 MBytes  52.8 Mbits/sec

b) Iperf logs for case with patch:

echo "1 64" > htt_max_amsdu_ampdu // disable A-MSDU
[  3] local 192.168.12.2 port 57512 connected with 192.168.12.1 port 5001
[ ID] Interval       Transfer     Bandwidth
[  3]  0.0- 5.0 sec  60.8 MBytes   102 Mbits/sec
[  3]  5.0-10.0 sec  62.2 MBytes   104 Mbits/sec
[  3] 10.0-15.0 sec  60.9 MBytes   102 Mbits/sec

echo "7 64" > htt_max_amsdu_ampdu  // set 7 A-MSDU subframes
[  3] 15.0-20.0 sec  68.1 MBytes   114 Mbits/sec
[  3] 20.0-25.0 sec  80.5 MBytes   135 Mbits/sec
[  3] 25.0-30.0 sec  83.0 MBytes   139 Mbits/sec
[  3] 30.0-35.0 sec  79.1 MBytes   133 Mbits/sec
[  3] 35.0-40.0 sec  77.1 MBytes   129 Mbits/sec
[  3] 40.0-45.0 sec  77.4 MBytes   130 Mbits/sec
Reported-by: NDenton Gentry <denton.gentry@gmail.com>
Signed-off-by: NJanusz Dziedzic <janusz.dziedzic@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

The 10.x firmware does not support IBSS mode at
all. It can't beacon and it crashes when trying to
scan.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

Firmware doesn't perform Rx reordering so it is
left to the host driver to do that.

Use mac80211 to perform reordering instead of
re-inventing the wheel.

This fixes TCP throughput issues in some
environments.
Reported-by: NDenton Gentry <denton.gentry@gmail.com>
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

It was possible to enter an endless loop while
processing a single pci copy engine pipe. This
could effectively render ath10k incapable of
responding to any requests.

An example case when this could happen is when
firmware generates a lot of events, e.g. spectral
scan phyerr via WMI.
Reported-by: NJanusz Dziedzic <janusz.dziedzic@tieto.com>
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

Apparently fw/hw generates a corrupted QoS Control
Field in Qos NullFunc frames. The only way to
workaround this is to downgrade frames to
NullFunc. This should be okay since powersave is
done by fw/hw and these frames are only used for
CQM purposes (e.g. from hostapd to check if
station is still connected).

This doesn't fix any user visible bug that I know
of. It just prevents from sending out funky frames
on the air.
Reported-by: NJanusz Dziedzic <janusz.dziedzic@tieto.com>
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

It always bugged me how tid is computed and stored
in a temporary var before written to the control
buffer. It was confusing and it made it difficult
to work with tx helpers.

While at it rename the qos workaround function as
it was misleading - it's not a workaround but
preparation for nwifi tx mode.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

Firmware could request inspection of some
submitted tx requests. Since the callback wasn't
implemented it was possible to bleed tx msdu_ids
which could translate to tx flushing timeouts.

There's nothing ath10k can do to help firmware
with tx processing now so just report all tx
frames as already inspected to prevent firmware
from sending up inspection events and force it to
report regular tx completion indications with
discard status.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

Apparently iomap writes that unmask CE irqs aren't
propagated properly sometimes. Before failing try
to poll for the control response message as it may
have been delivered without an interrupt.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

The tx ring index was immediately trimmed with a
bitmask. This discarded the 0xFFFFFFFF error case
(which theoretically can happen when a device is
abruptly disconnected) and led to using an invalid
tx ring index. This could lead to memory
corruption.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

It was possible for tx completion not to be
processed. In that case an old stack pointer was
left on copy engine tx ring. Next bmi exchange
would immediately pop it and use complete() on the
completion struct there causing corruption.

Make sure to wait for both tx and rx completions
properly.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

If firmware probing worker failed it called
device_release_driver() which synchronously called
remove() pci callback. The callback in turn waited
for the worker that called it to finish resulting
in a deadlock.

Waiting for a completion instead of a worker, like
some other drivers do, doesn't seem like the best
idea either:

  Syscall                 Worker

                          probe_fw()
  rmmod
  dev_lock()
  pci->remove()
  wait_for_completion()
                          complete_all()
                          device_release_driver()
                          dev_lock()
                          [sleep]
  free(ar)
  dev_unlock()
                          [resume]

There's no guarantee that Worker upon resuming can
still access any data/code of the module.

Leaving device bound to a driver is not as harmful
as deadlocking so remove the call to
device_release_driver() while a proper solution is
figured out.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

Allow to setup maximum subframes for AMSDU and AMPDU aggregation
via debugfs htt_max_amsdu_ampdu file.

Eg.
echo "2 64" > htt_max_amsdu_ampdu
will setup maximum amsdu subframes equal 2 and
maximum ampdu subframes equal to 64.
Signed-off-by: NJanusz Dziedzic <janusz.dziedzic@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

Some drivers (such as iwlmvm) can handle multiple bands in a single
HW scan request. Add a HW flag to indicate that the driver support
this. To hold the required data, create a separate structure for
HW scan request that holds cfg scan request and data about
different parts of the scan IEs.

As this changes the mac80211 API, update all drivers using it to
use the correct new function type/argument.
Signed-off-by: NDavid Spinadel <david.spinadel@intel.com>
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>

Firmware 10.x supports up to 8 virtual AP interfaces, but in a DFS
channel it was possible to create only 7 interfaces as ath10k internal
creates a monitor interface for DFS. Previous vdev map initialization
was missing enough space for 8 + 1 vdevs due to wrong define used and
that's why there was no space for 8th interface. Use the correct define
TARGET_10X_NUM_VDEVS with 10.x firmware to make it possible to create
the 8th virtual interface.
Signed-off-by: NBartosz Markowski <bartosz.markowski@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>