It was obseverd many times the short queue limit can be
easily exceeded in case of AP multiple BSSID scenarios.

"ath10k: wmi mgmt_tx queue limit reached."

This was leading to TX packets drops.

This patch expands the queue limit to 128. After many stress
experiments (high traffic and lot of management frames in
the air) the new value seems to fix the problem on platforms
with limited resources.
Reported-By: Tomasz Skapski <tomasz.skapski@tieto.com>
Signed-off-by: Bartosz Markowski <bartosz.markowski@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

With commit 0cfcefef ("mac80211: support reporting A-MSDU subframes
individually") there's no need to have the hack to clear the retry bit in
ath10k_htt_rx_amsdu(), mac80211 can handle this properly now.
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

Frames were never retransmitted with
different/lower bandwidths, e.g. only 80MHz
bandwidth was used when dealing with a VHT80
peer.

Allow HW rate control to try out different
bandwidths when retransmitting. This increases
robustness.

Also, document the parameter properly.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

HIF was not stopped properly in
ath10k_core_start() upon failure. This could cause
memory leaks of CE completions entries and
possibly other issues as well.

Move the HIF start/stop out of
ath10k_htc_wait_target(). The ctl_resp completion
is already prepared in ath10k_htc_init.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

This should make sure the device won't issue any
interrupts nor access any memory after the driver
is stopped/freed thus avoid memory corruption in
some cases.
Reported-By: Ben Greear <greearb@candelatech.com>
Reported-By: Janusz Dziedzic <janusz.dziedzic@tieto.com>
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

Add missing error reporting and adjust other
prints to make everything more consistent.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

This patch moves irq registering after necessary
structures have been allocated and initialized.
This should prevent interrupts from causing
tasklet access invalid memory pointers.
Reported-By: Ben Greear <greearb@candelatech.com>
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

In case firmware crashes it may report CE
completions for entries that were never
submitted/filled with meaningful data. This in
turn led to NULL dereferences.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

This shouldn't be silenced. This will be necessary
for PCI init code reordering.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

This removes some remaining direct use of the wake
register which could interfere with power state
tracking of the target device. This will allow
initialization code reordering.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

The check doesn't make much sense. If the address
were to be 0x0000 the check would fail. In this
case a 0 address isn't wrong.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

Make sure to put target back to sleep. This was a
minor issue as it didn't really matter if we put
target back to sleep at this point. It just looked
wrong.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

CE error interrupts were not disabled. This could
lead to invalid memory accesses / memory
corruption.

Also make sure CE watermark interrupts are also
disabled.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

What the function does is to actually wait for the
firmware indication bit to be set. Prerequisite
for this is having interrupts registered.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

The function will soon be called from more than 1
place.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

It was possible for FW error tasklet to be
executed during teardown. This could lead to
system crashes and/or memory corruption.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

It wasn't really useful to have it to begin with.
This makes it a little simpler to re-arrange PCI
init code as some function depended on
ar_pci->ce_count being set.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

Not all errors were properly notified, fix that.

kvalo: use ath10k_warn() for all messages, cosmetic changes to some of the messages
Signed-off-by: Ben Greear <greearb@candelatech.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

If the 'ar' registration fails early, the debug
objects were not initialized, and trouble ensued when
the object was later destroyed and the un-initialized
objects were accessed.

Per Kalle's suggestion, move the debug_destroy to the
core_unregister method.
Signed-off-by: Ben Greear <greearb@candelatech.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

memory is malloced in ath10k_pci_probe() and should be freed before
leaving from the error handling cases, otherwise it will cause
memory leak.
Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

If beacon tx command failed the wmi command buffer
was not freed. This led to OOM in the long run.
This became aparent when stress testing
multi-BSSID.
Reported-By: Tomasz Skapski <tomasz.skapski@tieto.com>
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

sk_buff was not freed in some cases. The patch
unifies the msdu freeing.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

This allows txpower limits other than regulatory
to be applied (e.g. user txpower limit provided
via iw/nl80211).
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

It was previously thought that FW expects tx power
in quarters of dBm.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

This will be needed for proper support of (user)
configurable tx power level.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

If the creation fails, this keeps us from crashing later when
trying to do a list_del(arvif->list).
Signed-off-by: Ben Greear <greearb@candelatech.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

ath10k_pci_wait() didn't notify any errors to callers, it
just printed a warning so add proper error handling. This fixes
a crash Ben reported:

ath10k: MSI-X interrupt handling (8 intrs)
ath10k: Unable to wakeup target
ath10k: target took longer 5000 us to wake up (awake count 1)
ath10k: Failed to get pcie state addr: -16
ath10k: early firmware event indicated
BUG: unable to handle kernel NULL pointer dereference at 0000000000000004
IP: [<ffffffffa06ae46c>] ath10k_ce_completed_send_next+0x47/0x122 [ath10k_pci]
Reported-by: Ben Greear <greearb@candelatech.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

10.1.389 firmware has some differences in
calculation of number of outstanding HTT TX
completions. This led to FW crashes of 10.1.389
while main firmware branch was unnaffected.

The patch makes sure ath10k doesn't queue up more
MSDUs than it should.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

To make it easier to access SOC registers. No functional
changes.
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

If firmware crashes during FW probing it would try
to perform FW recovery which uses mac80211
workqueue before registering to mac80211.

Using internal workqueue solves the problem.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

NSTS values reported in the VHT-SIG-A1 are 0
through 7 but they actually describe number of
streams 1 through 8.

1SS frames were dropped. This patch fixes this.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

This fixes incorrect values being printed on
big-endian hosts.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

Add a few checks and warnings to make it easier to
track any kind of monitor vdev mismanagement.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

This fixes issue with interface bridging.

Linux bridging sets promiscuous mode for all
interfaces that are in a bridge. This translates
to configure_filter() being called in a mac80211
driver.

Before the patch operational interface would be
started and upped again when promiscuous mode was
enabled causing all sorts of strange issues:

 * no HTT RX happening (i.e. no traffic)
 * FW crash upon driver reload/unload
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

Use the module_pci_driver() macro to make the code simpler
by eliminating module_init and module_exit calls.
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

The PCI and SoC specific drivers are using separate
code now so it is not reasonable to use the same
module for both drivers anymore.
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

The 'rt2800pci_hwcrypt_disabled' function is the
only PCI specific callback which is used by the
SoC driver. Create a clone of that to get rid of
the dependency.

Even though the two functions are using the same
variable, but the SoC specific code will be moved
into a separate module which will have its own
'modparam_nohwcrypt' variable.
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

Move the function into the rt2800mmio module, in order
to make it usable from other modules.
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

The function is called for PCI and SoC devices
however the MCU related part of the function
has no effect on SoC devices. Move the common
part of the function into a separate helper and
use that for the SoC devices.
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

The function contain code for SoC devices only.
Rename the function to 'rt2800soc_disable_radio'
and move it to the SoC specific section. Use
the renamed function in the SoC specific code
only and remove the 'if rt2x00_is_soc(rt2x00dev)'
condition from the function body.
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>