Provide detailed statistics for the Rx frames per MCS
Statistics printed in "stations" debugfs entry
Signed-off-by: NVladimir Kondratiev <qca_vkondrat@qca.qualcomm.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

Firmware defines new memory region, mac_rgf_ext
that need to be accessed from the host for debug purposes.

Add corresponded mapping
Signed-off-by: NVladimir Kondratiev <qca_vkondrat@qca.qualcomm.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

Control Bcast ring size in similar way as Rx and Tx ones,
through "bcast_ring_order" modparam, actual ring size is 1 << order
Signed-off-by: NVladimir Kondratiev <qca_vkondrat@qca.qualcomm.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

When one client is associated and connected to an ar6003 hw version
2.0 with firmware 3.1.1.149, and another client tries to connect, the
first client's MAC address is lost in the station list because the
"aid" is always "1".  The structure "wmi_connect_event" has the "aid"
as the second byte in the message, but it should be the first byte.

This patch has been tested with linux-3.10.40
Signed-off-by: NBrent Taylor <motobud@gmail.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

Found during code review. This was pretty much
impossible to happen but better safe than sorry.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

In the midst of chanctx patch review channel
switching became broken which I failed to notice
until now.

Function ath10k_mac_vif_chan() reports current
chandef which isn't updated until after
switch_vif_chanctx() is returned from.
Consequently the driver just restarted operation
on channels it was residing already instead of
switching to the new ones.

Fixes: 500ff9f9 ("ath10k: implement chanctx API")
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

In practice there's no point in having a copy of
chanctx_conf.

Most of the time the channel pointer (and band
along with it) is accessed and this can't change
after a chanctx is created because switching is
done using explicit chanctx swapping via
switch_vif_chanctx().

The only thing that can change within a
chanctx_conf and is used by the driver is
radar_enabled and channel width. These are however
always accessed in adequate mac80211 callback
context which guarantees safe access to the
chanctx data.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

Channel contexts aren't iterable until after
they've been added to the driver. The code assumed
otherwise.

This problem could result in:

 * rx_channel being NULL and forcing Rx path to go
   the slow way to get channel on QCA988X,

 * report incorrect channel when running
   multi-channel on QCA61X4 hw2.1,

 * report incorrect channel after AP channel
   switch.

Fixes: 500ff9f9 ("ath10k: implement chanctx API")
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

mete_data is extracted from ce descriptor and stored in variable 'id'.
later, id is not used anywhere in the same function.

Fixes: d84a512d ("ath10k: remove transfer_id from ath10k_hif_cb::tx_completion")
Signed-off-by: NRaja Mani <rmani@qti.qualcomm.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

When wmi mgmt event function fails to parse given skb,
it should be freed on failure condition to avoid memory
leaks. Found this during the code review.
Signed-off-by: NRaja Mani <rmani@qti.qualcomm.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

If probing failed pci sleep timer could remain
running and trigger after ath10k structures were
freed causing invalid pointer dereference:

 BUG: unable to handle kernel paging request at ffffc90001c80004
 IP: [<ffffffff81354728>] iowrite32+0x38/0x40
 ...
 Call Trace:
  <IRQ>
  [<ffffffffa00da048>] ? __ath10k_pci_sleep+0x48/0x60 [ath10k_pci]
  [<ffffffffa00da44e>] ath10k_pci_ps_timer+0x5e/0x80 [ath10k_pci]
  [<ffffffff810b210e>] call_timer_fn+0x3e/0x120
  [<ffffffffa00da3f0>] ? ath10k_pci_wake+0x150/0x150 [ath10k_pci]
  [<ffffffff810b3d11>] run_timer_softirq+0x201/0x2e0
  [<ffffffff8105d73f>] __do_softirq+0xaf/0x290
  [<ffffffff8105da95>] irq_exit+0x95/0xa0
  [<ffffffff81950406>] smp_apic_timer_interrupt+0x46/0x60
  [<ffffffff8194e77e>] apic_timer_interrupt+0x6e/0x80

Fixes: 77258d40 ("ath10k: enable pci soc powersaving")
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

Some of of qca988x solutions are having global reset issue
during target initialization. Bypassing PLL setting before
downloading firmware and letting the SoC run on REF_CLK is fixing
the problem. Corresponding firmware change is also needed to set
the clock source once the target is initialized. Since 10.2.4
firmware is having this ROM patch, applying skip_clock_init only
for 10.2.4 firmware versions.
Signed-off-by: NRajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

This could lead userspace initram images getting
built without necessary firmware files included
leading to probing failures of ath10k on boot with
QCA61X4.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

In some cases some channel survey data was
reported incorrect.

Channel info events were expected to come in pairs
without and with COMPLETE flag set respectively
for each channel visit during scan.

The known deviation from this is rule for last
scan chan info and first (next) scan chan info
both have COMPLETE flag set. This was either
programmed with the intent of providing BSS cycle
count info or this is an artefact of firmware scan
state machine. Either way this is useless due to
short wraparound time, wraparound quirks and no
overflow notification.

Survey dumps now include only data gathered during
scan channel visits that can be computed
correctly.

This should improve hostapd ACS a little bit.
Reported-by: NSrinivasa Duvvuri <sduvvuri@chromium.org>
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

When QCA988X cycle counter HW register wraps
around it resets to 0x7fffffff instead of 0. All
other cycle counter related registers are divided
by 2 so they never wraparound themselves. QCA61X4
has a uniform CC and it wraparounds in a regular
fashion though.

Worst case wraparound time is approx 24 seconds
(2**31 / 88MHz). Since scan channel visit times
are max 5 seconds (offchannel case) it is
guaranteed there's been at most 1 wraparound and
it is possible to compute survey active time
value. It is, however, impossible to determine the
point at which Rx Clear Count has been divided by
two so it is not reported upon wraparound.

This fixes some occasional incorrect survey data
on QCA988X as some channels (depending on how/when
scan/offchannel requests were requested) would
have approx 24 sec active time which wasn't
actually the case.

This should improve hostapd ACS a little bit.
Reported-by: NSrinivasa Duvvuri <sduvvuri@chromium.org>
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

The macro isn't WMI specific. Instead it is
related to hardware chip so move the macro
accordingly. While at it document the magic value.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

Add support for flashing 10GBaseT adapter with BCM 84834 PHY and
Aquantia AQ1202 PHY.

Updating of the PHY firmware must happen before the INITIALIZE_CMD.
Signed-off-by: NHariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Change more netdev_dbg statements over to netif_dbg and add some new
netif_dbg statements to the driver.
Signed-off-by: NTom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When the ethtool command is used to set the speed of the device while
the device is down, the check to set the initial mode may fail when
the device is brought up, causing failure to bring the device up.

Update the code to set the initial mode based on the desired speed if
auto-negotiation is disabled.
Signed-off-by: NTom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The device private data structure contains all the defined hardware
features for the device. However one of the features is not set. Even
though the feature is not currently used, set it to avoid future
issues of the feature being checked thinking it has been properly set.
Signed-off-by: NTom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Remove the limitation that disallows configuring single ported VFs
in the presence of IB ports, after addressing the issues that
prevented that to work.

SMI (QP0) requests/responses are still not supported for single
ported IB VFs.
Signed-off-by: NOr Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: NJack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

It's legal for drivers to provide the QP port through the
QPC schedule-queue field on the reset-to-init QP state change.

Add adjusting of the schedule queue port in the SRIOV wrapper
for that operation too.
Signed-off-by: NOr Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: NJack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Some VF drivers flow set the schedule queue in the QP context but
without setting none of OPTPAR_SCHED_QUEUE or OPTPAR_PRIMARY_ADDR_PATH.

To allow for such non-modified drivers to function as single ported
IB VFs, we must adjust the schedule queue port whenever being set,
e.g as currently done for single ported Eth VFs.
Signed-off-by: NOr Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: NJack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

As part of enabling single ported VFs over IB ports we need to handle
some of the flows for generting EQ events for VFs which don't come
into play under Eth ports.

This mainly includes port management events derived from changes of the
phyiscal port (lid change, client re-register, down/up, etc), VF pkey table
changes and VF guid changes initiated by the IB driver.

(1) make sure that events are generated only for VFs sitting on
    the relevant physical port (under the ALL_SLAVES flow).

(2) before generating the event, convert from physical (one or two)
    to VF port (always equals one).
Signed-off-by: NOr Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: NJack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When multiplexling a MAD sent from VF, we should convert the port used
by the guest to send the packet to the actual physical port which will be
used to transmit the packet, before building the relevant address-handle (AH).

This is needed under VPI for single ported VFs, since the code that builds
the AH (mlx4_ib_query_ah()) makes decisions based on the input port. If we
use the port number provided by the guest, it might have different protocol
vs. the one this packat has to go from, and hence the result could be wrong.

So far, the conversion was done after the AH was built and it worked for
single ported Eth VFs which were not enabled under VPI. When adding support
for single ported IB VFs and VPI, we hit that.

Fixes: 449fc488 ('net/mlx4: Adapt code for N-Port VF')
Signed-off-by: NOr Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Single port VFs always provide port = 1 (even if the actual physical
port used is port 2). As such, we need to convert the port provided
by the VF to the physical port before calling into the firmware.

It turns out that the Linux mlx4 VF RoCE driver maintains a copy of
the GID table and hence this change became critical only for single
ported IB VFs, but it could be needed for other RoCE VF drivers too.

Fixes: 449fc488 ('net/mlx4: Adapt code for N-Port VF')
Signed-off-by: NOr Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: NJack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NGeert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

According to the Zynq TRM, gigabit half duplex is not supported.  Add a
new cap and compatible string so Zynq can avoid advertising that mode.
Signed-off-by: NNathan Sullivan <nathan.sullivan@ni.com>
Acked-by: NNicolas Ferre <nicolas.ferre@atmel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The tx_curr_frame_payload field is u32. When we try to calculate a
small negative delta based on it, we end up with a positive integer
close to 2^32 instead.  So the tx_bytes pointer increases by about
2^32 for every transmitted frame.

Fix by calculating the delta as a signed long.

Cc: Ben Hutchings <ben.hutchings@codethink.co.uk>
Reported-by: NFlorian Bruhin <me@the-compiler.org>
Fixes: 7a1e890e ("usbnet: Fix tx_bytes statistic running backward in cdc_ncm")
Signed-off-by: NBjørn Mork <bjorn@mork.no>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Some firmware revisions pad 4th hw address in
Native Wifi Rx decap. This is the case with 10.x
and it was assumed that this is true for all
firmware images.

However QCA988X with 999.999.0.636 and QCA61X4
with WLAN.RM.2.0-00088 don't have the padding.
Hence add a feature flag indicating that the
padding isn't present so firmware images can
advertise it appropriately. This way driver will
behave as it was before with old firmware blobs
and doesn't cause any regressions from user
perspective.

Effectively this patch enables QCA988X with
999.999.0.636 and QCA61X4 with WLAN.RM.2.0-00088
to set up client bridging provided user has an
updated firmware blob.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

By using SOC_WAKE register it is possible to bring
down power consumption of QCA61X4 from 36mA to
16mA when associated and idle.

Currently the sleep threshold/grace period is at a
very conservative value of 60ms.

Contrary to QCA61X4 the QCA988X firmware doesn't
have Rx/beacon filtering available for client mode
and SWBA events are used for beaconing in AP/IBSS
so the SoC needs to be woken up at least every
~100ms in most cases. This means that QCA988X
is at a disadvantage and the power consumption
won't drop as much as for QCA61X4.

Due to putting irq-safe spinlocks on every MMIO
read/write it is expected this can cause a little
performance regression on some systems. I haven't
done any thorough measurements but some of my
tests don't show any extreme degradation.

The patch removes some explicit pci_wake calls
that were added in 320e14b8db51aa ("ath10k: fix
some pci wake/sleep issues"). This is safe because
all MMIO accesses are now wrapped and the device
is woken up automatically if necessary.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

mac80211 can update vif powersave state while
disconnected. Firmware doesn't behave nicely and
consumes more power than necessary if PS is
disabled on a non-started vdev. Hence
force-enable PS for non-running vdevs.

This reduces power drain on QCA61X4 from 88mA to
36mA when interface is up and not associated.
QCA988X wasn't measured.
Signed-off-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 is actually safe to enable ASPM after the
device is booted up.

This reduces power drain of QCA61X4 when driver is
simply loaded (no interface is up) from 31mA to
14mA. QCA988X wasn't measured but doesn't seem to
regress in any other way.
Signed-off-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>

Fix format specifier used for dma_addr_t, namely use %pad
Debug print virtual address for the same buffer as well.

Fixes: dc16427b ("wil6210: Add pmc debug mechanism memory management")
Signed-off-by: NVladimir Kondratiev <qca_vkondrat@qca.qualcomm.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

Wait a little bit longer, 50mS instead of 20mS, until the driver starts
polling for pen-up. The problematic behavior before this patch is applied
is as follows. The behavior was observed on the STMPE610QTR controller.

Upon a physical pen-down event, the touchscreen reports one set of x-y-p
coordinates and a pen-down event. After that, the pen-up polling is
triggered and since the controller is not ready yet, the polling mistakenly
detects a pen-up event while the physical state is still such that the pen
is down on the touch surface.

The pen-up handling flushes the controller FIFO, so after that, all the
samples in the controller are discarded. The controller becomes ready
shortly after this bogus pen-up handling and does generate again a pen-down
interrupt. This time, the controller contains x-y-p samples which all read
as zero. Since pressure value is zero, this set of samples is effectively
ignored by userland.

In the end, the driver just bounces between pen-down and bogus pen-up
handling, generating no useful results. Fix this by giving the controller a
bit more time before polling it for pen-up.
Signed-off-by: NMarek Vasut <marex@denx.de>
Reviewed-by: NViresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Use msecs_to_jiffies(20) instead of plain (HZ / 50), as the former is much
more explicit about it's behavior. We want to schedule the task 20 mS from
now, so make it explicit in the code.
Signed-off-by: NMarek Vasut <marex@denx.de>
Reviewed-by: NViresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Joydev is currently thinking some absolute mice are joystick, and that
messes up games in VMware guests, as the cursor typically gets stuck in
the top left corner.

Try to detect the event signature of a VMmouse input device and back off
for such devices. We're still incorrectly detecting, for example, the
VMware absolute USB mouse as a joystick, but adding an event signature
matching also that device would be considerably more risky, so defer that
to a later merge window.
Signed-off-by: NThomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Since the caller already check the presence of a of_node there
is no need to repeat the check in stmmac_probe_config_dt.

There is also no point in checking the return value of the
of_match_device function since if there wasn't match in the
first place we would never be in this function.
Signed-off-by: NJoachim Eastwood <manabian@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Since stmmac_dvr_probe takes care of setting driver data and
assign resources to the priv structure there is no need to
access the priv structure from the other probe functions.
This mean that this function can be changed into just return
an int and thus simplifying the callers.
Signed-off-by: NJoachim Eastwood <manabian@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Creat a struct that contain all the resources that needs to be
assigned to the priv struct in stmmac_dvr_probe. This makes it
possible to factor out more common code from the other probe
functions and also use this struct to hold the resources as
they are fetched.
Signed-off-by: NJoachim Eastwood <manabian@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>