A following patche will use trans_pcie->cmd_in_flight
for reference accounting as well. get ready for it.
Signed-off-by: Eliad Peller <eliadx.peller@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel 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: Emmanuel 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: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
This configuration is not needed for dvm, and it actually
broke it.
Reported-by: Oliver Hartkopp <socketcan@hartkopp.net>
Signed-off-by: Emmanuel 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: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel 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: Avri Altman <avri.altman@intel.com>
Signed-off-by: Emmanuel 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: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel 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: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel 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: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel 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: Avri Altman <avri.altman@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
I've never seen this happen, but it's useful to rule it out.
Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
In the 8000 HW family the register for forcing an NMI has
changed, so this allows to still be able to force an NMI
while taking into account the HW in order to write to the
correct register.
Signed-off-by: Liad Kaufman <liad.kaufman@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
When a firmware error occurs, capture the last 32 commands
(which are still in memory) in the error dump debugfs file.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
This variable always tracks a constant value (256) so there's
no need to have it. Removing it simplifies code generation,
reducing the .text size (by about 240 bytes on x86-64.)
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
The 'reset' argument is clearly a boolean, so use bool instead
of u8 with 0/1 values.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
7000 device series have a fix for this hardware feature.
Stop disabling it, and get an improvement in Tx throughput.
This feature allows the scheduler to fetch more frames on
the fly while an A-MPDU is being built - which means that
we can get larger A-MPDU. This, of course, give an
improvement in the Tx throughput.
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Similar to the recent RX queue patch, this changes the need_update
handling for the TX queues to be clearer and only done when needed.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Instead of using u8 to hold logic values, use bool.
Also fix a comment, the return value is no longer relevant.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
A hardware bug had been discovered on 7260 / 3160 and 7265
and the workaround for this bug is to force the NIC to stay
awake as long as we have host commands in flight. This
workaround has been introduced for all NICs in a previous
patch:
b943949105 ("iwlwifi: pcie: keep the NIC awake when commands are in flight")
In newer NICs, this bug is solved, so we can let the NIC go
to sleep even when we send commands. The hardware will wake
up when we increment the scheduler write pointer.
Make the workaround conditional to only use it on affected
hardware.
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
In case the firmware didn't assert but we want to restart
it, e.g. we didn't get the reply for a host command, or the
Tx queues are stuck, we should stop the firmware by
provoking an interrupt. This allows to better debug the
firmware in these bad scenarios.
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
The various code blocks in iwl_pcie_[rt]xq_inc_wr_ptr
finally do the same things, so just merge them
all and make the functions cleaner.
Signed-off-by: Eliad Peller <eliadx.peller@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
APMG HW block was removed in this NIC, hence, no need to
configure it.
Signed-off-by: Eran Harary <eran.harary@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
None of the devices supported by iwldvm have support for
shadow registers. This means that we wake the NIC
when we increment the write pointer on Tx ring.
This happened even before my bad commit mentionned below.
Since my commit below, we wake up the NIC when we put a
host command on the ring regardless of shadow register
support. This means that in iwldvm (when the NIC doesn't
support shadow register), we wake up the NIC twice:
pcie_enqueue_hcmd:
wake up the NIC
iwl_pcie_txq_inc_wr_ptr:
wake up the NIC - no shadow reg support
Since waking up the NIC means that we need to acquire a
spinlock, this obviously leads to a recursive spinlock
and hence a freeze.
Fixes: b943949105 ("iwlwifi: pcie: keep the NIC awake when commands are in flight")
Reported-by: Janusz Dziedzic <janusz.dziedzic@gmail.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Calling stop_device when start_fw wasn't called would issue:
Stopping tx queues that aren't allocated...
Also allow the op_mode to call stop_device and then to
disable the Tx queues - in that case just silently ignore
the disabling on the Tx queues, since the PRPH registers
aren't reachable any more.
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Under very specific circumstances, the firmware might
ignore a host command. This was debugged and we ended up
seeing that the power management hardware was faulty.
In order to workaround this issue, we keep the NIC awake
as long as we have host commands in flight. This will avoid
to put the hardware into buggy condition.
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Since we don't take this lock in the primary interrupt
handler, there is no pointin disabling the interrupt
in the critical section protected by trans_pcie->irq_lock.
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
In case a sync command timeouts or Tx is stuck while a FW error
interrupt arrives, we might call iwl_op_mode_nic_error twice before
a restart has been initiated. This will cause a reprobe. Unify calls
to this function at the transport level and only call it on the first
FW error in a given by checking the transport FW error flag.
While at it, remove the privately defined iwl_nic_error from PCIE code
and use the common callback instead.
Signed-off-by: Arik Nemtsov <arik@wizery.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Stop Tx and commands from arriving to the transport layer when a FW
error has occurred. A HW recovery should take place before. Remove
transport specific checks of the same nature (note that not all
transports were protected).
Signed-off-by: Arik Nemtsov <arik@wizery.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
The same bits are employed in all transport layers. Put the status
field in the common transport layer. This allows us to employ them
in common transport code.
Signed-off-by: Arik Nemtsov <arik@wizery.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
After wait_event_timeout(), the condition must still be
true if it returns >0, in fact almost the last thing in
it is checking the condition again. It's therefore not
useful to check yet again in our code, clean it up.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
The transport layer doesn't need to know the TX_CMD id.
It can be set by the op_mode.
The transport layer still needs to know the layout of the
Tx command because of alignment issues and because of the
scratch pointer.
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
If we call ieee80211_hw_restart, it means that the
firmware is in bad condition and will be reset soon.
Since the firmware will be reset, there is no good
reason to keep sending host commands.
Signed-off-by: Alexander Bondar <alexander.bondar@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Having a WARN_ON() followed by a printed message is
less useful than having the message in the warning
so move the message.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
In certain corner cases in the firmware implementation, powersave
transitions can cause the firmware to miss the fact that commands
were added to the queue/FIFO and thus never processes them. Since
the commands really are in the queue, try to poke the firmware in
such cases (by grabbing NIC access, which wakes up the NIC) so it
notices the new command and processes it.
Reviewed-by: Alexander Bondar <alexander.bondar@intel.com>
Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
This should really not happen. If it does, restarting is the
only way to recover since the driver and the firmware might
very well be out of sync. Moreover, iwl_op_mode_nic_error
will print data that might help debugging.
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
A few NICs can get into trouble if we reset the TX queue
counters in certain very rare situation. To be on the safe
side, simply avoid to reset the TX queue counter.
This is relevant for non-AMPDU queues only since on AMPDU
we have no choice - we must start the TX queue at the right
index.
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
There is a missing '-' character here so we return positive 'ENOMEM'
instead of negative. The caller doesn't care. All non-zero returns
are translated to '-ENOMEM' in iwl_pcie_nic_init().
This is just a cleanup.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Reduce the ambiguity spares to a single element if the window size is not
smaller than the queue size. If smaller, no spares are required at all.
Signed-off-by: Ido Yariv <ido@wizery.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
do some little cleanups in tx.c - eliminate duplicate checks,
use locally cached fields and predefined macros.
Signed-off-by: Eliad Peller <eliad@wizery.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Use the return value of WARN_ONCE() and add a message with
the queue ID that's getting used.
Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
When the queue is unmapped while it was so loaded that
mac80211's was stopped, we need to wake the queue after
having freed all the packets in the queue.
Not doing so can result in weird stuff like:
* run lots of traffic (mac80211's queue gets stopped)
* RFKILL
* de-assert RFKILL
* no traffic
Cc: stable@vger.kernel.org
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
When a queue is disabled, it frees all its entries. Later,
the op_mode might still get notifications from the firmware
that triggers to free entries in the tx queue. The transport
should be prepared for these races and know to ignore
reclaim calls on queues that have been disabled and whose
entries have been freed.
Cc: stable@vger.kernel.org
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
