Maya Erez [Tue, 24 Nov 2015 07:30:15 +0000 (09:30 +0200)]
wil6210: prevent external wmi commands during suspend flow
In __wmi_send we check if fw is ready at the beginning of the function.
While we wait for the completion of the previous command, system suspend
can be invoked and reset the HW, causing __wmi_send to read from HW
registers while it is not ready.
Taking the wmi_mutex in the reset flow when setting the FW ready bit
to zero will prevent the above race condition.
Signed-off-by: Maya Erez <qca_merez@qca.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Steve deRosier [Wed, 18 Nov 2015 22:51:28 +0000 (14:51 -0800)]
ath6kl: Don't print error message when recv is canceled
An error message ath6kl_htc_rxmsg_pending_handler isn't appropate for when
the error is ECANCELED. This could be the result of a perfectly appropriate
RX cancel due to shutdown or suspend. This allows the right cleanup to
continue, but without an alarming error message in this particular case.
Signed-off-by: Steve deRosier <steve.derosier@lairdtech.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
ath9k: Simplify and fix eeprom endianness swapping
The three eeprom implementations had quite some duplicate code when it
came to endianness swapping.
Additionally there was a bug in eeprom_4k and eeprom_9287 which
prevented the endianness swapping from working correctly, because the
swapping code was guarded within an "if (!ath9k_hw_use_flash(ah))". In
eeprom_def this check did not exist, so it seems that eeprom_def was the
only implementation where endianness swapping worked.
This patch takes the duplicate code and moves it from eeprom_* to
eeprom.c. The new code is derived from eeprom_def, while taking into
account the specifics from the other implementations.
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Felix Fietkau [Mon, 30 Nov 2015 18:32:01 +0000 (19:32 +0100)]
ath10k: do not use coherent memory for allocated device memory chunks
Coherent memory is more expensive to allocate (and constrained on some
architectures where it has to be pre-allocated). It is also completely
unnecessary, since the host has no reason to even access these allocated
memory spaces
Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Kalle Valo [Thu, 3 Dec 2015 15:10:25 +0000 (17:10 +0200)]
Merge ath-next from ath.git
Major changes:
ath10k
* support Manegement Frame Protection (MFP)
* add thermal throttling support for 10.4 firmware
* add support for pktlog in QCA99X0
* add debugfs file to enable Bluetooth coexistence feature
* use firmware's native mesh interface type instead of raw mode
Matti Gottlieb [Sun, 22 Nov 2015 08:08:56 +0000 (10:08 +0200)]
iwlwifi: mvm: ROC: Extend the ROC max delay duration & limit ROC duration
When associated to an AP and a ROC event with a long duration is scheduled
the FW may have a hard time scheduling a consecutive time event, since it
has to remain on the connection channel to hear the AP's DTIM.
In addition, when associated and a ROC is requested with a duration
greater than the DTIM interval, the FW will not be able to schedule
the ROC event, since it needs to wake up for the DTIM.
Increasing the "max delay" duration to the DTIM period will allow the FW to
wait until after the DTIM and then schedule the ROC time event.
Limiting the ROC to be less than the DTIM interval will assure that the
time event will be scheduled for at least part of the time (instead of
automatically failing)
Extend the ROC max delay duration to min(dtim_interval * 3, 600TU),
and limit the duration to be less than the DTIM interval.
Signed-off-by: Matti Gottlieb <matti.gottlieb@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Derek Basehore [Thu, 5 Nov 2015 01:37:52 +0000 (17:37 -0800)]
iwlwifi: mvm: report wakeup for wowlan
When the wifi wakes up the system, we need to report it via calling
pm_wakeup_event for lucid sleep. This is so userspace knowns that the
wifi woke up the system via the /sys/power/wakeup_type sysfs interface.
Johannes Berg [Sat, 21 Nov 2015 20:57:52 +0000 (21:57 +0100)]
iwlwifi: print index in api/capa flags parsing message
If the API or capabilities index is bigger than the driver expects,
an error message is printed. Make that message print the index and
distinguish between API and capabilities.
Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Dreyfuss, Haim [Mon, 1 Jun 2015 05:27:17 +0000 (08:27 +0300)]
iwlwifi: Add new PCI IDs for 9260 and 5165 series
Add 9000-family configuration to iwl_cfg struct
Add a new struct to define the 5165 series.
Rename the struct that defines the 9000 series to 9260.
Add some new sub-system IDs for the 9260 and 5165 series.
For 9260:
0x0A10, 0x0000, 0x0510, 0x0710, 0x0410, 0x0610.
For 5165:
0x2A10, 0x2010, 0x0310, 0x0210.
Eliad Peller [Thu, 12 Nov 2015 08:49:38 +0000 (10:49 +0200)]
iwlwifi: mvm: refactor d3 key update functions
We need to reuse the key update logic for d0i3
as well.
Add some parameters to deal with the constraints
implied by the d0i3 flow (specifically, support
non-SYNC commands, and don't take mutexes that
might deadlock).
Change some commands to be ASYNC, in order
to simplify locking a bit.
Johannes Berg [Thu, 24 Sep 2015 16:29:00 +0000 (18:29 +0200)]
iwlwifi: mvm: use firmware station lookup, combine code
In most cases, the firmware will already match the station that
we received a given frame from and tell us the station ID in the
RX status, so we can look up the station from that. This lets us
skip the (more expensive) hash table lookup in mac80211.
Also change the fallback case (no station info from the firmware)
to not attempt to look up a multicast source address.
While at it, also combine all the code using the station into a
single if block.
Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Nicholas Krause [Wed, 23 Sep 2015 00:24:19 +0000 (20:24 -0400)]
iwlwifi: mvm: fix incorrect fallthrough in iwl_mvm_check_running_scans()
In the iwl_mvm_check_running_scans() we were mistakenly ignoring the
value returned by iwl_mvm_scan_stop() for scheduled scans and falling
thorugh to the next case, which caused us to always return zero.
Signed-off-by: Nicholas Krause <xerofoify@gmail.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Johannes Berg [Wed, 11 Nov 2015 10:53:32 +0000 (11:53 +0100)]
iwlwifi: clean up transport debugfs handling
Transport code currently calls itself through the transport ops,
which is quite pointless. Clean up all of this. While at it,
remove the unnecessary dir argument and the redundant IDI code.
In slave transports, call both the common slave debugfs and the
transport's own. SDIO has no files, so remove it all there.
Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Golan Ben Ami [Mon, 27 Jul 2015 14:02:35 +0000 (17:02 +0300)]
iwlwifi: mvm: Support setting continuous recording debug mode
Add ability to set the continuous recording mode of the FW, while
the FW debug data is configured to be stored on the NIC.
This could be useful for storing large segments of FW usniffer
debug data on the host, while having small store space on the NIC.
The host receives the usniffer data through the regular RX path, and
the data can get extracted using trace-cmd.
Kalle Valo [Wed, 25 Nov 2015 13:38:27 +0000 (15:38 +0200)]
ath10k: split driver info messages during device initialisation
Earlier we printed all the info messages after a successful device
initialisation and firmware boot, but that's problematic if something goes
wrong and there's no easy way to know what firmware version was used and so on.
Split the info messages into smaller pieces and print them as soon as we have
the info available.
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Peter Oh [Tue, 24 Nov 2015 17:37:35 +0000 (09:37 -0800)]
ath10k: apply Mesh subtype when Mesh interface created.
QCA988X firmware starting from 10.2.4.70.12-2 has capability to support
Mesh Control Field Present bit in QoS field in native Wi-Fi mode.
Hence apply Mesh subtype according to the WMI service map.
Firmware will allows unicast, broadcast, multicast, and WDS frame
(FromDS = 1 and ToDS = 1) to be received via the interface, once Mesh
subtype is used.
The firmware and this patch together make native Wi-Fi mode comply to
IEEE802.11s Mesh frame in open mode, but the firmware doesn't yet
support secured Mesh mode.
Signed-off-by: Peter Oh <poh@qca.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Felix Fietkau [Tue, 24 Nov 2015 10:36:52 +0000 (11:36 +0100)]
ath10k: stop abusing GFP_DMA
Allocations from the DMA zone were originally added for legacy ISA
stuff, or PCI devices that have specific limitations in their DMA
addressing capabilities. It has no place in ath10k, which can do
full 32-bit DMA.
Fixes memory allocation errors on some platforms.
Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
ath10k: Enable AP + STA interface combination support for 10.x
Enable AP + STA interface combination support for 10.x and
this enables Repeater mode testing (with WDS Repeater combinations
as well). Fix indentation as well.
Based on the suggestions from Michal and Yanbo Li,
thanks to them
Signed-off-by: Mohammed Shafi Shajakhan <mohammed@qti.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Peter Oh [Fri, 20 Nov 2015 17:33:14 +0000 (09:33 -0800)]
ath10k: fix board data fetch error message
The error message order of board data fetch is board info,
directory, and then file name, hence place print arguments
in the order.
Fixes: 0a51b343abfe ("ath10k: add board 2 API support") Signed-off-by: Peter Oh <poh@qca.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Dan Carpenter [Thu, 26 Nov 2015 11:55:23 +0000 (14:55 +0300)]
rt2x00: type bug in _rt2500usb_register_read()
This code causes a static checker bug.
drivers/net/wireless/ralink/rt2x00/rt2500usb.c:232 _rt2500usb_register_read()
warn: passing casted pointer 'value' to 'rt2500usb_register_read()' 32 vs 16.
If the low 16 bits were initialized to zero then this code would only be
a problem on big endian systems. But in this case this is case the low
16 bits are never initialized. This is called from a function which is
created using a macro:
RT2X00DEBUGFS_OPS(csr, "0x%.8x\n", u32);
We end up copying uninitialized data to the user which is bogus and an
information leak.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Acked-by: Stanislaw Gruszka <sgruszka@redhat.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Hante Meuleman [Wed, 25 Nov 2015 10:32:46 +0000 (11:32 +0100)]
brcmfmac: Unify methods to define and map firmware files.
All bus drivers (sdio, usb and pcie) require firmware files which
needs to be downloaded to the device, The definitions and mapping
of device id and revision to firmware and nvram file is done by
each bus driver. This patch creates common functions and defines
to simplify and unify the definition of these firmware and nvram
files and mapping.
Reviewed-by: Arend Van Spriel <arend@broadcom.com> Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Signed-off-by: Hante Meuleman <meuleman@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Arend van Spriel [Wed, 25 Nov 2015 10:32:45 +0000 (11:32 +0100)]
brcmfmac: assure net_ratelimit() is declared before use
Under some kernel configuration we get build issue with implicit
declaration of net_ratelimit() function. Fix this by explicitly
including the file providing the prototype.
Reviewed-by: Hante Meuleman <meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Franky Lin [Wed, 25 Nov 2015 10:32:40 +0000 (11:32 +0100)]
brcmfmac: no retries on rxglom superframe errors
Aborting the current read attempt on the superframe also removes the
packet from the pipeline. Retries should not be attempted on the next
packet since it would not be a superframe(either a superframe descriptor
or other data packet) and should not be handled by brcmf_sdio_rxglom
Reviewed-by: Hante Meuleman <meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Reviewed-by: Arend Van Spriel <arend@broadcom.com> Signed-off-by: Franky Lin <frankyl@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Hante Meuleman [Wed, 25 Nov 2015 10:32:39 +0000 (11:32 +0100)]
brcmfmac: Change error print on wlan0 existence.
During initialization of the device, but also on some other
moments the driver prints an error that the netdev already exists.
This is a result of the way the driver is initializing the
firmware and not really an error. The code is not treating it as
an error either. This error print has resulted in many questions
by users and is confusing and incorrect. This patch changes the
error log into a debug info log.
Reviewed-by: Arend Van Spriel <arend@broadcom.com> Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Signed-off-by: Hante Meuleman <meuleman@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Hante Meuleman [Wed, 25 Nov 2015 10:32:38 +0000 (11:32 +0100)]
brcmfmac: Return actual error by fwil.
FWIL is always mapping back errors to EBADE. This is not very
conventient when trying to understand problems by reading logs.
Some callers print the error code, but that is quite useless
when the exact error code is not returned. It also makes it
impossible to differentiate based on error code. This patch
changes the return of EBADE into the actual error code.
Reviewed-by: Arend Van Spriel <arend@broadcom.com> Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Signed-off-by: Hante Meuleman <meuleman@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Andy Green [Tue, 10 Nov 2015 08:25:53 +0000 (03:25 -0500)]
wcn36xx: handle new hal response format
wcn3620 has a new message structure for the reply to some hal
commands. This patch adds the struct and helper routine that
uses it if the chip is wcn3620, or falls back to the old
helper routine.
We don't know what to do with the candidate list he sends back,
but we can at least accept and ignore it nicely instead of dying.
Signed-off-by: Andy Green <andy.green@linaro.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Luca Coelho [Fri, 2 Oct 2015 15:13:10 +0000 (18:13 +0300)]
iwlwifi: mvm: flush all used TX queues before suspending
There is a potential race condition when entering suspend with d0i3 in
PCIe. If there is a frame queued just before we suspend, it won't
complete and we will never clear the queue stuck timer. To solve
this, call TX_PATH_FLUSH to flush all queues (except the command
queue) as part of the d0i3 entry process. Add a new function that
returns all the flushable queues.
Golan Ben-Ami [Wed, 21 Oct 2015 13:15:38 +0000 (16:15 +0300)]
iwlwifi: dump prph registers in a common place for all transports
Currently the prph registers dump is in the transport layer,
and each bus needs an additional dump implementation.
Move the prph dump outside transport, and allow a common
implementation for all of the buses.
This is possible because prph base addresses are similar for
all buses.
Golan Ben-Ami [Wed, 21 Oct 2015 12:16:58 +0000 (15:16 +0300)]
iwlwifi: export the _no_grab version of PRPH IO functions
Expose _no_grab prph i/o functions that allow performing i/o
outside the transport, without requiring grab and release NIC access
for each operation. In addition, rename the functions so they reflect
their non-grabbing behavior.
This can be very useful for consecutive prph i/o operation that occur
outside trans, such as fw dumps.
Eyal Shapira [Thu, 22 Oct 2015 10:10:36 +0000 (13:10 +0300)]
iwlwifi: mvm: drop low_latency_agg_frame_cnt_limit
This was an old workaround for solving latency issues with
certain Miracast adapters like ActionTec. However this isn't
needed anymore and furthermore it hurts throughput in other
use cases.
iwlwifi: mvm: remove redundant d0i3 flag from the config struct
The d0i3 flag in the device configuration structure is redundant,
because the same information can be determined by checking the
firmware capability flag.
Golan Ben-Ami [Sun, 23 Aug 2015 14:57:33 +0000 (17:57 +0300)]
iwlwifi: mvm: add trigger for firmware dump upon TDLS events
This will allow to catch different TDLS events and get the
firmware data when they occur.
Add empty TX_LATENCY trigger on the way to avoid mismatch
with trees in which this trigger is implemented.
The PCIe transport will also need a d0i3_entry_timeout_ms parameter,
so move the existing one from the slave transports to iwlwifi, so it
can be reused. While at it, rename the parameter to something
shorter, namely d0i3_entry_delay.
Matti Gottlieb [Mon, 2 Nov 2015 11:01:54 +0000 (13:01 +0200)]
iwlwifi: mvm: check FW's response for nvm access write cmd
In case of using an external NVM file, the driver sends to the
FW the different nvm sections. In the response of the cmd, the
FW states the status of the writing of the chunk.
Currently the value is not checked by the driver.
Check FW's response for writing the nvm chunk in the NVM_ACCESS_CMD.
Signed-off-by: Matti Gottlieb <matti.gottlieb@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Avri Altman [Thu, 29 Oct 2015 14:50:57 +0000 (16:50 +0200)]
iwlwifi: mvm: Enable MPLUT only on supported hw
When there's a bt-wifi contention that requires arbitration,
we use a priority-based mechanism to decide which comm wins.
Over time, use cases become more and more complex, with multiple
concurrent active links with different traffic types and different QoS
requirements, on both WiFi and BT sides.
This, in turn, requires us to elaborate our prioritization mechanism.
However, our legacy products included hw that does not supports this,
so selectively enable this on specific hw - as signaled by the firmware.
Avraham Stern [Tue, 10 Nov 2015 12:56:24 +0000 (14:56 +0200)]
iwlwifi: mvm: Configure fragmented scan for scheduled scan
Configure the FW to use fragmented scan when the traffic load is high
or low latency traffic is on. This is useful for scans that are
managed by the FW (e.g. scheduled scan).
Johannes Berg [Thu, 12 Nov 2015 15:16:01 +0000 (16:16 +0100)]
iwlwifi: trans: make various conversion macros inlines
Make the various conversion functions typesafe, so we don't
accidentally try to call them with the wrong pointers and
cast them to something that will crash.
Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Johannes Berg [Fri, 13 Nov 2015 09:57:56 +0000 (10:57 +0100)]
iwlwifi: dvm: remove Kconfig default
The split of iwlwifi into DVM and MVM was a long time ago now,
so we can remove the "default IWLWIFI" that we had to keep all
existing .config files with working defaults during the split.
This is no longer necessary, practically nobody should now be
upgrading a .config that's older than the split.
Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
802.11ac allows A-MSDU that can be up to 12KB long. Since
an entire A-MSDU needs to fit into one single Receive
Buffer (RB), add support for big RBs.
Since this adds lots of pressure to the memory manager and
significantly increase the true_size of the RX buffers,
don't enable this by default.
Johannes Berg [Thu, 17 Sep 2015 13:54:47 +0000 (15:54 +0200)]
iwlwifi: nvm: fix up phy section when reading it
This is a workaround to an OTP bug. In Series 8000 1x1, the OTP
0xA052 defines 2x2 antenna configuration. This workaround overrides
the decision based on HW id and on MIMO disabled bit which is
correct in the OTP and set to disabled. This fixes the previous
workaround "force 1x1 antenna in Series 8000".
Dan Carpenter [Fri, 6 Nov 2015 10:01:20 +0000 (13:01 +0300)]
ath9k_htc: check for underflow in ath9k_htc_rx_msg()
We check for overflow here, but we don't check for underflow so it
causes a static checker warning.
Fixes: fb9987d0f748 ('ath9k_htc: Support for AR9271 chipset.') Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Markus Elfring [Fri, 6 Nov 2015 07:48:23 +0000 (08:48 +0100)]
brcm80211: Delete an unnecessary check before the function call "release_firmware"
The release_firmware() function tests whether its argument is NULL and then
returns immediately. Thus the test around the call is not needed.
This issue was detected by using the Coccinelle software.
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net> Acked-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Hante Meuleman [Thu, 29 Oct 2015 19:33:19 +0000 (20:33 +0100)]
brcmfmac: Add wowl wake indication report.
On wakeup of the system (resume) a wowl wakeup indication report
can be sent to cfg80211. This patch adds support for this. The
report specifies if the device was responsible for the wakeup
and if so, will specify the exact reason.
Reviewed-by: Arend Van Spriel <arend@broadcom.com> Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Signed-off-by: Hante Meuleman <meuleman@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Hante Meuleman [Thu, 29 Oct 2015 19:33:18 +0000 (20:33 +0100)]
brcmfmac: Use new methods for pcie Power Management.
Currently the legacy methods suspend and resume are used for pcie
devices. This is not the preferable method and is also causing
issues with some setups when doing hibernate. Changing this to
use the new PM methods.
Reviewed-by: Arend Van Spriel <arend@broadcom.com> Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Signed-off-by: Hante Meuleman <meuleman@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Hante Meuleman [Thu, 29 Oct 2015 19:33:17 +0000 (20:33 +0100)]
brcmfmac: Use consistent naming for bsscfgidx.
The variable bsscfgidx is used in different places with different
names, e.g. bsscfg, bssidx, bsscfg_idx. This patch cleans this up
by using bsscfgidx everywhere.
Reviewed-by: Arend Van Spriel <arend@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Signed-off-by: Hante Meuleman <meuleman@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Hante Meuleman [Thu, 29 Oct 2015 19:33:16 +0000 (20:33 +0100)]
brcmfmac: Add RSDB support.
Broadcom devices with a single 802.11 core can work on two band
concurrently using VSDB feature, ie. Virtual Simultaneous Dual-Band.
For devices that are fitted with two 802.11 cores and RF paths the
driver should support a firmware feature called RSDB, which stands
for Real Simultaneous Dual-Band. RSDB works almost autonomously in
firmware except for AP config. When the device supports RSDB then
the interface should not be brought down when configuring it,
otherwise the link (if configured) on the other interface will be
lost.
Reviewed-by: Arend Van Spriel <arend@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Signed-off-by: Hante Meuleman <meuleman@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com>
[kvalo@codeaurora.org: changed the commit log based on discussion] Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Hante Meuleman [Thu, 29 Oct 2015 19:33:15 +0000 (20:33 +0100)]
brcmfmac: Disable runtime pm for USB.
Currently runtime pm is enabled for USB, but it is not properly
supported by driver. This patch disables the runtime PM support
completely for USB, as it currently can result in problems on
some systems.
Reviewed-by: Arend Van Spriel <arend@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Signed-off-by: Hante Meuleman <meuleman@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Hante Meuleman [Thu, 29 Oct 2015 19:33:13 +0000 (20:33 +0100)]
brcmfmac: Remove unnecessary check from start_xmit.
The brcmf_netdev_start_xmit checks if the ndev is still valid by
checking if it still exists in database. This check is not needed
and therefor removed.
Reviewed-by: Arend Van Spriel <arend@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Signed-off-by: Hante Meuleman <meuleman@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Hante Meuleman [Thu, 29 Oct 2015 19:33:12 +0000 (20:33 +0100)]
brcmfmac: Simplify and fix usage of brcmf_ifname.
brcmf_ifname is a debug function to return a name related to an ifp,
but is using a rather complex implementation. It was also used
wrongly from bcdc as it did not use the bsscfgidx as it was supposed
to, but bssidx. This patch fixes that bug and simplifies
brcmf_ifname.
Reviewed-by: Arend Van Spriel <arend@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Signed-off-by: Hante Meuleman <meuleman@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Michal Kazior [Wed, 18 Nov 2015 05:59:23 +0000 (06:59 +0100)]
ath10k: store msdu_id instead of txbuf pointers
Txbuf is no longer a DMA pool and can be easily
tracked with a mere msdu_id. This saves 10 bytes
on 64bit systems and 6 bytes on 32bit systems of
precious sk_buff control buffer.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Michal Kazior [Wed, 18 Nov 2015 05:59:21 +0000 (06:59 +0100)]
ath10k: fix tx header parsing
Frames are not guaranteed to be 802.11 frames in
ath10k_htt_tx() and the tx completion handler.
In some cases, like TDLS, they can be Ethernet.
Hence checking, e.g. frame_control could yield
bogus results and behavior.
Fortunately this wasn't a real problem so far
because there's no FW/HW combination to encounter
this problem.
However it is good to fix this in advance.
Fixes: 75d85fd9993c ("ath10k: introduce basic tdls functionality") Fixes: eebc67fef3ee ("ath10k: fix pmf for wmi-tlv on qca6174") Fixes: 7b7da0a02192 ("ath10k: drop probe responses when too many are queued") Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
projects / linux-beck.git / log