No need to send both MCU_EXT_CMD_SET_RX_PATH and MCU_EXT_CMD_CHANNEL_SWITCH
together to MCU.
Split them out by passing the proper command in the corresponding flow.
Signed-off-by: Ryder Lee <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
MT7615 supports dual-wiphy which means that it can run on 2.4G and 5GHz
channels simultaneously, and driver should configure each band.
Add missing register settings, and refine band related definitions to
avoid duplicate codes.
Signed-off-by: Ryder Lee <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
Even if they are not currently used fix BK/BE endpoint definition order.
Fixes: b40b15e1521f ("mt76: add usb support to mt76 layer")
Signed-off-by: Lorenzo Bianconi <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
Fix the following compilation warning in mt76_eeprom_override routine
when CONFIG_OF is not set and label 'out' is not actually used
drivers/net/wireless/mediatek/mt76/eeprom.c: In function ‘mt76_eeprom_override’:
drivers/net/wireless/mediatek/mt76/eeprom.c:100:1: warning: label ‘out’ defined but not used [-Wunused-label]
Signed-off-by: Lorenzo Bianconi <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
Rely on FIELD_PREP macro for led register definitions and
remove open coding
Signed-off-by: Lorenzo Bianconi <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
Rely on FIELD_PREP macro for led register definitions and
remove open coding
Signed-off-by: Lorenzo Bianconi <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
Initialize brightness_set and blink_set callbacks to
mt7615_led_set_brightness and mt7615_led_set_blink in order to enable
LED support in mt7615 driver
Tested-by: Deng Qingfang <[email protected]>
Signed-off-by: Lorenzo Bianconi <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
We do not call mt76x02_mac_set_beacon() with NULL skb any longer and
we do not need to return error value.
Signed-off-by: Stanislaw Gruszka <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
With current implementation we do not cleanup beacon memory, so is not
needed to call mt76x02_mac_set_beacon() with NULL skb.
Signed-off-by: Stanislaw Gruszka <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
For transferring data over USB the optimal size is endpoint maxpacket.
For my hardware maxpaket for control endpoint is 64 bytes and changing
to this value from 128 bytes further shorten TBTT work time from
3ms to 1ms.
Signed-off-by: Stanislaw Gruszka <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
This patch enables channel switch support on mt76 usb interfaces.
Signed-off-by: Markus Theil <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
Use larger batches for usb copy to speed this operation up. Otherwise it
would be too slow for copying new beacons or broadcast frames over usb.
Assure, that always a multiple of 4 Bytes is copied, as outlined in
850e8f6fbd "mt76: round up length on mt76_wr_copy" from Felix Fietkau.
Signed-off-by: Markus Theil <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
This patch removes a mt76_wr_copy call from the beacon path to hw.
The skb which is used in this place gets therefore build with txwi
inside its data. For mt76 usb drivers, this saves one synchronuous
copy call over usb, which lets the beacon work complete faster.
In mmio case, there is not enough headroom to put the txwi into the
skb, it is therefore using an additional mt76_wr_copy, which is fast
over mmio. Thanks Stanislaw for pointing this out.
Signed-off-by: Markus Theil <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
On adding vifs the idx can become 1 + (7 & 7) = 8 for APs.
Check against that, as only AP vif idx 0-7 is possible.
Signed-off-by: Markus Theil <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
Sending beacons to the hardware always happens in batches. In order to
speed up beacon processing on usb devices, this patch splits out common
code an calls it only once.
Beacons are sequentially written into the beacon memory area, by
tracking its usage with the dev->beacon_data_count. For MBSS support
and buffered traffic dev->beacon_data_count is used to create the bypass
mask.
The code is also adapted for the mmio part of the driver, but should not
have any performance implication there.
MBSS tests were performed with AVM AC860 USB NIC with temporary support
for 5 BSS'. Different combinations of active vifs were created and
brought up. Afterwards connection and data transfer was tested for the
announced BSS'.
Signed-off-by: Markus Theil <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
mt76 hw does not send beacons from beacon slots, if the corresponding
bitmask is set accordingly. Therefore we can omit clearing the beacon
memory. Clearing uses many usb calls, if usb drivers are used. These
calls unnecessarily slow down the beacon tasklet. Thanks to Stanislaw
Gruzska for pointing this out.
Signed-off-by: Markus Theil <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
The dev_kfree_skb() function performs also input parameter validation.
Thus the test around the shown calls is not needed.
This issue was detected by using the Coccinelle software.
Signed-off-by: Markus Elfring <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>
Before extracting the tid out of the packet, check if it was qos-data.
Only accept tid values 0-7
Also, avoid accepting the hardware queue as skb queue mapping, it could
lead to an overrun. Instead, derive the hardware queue from the tid number,
in order to avoid issues with packets being filtered multiple times.
This also fixes a mismatch between hardware and software queue indexes.
Signed-off-by: Felix Fietkau <[email protected]>
MCU_EXT_CMD_EFUSE_BUFFER_MODE needs more time on initialization, and
with only 10 seconds it sometimes runs into timeouts.
Signed-off-by: Felix Fietkau <[email protected]>
Before this patch, mt76 handled rx traffic for all TIDs equally,
when released from reorder buffer early. This patch uses an AC specific
reorder timeout, in order to release partial aggregated frames for video
ACs earlier. Voice ACs are currently not aggregated (thanks to Felix for
this hint). For example, ath10k also uses AC specific reorder timeouts
(reported by firmware in that case).
Signed-off-by: Markus Theil <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
During the cleanup of the aggregation session, a rx handler (or release timer)
on another CPU might still hold a pointer to the reorder buffer and could
attempt to release some packets.
Clearing pointers during cleanup avoids a theoretical use-after-free bug here.
Signed-off-by: Felix Fietkau <[email protected]>
Add the capability to configure acktimeout for mt7615 driver. Moreover
configure slottime according to the value provided by mac80211
Signed-off-by: Lorenzo Bianconi <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
Set 0 as minimum configurable value for coverage_class since mt76 does
not support dynack
Signed-off-by: Lorenzo Bianconi <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
Fix coverage_class type in mt76x02_dev data structure since
coverage_class can be negative to enable dynack (just supported by
ath9k). Set 0 as minimum value for coverage_class
Fixes: 7bc04215a66b ("mt76: add driver code for MT76x2e")
Signed-off-by: Lorenzo Bianconi <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
Print fw_ver and build_date members of struct mt7615_fw_trailer
similarly to what appears in the output of 'dmesg' when the MCU firmware
is loaded.
Signed-off-by: Lorenzo Bianconi <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
Some compilers (tested with 4.8.5 from CentOS 7) fail properly process
FIELD_GET inside an inline function, which ends up in a BUILD_BUG_ON.
Convert inline function to a macro.
Fixes commit bf92e7685100 ("mt76: mt7615: add support for per-chain
signal strength reporting")
Reported in https://lkml.org/lkml/2019/9/21/146
Reported-by: kbuild test robot <[email protected]>
Signed-off-by: Pablo Greco <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
mt76x0e users reported some devices (e.g TP-Link Archer VR200v) have
been flashed with big endian radio partition. Add the possibility to
specify eeprom endianness using big-endian dts property and in case
covert eeprom data in little endian
Tested-by: Kevin Schmidt <[email protected]>
Signed-off-by: Lorenzo Bianconi <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
Reset MT_MIB_STAT_CCA after channel switch since it is used to track busy
time starting from 'commit dcff8d4dc301 ("mt76: mt7603: switch to a
different counter for survey busy time")'
Signed-off-by: Lorenzo Bianconi <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
The sband->bitrates[] array has "sband->n_bitrates" elements so this
check needs to be >= instead of > or we could read beyond the end of the
array.
These values come from when we call mt76_register_device():
ret = mt76_register_device(&dev->mt76, true, mt7603_rates,
ARRAY_SIZE(mt7603_rates));
Here sband->bitrates[] is mt7603_rates[] and ->n_bitrates is the
ARRAY_SIZE()
Fixes: 5ce09c1a7907 ("mt76: track rx airtime for airtime fairness and survey")
Signed-off-by: Dan Carpenter <[email protected]>
Make sure the MAC address is checked before function returns.
If CONFIG_OF is set and the device node is null, the function will
return directly, and an invalid MAC address will not be checked.
Signed-off-by: Shayne Chen <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
Based on 1 normalized pattern(s):
this program is free software you can redistribute it and or modify
it under the terms of the gnu general public license version 2 as
published by the free software foundation this program is
distributed in the hope that it will be useful but without any
warranty without even the implied warranty of merchantability or
fitness for a particular purpose see the gnu general public license
for more details
extracted by the scancode license scanner the SPDX license identifier
GPL-2.0-only
has been chosen to replace the boilerplate/reference in 655 file(s).
Signed-off-by: Thomas Gleixner <[email protected]>
Reviewed-by: Allison Randal <[email protected]>
Reviewed-by: Kate Stewart <[email protected]>
Reviewed-by: Richard Fontana <[email protected]>
Cc: [email protected]
Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>
To avoid having the hardware potentially write to memory behind stale
descriptors, set the dma-done flag on all of them before cleaning up
allocated rx buffers
Signed-off-by: Felix Fietkau <[email protected]>
The MT7615 firmware needs to know the association id at creation time,
which is unavailable during the transition from notexist to none in
.sta_state.
This can cause a number of issues, probably also breaking powersave
support.
Signed-off-by: Felix Fietkau <[email protected]>
Loading the mcu firmware and waiting for it to boot takes a long time,
which adds a significant amount to the system boot time.
Fix this by running the mcu init from a workqueue and waiting for it to
complete before starting the phy or issuing mcu commands via debugfs
Signed-off-by: Felix Fietkau <[email protected]>
This is only used for testing for now. In the future it will be possible to
enable the second PHY through an proper API via device tree
Signed-off-by: Felix Fietkau <[email protected]>
Beacons and management frames need to explicitly select the alternate tx
queues in order to be sent out on the right phy
Signed-off-by: Felix Fietkau <[email protected]>
Program the RMAC CHFREQ registers to properly indicate the band that the
frames are received on.
Add some sanity checks to the programmed values, because the firmware
programs these registers differently
Signed-off-by: Felix Fietkau <[email protected]>
Move chainmask to struct mt7615_phy and instead of needlessly making the
format similar to values for older chips, make it refer to the actual chain
bits used for the rx/tx path.
This is important for multiple wiphy support, where for a secondary phy,
antenna_mask will start at 0, and chainmask will start at the chain offset
(bit 2)
Signed-off-by: Felix Fietkau <[email protected]>
There are two DFS detectors on the chip. When using 160 MHz channel bandwidth
(not supported in dual-wiphy mode), both are used. Otherwise, one detector is
used per wiphy.
Rework the code to start/stop them separately per phy and to indicate the
radar event on the right phy based on the detector index
Signed-off-by: Felix Fietkau <[email protected]>
