- 11 Apr, 2021 35 commits
-
-
Ryder Lee authored
Set skb->ip_summed to CHECKSUM_UNNECESSARY if the hardware has validated the IP and TCP/UDP checksum Signed-off-by:
Ryder Lee <ryder.lee@mediatek.com> Signed-off-by:
Felix Fietkau <nbd@nbd.name>
-
Ryder Lee authored
Enable hw rx-amsdu de-aggregation support. This is a preliminary patch to enable rx checksum offload. Signed-off-by:
-
Felix Fietkau authored
With threaded NAPI, the rx handler function is no longer bound to the CPU that fired the interrupt, which significantly helps to spread the workload over multiple CPUs, especially when multiple devices are using threaded NAPI at the same time. Exclude the tx handler from threaded NAPI by using a separate dummy netdev. The work is small and short-lived enough that it makes more sense to run it in softirq instead of creating a dedicated thread Signed-off-by: -
Sean Wang authored
Update sched_scan command usage according to the current firmware submitted into linux-firmware.git. Fixes: 80fc1e37 ("mt76: mt7921: rely on mt76_connac_mcu module for sched_scan and hw_scan") Co-developed-by:
Soul Huang <Soul.Huang@mediatek.com> Signed-off-by:
Sean Wang <sean.wang@mediatek.com> Signed-off-by:
Lorenzo Bianconi <lorenzo@kernel.org> Signed-off-by:
-
Sean Wang authored
add flush operation to make sure cfg80211_mlme_deauth is able to flush the deauthtication frame into air immediately. Co-developed-by:
YN Chen <YN.Chen@mediatek.com> Signed-off-by:
-
Sean Wang authored
Since the related rx bitrate fields have been moved to group3 in Rxv, fix rx bitrate statistics in mt7921_mac_fill_rx routine. Fixes: 163f4d22 ("mt76: mt7921: add MAC support") Signed-off-by:
-
Sean Wang authored
Use proper value for ht40 mode configuration in mt76_connac_mcu_uni_add_bss routine and not ht20 one Fixes: d0e274af ("mt76: mt76_connac: create mcu library") Co-developed-by:
-
Sean Wang authored
Current firmware only supports new api for enabling hardware beacon filter. Fixes: 1d8efc74 ("mt76: mt7921: introduce Runtime PM support") Beacon filter cmd have to rely on the associatied access point's beacon interval and DTIM information. Signed-off-by:
-
Sean Wang authored
Fix possible memory leak in mt7921_coredump_work. Fixes: 1c099ab4 ("mt76: mt7921: add MCU support") Signed-off-by:
-
Sean Wang authored
Any pcie access should happen in pci D0 state and we should give ownership back to the device at the end of the suspend procedure. Fixes: 1d8efc74 ("mt76: mt7921: introduce Runtime PM support") Signed-off-by:
-
Ryder Lee authored
Use PCI standard defines. Signed-off-by:
-
Lorenzo Bianconi authored
As done for mt76_connac_mcu_alloc_wtbl_req, even if this is not a real bug since mt7915_mcu_alloc_wtbl_req routine can fails just if nskb is NULL, always check return value from mt7915_mcu_alloc_wtbl_req in order to avoid possible future mistake. Signed-off-by:
-
Lorenzo Bianconi authored
Even if this is not a real bug since mt76_connac_mcu_alloc_wtbl_req routine can fails just if nskb is NULL , always check return value from mt76_connac_mcu_alloc_wtbl_req in order to avoid possible future mistake. Reported-by:
Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
-
David Bauer authored
When operating two VAP on a MT7610 with encryption (PSK2, SAE, OWE), only the first one to be created will transmit properly encrypteded frames. All subsequently created VAPs will sent out frames with the payload left unencrypted, breaking multicast traffic (ICMP6 NDP) and potentially disclosing information to a third party. Disable GTK offloading and encrypt these frames in software to circumvent this issue. THis only seems to be necessary on MT7610 chips, as MT7612 is not affected from our testing. Signed-off-by:
David Bauer <mail@david-bauer.net> Signed-off-by:
-
Ryder Lee authored
Simplify WTBL defines. Signed-off-by:
-
Lorenzo Bianconi authored
Get rid of the following definitions in mt7921/mcu.h since they are not actually used: - MT7921_WTBL_UPDATE_MAX_SIZE - MT7921_STA_UPDATE_MAX_SIZE - MT7921_WTBL_UPDATE_BA_SIZE Signed-off-by:
-
Nigel Christian authored
In mt7921_pm_set() the variable "ret" is initialized to zero and then returned. Remove it and return zero. Signed-off-by:
Nigel Christian <nigel.l.christian@gmail.com> Signed-off-by:
-
Lorenzo Bianconi authored
Notify mac80211 mt7921 devices support random mac address during hw frequency scanning Signed-off-by:
-
Sander Vanheule authored
EEPROM blobs for MT7613BE radios start with (little endian) 0x7663, which is also the PCI device ID for this device. The EEPROM is required for the radio to work at useful power levels, otherwise only the lowest power level is available. Suggested-by:
Georgi Vlaev <georgi.vlaev@konsulko.com> Tested-by:
Stijn Segers <foss@volatilesystems.org> Signed-off-by:
Sander Vanheule <sander@svanheule.net> Signed-off-by:
-
Felix Fietkau authored
The first pointer in the txp needs to be unmapped as well, otherwise it will leak DMA mapping entries Reported-by:
Ben Greear <greearb@candelatech.com> Fixes: 27d5c528 ("mt76: fix double DMA unmap of the first buffer on 7615/7915") Signed-off-by:
-
Felix Fietkau authored
The first pointer in the txp needs to be unmapped as well, otherwise it will leak DMA mapping entries Fixes: 27d5c528 ("mt76: fix double DMA unmap of the first buffer on 7615/7915") Signed-off-by:
-
Felix Fietkau authored
There were multiple issues in the current key set/remove code: - deleting a key with the previous key index deletes the current key - BIP key would only be uploaded correctly initially and corrupted on rekey Rework the code to better keep track of multiple keys and check for the key index before deleting the current key Signed-off-by: -
Felix Fietkau authored
For AP and Client mode, the hardware can pass received packets as 802.3 frames that can be passed to the network stack as-is. Signed-off-by: -
Felix Fietkau authored
Set skb->ip_summed to CHECKSUM_UNNECESSARY if the hardware has validated the IP and TCP/UDP checksum Signed-off-by: -
Lorenzo Bianconi authored
Enable hw rx-amsdu de-aggregation support available in 7915 devices. This is a preliminary patch to enable rx checksum offload Signed-off-by:
-
Felix Fietkau authored
Do not try to access the header when receiving them Signed-off-by: -
Arnd Bergmann authored
There are a couple of warnings in this driver when building with W=1: drivers/net/wireless/intel/iwlegacy/common.c: In function 'il_power_set_mode': drivers/net/wireless/intel/iwlegacy/common.c:1195:60: error: suggest braces around empty body in an 'if' statement [-Werror=empty-body] 1195 | il->chain_noise_data.state); | ^ drivers/net/wireless/intel/iwlegacy/common.c: In function 'il_do_scan_abort': drivers/net/wireless/intel/iwlegacy/common.c:1343:57: error: suggest braces around empty body in an 'else' statement [-Werror=empty-body] Change the empty debug macros to no_printk(), which avoids the warnings and adds useful format string checks. Signed-off-by:Arnd Bergmann <arnd@arndb.de> Acked-by:
Stanislaw Gruszka <stf_xl@wp.pl> Signed-off-by:
Kalle Valo <kvalo@codeaurora.org> Link: https://lore.kernel.org/r/20210322104343.948660-3-arnd@kernel.org
-
Lorenzo Bianconi authored
Notify mac80211 the mt7601u chipset support 802.11 TDLS. The feature has been tested with a mt7610u peer. Signed-off-by:
Jakub Kicinski <kuba@kernel.org> Signed-off-by:
-
Colin Ian King authored
Currently the expression ~nic_conf1 is always true because nic_conf1 is a u16 and according to 6.5.3.3 of the C standard the ~ operator promotes the u16 to an integer before flipping all the bits. Thus the top 16 bits of the integer result are all set so the expression is always true. If the intention was to flip all the bits of nic_conf1 then casting the integer result back to a u16 is a suitabel fix. Interestingly static analyzers seem to thing a bitwise ! should be used instead of ~ for this scenario, so I think the original intent of the expression may need some extra consideration. Addresses-Coverity: ("Logical vs. bitwise operator") Fixes: c869f77d ("add mt7601u driver") Signed-off-by:Colin Ian King <colin.king@canonical.com> Acked-by:
Jakub Kicinski <kubakici@wp.pl> Signed-off-by:
-
Dan Carpenter authored
The sscanf() function returns the number of matches (0 or 1 in this case). It doesn't return error codes. We should return -EINVAL if the string is invalid Fixes: c376c1fc ("rtw88: add h2c command in debugfs") Signed-off-by:
-
Ching-Te Ku authored
While WL scan, WL is more high priority than BT. The packets from AP will be a big interference to A2DP. It will lead to A2DP stutters. Stop answering CTS to AP to decrease AP's packets Tx while WL scan + WL busy. Enable BT AFH feature to make BT leave away from WL channel. Desired BT firmware BT-COEX version: 0x1c Desired WL firmware version: 9.9.x Signed-off-by:
Ching-Te Ku <ku920601@realtek.com> Signed-off-by:
Ping-Ke Shih <pkshih@realtek.com> Signed-off-by:
-
Po-Hao Huang authored
Fix power tracking issue by replacing unnecessary IQ calibration with LC calibration. When thermal difference exceeds limitation, let RF circuit adjsut its characteristic to fit in current environment. Signed-off-by:
Po-Hao Huang <phhuang@realtek.com> Signed-off-by:
-
Zong-Zhe Yang authored
update tx power limits to RF v40 and apply fix on FCC's channel 12 and 13, so change the patch level to 1. Signed-off-by:
Zong-Zhe Yang <kevin_yang@realtek.com> Signed-off-by:
-
Zong-Zhe Yang authored
DIG min is expected to be set according to chip. And actually we assigned it under each chip information. However, we didn't use the setting when we did DIG process. It is unexpected, so we fix it. Signed-off-by:
-
Zong-Zhe Yang authored
Though mac queue flushing has been supported, sometimes data may be waiting on interface from host to chip. If it occurs, there may still be data that flows into mac just after we do flush. To avoid that, we add the hci part of flushing. Signed-off-by:
-
- 15 Mar, 2021 5 commits
-
-
Zong-Zhe Yang authored
Although FW crash logs are already supported for dumping in driver, the logs may not be sufficient to analyze field issues. To improve this part, we add a support to dump FW memory. When driver receives FW crash notifications, driver uses DDMA, which is a HW ability, to copy specified FW memory to FW fifo. Driver can then dump these information from FW fifo. With this support, not only FW crash log but also specified FW memory will be dumped while FW crash. Besides, specified registers are also dumped. This feature is implemeted on 8822C first due to difference of FW layouts between ICs. In addition, we add a debugfs to trigger FW crash. It can simulate the process of crash, dump and reset. Through it, we can also check if a reset is ongoing. Signed-off-by:
-
Ajay Singh authored
Use same cookie for request_threaded_irq() & free_irq() to properly free IRQ during module unload. free_irq() already uses *wilc* handler so the changes are required for request_threaded_irq(). Signed-off-by:
Ajay Singh <ajay.kathat@microchip.com> Signed-off-by:
-
Marcus Folkesson authored
Write the value instead of reading it twice. Fixes: c5c77ba1 ("staging: wilc1000: Add SDIO/SPI 802.11 driver") Signed-off-by:
Marcus Folkesson <marcus.folkesson@gmail.com> Signed-off-by:
-
David Mosberger-Tang authored
chip_allow_sleep() only supported wakeup via SDIO, which made the driver unusable over SPI. This code is a straight forward port from the driver in the linux-at91 repository. Signed-off-by:
David Mosberger-Tang <davidm@egauge.net> Acked-by:
-
wengjianfeng authored
The variable ret is defined at the beginning and initialized to 0 until the function returns ret, and the variable ret is not reassigned.Therefore, we do not need to define the variable ret, just return 0 directly at the end of the function. Signed-off-by:
wengjianfeng <wengjianfeng@yulong.com> Signed-off-by:
-
