- 13 Sep, 2019 32 commits
-
-
git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/ath.gitKalle Valo authored
ath.git patches for 5.4. Major changes: wil6210 * add support for Enhanced Directional Multi-Gigabit (EDMG) channels 9-11 * add debugfs file to show PCM ring content * report boottime_ns in scan results ath9k * add a separate loader for AR92XX (and older) pci(e) without eeprom, enabled with the new ATH9K_PCI_NO_EEPROM Kconfig option
-
Yan-Hsuan Chuang authored
Report chains and chain_signal in ieee80211_rx_status. It is useful for program such as tcpdump to see if the antennas are well connected/placed. 8822C is able to receive CCK rates with 2 antennas, while 8822B can only use 1 antenna path to receive CCK rates. Signed-off-by:
Yan-Hsuan Chuang <yhchuang@realtek.com> Reviewed-by:
Chris Chiu <chiu@endlessm.com> Signed-off-by:
Kalle Valo <kvalo@codeaurora.org>
-
Tsang-Shian Lin authored
Fix the wrong RF path for CCK rx power calculation. Fixes: e3037485 ("rtw88: new Realtek 802.11ac driver") Signed-off-by:
Tsang-Shian Lin <thlin@realtek.com> Signed-off-by:
-
Michael Straube authored
Function _rtl92d_evm_db_to_percentage is functionally identical to the generic version rtl_evm_db_to_percentage, so remove _rtl92d_evm_db_to_percentage and use the generic version instead. Signed-off-by:
Michael Straube <straube.linux@gmail.com> Signed-off-by:
-
Michael Straube authored
Function _rtl92c_evm_db_to_percentage is functionally identical to the generic version rtl_evm_db_to_percentage, so remove _rtl92c_evm_db_to_percentage and use the generic version instead. Signed-off-by:
-
Michael Straube authored
Function _rtl92c_evm_db_to_percentage is functionally identical to the generic version rtl_evm_db_to_percentage, so remove _rtl92c_evm_db_to_percentage and use the generic version instead. Signed-off-by:
-
Yan-Hsuan Chuang authored
The hardware has enough buffer to receive like 8K for an MPDU. So tell mac80211 that we can receive AMSDU in AMPDU. Signed-off-by:
-
Tzu-En Huang authored
This mechanism reduces the numbers of false alram in cck rate by dynamically adjusting the value of power threshold and cs_ratio. We determine the new value by three factors, which are rssi, false alarm count and igi. Based on these factors, we define the current condition into five levels. Compared to the previous level, if the level is changed, we set the new values for power threshold and cs_ratio. Signed-off-by:
Tzu-En Huang <tehuang@realtek.com> Signed-off-by:
-
Yan-Hsuan Chuang authored
Since 8822c requires to do not only IQK, but also DPK. Move these calibrations that need to be done once the channel is determined, into phy_calibration. And note that the order of the calibrations matters, 8822c should do IQK first, then DPK. Signed-off-by:
-
Tzu-En Huang authored
Power amplifiers are not linear components, and require DPK to reduce its nonlinearity. DPK is called Digital Pre-distortion Calibration, can be used to compensate the output of power. DPK tracking is in charge of tracking the thermal changes. And it then shifts the power curve accordingly, which makes the power output remains linear even if the PA works in different temperature. To perform DPK, the parameter table should also be updated. And the table will be applied when device is powered on. Then DPK will reference the values to calibrate. Signed-off-by:
-
Yan-Hsuan Chuang authored
Add support for doing IQK in firmware Ideally the RF component's I/Q vectors should be orthogonal, but usually they are not. So we need to calibrate for the RF components, ex. PA/LNA, ADC/DAC. And if the I/Q vectors are more orthogonal, the mixed signal will have less deviation. This helps with those rates with higher modulation (MCS8-9), because they have more strict EVM/SNR requirement. Also the better of the quality of the signal, the longer it can propagate, and the better throughput performance we can get. Signed-off-by:
-
Tsang-Shian Lin authored
Enable 8822C Tx/Rx interrupt migration. In some platforms, performance test may cause heavy cpu loading and get bad results. Interrupt migration can decrease the amount of interrupts, and lower cpu loading. Signed-off-by:
-
Chin-Yen Lee authored
update sequence to v13 to reduce power consumption when MAC power off Signed-off-by:
Chin-Yen Lee <timlee@realtek.com> Signed-off-by:
-
Yan-Hsuan Chuang authored
Update PHY hardware parameters to v38. Signed-off-by:
-
Larry Finger authored
rtlwifi: rtl8188ee: rtl8192ce: rtl8192de: rtl8723ae: rtl8821ae: Remove some unused bit manipulation macros Each of these drivers defines some device to host macros that are never used, thus they can be removed. Signed-off-by:
Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by:
-
Larry Finger authored
In this step, the read/write routines for the descriptors are converted to use __le32 quantities, thus a lot of casts can be removed. Callback routines still use the 8-bit arrays, but these are changed within the specified routine. The macro that cleared a descriptor has now been converted into an inline routine. Signed-off-by:
-
Larry Finger authored
As a first step in the conversion, the macros that set the RX and TX descriptors are converted to static inline routines, and the names are changed from upper to lower case. To minimize the changes in a given step, the input descriptor information is left as as a byte array (u8 *), even though it should be a little-endian word array (__le32 *). That will be changed in the next patch. Several places where checkpatch.pl complains about a space after a cast are fixed. Signed-off-by:
-
Larry Finger authored
This driver uses a set of local macros to manipulate the RX and TX descriptors, which are all little-endian quantities. These macros are replaced by the bitfield macros le32p_replace_bits() and le32_get_bits(). In several places, the macros operated on an entire 32-bit word. In these cases, a direct read or replacement is used. Signed-off-by:
-
Larry Finger authored
iAs the first step in converting from macros that get/set information in the RX and TX descriptors, unused macros are being removed. Signed-off-by:
-
Larry Finger authored
In this step, the read/write routines for the descriptors are converted to use __le32 quantities, thus a lot of casts can be removed. Callback routines still use the 8-bit arrays, but these are changed within the specified routine. The macro that cleared a descriptor has now been converted into an inline routine. Signed-off-by:
-
Larry Finger authored
As a first step in the conversion, the macros that set the RX and TX descriptors are converted to static inline routines, and the names are changed from upper to lower case. To minimize the changes in a given step, the input descriptor information is left as as a byte array (u8 *), even though it should be a little-endian word array (__le32 *). That will be changed in the next patch. Several places where checkpatch.pl reports lines too long are fixed. Signed-off-by:
-
Larry Finger authored
This driver uses a set of local macros to manipulate the RX and TX descriptors, which are all little-endian quantities. These macros are replaced by the bitfield macros le32p_replace_bits() and le32_get_bits(). In several places, the macros operated on an entire 32-bit word. In these cases, a direct read or replacement is used. Signed-off-by:
-
Larry Finger authored
As the first step in converting from macros that get/set information in the RX and TX descriptors, unused macros are being removed. Signed-off-by:
-
Lubomir Rintel authored
With the commit e86dc1ca ("Libertas: cfg80211 support") we've lost the ability to actually set the Mesh SSID from userspace. NL80211_CMD_SET_INTERFACE with NL80211_ATTR_MESH_ID sets the mesh point interface's ssid field. Let's use that one for the Libertas Mesh operation Signed-off-by:
Lubomir Rintel <lkundrak@v3.sk> Signed-off-by:
-
Colin Ian King authored
Don't populate the array pwr_info_offset on the stack but instead make it static const. Makes the object code smaller by 207 bytes. Before: text data bss dec hex filename 26066 3000 64 29130 71ca drivers/ssb/pci.o After: text data bss dec hex filename 25763 3096 64 28923 70fb drivers/ssb/pci.o (gcc version 9.2.1, amd64) Signed-off-by:
Colin Ian King <colin.king@canonical.com> Acked-by:
Michael Büsch <m@bues.ch> Signed-off-by:
-
Luis Correia authored
Signed-off-by:
Luis Correia <luisfcorreia@gmail.com> Signed-off-by:
-
zhong jiang authored
The kernel.h macro DIV_ROUND_CLOSEST performs the computation (x + d/2)/d but is perhaps more readable. Signed-off-by:
zhong jiang <zhongjiang@huawei.com> Signed-off-by:
-
Colin Ian King authored
Arrays pwr_info_offset and sprom_sizes can be make static const rather than populating them on the stack. Shrinks object size by 236 bytes. Before: text data bss dec hex filename 11300 1320 64 12684 318c drivers/bcma/sprom.o After: text data bss dec hex filename 10904 1480 64 12448 30a0 drivers/bcma/sprom.o (gcc version 9.2.1, amd64) Signed-off-by:
-
Colin Ian King authored
The array channel_all can be make static const rather than populating it on the stack, this makes the code smaller. Also, variable place is being initialized with a value that is never read, so this assignment is redundant and can be removed. Before: text data bss dec hex filename 118537 9591 0 128128 1f480 realtek/rtlwifi/rtl8821ae/phy.o After: text data bss dec hex filename 118331 9687 0 128018 1f412 realtek/rtlwifi/rtl8821ae/phy.o Saves 110 bytes, (gcc version 9.2.1, amd64) Signed-off-by:
-
Rafał Miłecki authored
Providing a new wiphy on every PCIe reset was confusing and was causing configuration problems for some users (supplicant and authenticators). Sticking to the existing wiphy should make error recovery much simpler and more reliable. Signed-off-by:
Rafał Miłecki <rafal@milecki.pl> Acked-by:
Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by:
-
Rafał Miłecki authored
Move code allocating/freeing wiphy out of above functions. This will allow reinitializing the driver (e.g. on some error) without allocating a new wiphy. Signed-off-by:
-
Rafał Miłecki authored
This moves "ops" pointer from "struct brcmf_cfg80211_info" to the "struct brcmf_pub". This movement makes it possible to allocate wiphy without attaching cfg80211 (brcmf_cfg80211_attach()). It's required for later separation of wiphy allocation and driver initialization. While at it fix also an unlikely memory leak in the brcmf_attach(). Signed-off-by:
-
- 12 Sep, 2019 8 commits
-
-
Lior David authored
There are special kinds of FW such as WMI only which are used for testing, diagnostics and other specific scenario. Such FW is loaded during driver probe and the driver disallows enabling any network interface, to avoid operational issues. In many cases it is used to debug early versions of FW with new features, which sometimes fail on startup. Currently when such FW fails to load (for example, because of init failure), the driver probe would fail and shutdown the device making it difficult to debug the early failure. To fix this, ignore load failures in WMI only FW and allow driver probe to succeed, making it possible to continue and debug the FW load failure. Signed-off-by:
Lior David <liord@codeaurora.org> Signed-off-by:
Maya Erez <merez@codeaurora.org> Signed-off-by:
-
Lior David authored
The short frame check in wil_sring_reap_rx_edma uses skb->len which store the maximum frame length. Fix this to use dmalen which is the actual length of the received frame. Signed-off-by:
-
Lior David authored
writel_relaxed can be used in wil_debugfs_iomem_x32_set since there is a wmb call immediately after. Signed-off-by:
-
Maya Erez authored
Call cfg80211_inform_bss_frame_data to report cfg80211 on the boottime_ns in order to prevent the scan results filtering due to aging. Signed-off-by:
-
Dedy Lansky authored
Upon driver rmmod, cancel_work_sync() can be invoked on p2p.discovery_expired_work before this work struct was initialized. This causes a WARN_ON with newer kernel version. Add initialization of discovery_expired_work inside wil_vif_init(). Signed-off-by:
Dedy Lansky <dlansky@codeaurora.org> Signed-off-by:
-
Alexei Avshalom Lazar authored
cid value is not being verified in wmi_evt_delba(), verification is added. Signed-off-by:
Alexei Avshalom Lazar <ailizaro@codeaurora.org> Signed-off-by:
-
Dedy Lansky authored
Due to compiler optimization, it's possible that dr_bit (descriptor ready) is read last from the status message. Due to race condition between HW writing the status message and driver reading it, other fields that were read earlier (before dr_bit) could have invalid values. Fix this by explicitly reading the dr_bit first and then using rmb before reading the rest of the status message. Signed-off-by:
-
Ahmad Masri authored
Fix a race between cfg80211 add_key call and transmitting of 4/4 EAP packet. In case the transmit is delayed until after the add key takes place, message 4/4 will be encrypted with the new key, and the receiver side (AP) will drop it due to MIC error. Wil6210 will monitor and look for the transmitted packet 4/4 eap key. In case add_key takes place before the transmission completed, then wil6210 will let the FW store the key and wil6210 will notify the FW to use the PTK key only after 4/4 eap packet transmission was completed. Signed-off-by:
Ahmad Masri <amasri@codeaurora.org> Signed-off-by:
-
