- 26 Jan, 2018 1 commit
-
-
Felix Fietkau authored
Sending frames to mac80211 needs time, which could allow for more rx packets to end up in the DMA ring. Retry polling until there are no more frames left. Improves rx latency under load. Signed-off-by:
Felix Fietkau <nbd@nbd.name> Signed-off-by:
Kalle Valo <kvalo@codeaurora.org>
-
- 25 Jan, 2018 16 commits
-
-
git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/ath.gitKalle Valo authored
ath.git patches for 4.16. Major changes: wil6210 * add PCI device id for Talyn * support flashless device ath9k * improve RSSI/signal accuracy on AR9003 series
-
Ping-Ke Shih authored
Add 'static' or declaration to resolve the warnings, and remove two unused functions halbtc_set_macreg() and halbtc_get_macreg() exposed when they were made static. Signed-off-by:
Ping-Ke Shih <pkshih@realtek.com> Acked-by:
Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by:
-
Ramon Fried authored
wcn36xx_dxe_init() doesn't check for the return value of wcn36xx_dxe_init_descs(), release the resources in case an error ocurred. Signed-off-by:
Ramon Fried <rfried@codeaurora.org> Signed-off-by:
-
Wojciech Dubowik authored
Display per frequency calibration data in dump_modal debugfs entry including reference power, voltage, tx temperature and noise floor. Example of chain 0 of OEM card (dump from modal_eeprom): Chain 0 Freq ref volt temp nf_Cal nf_Pow rx_temp 5180 -30 0 137 0 0 0 5320 -24 0 137 0 0 0 5500 -15 0 137 0 0 0 5620 -10 0 137 0 0 0 5700 -15 0 137 0 0 0 5745 -16 0 135 0 0 0 5785 -19 0 136 0 0 0 5825 -22 0 136 0 0 0 Example of a card with calibrated noise floor. Chain 0 Freq ref volt temp nf_Cal nf_Pow rx_temp 4890 -49 0 128 -107 -97 124 5100 -23 0 128 -101 -96 124 5180 -18 0 128 -101 -96 124 5300 -12 0 128 -102 -97 124 5500 -9 0 128 -101 -97 125 5640 -17 0 128 -101 -98 124 5785 -25 0 128 -101 -98 124 5940 -33 0 128 -106 -99 124 Signed-off-by:
Wojciech Dubowik <Wojciech.Dubowik@neratec.com> Signed-off-by:
-
Wojciech Dubowik authored
AR9003 series allows to calibrate noise floor for different frequency bins. Once it's done it's possible to get more accurate rssi/signal values over whole frequency band at a given temperature. The RSSI/signal accuracy reported by calibrated RF cards improves from 6 to up to 2dB. This could be interesting for application which require good signal accuracy like roaming or mesh protocols. Signed-off-by:
-
Wojciech Dubowik authored
AR9003 devices can have calibrated noise floor values which can be used instead of hard coded one. Read them from eeprom and save interpolated value in nf limits for the current channel. Signed-off-by:
-
Wojciech Dubowik authored
AR9003 factory calibration allows to use bigger EEPROM than standard 1k without changing the default layout. Allow probing of EEPROM at alternative address if initial check for default fails. The original ar9003 eeprom ops are still be used. Signed-off-by:
-
Maya Erez authored
In order to allow using the same brd file across different 11ad chips, the address for loading the brd file should be part of the FW file, instead of the brd file. The brd file is expected to include only one section. To allow backward compatibility the driver reads the address from the brd file in case it is not included in the FW file. Signed-off-by:
Maya Erez <merez@codeaurora.org> Signed-off-by:
-
Lior David authored
Sparrow D0 chips have slightly larger mac_rgf_ext section in order to support extra features such as multiple virtual interfaces. Signed-off-by:
Lior David <liord@codeaurora.org> Signed-off-by:
-
Lior David authored
Currently when we want to bring the interface up, we first reset the device which causes the boot loader to run. Then we halt the device CPU, load FW image and resume the device CPU. There are some boot loader versions which perform redundant memory accesses even when idle. Halting the device CPU while boot loader access memory can cause the device memory controller to get stuck, the FW will fail to load and the network interface will not come up. For such boot loaders implement a workaround where we freeze the boot loader before halting the device CPU, so it will not perform any memory accesses. Signed-off-by:
-
Maya Erez authored
OTP HW vectors values are different for 40MHz and 330MHz AHB clocks. In SW reset flow, the host driver sets the clock frequency to 40MHz. In order to allow reading from the OTP, the host driver should configure the OTP HW vectors with the values that fits this clock frequency. Signed-off-by:
-
Dedy Lansky authored
Talyn device supports boot without flash. Driver detects flashless device and in this case waits for ready indication from HW machine (instead of bootloader ready indication). Also in this case, MAC address is retrieved from OTP. Signed-off-by:
Dedy Lansky <dlansky@codeaurora.org> Signed-off-by:
-
Maya Erez authored
Talyn memory has changed, areas were increased and shifted to new locations. Use the appropriate address map according to the device JTAG ID. Signed-off-by:
-
Maya Erez authored
Add Talyn JTAG ID to the list of valid IDs and identify this device as Talyn. Signed-off-by:
-
Maya Erez authored
PCIe device ID has changed in Talyn. Add this ID to wil6210_pcie_ids list to allow recognition of Talyn device. Signed-off-by:
-
Felix Fietkau authored
Sometimes the hardware will push small packets that trigger a WARN_ON in mac80211. Discard them early to avoid this issue. Reported-by:
Stijn Tintel <stijn@linux-ipv6.be> Signed-off-by:
-
- 24 Jan, 2018 23 commits
-
-
Arend Van Spriel authored
When using the firmware api it can fail simply because firmware does not like the request or it fails due to issues in the host interface. Currently, there is only a single error code which is confusing. So adding a parameter to pass the firmware error separately and in case of a firmware error always return -EBADE to user-space. Reviewed-by:
Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by:
Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by:
Franky Lin <franky.lin@broadcom.com> Signed-off-by:
Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by:
-
Arend Van Spriel authored
The protocol layer api defines callbacks for dongle commands. Although not really well documented these should only return an error code in case of an error, or 0 upon success. In the bcdc protocol it can return value above 0 and we carry a fix in the caller of the protocol layer api. This patch makes it adhere to the intent of the api as described above. Reviewed-by:
-
Sergey Matyukevich authored
Signed-off-by:
Sergey Matyukevich <sergey.matyukevich.os@quantenna.com> Signed-off-by:
-
Igor Mitsyanko authored
Events processing requires locking of bus mutex, which is also used by cfg80211 layer before calling several of cfg80211 callbacks. Since all cfg80211 callbacks in qtnfmac driver also lock bus mutex, this potentially may lead to a deadlock. Do not use bus lock for event processing. Use RTNL lock instead to serialize events and commands processing threads. Signed-off-by:
Igor Mitsyanko <igor.mitsyanko.os@quantenna.com> Signed-off-by:
-
Igor Mitsyanko authored
The function qtnf_scan_done makes use of mutexes which is wrong since it may be called from timer context. Move scan timeout handler from timer to deferred work. Signed-off-by:
-
Sergey Matyukevich authored
STA does not reconnect to the same AP after disconnect. The reason is that STA is marked as disconnected in cfg80211 disconnect callback. This is too early since in this case qtnfmac event handler skips cfg80211_disconnected call when processing disconnect event from the card. As a result, wdev is left in an inconsistent state. Signed-off-by:
-
Sergey Matyukevich authored
Validate new interface combinations using wireless core checks when new interface is added or when the type of existing interface is modified. This is performed to make sure that new interface combination is supported by hardware. As a result, invalid interface combinations are rejected early, rather than passed to hardware with sometimes unpredictable results. Signed-off-by:
-
Sergey Matyukevich authored
Update existing code handling configuration of supported interface combinations. Current implementation is not complete since it does not report multiple interface combinations which are incompatible with each other. Instead current implementation packs all the supported combinations into single entry. In fact currently qsr10g wireless card supports the following two distinct interface combinations: 1. STA/repeater: 1 STA and/or 1 AP { { .max = 1, .types = NL80211_IFTYPE_AP}, { .max = 1, .types = NL80211_IFTYPE_STA}, } 2. AP/mBSS { { .max = 8, .types = NL80211_IFTYPE_AP}, } The list of supported configuration is reported by firmware during wireless card bring-up. Communication protocol between firmware and host has been updated accordingly in order to accommodate passing multiple interface combination entries. Signed-off-by: -
Vasily Ulyanov authored
Enable reporting of qtnfmac hardware and firmware details using ethtool command. Signed-off-by:
Vasily Ulyanov <vulyanov@quantenna.com> Signed-off-by:
-
Vasily Ulyanov authored
Various bits of hardware and firmware versions are useful for debug and troubleshooting. Get more information from the wireless card. Signed-off-by:
-
Vasily Ulyanov authored
On 32-bit platforms packet counters are stored in a net_device_stats struct as unsigned long integers. As a result, after some time of network activity an overflow takes place in network packet counters. This patch makes use of new structs for holding interface statistics. Signed-off-by:
-
Vasily Ulyanov authored
Virtual interface should be deleted after calling unregister_netdevice since this function ends up with sending updown_intf command to card. Signed-off-by:
-
Vasily Ulyanov authored
TLV is used to pass ACL data to firmware in start_ap cfg80211 callback. Use the same approach in set_mac_acl cfg80211 callback. Signed-off-by:
-
Luis de Bethencourt authored
The trailing semicolon is an empty statement that does no operation. Removing it since it doesn't do anything. Signed-off-by:
Luis de Bethencourt <luisbg@kernel.org> Acked-by:
Jes Sorensen <Jes.Sorensen@gmail.com> Signed-off-by:
-
Ping-Ke Shih authored
The field spec_ver is IC's specification mask for common code to do proper process to specified IC. This commit add a field new rate ID for new generation IC. Signed-off-by:
-
Ping-Ke Shih authored
New generation ICs will support 11ac, 5G, n-NSS, etc, so we define a set of rate ID. Signed-off-by:
-
Ping-Ke Shih authored
8822be is 2x2 11ac wifi chip, so report VHT capability to mac80211. Signed-off-by:
-
Ping-Ke Shih authored
In files halbtc8822b1ant.c and halbtc8822b2ant.c that I will submit later, two undesired directives named BT_8822B_1ANT_COEX_DBG and BT_8822B_2ANT_COEX_DBG will be replaced by boolean variables. Signed-off-by:
-
Ping-Ke Shih authored
Originally, btcoex controls the antenna of combo card, but solo card is also needed to setup properly. The new ops are named with suffix '_wifi_only' opposited to original btc_ops, and new structures and definitions are also introduced. The wifi_only oly contains four ops that are initial variable, hw config, scan notify, and switch band notify. Signed-off-by:
-
Ping-Ke Shih authored
Three steps of connection procedure are scan, enter/leave IPS, auth. There is no scan between leaving IPS and sending auth, but btcoex use scan as an important clue that indicates user is going to connect. So add scan notifications in ips_notify to correct btcoex's state. Signed-off-by:
-
Ping-Ke Shih authored
The variable will be used by btcoex of 8822be, so we prepare this variable in advance. Signed-off-by:
-
Ping-Ke Shih authored
BT shares 2.4G band but not 5G band, so inform current band to btcoex to setup antenna properly. Signed-off-by:
-
Ping-Ke Shih authored
The rtlwifi newer ICs support 80M bandwidth in 5G band, so extend get_wifi_bw() to know bandwidth 80M that helps btcoex to make correct decisions. Signed-off-by:
-
