- 31 Jul, 2018 18 commits
-
-
Colin Ian King authored
Pointer 'p' is being assigned but is never used hence it is redundant and can be removed. Also re-work if statement to remove redundant assignment of p. Cleans up clang warning: warning: variable 'p' set but not used [-Wunused-but-set-variable] Signed-off-by:
Colin Ian King <colin.king@canonical.com> Signed-off-by:
Kalle Valo <kvalo@codeaurora.org>
-
Colin Ian King authored
Variables id, unicast, write, conf, a_band, accum_tx and ucode are assigned a value but it is never read, hence they are redundant and can be removed. Cleans up clang warnings: warning: variable 'id' set but not used [-Wunused-but-set-variable] warning: variable 'unicast' set but not used [-Wunused-but-set-variable] warning: variable 'write' set but not used [-Wunused-but-set-variable] warning: variable 'conf' set but not used [-Wunused-but-set-variable] warning: variable 'a_band' set but not used [-Wunused-but-set-variable] warning: variable 'tx' set but not used [-Wunused-but-set-variable] warning: variable 'accum_tx' set but not used [-Wunused-but-set-variable] warning: variable 'ucode' set but not used [-Wunused-but-set-variable] Signed-off-by:
Stanislaw Gruszka <sgruszka@redhat.com> Signed-off-by:
-
Colin Ian King authored
Variables len, ret, reason and crypt are assigned values that are never read, hence they are redundant and can be removed. Note: For the variable ret, a return code is being assigned, but this is not returned and 0 is currently being returned, I believe this is OK. Cleans up clang warnings: warning: variable 'len' set but not used [-Wunused-but-set-variable] variable 'ret' set but not used [-Wunused-but-set-variable] warning: variable 'reason' set but not used [-Wunused-but-set-variable] warning: variable 'crypt' set but not used [-Wunused-but-set-variable] Signed-off-by:
-
Colin Ian King authored
Variables len and dev assigned values but are never used hence they are redundant and can be removed. Also add in white space in for-loop and remove some { } braces on if statement. Cleans up clang warnings: warning: variable 'len' set but not used [-Wunused-but-set-variable] warning: variable 'dev' set but not used [-Wunused-but-set-variable] Signed-off-by: -
Amol Hanwate authored
common->init_done flag should set after basic initialization. Hence, moving init_done flag at end of rsi_91x_init(). Signed-off-by:
Amol Hanwate <amol.hanwate@redpinesignals.com> Signed-off-by:
Siva Rebbagondla <siva.rebbagondla@redpinesignals.com> Signed-off-by:
-
Amol Hanwate authored
rx_params are getting updated two times in driver, which is not required. Hence, removing duplicate updation of rx_params from rsi_prepare_skb(). Signed-off-by:
-
Amol Hanwate authored
Currently, RSI_NEEDED_HEADROOM is '80' for rsi driver, which is wrong. As per rsi internal frame format, the RSI_NEEDED_HEADROOM shall be '84', which is 64(dword_align) + 4(extended_desc) + 16(frame_desc). Hence, corrected the needed headroom. Signed-off-by:
-
Siva Rebbagondla authored
Currently, AP mode will work on only WLAN alone firmware. To give support for AP and BT dual mode, adding firmware entry in 'struct ta_metadata'. The firmware entry is based on what coex_mode is used in driver and coex mode '4' for all AP+BT related functionalities. Hence, added the same. Signed-off-by:
-
Siva Rebbagondla authored
while cleaning up the driver, observed that flash_content pointer is not necessary in rsi_load_firmware(). Instead of this, driver can use 'fw_entry->data' directly.Hence, removed redundant flash_content pointer. Signed-off-by:
-
Siva Rebbagondla authored
Removing redundant device id's from both usb and sdio idtables, as rsi driver currently supporting only one module(RS9113). Also, replaced ids with specific defines. Signed-off-by:
-
Siva Rebbagondla authored
Due to multiple calls of add interface routine, vif is getting duplicated and at certain instance, we are out of vifs, causing the driver to behave abnormal. Fix: Every vif has a unique mac-id, when we got a vif with same mac-id as the previous id's, we will override the respective vif. Signed-off-by:
Sushant Kumar Mishra <sushant.mishra@redpinesignals.com> Signed-off-by:
-
Siva Rebbagondla authored
System is unable to wake-up through magic-packet in secured connections. Because key descriptor is getting corrupted and firmware is unable to decrypt the magic packet. Fixed the issue by properly preparing it before sending it to firmware. Signed-off-by:
-
Ganapathi Raju authored
Radio capabilities packet is not prepared properly for 40MHZ case, ppe_ack_rate is a two byte variable which is initialized wrongly, hence we are unable to connect, resolved by assigning it properly. Signed-off-by:
Ganapathi Raju <ganapathi.kondraju@redpinesignals.com> Signed-off-by:
-
Siva Rebbagondla authored
observed low throughput rates during verification. This is because, QoS enable flag is overridden by sequence number in the data descriptor frame. Hence, added the fix for same. Signed-off-by:
-
Roman Stratiienko authored
Mwifiex firmware inserts only Microsoft information element Allow other vendor specific IEs to pass from userspace Signed-off-by:
Roman Stratiienko <roman.stratiienko@globallogic.com> Signed-off-by:
-
Ganapathi Bhat authored
Driver is using boolean variables to maintain vairous status information of adapter. These status variables are accessed by multiple threads and there is a possibility of a race. To avoid this, convert these variables to a set of bitops flags, to be operated atomically. Below variables of mwifiex_adapter are converted to bitop flags: surprise_removed is_cmd_timedout is_suspended is_hs_configured hs_enabling Signed-off-by:
Ganapathi Bhat <gbhat@marvell.com> Signed-off-by:
-
Ganapathi Bhat authored
Driver must ensure that whenever it holds a pointer to the list entry mwifiex_rx_reorder_tbl, it must protect the same with rx_reorder_tbl_lock. At present there are many places where driver does not ensure this. To cover all cases, spinlocks in below funcions are moved out and made sure that the caller will hold the spinlock: mwifiex_11n_dispatch_pkt_until_start_win() mwifiex_11n_scan_and_dispatch() mwifiex_del_rx_reorder_entry() mwifiex_11n_get_rx_reorder_tbl() mwifiex_11n_find_last_seq_num() Signed-off-by:
-
Ganapathi Bhat authored
At present driver spinlock protects iteration of list rx_reorder_tbl_ptr with rx_reorder_tbl_lock. To protect the individual items in this list, it uses rx_pkt_lock. But, we can use a single rx_reorder_tbl_lock for both purposes. This patch replaces rx_pkt_lock by rx_reorder_tbl_lock. Signed-off-by:
-
- 30 Jul, 2018 7 commits
-
-
Lorenzo Bianconi authored
Use 11M as tx rate for CCK protection control frames Fixes: 7bc04215 ("mt76: add driver code for MT76x2e") Signed-off-by:
Lorenzo Bianconi <lorenzo.bianconi@redhat.com> Signed-off-by:
-
Lorenzo Bianconi authored
Introduce mac80211 rts threshold handler in order to add frame protection support to mt76x2 driver Signed-off-by:
-
Davide Caratti authored
Set MFP_CAPABLE bit in hw flag capabilities exported by the driver Signed-off-by:
Davide Caratti <dcaratti@redhat.com> Signed-off-by:
-
Lorenzo Bianconi authored
Fall back to software encryption for hw unsupported ciphers in order to enable 802.11w Signed-off-by:
-
Andrey Shevchenko authored
Enable support for multiple SSIDs scans. Get max number of supported SSIDs from firmware and report to cfg80211 core. Signed-off-by:
Andrey Shevchenko <ashevchenko@quantenna.com> Signed-off-by:
-
Sergei Maksimenko authored
Implement set_power_mgmt() callback that forwards power saving settings to the device firmware. Signed-off-by:
Sergei Maksimenko <smaksimenko@quantenna.com> Signed-off-by:
-
Kalle Valo authored
Merge tag 'iwlwifi-next-for-kalle-2018-07-26' of git://git.kernel.org/pub/scm/linux/kernel/git/iwlwifi/iwlwifi-next First batch of iwlwifi patches for 4.19 * Implement 802.11ax D2.0; * Changes in the base code to support the new 22560 devices; * Support for the new 22560 device family;
-
- 26 Jul, 2018 15 commits
-
-
Golan Ben Ami authored
The smallest rb size supported today is 4k rx buffers. 22560 devices use 2k rxb's, so allow using 2k buffers. Signed-off-by:
Golan Ben Ami <golan.ben.ami@intel.com> Signed-off-by:
Luca Coelho <luciano.coelho@intel.com>
-
Golan Ben Ami authored
In 22560 devices the firmware will do all the hw configurations, but that's not ready yet. Update the correct registers in the driver until the FW is ready and does it by itself. Signed-off-by:
-
Golan Ben Ami authored
22560 devices RFH uses different structures, which act similar to the legacy rxq management lists - free and used list. The iwl_rx_transfer_desc struct is part of the free list, and consists of pointers to the empty rb's the driver wants to pass to the fw. The iwl_rx_completion_desc struct is part of the used list, and consists of pointers to the buffer the fw filled up with new rx, both commands and data, for the host. Signed-off-by:
-
Golan Ben Ami authored
22560 devices use a new tx cmd api. Update the code to use the new api. Signed-off-by:
-
Golan Ben Ami authored
22560 devices tfd queue max size is 2^16. Allow a configurable max size in the driver for supporting different devices. Signed-off-by:
-
Golan Ben Ami authored
Nowadays, the tfd queue max size is 2^8, and the reserved size in the command header sequence field for the tfd entry index is 8 bits, allowing an injective function from the hw pointers to the tfd entry index in the sequence field. In 22560 devices the tfd queue max size is 2^16, meaning that the hw pointers are 16 bit long (allowing to point to each entry in the tfd queue). However, the reserved space in the sequence field for the tfd entry doesn't change, and we are limited to 8 bit. This requires cancelling the injective function from hw pointer to tfd entry in the sequence number. Use iwl_pcie_get_cmd_index to wrap the hw pointer's to the n_window size, which is maximum 256 in tx queues, and so, keep the injective function between the window wrapped hw pointers to tfd entry index in the sequence. Signed-off-by:
-
Golan Ben Ami authored
In 22560 devices the ROM sendis an interrupt to the host once the IML reading is done. Handle this interrupt, and indicate sw error in case the value is fail. Additionally, the cause for sw error in 22560 devices have been changed, so update the cause list. Signed-off-by:
-
Golan Ben Ami authored
In 22560 devices we can start debug using context info gen3. Configure the fw to start collecting logs to the dram before init. Signed-off-by:
-
Golan Ben Ami authored
For devices which use the image loader image, the length of the frame must be updated in the byte count in bytes, and not dwords as today. Avoid dividing the input length by 4. Signed-off-by:
-
Golan Ben Ami authored
Context information structure was added to 22000 devices for firmware self init. In the next generation of devices the context information changes significantly, and the original context information is divided roughly to three data structures: context information gen3, prph information and prph scratch. In addition, the init flow changes so the firmware is loaded by the IML, and so we must allocate the IML on the DRAM and give the ROM the IML's address before kicking the firmware's self init. Signed-off-by:
-
Golan Ben Ami authored
The hw now refers to two new blocks: * rx tr tail - The Tail index on the free buffers queue TR, which is update by the device after reading the free buffer from the tr. * rx cr tail - Updated by the driver when completing processing a new completion descriptor in the cr. Add these two new struct to the rxq, allocate and free them when needed. In addition, the register for rx write pointer had been changed to HBUS_TARG_WRPTR. The way to differentiate tx from rx is the queue number. TX range is 0-511, and RX's is 512-527. Signed-off-by:
-
Golan Ben Ami authored
Device 22560 have many different hw and sw features than 22000 family, so introduce a new family of devices - 22560. Signed-off-by:
-
Golan Ben Ami authored
Different device families may have different flag values for passing a message to the fw (i.e. SW_RESET). In order to keep the code readable, and avoid conditioning upon the family, store a value for each flag, which indicates the bit that needs to be enabled. Additionally, support 22560 device csr flags and addresses. Signed-off-by:
-
Luca Coelho authored
Add support for the new 22560 family of devices and, while at it, reorganize the 22000 family so it fits better with the new one. Signed-off-by:
-
Johannes Berg authored
Our current HE devices don't support BlockAck with the large bitmap, so can't do TX aggregation with 256 frames. Restrict to the lower HT size. Signed-off-by:
Johannes Berg <johannes.berg@intel.com> Signed-off-by:
-
