- 14 Jul, 2014 1 commit
-
-
Janusz Dziedzic authored
Allow to setup maximum subframes for AMSDU and AMPDU aggregation via debugfs htt_max_amsdu_ampdu file. Eg. echo "2 64" > htt_max_amsdu_ampdu will setup maximum amsdu subframes equal 2 and maximum ampdu subframes equal to 64. Signed-off-by:
Janusz Dziedzic <janusz.dziedzic@tieto.com> Signed-off-by:
Kalle Valo <kvalo@qca.qualcomm.com>
-
- 02 Jun, 2014 2 commits
-
-
Janusz Dziedzic authored
Print Kconfig options enabled/disabled in the build. Signed-off-by:
-
Kalle Valo authored
These len variables are not used anywhere. Signed-off-by:
-
- 30 May, 2014 3 commits
-
-
Vladimir Kondratiev authored
Print message if no events received. This should not happen. If it is, it points to the problem in firmware. Track also cases when multiple events processed in one IRQ Print information as soon as possible - mbox pointers and event header right after reading it. This helps to identify potential problem with memory allocation for the event buffer. Signed-off-by:
Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com> Signed-off-by:
John W. Linville <linville@tuxdriver.com>
-
-
git://git.kernel.org/pub/scm/linux/kernel/git/sameo/nfc-nextJohn W. Linville authored
Samuel Ortiz <sameo@linux.intel.com> says: "NFC: 3.16: Second pull request This is the 2nd NFC pull request for 3.16. We have: - Felica (Type3) tags support for trf7970a - Type 4b tags support for port100 - st21nfca DTS typo fix - A few sparse warning check fixes" Signed-off-by:
-
- 29 May, 2014 34 commits
-
-
Paul Bolle authored
A check for CONFIG_AUTOSUSPEND was included in this driver when it was added in v2.6.39. But that Kconfig symbol doesn't exist. Remove that check and the single line it hides. Signed-off-by:
Paul Bolle <pebolle@tiscali.nl> Signed-off-by:
-
Sachin Kamat authored
fw_common.h was included twice. Signed-off-by:
Sachin Kamat <sachin.kamat@linaro.org> Signed-off-by:
-
Sachin Kamat authored
phy.h was included twice. Signed-off-by:
-
Rajkumar Manoharan authored
Currently mac80211 does not support WDS and DFS with channel context drivers. So advertise these features only when the driver is not supporting channel context and modparam "use_chanctx" is introduced for preparing channel context support in ath9k. Signed-off-by:
Rajkumar Manoharan <rmanohar@qti.qualcomm.com> Signed-off-by:
-
Bing Zhao authored
If host sleep parameter gap is set to 0xff, firmware will wait for an ack from host to confirm the success of host wakeup. This prevents firmware from uploading data packet before host actually wakes up. Signed-off-by:
Bing Zhao <bzhao@marvell.com> Signed-off-by:
-
Avinash Patil authored
Current implementation sets tdls_link flag only while restoring packets from TDLS queue. If traffic to peer starts after TDLS is setup, there is no way to set TDLS link flag to true. Do this while creating RA list and we confirm that there exist a TDLS peer for which setup is complete. Signed-off-by:
Avinash Patil <patila@marvell.com> Signed-off-by:
-
Avinash Patil authored
[113.967694] Unable to handle kernel NULL pointer dereference at virtual address 00000020 ............ [113.967859] PC is at mwifiex_update_rxreor_flags+0xfc/0x430 ............ [113.968110] mwifiex_update_rxreor_flags+0xfc/0x430 [113.968129] mwifiex_handle_event_ext_scan_report+0x1e4/0x21c [113.968148] mwifiex_process_sta_event+0x410/0x508 [113.968165] mwifiex_process_event+0x184/0x1e0 [113.968181] mwifiex_main_process+0x220/0x48c [113.968197] mwifiex_sdio_interrupt+0xc8/0x1cc [113.968210] sdio_irq_thread+0x11c/0x290 In case of legacy scan, adapter->curr_cmd is guranteed to be non-NULL in check_next_scan_cmd. This may not be case in extended scan where scan command response would come earlier and set curr_cmd to NULL. Extended scan event comes later and while trying to complete IOCTL for scan, driver would crash in dereferencing adapter->curr_cmd->wait_q_enabled. Avoid this by completing IOCTL in case of legacy scans only. Internal scan would be completed while handling extended scan command response. Signed-off-by: -
Rafał Miłecki authored
Broadcom's wl 6.30.223.141 has some optimizations for radios 0x205[67]. Signed-off-by:
Rafał Miłecki <zajec5@gmail.com> Signed-off-by:
-
Rafał Miłecki authored
Signed-off-by:
-
Rafał Miłecki authored
PHY has to be often re-initialized (e.g. during band switching after PHY reset), however some operations have to be performed only once (only power reset affects them). Signed-off-by:
-
Vladimir Kondratiev authored
In case of receiving frame with sequence number far greater than current, wil_release_reorder_frames() will iterate many times over empty buffer. Optimize this case by checking buffer emptiness and simply update head_seq_num without iterating. Suggested-by:
Vladimir Shulman <Vladimir.Shulman@Wilocity.com> Signed-off-by:
-
Vladimir Kondratiev authored
use proper format %pad for the dma_addr_t arguments; prefix %p with 0x, as %p don't print is by itself Signed-off-by:
-
Vladimir Kondratiev authored
If scan has not finished in some reasonable time (10sec), interpret it as if firmware error occurs but was not reported. Firmware should report scan completion for every scan request, so it is error condition indeed. Perform firmware recovery procedure. Signed-off-by:
-
Vladimir Kondratiev authored
Provide clear definition of the watermarks for the vring descriptor space. Signed-off-by:
-
Vladimir Kondratiev authored
In case there is something fundamentally wrong with the firmware (example: RF cable disconnected), FW will always crash immediately after reset. This leads to infinite fw error recovery loop. Count consecutive unsuccessful error recovery attempts in a short period of time, and stop doing recovery after some reasonable count. It is still possible to manually reset fw doing interface down/up sequence. Signed-off-by:
-
Vladimir Kondratiev authored
Obey 802.11 spec that defines max. data size 7920 bytes Signed-off-by:
-
Rajkumar Manoharan authored
pm_lock is taken twice while syncing HW TSF of p2p vif. Fix this by taking the lock at caller side. Cc: Felix Fietkau <nbd@openwrt.org> Signed-off-by:
-
Arend van Spriel authored
The USB driver was using a list for firmware info that was used in suspend/resume scenario. Now that brcmfmac is using the asynchronous firmware request this is no longer needed. Reviewed-by:
Hante Meuleman <meuleman@broadcom.com> Reviewed-by:
Franky (Zhenhui) Lin <frankyl@broadcom.com> Reviewed-by:
Daniel (Deognyoun) Kim <dekim@broadcom.com> Reviewed-by:
Pieter-Paul Giesberts <pieterpg@broadcom.com> Signed-off-by:
Arend van Spriel <arend@broadcom.com> Signed-off-by:
-
Hante Meuleman authored
The USB bus driver always configured an USB intr EP urb. The driver did not use the result at all and with newer firmware it is causing continues errors on this EP. Reviewed-by:
-
Daniel Kim authored
The max buffer size for receiving control message from dongle needs to be increased considering possible block padding. Otherwise some big control message can't be received due to buffer overrun check. Reviewed-by:
-
Arend van Spriel authored
The firmware channel specification is a bitfield using a 16-bit integer, but only 14 lsb are used. Upon encoding this value assure all 16 bits are cleared. Reviewed-by:
-
Arend van Spriel authored
The function brcmf_fw_nvram_strip() is no longer called so it does not need to be exposed. Reviewed-by:
-
Arend van Spriel authored
This patch adds use of asynchronous firmware request to the driver USB layer. Reviewed-by:
-
Arend van Spriel authored
This patch adds use of asynchronous firmware request to the driver SDIO layer. Reviewed-by:
-
Arend van Spriel authored
The driver needs firmware to be loaded to the device, which is done through the firmware class API. The synchronous call request_firmware() need root filesystem to be mounted and/or user-mode helper. These may not be avaliable on the moment it is called. Instead use request_firmware_nowait(). Reviewed-by:
-
Hante Meuleman authored
Reviewed-by:
-
Arend van Spriel authored
The resume callbacks do partly the same a the probe callback so put common code in separate function for use in the callbacks. This also fixes suspend/resume regression introduced by brcmfmac: remove .init() callback for internal bus interface The .init() callback was the first function called by the common bus function brcmf_bus_start(). Given that it is not really necessary and the bus layer can call it before calling the brcmf_bus_start() function. Reviewed-by: -
Arend van Spriel authored
The function brcmf_detach() checks whether it needs to do his stuff or can return immediately. No need to have the same check in the calling code. Reviewed-by:
-
Arend van Spriel authored
The firmware processing will be modified to use asynchronous request firmware api. In preparation this patch is simple rename of source and header file to which the functionality will be added. Reviewed-by:
-
Arend van Spriel authored
The .init() callback was the first function called by the common bus function brcmf_bus_start(). Given that it is not really necessary and the bus layer can call it before calling the brcmf_bus_start() function. Reviewed-by:
-
Arend van Spriel authored
The function chandef_to_chanspec() was added by brcmfmac: determine chanspec from struct cfg80211_chan_def info The struct cfg80211_chan_def contains additional info to derive the bandwidth and side-band information of the chanspec. This patch adds chandef_to_chanspec() function used in IBSS join and starting AP operation. However, it introduced a sparse warning because the function is only called from within the source file wl_cfg80211.c. Reported-by:Fengguang Wu <fengguang.wu@intel.com> Reviewed-by:
-
Arend van Spriel authored
Before informing cfg80211 about the scan status the device should be put back in mpc state. If done after user-space may initiate another (scheduled) scan and fail because scan is still busy as shown in logging below: [ 3301.367376] brcmfmac: brcmf_fweh_event_worker event ESCAN_RESULT (69) [ 3301.377305] brcmfmac: brcmf_fweh_event_worker version 2 flags 0 status 0 [ 3301.384993] brcmutil: event payload, len=12 [ 3301.389208] 00000000: 0c 00 00 00 6d 00 00 00 34 12 00 00 [ 3301.389214] brcmfmac: brcmf_sdio_kso_control Enter: on=0 [ 3301.402196] brcmfmac: brcmf_inform_bss scanned AP count (0) [ 3301.407808] brcmfmac: brcmf_notify_escan_complete Enter [ 3301.413064] brcmfmac: brcmf_notify_escan_complete ESCAN Completed scan: Done [ 3301.420137] brcmfmac: brcmf_sdio_bus_txctl Enter [ 3301.420368] brcmfmac: brcmf_cfg80211_sched_scan_start Enter [ 3301.420370] brcmfmac: brcmf_cfg80211_sched_scan_start: Scanning already: status (1) [ 3301.440190] brcmfmac: brcmf_sdio_kso_control Enter: on=1 [ 3301.448695] brcmfmac: brcmf_sdio_tx_ctrlframe Enter [ 3301.453662] brcmfmac: brcmf_sdio_bus_rxctl Enter [ 3301.458326] brcmfmac: brcmf_sdio_isr Enter [ 3301.462523] brcmfmac: brcmf_sdio_dpc Enter [ 3301.466632] brcmfmac: brcmf_sdio_readframes Enter [ 3301.471431] brcmfmac: brcmf_sdio_read_control Enter [ 3301.476340] brcmfmac: brcmf_set_mpc MPC : 1 Reviewed-by:
-
Arend van Spriel authored
When a regular scan does not return any networks user-space does request a scheduled scan without any matchset or ssid. This can not be handled by the firmware so we return -EINVAL. However, as this request is done let us not add an error message to the log. Reviewed-by:
-
Benoit Taine authored
This issue was reported by coccicheck using the semantic patch at scripts/coccinelle/api/memdup.cocci Signed-off-by:
Benoit Taine <benoit.taine@lip6.fr> Signed-off-by:
-
