- 19 May, 2014 9 commits
-
-
Julia Lawall authored
Delete unnecessary local variable whose value is always 0 and that hides the fact that the result is always 0. A simplified version of the semantic patch that fixes this problem is as follows: (http://coccinelle.lip6.fr/) // <smpl> @r exists@ local idexpression ret; expression e; position p; @@ -ret = 0; ... when != ret = e return - ret + 0 ; // </smpl> Signed-off-by:
Julia Lawall <Julia.Lawall@lip6.fr> [arend@broadcom.com: make brcms_b_detach() a void function] Signed-off-by:
Arend van Spriel <arend@broadcom.com> Signed-off-by:
John W. Linville <linville@tuxdriver.com>
-
Rajkumar Manoharan authored
The data transmission to the power save station should be aborted immediately, whenever the station informs sleep state. Right now the frames queued into into hardware are being transmitted until the hardware detects the power save station based excessive retries of the data frames due to unacknowlegdement. Then remaining frames are returned with filetered status and might be retried later by driver or mac80211. Per WFA certification testing, AP should not send out more than two frames after processing nullfunc with PM bit set from associated station. To speed up tx filtering, the pending frames in hardware queues for given station will be aborted immediately via tx filter registers. This transmit filters can be ignored if the descriptor is having invalid destination index or clear destination mask set. Signed-off-by:
Rajkumar Manoharan <rmanohar@qti.qualcomm.com> Signed-off-by:
-
Rafał Miłecki authored
These are extra configs specific to the SSB. Lack of this dependency and CONFIG_B43=y CONFIG_B43_BUSES_BCMA=y CONFIG_SSB=m would result in: > drivers/built-in.o: In function `b43_sdio_remove': > sdio.c:(.text+0x14657f): undefined reference to `ssb_bus_unregister' > drivers/built-in.o: In function `b43_sdio_probe': > sdio.c:(.text+0x14672f): undefined reference to `ssb_bus_sdiobus_register' Reported-by:
Fengguang Wu <fengguang.wu@intel.com> Signed-off-by:
Rafał Miłecki <zajec5@gmail.com> Signed-off-by:
-
Rafał Miłecki authored
Some devices may have different features despite sharing the same ID (e.g. PCI ID). For example 14e4:4331 is usually a dual band, but this can be "limited". Device with "pci/x/y/devid=0x4332" supports 2.4 GHz only. Similarly 0x4333 will mean support for 5 GHz only. Add entry in SPROM so info described above can be extracted and stored. Signed-off-by:
Hauke Mehrtens <hauke@hauke-m.de> Signed-off-by:
-
Rafał Miłecki authored
We always operate on one core and simple band switch doesn't require full core reset. Simply reset the PHY. Signed-off-by:
-
Rafał Miłecki authored
Use separated function for taking PHY out of reset and implement reset for BCMA. Signed-off-by:
-
Rafał Miłecki authored
Signed-off-by:
-
Rafał Miłecki authored
There are two groups of init values. The first one has to be uploaded once per wireless core reset but the second one on every band switch. To implement band switching in an optimal way allow uploading band init values only (by using a separated function). Signed-off-by:
-
Peter Wu authored
Unused as configure_filter takes care of setting/clearing RCR_AAP. In commit "rtlwifi: rtl8723be: rtl8723com: Remove unused allow_all_destaddr functions", Larry Finger removed allow_all_destaddr from the struct. This commit removes the related function too. Signed-off-by:
Peter Wu <peter@lekensteyn.nl> Signed-off-by:
-
- 16 May, 2014 10 commits
-
-
Joe Perches authored
These should not have trailing semicolons so remove them. Signed-off-by:
Joe Perches <joe@perches.com> Signed-off-by:
-
Avinash Patil authored
While sending null packet from driver we are passing NULL tx_param pointer to indicate there are no more packets in queue. PCIe send routine assumes caller has done sanity check on tx_param and may cause crash while dereferencing next_pkt_len from tx_param. Avoid this by passing tx_param structure with next_pkt_len as zero instead of NULL pointer. Signed-off-by:
Avinash Patil <patila@marvell.com> Signed-off-by:
Bing Zhao <bzhao@marvell.com> Signed-off-by:
-
Avinash Patil authored
During SDIO MP aggregation, we at first acquire current write port to write data onto and then proceed with using this port for SDIO write. If error occurs later in mwifiex_write_data_sync because device is suspended or SDIO write failure, we do not restore current write port and write bitmap. This results into leaking one port and hole in SDIO write port bitmap. Restore current write port and reset bitmap accordingly in failure cases to avoid this. Reported-by:
James Cameron <quozl@laptop.org> Signed-off-by:
-
Janusz Dziedzic authored
Add initial values for JP DFS pattern detector. Signed-off-by:
Janusz Dziedzic <janusz.dziedzic@tieto.com> Signed-off-by:
-
Janusz Dziedzic authored
Add initial values for DFS FCC pattern detector. Signed-off-by:
Bartosz Markowski <bartosz.markowski@tieto.com> Signed-off-by:
-
George Spelvin authored
These devices require commands stored in buffers in an odd order, different from that in which the CRC is computed. Rather than make two copies of the commands in two different orders, form the commands in logical (CRC) order, append the CRC, then byte-swap in place to the desired order. The old code worked fine, I'm just scratching an "ugh, that's ugly" itch. Signed-off-by:
George Spelvin <linux@horizon.com> Reviewed-by:
Pavel Machek <pavel@ucw.cz> Signed-off-by:
-
George Spelvin authored
Very minor source and binary size reduction. Signed-off-by:
Geert Uytterhoeven <geert@linux-m68k.org> Acked-by:
Ulf Hansson <ulf.hansson@linaro.org> Signed-off-by:
-
George Spelvin authored
This eliminates a 1-bit left shift in every single caller, and makes the inner loop of the CRC computation more efficient. Renamed crc7 to crc7_be (big-endian) since the interface changed. Also purged #include <linux/crc7.h> from files that don't use it at all. Signed-off-by:
-
-
-
- 15 May, 2014 5 commits
-
-
Johannes Berg authored
If DVM or MVM are built-in but LEDS_CLASS isn't then the current Kconfig will enable LED support and fail the build. Fix this by making the LED support depend on LEDS_CLASS being built-in or, if it is modular, only enabling it if iwlwifi also is. Signed-off-by:
Johannes Berg <johannes.berg@intel.com> Signed-off-by:
Emmanuel Grumbach <emmanuel.grumbach@intel.com>
-
Liad Kaufman authored
In the 8000 HW family the register for forcing an NMI has changed, so this allows to still be able to force an NMI while taking into account the HW in order to write to the correct register. Signed-off-by:
Liad Kaufman <liad.kaufman@intel.com> Reviewed-by:
-
Luciano Coelho authored
Instead of having two nearly identical functions to send the mac context commands, use a single way that can handle both the p2p and !p2p cases. Signed-off-by:
Luciano Coelho <luciano.coelho@intel.com> Reviewed-by:
-
Emmanuel Grumbach authored
The firmware needs to know on what channel we run before we set the association bit in the MAC context. Change a bit the flow to achieve this. Signed-off-by: -
Matt Chen authored
Some APs (e.g. TP-LINK TL-WA801N) are disabling aggregation (downlink to station) when U-APSD is enabled, resulting in low throughput. Add a module parameter to allow disabling U-APSD support in the driver. Also re-enable U-APSD for -9 firmware since the firmare issues were fixed in this release. There are devices that won't support U-APSD even with newer firmware, so bring the TLV flag back to detect those. Signed-off-by:
Matt Chen <matt.chen@intel.com> Signed-off-by:
-
- 14 May, 2014 1 commit
-
-
Eran Harary authored
New transport need to configure internal memory based on the data in the (enlarged) alive notification from the firmware. Add a transport API for this. Signed-off-by:
Eran Harary <eran.harary@intel.com> Signed-off-by:
-
- 13 May, 2014 15 commits
-
-
Jahnavi Meher authored
Signed-off-by:
Jahnavi Meher <jahnavi.meher@redpinesignals.com> Signed-off-by:
-
Arend van Spriel authored
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. Reviewed-by:
Franky Lin <frankyl@broadcom.com> Reviewed-by:
Hante Meuleman <meuleman@broadcom.com> Signed-off-by:
-
Arend van Spriel authored
The device is queried about the usability of channels, but it did not take 80MHz channels into consideration. This patch adds processing those chanspecs and clear the NO_80MHZ flag for those control/primary channels. Reviewed-by:
Daniel (Deognyoun) Kim <dekim@broadcom.com> Reviewed-by:
Pieter-Paul Giesberts <pieterpg@broadcom.com> Signed-off-by:
-
Arend van Spriel authored
The function does not provide any additional functionality and is used only once so just get rid of it. Reviewed-by:
-
Arend van Spriel authored
By default allow 80Mhz in custom regulatory rules of the 5G band so the channels will not be flagged with N0_80MHZ. Reviewed-by:
-
Arend van Spriel authored
Although brcmfmac support several 11ac devices it did not advertise VHT related information to cfg80211. Reviewed-by:
-
Arend van Spriel authored
The channel values used by firmware is handled using conversion functions depending on the type of chip. These functions were already in place but lacked proper support for 80MHz channel definitions. This patch adds the support for that. Reviewed-by:
-
Arend van Spriel authored
In the .start_ap callback cfg80211_get_chandef_type() was used to provide debug log info. However, this causes a warning when the chandef contains VHT channel with 80MHz bandwidth. Avoid the warning by just printing the channel and bandwidth instead. Reviewed-by:
-
Daniel Kim authored
The DISASSOC command needs to be sent to firmware when a connection loss is detected by firmware (e.g., beacon timeout). Otherwise the next connect request fails due to a lingering LINK(down) event from firmware. This patch resolves the issue by using brcmf_link_down() handler, instead of the incomplete duplicated codes. Reviewed-by:
-
Arend van Spriel authored
The driver serializes the nvram firmware file before sending it to the device. This patch enhances this to assure serialized data is properly formatted and provide warnings on syntax failures. Reviewed-by:
-
Hante Meuleman authored
FWS is always queuing frames and using a worker for de-queueing, this is not always efficient for all bus layer. For example SDIO has an internal queue and worker making the queueing of FWS unnecessary. Make it possible to bypass the worker if fws mode is none using a bus interface configuration. For USB bus layer this configuration is set true to have fws provide queueing regardless the fws mode. Reviewed-by:
-
Hante Meuleman authored
Tracking and handling of 802.1x frames is done in two modules, it is more logical and clear to move this to dhd_linux module. Reviewed-by:
-
Hante Meuleman authored
In tx_finalize the hdrpull is performed. For the new protocol msgbuf this is complex, because it does not use protocol headers in front of payload anymore and therefor can not determine interface index in the hdr pulll operation. Move out the hdrpull operation from tx_finalize to make msgbuf implementation easier. Reviewed-by:
-
Daniel Kim authored
Currently firmware roaming support is not reported to cfg80211. This patch reports the support of firmware based roaming when it is enabled. Reviewed-by:
-
Daniel Kim authored
When a BSS provides both 2.4GHz and 5GHz bands, in many cases it makes sense to choose 5GHz. Typically a 5GHz channel is less crowded and has less interference and therefore its performance will be better than a crowded 2.4 GHz channel. This patch configures 'join_pref' to induce firmware to preferably select 5GHz BSS. Reviewed-by:
-
