- 12 Feb, 2014 40 commits
-
-
Aaron Durbin authored
Depending on the underlying DMA implementation its not possible to partially unmap DMA buffers. Moreover its not possible to understand the intent of passing 0 as the size to dma unmap. The intent of this driver is unmap the entire skb buffer. The only way to ensure that the size matches on unmap is to store both the dma address and the size in the skb ca field. Introduce a mwifiex_dma_mapping structure which tracks the dma address and the size. Additionally, provide a mwifiex_unmap_pci_memory() that utilizes the new structure. This also provide symmetry within the internal API. Signed-off-by:
Aaron Durbin <adurbin@chromium.org> Reviewed-by:
Paul Stewart <pstew@chromium.org> Reviewed-by:
Avinash Patil <patila@marvell.com> Signed-off-by:
Bing Zhao <bzhao@marvell.com> Signed-off-by:
John W. Linville <linville@tuxdriver.com>
-
Amitkumar Karwar authored
In extended scan, host gets scan results through one or multiple events instead of scan command response. Host will send next scan command when all the events are received. Legacy scan sometimes truncates scan results in a noisy environment due to buffer length limitation. This issue is addressed in extended scan. Signed-off-by:
Amitkumar Karwar <akarwar@marvell.com> Signed-off-by:
-
Amitkumar Karwar authored
This new function will be useful later for extended scan feature. Signed-off-by:
-
Amitkumar Karwar authored
This new function will be useful later for extended scan feature. Signed-off-by:
-
Amitkumar Karwar authored
'mwifiex_bcn_param' structure contains five parameters which are present in beacon buffer in case of legacy scan. 'rssi' field won't be there in this buffer for extended scan. Hence 'bssid' and 'rssi' are removed from the structure and it is renamed as 'mwifiex_fixed_bcn_param' so that we can have common parsing logic later for both. Signed-off-by:
-
Bing Zhao authored
bit12 in fw_cap_info is for testing only. Remove all related code. Signed-off-by:
-
Bing Zhao authored
1) rename max_mcs to mcs; 2) initialize 'i' and 'nss' as 1 instead of 0 in nss lookup; 3) use GET_VHTNSSMCS(mcs_map, nss) macro; 4) use IEEE80211_VHT_MCS_* definitions instead of hard coding Reported-by:
-
Bing Zhao authored
Remove these local array variables and define them as static const array in global space. The duplicated mcs_rate table is removed automatically with this change. Reported-by:
-
Bing Zhao authored
Remove driver's macro and use ieee80211's definition instead Signed-off-by:
-
Sujith Manoharan authored
This patch adds a routine to calculate the median IQ correction values for AR955x, which is used for outlier detection. The normal method which is used for all other chips is bypassed for AR955x. Signed-off-by:
Sujith Manoharan <c_manoha@qca.qualcomm.com> Signed-off-by:
-
Sujith Manoharan authored
This will be used for storing data for mutiple IQ calibration runs, for AR955x. Signed-off-by:
-
Sujith Manoharan authored
IQ calibration post-processing for AR955x is different from other chips - instead of just doing it as part of AGC calibration once, it is triggered 3 times and a median is determined. This patch adds initial support for changing the calibration behavior for AR955x. Also, to simplify things, a helper routine to issue/poll AGC calibration is used. For non-AR955x chips, the iqcal_idx (which will be used in subsequent patches) is set to zero. Signed-off-by:
-
Sujith Manoharan authored
Incorrect values are programmed in the registers containing the IQ correction coefficients by the IQ-CAL post-processing code. Fix this. Signed-off-by:
-
Sujith Manoharan authored
Use ar9003_hw_tx_iq_cal_outlier_detection instead. Signed-off-by:
-
Sujith Manoharan authored
In chips like AR955x, the initvals contain the information whether IQ calibration is to be done in the HW when an AGC calibration is triggered. Check if IQ-CAL is enabled in the initvals before flagging 'txiqcal_done' as true. Signed-off-by:
-
Sujith Manoharan authored
Calibration data is not reused for SoC chips, so call ar9003_hw_tx_iq_cal_post_proc() with the correct argument. The 'is_reusable' flag is currently used only for PC-OEM chips, but it makes things clearer to specify it explicity. Signed-off-by:
-
Masaki TAGAWA authored
Buffalo WLI-UV-AG300P is almost the same as Sony UWA-BR100. Signed-off-by:
Masaki TAGAWA <masaki@club.kyutech.ac.jp> Signed-off-by:
-
Sujith Manoharan authored
There is no benefit in retaining the legacy rate control module in the driver codebase. It is known to be buggy and has less than optimal performance in real-world environments compared with minstrel. The only reason that it was kept when we made the switch to minstrel as default was that it showed higher throughput numbers in a clean/ideal environment. This is no longer the case and minstrel can push ath9k to the same throughput levels. In TCP, with 3-stream cards, more than 295 Mbps can be obtained in open air, with 2-stream cards, 210 Mbps is easily reached. To test performance issues, instead of using a broken rate control module, it is better to use the fixed-rate interface provided by mac80211 anyway. The ath9k RC has not received any bug fixes in years and is just bit-rotting away - this patch removes it. Signed-off-by:
-
Oleksij Rempel authored
... and print what we get. Signed-off-by:
Oleksij Rempel <linux@rempel-privat.de> Signed-off-by:
-
Oleksij Rempel authored
after switch to common fucntions we do not need this memcpy any more. Signed-off-by:
-
Oleksij Rempel authored
Signed-off-by:
-
Oleksij Rempel authored
and rename it to ath9k_cmn_rx_skb_postprocess. We will use it on ath9k_htc. Signed-off-by:
-
Oleksij Rempel authored
Signed-off-by:
-
Oleksij Rempel authored
Signed-off-by:
-
Oleksij Rempel authored
Signed-off-by:
-
Oleksij Rempel authored
Signed-off-by:
-
Oleksij Rempel authored
Signed-off-by:
-
Oleksij Rempel authored
we can reuse it on ath9k_htc Signed-off-by:
-
Oleksij Rempel authored
we can reuse this function in ath9k_htc Signed-off-by:
-
Oleksij Rempel authored
we can reuse this fucntion on ath9k_htc. Now we will need to use common version last_rssi, so switch it too. Signed-off-by:
-
Oleksij Rempel authored
we need access to this variable from common functions. Signed-off-by:
-
Stanislaw Gruszka authored
On commit 28f2bce9 I make change that print various messages as default. This can cause flood of messages related to TX status timeout on some environments. I partially fixed problem on commit bb9c298f, but forgot to move two more messages to debug level. Signed-off-by:
Stanislaw Gruszka <sgruszka@redhat.com> Acked-by:
Gertjan van Wingerde <gwingerde@gmail.com> Signed-off-by:
-
Arend van Spriel authored
In the enter and exit download sequence the chip core info was checked for presence of CM3 ARM core. If found it would enter download state for the CM3. However, on devices that have a CM3 and CR4 this is not correct and the CR4 should be used to enter download state. This patch changes the ARM core lookup giving CR4 precedence. Reviewed-by:
Hante Meuleman <meuleman@broadcom.com> Reviewed-by:
Pieter-Paul Giesberts <pieterpg@broadcom.com> Signed-off-by:
Arend van Spriel <arend@broadcom.com> Signed-off-by:
-
Daniel Kim authored
The brcmf_add_keyext() is for setting per-station key for cipher algorithms such as WPA1/WPA2 and should not be used to set WEP broadcast/unicast keys. This patch fixes connect failure problem with AP using 802.1x-WEP. Reviewed-by:
Franky (Zhenhui) Lin <frankyl@broadcom.com> Signed-off-by:
Daniel Kim <dekim@broadcom.com> Signed-off-by:
-
Arend van Spriel authored
Instead of instantiating core info structs based upon the chip identifier it is now done parsing information provided on the device. Reviewed-by:
-
Arend van Spriel authored
In brcmf_contstruct_reginfo() some error logging was added by: commit f7c51a1a72f50870f80001ddf528a6f7f992bc16 Author: Arend van Spriel <arend@broadcom.com> Date: Wed Dec 11 16:21:21 2013 +0100 brcmfmac: correct reporting HT40 support in wiphy htcap This logging was not intended to be delivered and adds a lot of messages in the log. The patch removes this logging statement. Reviewed-by: -
Arend van Spriel authored
The function brcmf_sdio_bus_sleep() function is called rather frequently, which fills the log when TRACE level is enabled. Reduced the level to SDIO. Reviewed-by:
-
Arend van Spriel authored
This patch make several chip related functions host interface independent by defining callback interface struct brcmf_buscore_ops. Reviewed-by:
-
Hante Meuleman authored
Currently the function sdio_remove will first destroy the datawork workqueue and then detach the bus. This can create the situation where work gets added on non-existing work queue resulting in panic. Reviewed-by:
-
Hante Meuleman authored
brcmf_sdio_download_code_file is using a loop to send small blobs of data. This is unnecessarily complex and was simplified with this patch. Reviewed-by:
-
