summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* airo: clean up and clarify micinit()Dan Williams2009-01-291-41/+41
| | | | | | | Fix some endian issues too. Signed-off-by: Dan Williams <dcbw@redhat.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* airo: use __attribute__ ((packed)) not #pragmaDan Williams2009-01-291-41/+49
| | | | | Signed-off-by: Dan Williams <dcbw@redhat.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* b43: Fix phy_g.c compiler warningMichael Buesch2009-01-291-0/+3
| | | | | | | | | | Fix compile warning for non-debug builds: drivers/net/wireless/b43/phy_g.c: In function ‘b43_gphy_op_recalc_txpower’: drivers/net/wireless/b43/phy_g.c:3195: warning: unused variable ‘dbm’ Signed-off-by: Michael Buesch <mb@bu3sch.de> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* mac80211: flush workqueue a second time in suspend()Bob Copeland2009-01-291-0/+3
| | | | | | | | | Drivers can theoretically queue more work in one of their callbacks from mac80211 suspend, so let's flush it once more to be on the safe side, just before calling ->stop(). Signed-off-by: Bob Copeland <me@bobcopeland.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* mac80211: change workqueue back to non-freezeableBob Copeland2009-01-291-1/+1
| | | | | | | | | | | | | | | | | | | | | "mac80211: make workqueue freezable" made the mac80211 workqueue freezeable to prevent us from doing any work after the driver went away. This was fine before mac80211 had any suspend support. However, now we want to flush this workqueue in suspend(). Because the thread for a freezeable workqueue is stopped before the device class suspend() is called, flush_workqueue() will hang in the suspend-to-disk case. Converting it back to a non-freezeable queue will keep suspend from hanging. Moreover, since we flush the workqueue under RTNL and userspace is stopped, there won't be any new work in the workqueue until after resume. Thus we still don't have to worry about pinging the AP without hardware. Signed-off-by: Bob Copeland <me@bobcopeland.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* mac80211: Fixed BSSID handling revisitedAlina Friedrichsen2009-01-291-39/+33
| | | | | | | | | | | | | | | | | | | | This patch cleanup the fixed BSSID handling, that ieee80211_sta_set_bssid() works like ieee80211_sta_set_ssid(). So that the BSSID is only a second selection criterion besides the SSID. This allows us to create new IBSS networks with fixed BSSIDs, which was broken before. In the second version of this patch the handling of the stupid merges to the same BSSID is moved out to get reworked into an other patch. And this version hopefully solves the problems with some low-level drivers and re-adds the config BSSID warning to help debugging the low-level drivers. Much thanks to all who have helped testing! :) Signed-off-by: Alina Friedrichsen <x-alina@gmx.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* airo: clean up WEP key operationsDan Williams2009-01-291-56/+146
| | | | | | | | | | get_wep_key() and set_wep_key() combind both get/set of the actual WEP key and get/set of the transmit index into the same functions. Split those out so it's clearer what is going one where. Add error checking to WEP key hardware operations too. Signed-off-by: Dan Williams <dcbw@redhat.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* airo: simplify WEP index and capability checksDan Williams2009-01-291-32/+27
| | | | | | | | Do the computation once at init time; don't ask the hardware every time. Signed-off-by: Dan Williams <dcbw@redhat.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* airo: re-arrange WPA capability checksDan Williams2009-01-291-34/+32
| | | | | | | | | | The capability register has to be read for other (upcoming) stuff, so fold the WPA test function back into _init_airo_card() and move the netdevice registration stuff above it so that the netdevice has a name by the time the card's capabilities are printed out. Signed-off-by: Dan Williams <dcbw@redhat.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* airo: clean up and clarify interrupt-time task handlingDan Williams2009-01-291-286/+327
| | | | | | | | Split each specific interrupt-time task out into its own function to make airo_interrupt() actually readable. Signed-off-by: Dan Williams <dcbw@redhat.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* iwlwifi: fix probe mask for 39 scan APIWinkler, Tomas2009-01-293-3/+8
| | | | | | | | | This pach make use of 39 own scan probe mask the variables or of different types Signed-off-by: Tomas Winkler <tomas.winkler@intel.com> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* iwl3945: Remaining host command cleanupsSamuel Ortiz2009-01-299-127/+45
| | | | | | | | | With the recent host command routines merge, we can now look at the various host command helpers and get rid of the duplicated ones. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* iwl3945: Add restart_fw module parameterSamuel Ortiz2009-01-291-1/+6
| | | | | | | | | In order to be in sync with the agn code, we're ading a fw_restart3945 module parameter to iwl3945. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* iwl3945: Getting rid of priv->antennaSamuel Ortiz2009-01-293-41/+39
| | | | | | | | | The iwl_priv antenna field is useless as we can simply use the corresponding mod_params antenna field. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* iwlwifi: Remove IWL3945_DEBUGSamuel Ortiz2009-01-294-52/+31
| | | | | | | | | | | IWL3945_DEBUG is pointless and obsolete. We already have an IWLWIFI_DEBUG symbol, that needs to be set if we actually want to get 3945 debug (see iwl-debug.h). Thus, we can simply get rid of this symbol. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* iwl3945: Use the iwlcore geos routinesSamuel Ortiz2009-01-293-138/+21
| | | | | | | | | By removing the init_rates() routine outside of the init_geos() one, we can share the geos routines between 3945 and agn. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* iwl3945: Use iwl-eeprom.c routinesSamuel Ortiz2009-01-296-367/+115
| | | | | | | | | | | By adding the eeprom ops to the 3945 code, we can now use the iwlcore eeprom routines (defined in iwl-eeprom.c). We also removed the heavy eeprom39 reference from iwl_priv and use the eeprom pointer instead. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* iwl3945: Use iwlcore TX queue management routinesSamuel Ortiz2009-01-298-246/+60
| | | | | | | | | | | By adding an additional hw_params (tfd_size) and a new iwl_lib ops (txq_init), we can now use the iwlcore TX queue management routines. We had to add a new hw_params because we need to allocate the right DMA buffer for TFDs, and those have a different sizes depending if you're on 3945 or agn. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* iwlwifi: make iwl_tx_queue->tfds void*Samuel Ortiz2009-01-295-16/+17
| | | | | | | | | | Instead of having both tfds and tfds39, we can just have a void *tfds. It makes the tx_queue structure nicer, and the code cleaner. It also helps with further TX queues management code merging. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* iwl3945: Use iwlcore scan codeSamuel Ortiz2009-01-293-301/+51
| | | | | | | | | A lot of the scanning related code is duplicated between 3945 and agn. Let's use the iwlcore one and get rid of the 3945. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* iwl3945: Define send_tx_powerSamuel Ortiz2009-01-292-5/+6
| | | | | | | | | We can define this iwl_lib ops for 3945, as this would help us cleaning the scanning code. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* iwlagn: reduce off channel reception for 4965Rick Farrington2009-01-293-0/+17
| | | | | | | | | | | Force use of chains B and C (0x6) for Rx for 4965 Avoid A (0x1) because of its off-channel reception on A-band. Signed-off-by: Rick Farrington <rickdic@hotmail.com> Signed-off-by: Ben Cahill <ben.m.cahill@intel.com> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com> Signed-off-by: Tomas Winkler <tomas.winkler@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* iwlwifi: add test to determine if interface in monitor modeChatre, Reinette2009-01-291-0/+12
| | | | | | | | | | | mac80211 sets driver in monitor mode through configuring the RX filters. We cannot trust priv->iw_mode to be accurate regarding monitor mode as iw_mode is only set in add_interface, which is not called by mac80211 when in monitor mode. Signed-off-by: Reinette Chatre <reinette.chatre@intel.com> Signed-off-by: Tomas Winkler <tomas.winkler@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* b43: Automatically probe for opensource firmwareMichael Buesch2009-01-293-44/+149
| | | | | | | | | | | | | | | | | First probe for proprietary firmware and then probe for opensource firmware. This way around it's a win-win situation. 1) If proprietary fw is available, it will work. 2) If opensource firmware is available, but no proprietary (Distros can only ship open fw) it might work. 3) If both open and proprietary are available, it will work, because it selects the proprietary. We currently don't prefer the open fw in this case, because it doesn't work on all devices. It would introduce a regression otherwise. The remaining FIXMEs in this patch are harmless, because they only matter on multiband devices, which are not implemented yet anyway. Signed-off-by: Michael Buesch <mb@bu3sch.de> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* rtl8187: Fix driver to return TX retry info for RTL8187LLarry Finger2009-01-292-19/+58
| | | | | | | | | | | | | | | | | | | | Current code for the RTL8187 is not returning valid retry information, thus the rate-setting mechanism is not functioning. As a further complication, this info is only obtained by reading a register, which cannot be read while in interrupt context. This patch implements the TX status return to mac80211 through the use of a work queue. One additional problem is that the driver currently enables the rate fallback mechanism of the device, which conflicts with the mac80211 rate-setting algorithm. This version of the patch disables rate fallback. Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Tested-by: Herton Ronaldo Krzesinski <herton@mandriva.com.br> Tested-by: Martín Ernesto Barreyro <barreyromartin@gmail.com> Acked-by: Hin-Tak Leung <htl10@users.sourceforge.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* rtl8187: Increase receive queue depthLarry Finger2009-01-291-1/+1
| | | | | | | | The receive queue depth in rtl8187 may not be long enough to keep the pipe full. Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* rtl8187: Fix locking of private dataLarry Finger2009-01-291-4/+7
| | | | | | | | In rtl8187_add_interface(), the mutex that protects the data in struct rtl8187_priv does not include all references to that structure. Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* libertas: fix CF firmware loading for some cardsDan Williams2009-01-291-1/+1
| | | | | | | | | | | | | if_cs_poll_while_fw_download() returned the number of iterations remaining on success, which in turn got returned as the value from if_cs_prog_real() and if_cs_prog_helper(). But since if_cs_probe() interprets non-zero return values from firmware load functions as an error, this sometimes caused spurious firmware load failures. Signed-off-by: Dan Williams <dcbw@redhat.com> Tested-by: Ryan Mallon <ryan@bluewatersys.com> Acked-by: Ryan Mallon <ryan@bluewatersys.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* rt2x00: Simplify suspend/resume handlingIvo van Doorn2009-01-292-89/+5
| | | | | | | | | | | | | With mac80211 handling all open interfaces during suspend and resume we can simplify suspend/resume within rt2x00lib. The only thing rt2x00 needs to do is free up memory during suspend and bring back the minimal required components during resume. Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* rt2x00: Restrict firmware file lengthsIvo van Doorn2009-01-292-0/+10
| | | | | | | | | | | Add extra security to the drivers for firmware loading, check the firmware file length before uploading it to the hardware. Incorrect lengths might indicate a firmware upgrade (which is not yet supported by the driver) or otherwise incorrect firmware. Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* rt2x00: rt2x00_rev() should return u32Ivo van Doorn2009-01-291-1/+1
| | | | | | | | | The "rev" field in chipset definition is an u32, which means that rt2x00_rev() which returns that field should be of the same type. Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* ath9k: Clean up the way the eeprom antenna configuration is readVasanthakumar Thiagarajan2009-01-292-51/+17
| | | | | Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* ath9k: Remove unused ath9k_hw_select_antconfig() from hw.cVasanthakumar Thiagarajan2009-01-292-22/+0
| | | | | Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* ath9k: Fix bug in NF calibrationSujith2009-01-291-1/+3
| | | | | | | | The number of chainmasks for AR9285 weren't being setup when running NF calibration. Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* ath9k: Fix bug in rate control capability registrationSujith2009-01-291-1/+3
| | | | | | | | Dual stream capability must be registered only when the hardware supports it. Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* ath9k: Fix MCS rates registration bug for AR9285Sujith2009-01-291-5/+18
| | | | | | | | | AR9285 based devices support only single stream MCS rates. This patch fixes a bug where dual stream stream rates were also being registered. Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* ath9k: Read and write the TSF via debugfsAlina Friedrichsen2009-01-294-0/+58
| | | | | | | This patch adds an ath9k specific entry to read, write and reset the TSF into the debugfs, like in ath5k. This makes debugging the IBSS handling of wifi drivers _much_ easier. Signed-off-by: Alina Friedrichsen <x-alina@gmx.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* ath5k: Read and write the TSF via debugfsAlina Friedrichsen2009-01-293-2/+27
| | | | | | | This patch updates the ath5k specific entry in the debugfs to read and reset the TSF value, to allowing write it, too. This makes debugging the IBSS handling of wifi drivers _much_ easier. Signed-off-by: Alina Friedrichsen <x-alina@gmx.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* mac80211: Read the TSF via debugfsAlina Friedrichsen2009-01-292-0/+5
| | | | | | | This patch adds an low-level driver independent entry to read the TSF value into the debugfs of mac80211. This makes debugging the IBSS handling of wifi drivers easier. Signed-off-by: Alina Friedrichsen <x-alina@gmx.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* ath9k: simplify regulatory codeLuis R. Rodriguez2009-01-298-3030/+809
| | | | | | | | | | | | | | | | | | | | Now that cfg80211 has its own regulatory infrastructure we can condense ath9k's regulatory code considerably. We only keep data we need to provide our own regulatory_hint(), reg_notifier() and information necessary for calibration. Atheros hardware supports 12 world regulatory domains, since these are custom we apply them through the the new wiphy_apply_custom_regulatory(). Although we have 12 we can consolidate these into 5 structures based on frequency and apply a different set of flags that differentiate them on a case by case basis through the reg_notifier(). If CRDA is not found our own custom world regulatory domain is applied, this is identical to cfg80211's except we enable passive scan on most frequencies. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* cfg80211: allow users to help a driver's complianceLuis R. Rodriguez2009-01-291-1/+2
| | | | | | | Let users be more compliant if so desired. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* mac80211: allow mac80211 drivers to get to struct ieee80211_hw from wiphyLuis R. Rodriguez2009-01-292-0/+22
| | | | | | | | | | | | If a driver is given a wiphy and it wants to get to its private mac80211 driver area it can use wiphy_to_ieee80211_hw() to get first to its ieee80211_hw and then access the private structure via hw->priv. The wiphy_priv() is already being used internally by mac80211 and drivers should not use this. This can be helpful in a drivers reg_notifier(). Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* cfg80211: Allow for strict regulatory settingsLuis R. Rodriguez2009-01-292-1/+37
| | | | | | | | | | | | | | | | | This allows drivers to request strict regulatory settings to be applied to its devices. This is desirable for devices where proper calibration and compliance can only be gauranteed for for the device's programmed regulatory domain. Regulatory domain settings will be ignored until the device's own regulatory domain is properly configured. If no regulatory domain is received only the world regulatory domain will be applied -- if OLD_REG (default to "US") is not enabled. If OLD_REG behaviour is not acceptable to drivers they must update their wiphy with a custom reuglatory prior to wiphy registration. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* cfg80211: pass more detailed regulatory request information on reg_notifier()Luis R. Rodriguez2009-01-294-41/+48
| | | | | | | | | | Drivers may need more information than just who set the last regulatory domain, as such lets just pass the last regulatory_request receipt. To do this we need to move out to headers struct regulatory_request, and enum environment_cap. While at it lets add documentation for enum environment_cap. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* cfg80211: move check for ignore_reg_update() on wiphy_update_regulatory()Luis R. Rodriguez2009-01-291-2/+4
| | | | | | | | | | | This ensures that the initial REGDOM_SET_BY_CORE upon wiphy registration respects the wiphy->custom_regulatory setting. Without this and if OLD_REG is disabled (which will be default soon as we remove it) the wiphy->custom_regulatory is simply ignored. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* cfg80211: rename fw_handles_regulatory to custom_regulatoryLuis R. Rodriguez2009-01-294-6/+6
| | | | | | | | Drivers without firmware can also have custom regulatory maps which do not map to a specific ISO / IEC alpha2 country code. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* cfg80211: ignore consecutive equal regulatory hintsLuis R. Rodriguez2009-01-291-2/+11
| | | | | | | | | | | We ignore regulatory hints for the same alpha2 if we already have processed the same alpha2 on the current regulatory domain. For a driver regulatory_hint() this means we copy onto its wiphy->regd the previously procesed regulatory domain from CRDA without having to call CRDA again. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* cfg80211: process user requests only after previous user/driver/core requestsLuis R. Rodriguez2009-01-291-0/+10
| | | | | | | | This prevents user regulatory changes to be considered prior to previous pending user, core or driver requests which have not be applied. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* cfg80211: export freq_reg_info()Luis R. Rodriguez2009-01-292-22/+26
| | | | | | | This can be used by drivers on the reg_notifier() Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* cfg80211: add wiphy_apply_custom_regulatory()Luis R. Rodriguez2009-01-292-24/+108
| | | | | | | | | | This adds wiphy_apply_custom_regulatory() to be used by drivers prior to wiphy registration to apply a custom regulatory domain. This can be used by drivers that do not have a direct 1-1 mapping between a regulatory domain and a country. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
OpenPOWER on IntegriCloud