summaryrefslogtreecommitdiff
path: root/drivers/net/wireless/iwlwifi
AgeCommit message (Collapse)AuthorFilesLines
2013-03-20iwlwifi: mvm: Add beacon notification handlerIlan Peer5-0/+31
Mostly for debugging purposes Signed-off-by: Ilan Peer <ilan.peer@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-20iwlwifi: mvm: suppress key error messages in AP modeJohannes Berg1-0/+13
In AP mode, don't attempt to program GTKs into the device, they're used for TX only so not needed and programming them causes error messages. Also, in this case and if key programming fails, avoid trying to remove the key that isn't present later. Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-20iwlwifi: mvm: MVM op_mode is supported on 7000 onlyEmmanuel Grumbach5-162/+38
The code removed in this patch was used for bring up on older NICs. No MVM capable fw will ever be released for older NICs, so remove that code. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-18mac80211: pass queue bitmap to flush operationJohannes Berg1-1/+1
There are a number of situations in which mac80211 only really needs to flush queues for one virtual interface, and in fact during this frames might be transmitted on other virtual interfaces. Calculate and pass a queue bitmap to the driver so it knows which queues to flush. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-15drivers:net: Remove dma_alloc_coherent OOM messagesJoe Perches1-3/+1
I believe these error messages are already logged on allocation failure by warn_alloc_failed and so get a dump_stack on OOM. Remove the unnecessary additional error logging. Around these deletions: o Alignment neatening. o Remove unnecessary casts of dma_alloc_coherent. o Hoist assigns from ifs. Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-03-09Merge branch 'for-john' of ↵John W. Linville85-696/+2000
git://git.kernel.org/pub/scm/linux/kernel/git/iwlwifi/iwlwifi-next
2013-03-06iwlwifi: mvm: the SCD byte count is a TLV flagEmmanuel Grumbach2-2/+3
The SCD byte count layout is decided by the configuration done in fw, it is then logical to export it as a TLV flag and not per HW SKU. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: fix indirect write bugAmnon Paz1-2/+3
Fix a bug in writing to indirect (periphery) registers; although writes seem successful the data is not written to the desired address). Also fix address mask for HBUS_TARG_PRPH_RADDR and HBUS_TARG_PRPH_WADDR registers. Signed-off-by: Amnon Paz <amnonX.paz@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: move firmware restart debugfs hook to op_modeEmmanuel Grumbach5-22/+52
This allows to test fw restart flow. The hook in transport layer doesn't really make the fw assert. Moving this hook to the op_mode allows to use the fw API to actually send a host command that will make the fw assert. Change the restart_fw module parameter to be a boolean on the way. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: mvm: update firmware API - MAC ID in RXEmmanuel Grumbach1-1/+2
The firmware tells the driver to what MACs the received frame belongs (based on the time slot in which it was received). Note that there can be several MACs if they share the same binding. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: mvm: export last bt_notif through debugfsEmmanuel Grumbach3-0/+104
This will allow to track how BT core updates the driver. This is required to debug the BT Coexistence mechanism. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: mvm: update SMPS when BT gets activeEmmanuel Grumbach1-0/+52
When BT traffic load gets higher, we want to avoid using the shared antenna. In order to do so, we need to tell the AP that we don't support MIMO any more, or at least not all the time: in short, use the SMPS to achieve this. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: mvm: handle BT-coex notificationEmmanuel Grumbach3-0/+58
The BT-Coex notification is sent by the fw when there are updates wrt. BT activity. Driver action might be taken based on the info in this notification. For now, update the Ack/Cts_kill_msk if HID / SCO / A2DP profiles are active. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: mvm: begin basic BT-Coex implementationEmmanuel Grumbach4-1/+259
Send the PRIO table before the calibrations. This table tells the fw what priority to give to what (WiFi / BT) according to events. Send a hardcoded BT_COEX command to the fw to enable basic BT coexistence. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: mvm: add BT Coex FW APIEmmanuel Grumbach3-0/+330
This is the API to tell the fw to handle the BT Coexistence. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: mvm: add CARD_STATE_NOTIFICATION to the cmd stringsEmmanuel Grumbach1-0/+1
Then the transport can print it nicely in its debug prints. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: export symbols only conditionallyJohannes Berg11-35/+64
If all the pieces of iwlwifi are built into the kernel then there's no need for it to export its symbols to other modules, so prevent that. Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: allow selecting only MVM driverJohannes Berg1-1/+4
Now that we have two drivers (DVM and MVM) stop selecting the DVM one (but make it default) and allow enabling only the MVM driver if so desired. Add a warning for the case of having neither DVM nor MVM enabled -- that's useless. Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: mvm: Change NVM default section read sizeDor Shaish1-3/+6
Signed-off-by: Dor Shaish <dor.shaish@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: mvm: don't read system time when modifying AP/GO MACJohannes Berg2-7/+20
When modifying a MAC, we update its beacon system time which is taken as a base to calculate TBTT. The firmware doesn't use the new timestamp because the time is never used after the MAC and broadcast station were added, but it is safer to not rely on this and avoids the overhead of reading the register every time the MAC is updated. Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: make device configuration bus agnosticEmmanuel Grumbach9-124/+48
Newer devices can work on different buses. This means that their configuration can be shared between different buses. Hence the configuration structures should exported to all the buses and not only to PCIE. Change this. Note that this requires all the fields to be the same amongst the buses. If differences will appear, we can always define a part that is bus dependent. Today, this is not needed. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: dvm: pad iwl_compressed_ba_respEmmanuel Grumbach1-0/+1
All the data coming from the fw must have a length that is multiple of 4. This doesn't change anything to the way we handle the notification. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: mvm: advertise VHT capabilitiesEytan Lifshitz1-0/+46
Update the NVM parsing functions to add VHT capabilities; they are only added for 5 GHz, of course. This assumes that all devices with NVM reading (rather than EEPROM) that support 5 GHz have VHT, which is true right now. Signed-off-by: Eytan Lifshitz <eytan.lifshitz@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: mvm: set up initial SMPS/NSS station infoJohannes Berg1-2/+49
When a station is added, we need to tell the firmware what the SMPS settings and number of streams are. After having the initial data, the firmware will track future changes by itself. Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: mvm: implement remote wakeJohannes Berg3-1/+334
With remote wake, the firmware creates a TCP connection and sends some configurable data on it, until a special TCP data packet from the server is received that triggers a wakeup. The configuration is a bit tricky because it is based on packet pattern matching but this is hidden in the driver and the exposed API in cfg80211 is just based on the required TCP connection parameters. Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: convert to use simple_open()Wei Yongjun1-9/+3
This removes an open coded simple_open() function and replaces file operations references to the function with simple_open() instead. Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: use __get_str in tracingJohannes Berg1-1/+1
Instead of using (char *)__get_dynamic_array use __get_str. The latter is actually a macro that expands to the former in the code, but trace-cmd in userspace can parse __get_str only. Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: support DSSS/CCK mode in 40 MHzJohannes Berg1-1/+1
All hardware after 4965 supports this. It's likely that it wasn't set because for 4965 it was irrelevant (HT is only supported on 5 GHz there) and then never updated. Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: mvm: support IEEE80211_AMPDU_TX_STOP_FLUSHEmmanuel Grumbach3-1/+33
mac80211 tells us when we need to dump the frames from the AGG queue instead of releasing them as single MPDUs. Being able to differentiate between the different cases (IEEE80211_AMPDU_TX_STOP_*) allows us to handle races better. When the station is removed, mac80211 asks to flush and removes the station right away. This allows to avoid a case where we still have frames in AGG queues, but the station has been remove already. Note that we can have frames on the shared queues, but this is not a problem: the station in the fw will be kept until all the frames on the shared queues have been drained. AGG queues are a special case since they are dynamically allocated. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: mvm: Add support for additional addressesIlan Peer2-6/+10
Use the number of addresses (max 5) from the NVM instead of limiting to 2 artificially. Signed-off-by: Ilan Peer <ilan.peer@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: disable greenfield transmissions as a workaroundBeni Lev1-6/+8
There's a bug that causes the rate scaling to get stuck when it has to use single-stream rates with a peer that can do GF and SGI; the two are incompatible so we can't use them together, but that causes the algorithm to not work at all, it always rejects updates. Disable greenfield for now to prevent that problem. The MVM driver currently only works on devices that don't support greenfield anyway, but better be safe and not allow us to forget about this. Signed-off-by: Beni Lev <beni.lev@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: 7000: disable HT greenfield supportBeni Lev1-1/+0
The 7000 series devices don't support HT greenfield mode so don't advertise or use it. Signed-off-by: Beni Lev <beni.lev@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: mvm: Update MAC context filter flagsIlan Peer1-1/+13
1. For P2P Device filter in only probe requests. 2. For station mode filter in all group cast frames, and in addition beacons as long as we are not associated. 3. For AP/GO filter in all group cast and in addition probe requests. Signed-off-by: Ilan Peer <ilan.peer@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: mvm: don't use cts to selfDor Shaish1-2/+2
The current fw doesn't currently support cts to self. There is a bug in the fw that prevents us from using cts to self. Use full protection (including RTS) for now. Signed-off-by: Dor Shaish <dor.shaish@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: mvm: free AGG queue when we STA is removedEmmanuel Grumbach1-3/+3
When we stop an AGG session, we need to look at the sequence numbers in in the private area of the ieee80211_sta struct. This allows us to know is the queue is empty. To get access to this private area, we use fw_id_to_mac_id that maps sta_id (index of the STA in fw table) to ieee80211_sta. When the STA exists in fw, but not in mac80211, we set an ERR ptr in fw_id_to_mac_id. But if we first set an ERR ptr to fw_id_to_mac_id, and only then flush the queues, then we won't be able to access the sequence numbers in ieee80211_sta from the reclaim flow. This means that we will never be able to release an AGG queue when a station is deleted. So first, flush the queue. That will let the reclaim flow call iwl_mvm_check_ratid_empty which will disable the AGG queue as needed, and only then, remove the mapping in fw_id_to_mac_id. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: mvm: respect disable Tx AGG parameterEmmanuel Grumbach1-0/+4
We didn't check that we allowed to start Tx AGG. This can possibly be avoided by a module parameter. Fix that. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06iwlwifi: a few fixes in licenseEmmanuel Grumbach69-80/+117
7000.c was released as GPL only by mistake: it should be dual licensed - GPL / BSD. The file that contains the license in the kernel is COPYING and not LICENSE.GPL. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06Merge remote-tracking branch 'iwlwifi-fixes/master' into HEADJohannes Berg18-385/+370
This is needed to resolve some conflicts that would otherwise happen between wireless-next and the code here. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06mac80211: Allow drivers to differentiate between ROC typesIlan Peer2-4/+6
Some devices can handle remain on channel requests differently based on the request type/priority. Add support to differentiate between different ROC types, i.e., indicate that the ROC is required for sending managment frames. Signed-off-by: Ilan Peer <ilan.peer@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06wireless: move sequence number arithmetic to ieee80211.hJohannes Berg5-12/+10
Move the sequence number arithmetic code from mac80211 to ieee80211.h so others can use it. Also rename the functions from _seq to _sn, they operate on the sequence number, not the sequence_control field. Also move macros to convert the sequence control to/from the sequence number value from various drivers. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-06Merge branch 'master' of ↵John W. Linville15-339/+229
git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless into for-davem
2013-03-04iwlwifi: mvm: don't warn on normal BAR sendingEmmanuel Grumbach1-5/+1
This flow happens when we get a failed single Tx response on an AMPDU queue. In this case, the frame won't be sent any more. So we need to move the window on the recipient side. This is done by a BAR. Now if we are in the following case: 10, 12 and 13 are ACKed and 11 isn't. 10 11 12 13. V X V V Then, 11 will be sent 16 times as an MPDU (as oppsed to A-MPDU). If this failed, we are entering the flow described above. So we need to send a BAR with ssn = 12. But in this case, the scheduler will tell us to free frames up to 13 (included). So, it is perfectly possible to get a failed single Tx response on an AMPDU queue that makes the scheduler's ssn jump by more than 1 single packet. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-04iwlwifi: mvm: update the rssi calculationEmmanuel Grumbach3-23/+35
Make the rssi more accurate by taking in count per-chain AGC values. Without this, the RSSI reports inaccurate values. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-04iwlwifi: mvm: ignore STOP_AGG when restartingEmmanuel Grumbach1-0/+10
Since the device is being restarted, all the Rx / Tx Block Ack sessions are been wiped out by the driver. So ignore the requests from mac80211 that stops Tx agg while reconfiguring the device. Note that stopping a non-existing Rx BA session is harmless, so just honor mac80211's request. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-04iwlwifi: mvm: Remove overriding calibrations for the 7000 familyDor Shaish1-114/+0
This fix removes the override of calibration request values sent to the FW. Due to that, the sending of default values to now implemented calibrations is removed. Signed-off-by: Dor Shaish <dor.shaish@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-04iwlwifi: mvm: Take the phy_cfg from the TLV valueDor Shaish1-15/+1
The phy_cfg is given from the TLV value and does not have to be built by us. Signed-off-by: Dor Shaish <dor.shaish@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-04iwlwifi: mvm: Set valid TX antennas value before calib requestDor Shaish1-0/+5
We must set the valid TX antennas number in the ucode before sending the phy_cfg_cmd and request for calibrations. Signed-off-by: Dor Shaish <dor.shaish@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-04iwlwifi: mvm: restart the NIC of the cmd queue gets fullEmmanuel Grumbach1-5/+13
This situation is clearly an error situation and the only way to recover is to restart the driver / fw. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-01Merge branch 'master' of ↵John W. Linville7-71/+166
git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless into for-davem
2013-02-28iwlwifi: use coherent DMA memory for command headerJohannes Berg5-142/+137
Recently in commit 8a964f44e01ad3bbc208c3e80d931ba91b9ea786 ("iwlwifi: always copy first 16 bytes of commands") we fixed the problem that the hardware writes back to the command and that could overwrite parts of the data that was still needed and would thus be corrupted. Investigating this problem more closely we found that this write-back isn't really ordered very well with respect to other DMA traffic. Therefore, it sometimes happened that the write-back occurred after unmapping the command again which is clearly an issue and could corrupt the next allocation that goes to that spot, or (better) cause IOMMU faults. To fix this, allocate coherent memory for the first 16 bytes of each command, containing the write-back part, and use it for all queues. All the dynamic DMA mappings only need to be TO_DEVICE then. This ensures that even when the write-back happens "too late" it can't hit memory that has been freed or a mapping that doesn't exist any more. Since now the actual command is no longer modified, we can also remove CMD_WANT_HCMD and get rid of the DMA sync that was necessary to update the scratch pointer. Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
generated by cgit v1.2.3 (git 2.39.0) at 2025-10-02 11:33:29 +0300