- 10 Dec, 2020 1 commit
-
-
Kalle Valo authored
Merge tag 'iwlwifi-next-for-kalle-2020-12-09' of git://git.kernel.org/pub/scm/linux/kernel/git/iwlwifi/iwlwifi-next iwlwifi patches intended for v5.11 * A few fixes and improvements in the FW debugging framework; * A fix in the HE capabilities settings; * Small fix in the logs for SAR; * Move queue code to a common place; * Support for 6GHz; * Improve validation of firmware notifications; * Debugging improvements and fixes; * Replace terminology with inclusive words determined by the guidelines; * Convert copyright notices to SPDX tags; * Added more validations for firmware notifications; * A couple of debugging/recovery fixes; * Added some more validations for firmware notifications; * Support for a new type of HW; * A couple of channel switch fixes; * Support new FW reset handshake; * Add a couple of RX handlers that were accidentally left out; * Some other clean-ups and small fixes; # gpg: Signature made Thu 10 Dec 2020 12:19:59 AM EET using RSA key ID 1A3CC5FA # gpg: Good signature from "Luciano Roth Coelho (Luca) <luca@coelho.fi>" # gpg: aka "Luciano Roth Coelho (Intel) <luciano.coelho@intel.com>"
-
- 09 Dec, 2020 39 commits
-
-
Johannes Berg authored
When waiting for a notification and then processing it, we also need to check the size of the data before we use it. Most places do that already, but fix the remaining ones to do it as well. Signed-off-by:
Johannes Berg <johannes.berg@intel.com> Signed-off-by:
Luca Coelho <luciano.coelho@intel.com> Link: https://lore.kernel.org/r/iwlwifi.20201209231352.b29573bcba39.I4b7e72824d06dc0719a40021d933e29edfc14713@changeidSigned-off-by:
-
Emmanuel Grumbach authored
When the firmware is loaded, its internal timer (GP2) is reset and all the time_sync in the BSS table is now unusable. Ask cfg80211 to purge the BSS table. Signed-off-by:
Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by:
-
Mordechay Goodstein authored
FW now puts in the struct version, the TLV version so we also check it to make sure it matches. Signed-off-by:
Mordechay Goodstein <mordechay.goodstein@intel.com> Signed-off-by:
-
Avraham Stern authored
The range response notification has several versions. Check the notification size according to the expected notification version. Notifications with incorrect size will be ignored. Signed-off-by:
Avraham Stern <avraham.stern@intel.com> Signed-off-by:
-
Johannes Berg authored
The RX handlers for probe response data and channel switch weren't hooked up properly, fix that. Fixes: 86e177d8 ("iwlwifi: mvm: add NOA and CSA to a probe response") Fixes: d3a108a4 ("iwlwifi: mvm: Support CSA countdown offloading") Signed-off-by:
-
Johannes Berg authored
Previously, we added checks that the contained MPDU size is long enough, but really we should also check that the notification itself fits into the data. Add some checks for that. Also add unlikely() annotations on the previously added checks. Signed-off-by:
-
Matti Gottlieb authored
Add a PCI ID for snj with mr in AX family. Signed-off-by:
Matti Gottlieb <matti.gottlieb@intel.com> Signed-off-by:
-
Johannes Berg authored
If the channel switch delay that we would incur after the channel switch actually happens is longer than the quiet time we're willing to tolerate, disconnect as well. Signed-off-by:
-
Johannes Berg authored
There are some races in the hardware that can possibly lead to a bus lockup later during a restart when we manage to kill the firmware at a bad time (while it's accessing the bus). To work around this, add support for a new handshake between firmware and driver to ensure that the firmware is in a well- known state before we kill it. Signed-off-by:
-
Johannes Berg authored
Start tracking not just if the firmware is dead or alive, but also if it's starting. Signed-off-by:
-
Sara Sharon authored
When we get a channel switch with a very long quiet period, we schedule a work to disconnect after a while. This work runs in background. In the meanwhile, we keep getting beacons and sending FW modify command for each. This has a potential race, where we modify the CSA after we aborted it. Protect the flow by setting csa_failed to true in case we abort, and check it before sending the modify command. This required a modification to the way we treat csa_failed in iwl_mvm_post_channel_switch: 1. The variable isn't being reset anymore, so we can still look at it in iwl_mvm_channel_switch_rx_beacon. This is fine, since we reset it when starting a new CSA. 2. There is no more early return in case of csa_failed. This is fine, since before this patch csa_failed was set only for GO, and for GO the function is only resetting the power settings, which we want to restore even in case of failure. Signed-off-by:
Sara Sharon <sara.sharon@intel.com> Signed-off-by:
-
Johannes Berg authored
We send some data to the firmware and expect to get it back, but we shouldn't really trust the firmware on this. Check the size of all the data we send down to avoid using bad or just uninitialized data when the firmware doesn't respond right. Signed-off-by:
-
Johannes Berg authored
We don't need the sequence/index/cmd_index unless we're doing reclaim, they're not even valid in the other cases. Move the variables and their assignments into the right if statement and combine the two if statements into a single one as well. Signed-off-by:
-
Johannes Berg authored
Use SPDX tags instead of the long copyright notices. Also cleanup some duplicate copyright notices and combine the years where possible. Signed-off-by:
-
Johannes Berg authored
There's a bug in the lengths - the 'old length' needs to be calculated using the 'old' pointer, of course, likely a copy/paste mistake. Fix this. Reported-by:
Daniel Gabay <daniel.gabay@intel.com> Signed-off-by:
-
Johannes Berg authored
If we get an error, no longer consider the firmware to be in IWL_TRANS_FW_ALIVE state. Signed-off-by:
-
Johannes Berg authored
We set this here, but don't really use it until we've enabled interrupts. But when enabling interrupts we always overwrite this value anyway, so remove setting it here, mostly in order not to have some additional code duplicated later. Signed-off-by:
-
Johannes Berg authored
This was mistakenly tagged with CAPA instead of API, fix it. Signed-off-by:
-
Johannes Berg authored
This is actually wrong, the bit used here by the image loader is BIT(1), not BIT(2). The latter will be reused by the new reset flow soon. However, as we never had any complaints about not printing the IML status or not handling the IML error interrupt (and I suspect the code handling it was incorrectly anyway) just remove the code for it. Signed-off-by:
-
Johannes Berg authored
We keep the scan status per UID in scan_uid_status field when the iwl_mvm_build_scan_cmd() function is called. If we error out after this, e.g. due to FW restart being in progress, we're not cleaning up properly, and can run into warnings later. Clean up internal variables when starting fails after calling the iwl_mvm_build_scan_cmd() function. Signed-off-by:
-
Johannes Berg authored
We're currently doing accounting on the queue sync with an atomic variable that counts down the number of remaining notifications that we still need. As we've been hitting issues in this area, modify this to track a bitmap of queues, not just the number of queues, and print out the remaining bitmap in the warning. Signed-off-by:
-
Johannes Berg authored
For triggering an NMI in the firmware, we should only set BIT(24) in the corresponding register, not the entire mask that's usable by the driver. This currently doesn't matter because the firmware only enables BIT(24), but we'll start using BIT(25) for other purposes with an upcoming API change. Signed-off-by:
-
Johannes Berg authored
If 12k A-MSDU size is requested, we will actually allocate 16k due to page allocation. Thus, change it to actually mean 16k, which is useful for certain sniffer use cases. Signed-off-by:
-
Johannes Berg authored
This function has grown quite a bit over time, move it out of line before we grow it even further. Signed-off-by:
-
Johannes Berg authored
The argument type to iwl_mvm_notify_rx_queue() is currently just a u8 *, but that's misleading because we actually need the inner data to be of type struct iwl_mvm_internal_rxq_notif, because we interpret it when we get it back from the device (to check the sync bool and possibly the cookie.) Therefore, clear up any potential confusion and require that the data passed is of type struct iwl_mvm_internal_rxq_notif *. Also, while at it, rename the "count" to "notif_size" as "count" doesn't really clearly say what it's counting. Signed-off-by:
-
Emmanuel Grumbach authored
This was a hack done to test the data path of devices that didn't support well CSUM offload in Tx. This is not needed anymore. Signed-off-by:
-
Johannes Berg authored
Validate the maximum RX descriptor length against the size of the buffers we gave the device - if it doesn't fit then the hardware messed up. Signed-off-by:
-
Johannes Berg authored
The MPDU contained in a notification shouldn't be larger than the notification size itself is, validate this. Reported-by:
Haggai Abramovsky <haggai.abramovsky@intel.com> Signed-off-by:
-
Mordechay Goodstein authored
Instead of enumerating all possible stations iterate only active ones. Signed-off-by:
-
Emmanuel Grumbach authored
The comments about the NVM offsets were inaccurate. Fix the enum so that the right values appear under the right comment. Signed-off-by:
-
Mordechay Goodstein authored
Change "anntena" to "antenna" in a comment in stats.h. Signed-off-by:
-
Emmanuel Grumbach authored
The new inclusive terminology requires to change a few terms that were used in iwlwifi. Signed-off-by:
-
Emmanuel Grumbach authored
Similarly to what I did to iwl_run_init_mvm_ucode, there is no need to pass the read_nvm parameter. Either we have an NVM and we don't need to read it, or we don't and we need to read it. Signed-off-by:
-
Mordechay Goodstein authored
Curretly we only mark HW error state "after" trying to collect HW data, but if any HW error happens while colleting HW data we go into endless loop. avoid this by setting HW error state "before" collecting HW data. Signed-off-by:
-
Mordechay Goodstein authored
This is used for BT node and for any user that wants to control what events would be send from FW to the driver. Signed-off-by:
-
Luca Coelho authored
Add support to the 6GHz band (aka. Ultra High Band or UHB). This allows us to scan and connect to channels in that band, including all the relevant features, such as preferred scan channels, colocated channels etc. Co-developed-by:
Haim Dreyfuss <haim.dreyfuss@intel.com> Signed-off-by:
Ilan Peer <ilan.peer@intel.com> Signed-off-by:
Tova Mussai <tova.mussai@intel.com> Signed-off-by:
Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by:
Tali Levi Rovinsky <Tali.Levi-rovinsky@intel.com> Signed-off-by:
Ayala Beker <ayala.beker@intel.com> Signed-off-by:
-
Johannes Berg authored
This data is not necessarily the same across devices as we may modify it due to the number of antennas and for overrides (though in practice overrides are likely to be identical), so modifying the global data is wrong. Make a copy of it in the NVM data and modify it there instead. Signed-off-by:
-
Emmanuel Grumbach authored
We no longer need code that was introduced to differentiate between two early versions of 8260. We can remove this convoluted way to get the hardware version that was needed because of a bug in the register's configuration. Moreover, since we no longer need to access the PRPH registers, we no longer need to wake up the device, request ownership, etc... Remove all that. This allows us to get the rid of the obsolete comment about the AUX bus MISC address space which should have been moved when this code was moved away from here. Signed-off-by:
-
Emmanuel Grumbach authored
If we already read the NVM, we don't need to read it again. Make sure that this is the case with a WARNING. Signed-off-by:
-
