- 14 Feb, 2020 3 commits
-
-
John Crispin authored
Currently a mac80211 driver can only set the txq_limit when using wake_tx_queue. Not all drivers use wake_tx_queue. This patch adds a new element to wiphy allowing a driver to set a custom tx_queue_len and the code that will apply it in case it is set. The current default is 1000 which is too low for ath11k when doing HE rates. Signed-off-by:
John Crispin <john@phrozen.org> Link: https://lore.kernel.org/r/20200211122605.13002-1-john@phrozen.orgSigned-off-by:
Johannes Berg <johannes.berg@intel.com>
-
Sergey Matyukevich authored
Add the definition for Opportunistic Wireless Encryption AKM selector. Signed-off-by:
Sergey Matyukevich <sergey.matyukevich.os@quantenna.com> Link: https://lore.kernel.org/r/20200213131608.10541-3-sergey.matyukevich.os@quantenna.comSigned-off-by:
-
Ben Greear authored
With multiple VIFS ath10k, and probably others, tries to find the minimum txpower for all vifs and uses that when setting txpower in the firmware. If a second vif is added and starts to scan, it's txpower is not initialized yet and it set to zero. ath10k had a patch to ignore zero values, but then it is impossible to actually set txpower to zero. So, instead initialize the txpower to INT_MIN in mac80211, and let drivers know that means the power has not been set and so should be ignored. This should fix regression in: commit 88407beb Author: Ryan Hsu <ryanhsu@qca.qualcomm.com> Date: Tue Dec 13 14:55:19 2016 -0800 ath10k: fix incorrect txpower set by P2P_DEVICE interface Tested on ath10k 9984 with ath10k-ct firmware. Signed-off-by:
Ben Greear <greearb@candelatech.com> Link: https://lore.kernel.org/r/20191217183057.24586-1-greearb@candelatech.comSigned-off-by:
-
- 07 Feb, 2020 20 commits
-
-
Markus Theil authored
This is now a trivial patch, but for seeing the actual changes I (Johannes) split it out from the original. Signed-off-by:
Markus Theil <markus.theil@tu-ilmenau.de> Link: https://lore.kernel.org/r/20200115125522.3755-1-markus.theil@tu-ilmenau.de [split into separate cfg80211/mac80211 patches] Signed-off-by:
-
Markus Theil authored
When using control port over nl80211 in AP mode with pre-authentication, APs need to forward frames to other APs defined by their MAC address. Before this patch, pre-auth frames reaching user space over nl80211 control port have no longer any information about the dest attached, which can be used for forwarding to a controller or injecting the frame back to a ethernet interface over a AF_PACKET socket. Analog problems exist, when forwarding pre-auth frames from AP -> STA. This patch therefore adds the NL80211_ATTR_DST_MAC and NL80211_ATTR_SRC_MAC attributes to provide more context information when forwarding. The respective arguments are optional on tx and included on rx. Therefore unaware existing software is not affected. Software which wants to detect this feature, can do so by checking against: NL80211_EXT_FEATURE_CONTROL_PORT_OVER_NL80211_MAC_ADDRS Signed-off-by:
-
Shaul Triebitz authored
Parse also the RSN Extension IE when parsing the rest of the IEs. It will be used in a later patch. Signed-off-by:
Shaul Triebitz <shaul.triebitz@intel.com> Signed-off-by:
Luca Coelho <luciano.coelho@intel.com> Link: https://lore.kernel.org/r/20200131111300.891737-21-luca@coelho.fiSigned-off-by:
-
Ilan Peer authored
To support Pre Association Security Negotiation (PASN) while already associated to one AP, allow user space to register to Rx authentication frames, so that the user space logic would be able to receive/handle authentication frames from a different AP as part of PASN. Note that it is expected that user space would intelligently register for Rx authentication frames, i.e., only when PASN is used and configure a match filter only for PASN authentication algorithm, as otherwise the MLME functionality of mac80211 would be broken. Additionally, since some versions of the user space daemons wrongly register to all types of authentication frames (which might result in unexpected behavior) allow such registration if the request is for a specific authentication algorithm number. Signed-off-by:
Ilan Peer <ilan.peer@intel.com> Signed-off-by:
-
Zvika Yehudai authored
Remove redundant 'the' where 'the the' was written. Signed-off-by:
Zvika Yehudai <zvikayeh@gmail.com> Link: https://lore.kernel.org/r/20200203080823.24949-1-zvikayeh@gmail.comSigned-off-by:
-
Ilan Peer authored
The SMPS feature is defined in the specification only to be used by non-AP stations and not by APs, so remove the support for changing the AP's SMPS mode dynamically. Signed-off-by:
-
Ilan Peer authored
According to IEEE802.11 specifications the SM power save field in the HT capability IE and the HE extended capability IE is valid only in (re)association frames and should be ignored otherwise. Remove code paths that handled this also for non AP modes. Signed-off-by:
-
Andrei Otcheretianski authored
An AP that operates on 6GHz may respond with a broadcast probe response. Don't ignore such frames. Signed-off-by:
Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by:
-
Tova Mussai authored
In case of HE, the RX NSS is taken from the HE capabilities. If the supported NSS capabilities that are reported by AP for HE mode in the HE Capabilities element are different from the NSS capabilities that are reported by AP for the VHT mode in the VHT Capabilities element, use the lowest supported NSS to not get all the values confused. Signed-off-by:
Tova Mussai <tova.mussai@intel.com> Signed-off-by:
-
Johannes Berg authored
Set the station bandwidth in HE capability parsing and from HE capability as the HT/VHT information will not be present on the 6 GHz band. Signed-off-by:
-
Johannes Berg authored
We should not include the supported channels element if we have (advertise) support for extended channel switching. To avoid any interop issues because we always added it in the past, obey this restriction only in the (new) 6 GHz band. Signed-off-by:
-
Daniel Gabay authored
Disable HE if the beacon does not contain an HE operation IE. Signed-off-by:
Daniel Gabay <daniel.gabay@intel.com> Signed-off-by:
-
Luca Coelho authored
This function always returns 0, so there's no point in returning int. Make it void and remove the impossible error-path when calling it. Signed-off-by:
-
Johannes Berg authored
Set ndev->hw_features as well as ndev->features to allow changing the TX-related features with ethtool. We cannot (yet) change RX-related features since that requires telling the driver about it and we have no API for that yet. Signed-off-by:
-
Haim Dreyfuss authored
The wireless device might be capable to connect HE as well as the AP. However, the regulatory domain might forbid it. Check whether the regulatory domain allows HE connection when considering if HE IE should be added. Also, add it when setting our peer capability. Signed-off-by:
Haim Dreyfuss <haim.dreyfuss@intel.com> Signed-off-by:
-
Johannes Berg authored
This code was really ugly, refactor it a bit to make it more readable. While at it, use sizeof() and fix the UORA element length check bug. Signed-off-by:
-
Johannes Berg authored
Check early on that a device has support for QoS (at least 4 queues) when it supports HT/VHT/HE, so we don't have to check this while connecting. This lets us clean up the code there: move some of it into channel preparation to clean up a bit more, and then change the logic to only check the "wmm_used" flag. Additionally, disable HE consistently when VHT is disabled. Signed-off-by:
-
Veerendranath Jakkam authored
Commit ab4dfa20 ("cfg80211: Allow drivers to advertise supported AKM suites") introduces the support to advertize supported AKMs to userspace. This needs an enhancement to advertize the AKM support per interface type, specifically for the cfg80211-based drivers that implement SME and use different mechanisms to support the AKM's for each interface type (e.g., the support for SAE, OWE AKM's take different paths for such drivers on STA/AP mode). This commit aims the same and enhances the earlier mechanism of advertizing the AKMs per wiphy. Add new nl80211 attributes and data structure to provide supported AKMs per interface type to userspace. the AKMs advertized in akm_suites are default capabilities if not advertized for a specific interface type in iftype_akm_suites. Signed-off-by:
Veerendranath Jakkam <vjakkam@codeaurora.org> Link: https://lore.kernel.org/r/20200126203032.21934-1-vjakkam@codeaurora.orgSigned-off-by:
-
Haim Dreyfuss authored
The regulatory domain might forbid HE operation. Certain regulatory domains may restrict it for specific channels whereas others may do it for the whole regulatory domain. Add an option to indicate it in the channel flag. Signed-off-by:
-
John Crispin authored
The 802.11 encapsulation returned early when setting up the keys in case offloading was enabled. This causes ieee802.11w to not work anymore. Fix this by moving the check for offloading into the switch/case construct and allowing CCMP/GCMP keys. With this patch applied ieee80211w works again when enabling offloading. Fixes: 50ff477a ("mac80211: add 802.11 encapsulation offloading support") Reported-by:
Maharaja Kennadyrajan <mkenna@qti.qualcomm.com> Signed-off-by:
-
- 15 Jan, 2020 6 commits
-
-
Thomas Pedersen authored
commit 08a5bdde ("mac80211: consider QoS Null frames for STA_NULLFUNC_ACKED") Fixed a bug where we failed to take into account a nullfunc frame can be either non-QoS or QoS. It turns out there is at least one more bug in ieee80211_sta_tx_notify(), introduced in commit 7b6ddeaf ("mac80211: use QoS NDP for AP probing"), where we forgot to check for the QoS variant and so assumed the QoS nullfunc frame never went out Fix this by adding a helper ieee80211_is_any_nullfunc() which consolidates the check for non-QoS and QoS nullfunc frames. Replace existing compound conditionals and add a couple more missing checks for QoS variant. Signed-off-by:
Thomas Pedersen <thomas@adapt-ip.com> Link: https://lore.kernel.org/r/20200114055940.18502-3-thomas@adapt-ip.comSigned-off-by:
-
Thomas Pedersen authored
Advertise the correct flags to mac80211 to indicate PS trigger frames and frame buffering should be handled by mac80211. This means mac80211_hwsim will now also have to release buffered multicast frames after a (DTIM) beacon. Leave STA power save disabled by default since it causes several p2p, scanning, and rrm hostap tests to fail. Signed-off-by:
-
Markus Theil authored
Before this patch, frames which where successfully transmitted without requiring acks where accounted as lost frames. Signed-off-by:
-
John Crispin authored
It is now possible to propagate BSS color settings into the subsystem. Lets make mac80211 also handle them so that we can send them further down the stack into the drivers. We drop the old bss_color field and change iwlwifi to use the new he_bss_color struct. Signed-off-by:
-
John Crispin authored
This patch adds the attributes, policy and parsing code to allow userland to send the info about the BSS coloring settings to the kernel. Signed-off-by:
-
Johannes Berg authored
This is too low, at least ETSI regulatory domain would allow 36 dBm on 5 GHz in (some? most?) cases. But there's no reason hwsim should have a limit at all, so remove it. See also https://lore.kernel.org/r/20191108152013.13418-1-ramonreisfontes@gmail.com Change-Id: I6edf81d2e82bc8ce7baa40bfc0f210b7b56ef275 Reported-by:
Ramon Fontes <ramonreisfontes@gmail.com> Signed-off-by:
-
- 02 Jan, 2020 2 commits
-
-
Lorenzo Bianconi authored
Improve airtime_flags debugfs handler readability reporting configured airtime flags in both numeric and human readable manner Signed-off-by:
Lorenzo Bianconi <lorenzo@kernel.org> Acked-by:
Toke Høiland-Jørgensen <toke@redhat.com> Link: https://lore.kernel.org/r/9df7e40b45e95bb0b320317831455beaed1ee3ee.1574872357.git.lorenzo@kernel.org [remove AQL since it's no longer there] Signed-off-by:
-
Johannes Berg authored
Merging to get the mac80211 updates that have since propagated into net-next. Signed-off-by:
-
- 01 Jan, 2020 1 commit
-
-
git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queueDavid S. Miller authored
Jeff Kirsher says: ==================== 1GbE Intel Wired LAN Driver Updates 2019-12-31 This series contains updates to e1000e, igb and igc only. Robert Beckett provide an igb change to assist in keeping packets from being dropped due to receive descriptor ring being full when receive flow control is enabled. Create a separate function to setup SRRCTL to ease in reuse and ensure that setting of the drop enable bit only if receive flow control is not enabled. Sasha adds support for scatter gather support in igc. Improve the direct memory address mapping flow by optimizing/simplifying and more clear. Update igc to use pci_release_mem_regions() instead of pci_release_selected_regions(). Clean up function header comments to align with the actual code. Adds support for 64 bit DMA access, to help handle socket buffer fragments in high memory. Adds legacy power management support in igc by implementing suspend, resume, runtime_suspend/resume, and runtime_idle callbacks. Clean up references to Serdes interface in igc since that interface is not supported for i225 devices. Alex replaces the pr_info calls with netdev_info in all cases related to netdev link state, as suggested by Joe Perches. ==================== Signed-off-by:David S. Miller <davem@davemloft.net>
-
- 31 Dec, 2019 8 commits
-
-
Sasha Neftin authored
Serdes interface is not applicable for i225 devices. Remove this from comments and make comments more clearly. Signed-off-by:
Sasha Neftin <sasha.neftin@intel.com> Tested-by:
Aaron Brown <aaron.f.brown@intel.com> Signed-off-by:
Jeff Kirsher <jeffrey.t.kirsher@intel.com>
-
Alexander Duyck authored
Replace the pr_info calls with netdev_info in all cases related to the netdevice link state. As a result of this patch the link messages will change as shown below. Before: e1000e: ens3 NIC Link is Down e1000e: ens3 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: Rx/Tx After: e1000e 0000:00:03.0 ens3: NIC Link is Down e1000e 0000:00:03.0 ens3: NIC Link is Up 1000 Mbps Full Duplex, Flow Control: Rx/Tx Suggested-by:
Joe Perches <joe@perches.com> Signed-off-by:
Alexander Duyck <alexander.h.duyck@linux.intel.com> Acked-by:
Neil Horman <nhorman@tuxdriver.com> Tested-by:
-
Sasha Neftin authored
Add suspend, resume, runtime_suspend, runtime_resume and runtime_idle callbacks implementation. Reported-by:
kbuild test robot <lpk@intel.com> Signed-off-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/netdev/netDavid S. Miller authored
Simple overlapping changes in bpf land wrt. bpf_helper_defs.h handling. Signed-off-by: -
Sasha Neftin authored
On relevant platforms ndo_start_xmit can handle socket buffer fragments in high memory Signed-off-by:
-
Sasha Neftin authored
igc_watchdog, igc_set_interrupt_capability, igc_init_interrupt_scheme, __igc_open and __igc_close parameter descriptions has not reflected functions meaning. Add meaningful description. Signed-off-by:
-
Sasha Neftin authored
The function description for igc_alloc_rx_buffers has not reflected the function meaning. Add meaningful description. Signed-off-by:
-
Sasha Neftin authored
The function description for igc_is_non_eop includes an extra @skb parameter description. This parameter doesn't exist on the function, so remove it. Suggested-by:
Jacob Keller <jacob.e.keller@intel.com> Signed-off-by:
-
