- 23 Oct, 2014 5 commits
-
-
Arik Nemtsov authored
During reconfig the station list is traversed in order and station are added back to the driver. Make sure the stations are added to the driver in the same order they were added to mac80211. This has a real side effect - some drivers (iwlwifi) require TDLS stations to be added only after the AP station for the same network. Signed-off-by:
Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by:
Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by:
Johannes Berg <johannes.berg@intel.com>
-
Arik Nemtsov authored
Some drivers need to know which station is the TDLS link initiator. Expose this value via the mac80211 ieee80211_sta structure. Signed-off-by:
-
Arik Nemtsov authored
When an IV is generated, only the MAC header is moved back. The network header location remains the same relative to the skb head, as the new IV is using headroom space that was reserved in advance. Signed-off-by:
-
Andrei Otcheretianski authored
Export ieee80211_ie_split function, so it can be reused by drivers which need to insert additional elements. Signed-off-by:
Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by:
-
Karl Beldan authored
When displaying a rate through debugfs minstrel_ht guesses its flags comparing group indexes. Since 3ec373c4 ("mac80211: minstrel_ht: include type (cck/ht) in rates flag"), the rate flags of interest are present in the mcs_group-s, so use it. While improving the code, this also fixes a smatch false positive "error: testing array offset 'i' after use" in minstrel_ht_stats_dump. This warning only triggers after 9208247d ("mac80211: minstrel_ht: add basic support for VHT rates <= 3SS@80MHz") with CONFIG_MAC80211_RC_MINSTREL_VHT unset because then MINSTREL_VHT_GROUP_0 is above MINSTREL_GROUPS_NB and smatch only barks when the "testing array offset" seems to prevent possible out of bonds accesses (which does not happen here since i < ARRAY_SIZE(mi->groups)). Signed-off-by:
Karl Beldan <karl.beldan@rivierawaves.com> Cc: Felix Fietkau <nbd@openwrt.org> Signed-off-by:
-
- 22 Oct, 2014 4 commits
-
-
Johannes Berg authored
When a key is tainted during resume, it is no longer programmed into the device; however, it's uploaded flag may (will) be set. Clear the flag when not programming it because it's tainted to avoid attempting to remove it again later. Signed-off-by: -
Johannes Berg authored
Use the currently existing APIs between mac80211 and the low level driver to implement WMM admission control. The low level driver needs to report the media time used by each transmitted packet in ieee80211_tx_status. Based on that information, mac80211 will modify the QoS parameters of the admission controlled Access Category when the limit is reached. Once the original QoS parameters can be restored, mac80211 will do so. One issue with this approach is that management frames will also erroneously be downgraded, but the upside is that the implementation is simple. In the future, it can be extended to driver- or device-based implementations that are better. Signed-off-by:
-
Johannes Berg authored
There's no reason to ever set invalid CW_min/CW_max to the drivers, we should catch it in higher layers. However, the consequences of setting it wrong can be quite severe, so double-check at a low level and error out for invalid data. Signed-off-by: -
Johannes Berg authored
During the review of the corresponding wpa_supplicant patches we noticed that the only way for it to detect that this functionality is supported currently is to check for the command support. This can be misleading though, as the command was also designed to, in the future, support pure 802.11 TSPECs. Expose the WMM-TSPEC feature flag to nl80211 so later we can also expose an 802.11-TSPEC feature flag (if needed) to differentiate the two cases. Note: this change isn't needed in 3.18 as there's no driver there yet that supports the functionality at all. Signed-off-by:
-
- 21 Oct, 2014 2 commits
-
-
Fabian Frederick authored
The debugfs_remove() function can safely take NULL parameters so the additionally null test isn't required, and there's no other reason to have it here, so remove it. Signed-off-by:
Fabian Frederick <fabf@skynet.be> [rewrite commit message, re-introduce blank line after assert] Signed-off-by:
-
Karl Beldan authored
When the new CONFIG_MAC80211_RC_MINSTREL_VHT is not set (default 'N'), there is no behavioral change including in sampling and MCS_GROUP_RATES remains 8. Otherwise MCS_GROUP_RATES is 10, and a module parameter *vht_only* (default 'true'), restricts the rates selection to VHT when VHT is supported. Regarding the debugfs stats buffer: It is explicitly increased from 8k to 32k to fit every rates incl. when both HT and VHT rates are enabled, as for the format, before: type rate tpt eprob *prob ret *ok(*cum) ok( cum) HT20/LGI ABCDP MCS0 0.0 0.0 0.0 1 0( 0) 0( 0) after: type rate tpt eprob *prob ret *ok(*cum) ok( cum) HT20/LGI ABCDP MCS0 0.0 0.0 0.0 1 0( 0) 0( 0) VHT40/LGI MCS5/2 0.0 0.0 0.0 0 0( 0) 0( 0) Signed-off-by:
-
- 20 Oct, 2014 8 commits
-
-
Karl Beldan authored
ATM, we grep cck rates idx with idx / MCS_GROUP_RATES == MINSTREL_CCK_GROUP. Matching neither-cck-non-ht rates could be done by replacing '==' with '>', however it would be less versatile or explicit. This will allow to match VHT rates with IEEE80211_TX_RC_VHT_MCS. Signed-off-by:
-
Karl Beldan authored
No functional change. Signed-off-by:
-
Karl Beldan authored
Since 5935839a ("mac80211: improve minstrel_ht rate sorting by throughput & probability"), the rate indexes are manipulated via u8's and hence allow for a maximum of 256 mcs_group entries in minstrel_mcs_groups. ATM, minstrel_ht advertizes support up to 3HTSS@40MHz, consuming: 8(MCS_GROUP_RATES) * (3(SS)*2(GI)*2(BW)+1(CCK)), i.e. 104 entries. Support for 3VHTSS@80MHz will require: 10(MCS_GROUP_RATES) * (3(SS)*2(GI)*2(BW)+1(CCK)) + 10(MCS_GROUP_RATES) * (3(SS)*2(GI)*3(BW)), i.e. 130 + 180 entries. This change moves from u8s to u16s where necessary. Signed-off-by:
-
Johannes Berg authored
This was needed to avoid conflicts in the minstrel changes. Signed-off-by: -
Johannes Berg authored
This showed up as a sparse warning (with higher verbosity) and is certainly correct - the change flags should be unsigned. It's not that important since high flag numbers aren't used and bitwise operations would still work. Signed-off-by: -
Jouni Malinen authored
The optional NL80211_ATTR_MGMT_SUBTYPE and NL80211_ATTR_REASON_CODE attributes can now be included in NL80211_CMD_DEL_STATION to indicate to the driver which frame (Deauthentication/Disassociation) and reason code in that frame should be used to indicate removal to the specific station. This is used by drivers that implement AP SME and generate those frames internally. Signed-off-by:
Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by:
-
Karl Beldan authored
ATM an HT rc_stats line is 106 chars. Times 8(MCS_GROUP_RATES)*3(SS)*2(GI)*2(BW) + CCK(4), i.e. x100, this is well above the current 8192 - sizeof(*ms) currently allocated. Fix this by squeezing the output as follows (not that we're short on memory but this also improves readability and range, the new format adds one more digit to *ok/*cum and ok/cum): - Before (HT) (106 ch): type rate throughput ewma prob this prob retry this succ/attempt success attempts CCK/LP 5.5M 0.0 0.0 0.0 0 0( 0) 0 0 HT20/LGI ABCDP MCS0 0.0 0.0 0.0 1 0( 0) 0 0 - After (75 ch): type rate tpt eprob *prob ret *ok(*cum) ok( cum) CCK/LP 5.5M 0.0 0.0 0.0 0 0( 0) 0( 0) HT20/LGI ABCDP MCS0 0.0 0.0 0.0 1 0( 0) 0( 0) - Align non-HT format Before (non-HT) (83 ch): rate throughput ewma prob this prob this succ/attempt success attempts ABCDP 6 0.0 0.0 0.0 0( 0) 0 0 54 0.0 0.0 0.0 0( 0) 0 0 - After (61 ch): rate tpt eprob *prob *ok(*cum) ok( cum) ABCDP 1 0.0 0.0 0.0 0( 0) 0( 0) 54 0.0 0.0 0.0 0( 0) 0( 0) *This also adds dynamic checks for overflow, lowers the size of the non-HT request (allowing > 30 entries) and replaces the buddy-rounded allocations (s/sizeof(*ms) + 8192/8192). Signed-off-by:Felix Fietkau <nbd@openwrt.org> Signed-off-by:
-
Jouni Malinen authored
This makes it easier to add new parameters for the del_station calls without having to modify all drivers that use this. Signed-off-by:
-
- 14 Oct, 2014 1 commit
-
-
Karl Beldan authored
It affects non-(V)HT rates and can lead to selecting an rts_cts rate that is not a basic rate or way superior to the reference rate (ATM rates[0] used for the 1st attempt of the protected frame data). E.g, assuming drivers register growing (bitrate) sorted tables of ieee80211_rate-s, having : - rates[0].idx == d'2 and basic_rates == b'10100 will select rts_cts idx b'10011 & ~d'(BIT(2)-1), i.e. 1, likewise - rates[0].idx == d'2 and basic_rates == b'10001 will select rts_cts idx b'10000 The first is not a basic rate and the second is > rates[0]. Also, wrt severity of the addressed misbehavior, ATM we only have one rts_cts_rate_idx rather than one per rate table entry, so this idx might still point to bitrates > rates[1..MAX_RATES]. Fixes: 5253ffb8 ("mac80211: always pick a basic rate to tx RTS/CTS for pre-HT rates") Cc: stable@vger.kernel.org Signed-off-by:
-
- 10 Oct, 2014 2 commits
-
-
Karl Beldan authored
ATM, specifying the frequency when connecting sends a void 'supported rates' EID. Signed-off-by:
-
Michal Kazior authored
It is okay to enable DFS for channel contexts based drivers as long as no combination advertises radar detection and multi-channel operation at the same time. Signed-off-by:
Michal Kazior <michal.kazior@tieto.com> Signed-off-by:
-
- 09 Oct, 2014 18 commits
-
-
Fabian Frederick authored
No need to store ieee80211_vif_use_reserved_context() result and test it before returning. Signed-off-by:
-
Jukka Rissanen authored
Add support for new attribute HWSIM_ATTR_DESTROY_RADIO_ON_CLOSE which can be set by the user space component. The attribute will cause the kernel to destroy all those radios that were created by a process if that process dies. The old behaviour i.e., radios are persistent is still the default. Signed-off-by:
Jukka Rissanen <jukka.rissanen@linux.intel.com> Signed-off-by:
-
Luciano Coelho authored
Channel switch with multiple channel contexts should now work fine. Remove check that disallows switches when multiple contexts are in use. Signed-off-by:
Luciano Coelho <luciano.coelho@intel.com> Signed-off-by:
-
Luciano Coelho authored
Instead of immediately reopening the queues (in case of block_tx), calling the post_channel_switch operation and sending the notification, wait for the first beacon on the new channel. This makes sure that we don't lose packets if the AP/GO is not on the new channel yet. Signed-off-by:
-
Luciano Coelho authored
As a counterpart to the pre_channel_switch operation, add a post_channel_switch operation. This allows the drivers to go back to a normal configuration after the channel switch is completed. Signed-off-by:
-
Luciano Coelho authored
Some drivers may need to prepare for a channel switch also when it is initiated from the remote side (eg. station, P2P client). To make this possible, add a generic callback that can be called for all interface types. Signed-off-by:
-
Luciano Coelho authored
The Extended Channel Switching capability bit in the extended capabilities element must be set if the driver supports CSA on non-beaconing interfaces. Since this capability needs to be set during driver registration, the extended_capabiliities global variable needs to be moved to the local structure so that it can be modified. Signed-off-by:
-
Luciano Coelho authored
Some devices may need the device timestamp in order to synchronize the channel switch. To pass this value back to the driver, add it to the channel switch structure and copy the device_timestamp value received in the rx info structure into it. Signed-off-by:
-
Luciano Coelho authored
The nl80211 channel switch count attribute (NL80211_ATTR_CH_SWITCH_COUNT) is specified as u32, but the specification uses u8 for the counter. To make sure strange things don't happen without informing the user, sanity check the value and return -EINVAL if it doesn't fit in u8. Signed-off-by:
-
Henning Rogge authored
Implement get_mpp and dump_mpp cfg80211_ops to export the content of the 802.11s mesh proxy path table to userspace. Signed-off-by:
Henning Rogge <henning.rogge@fkie.fraunhofer.de> Signed-off-by:
-
Henning Rogge authored
Add two new cfg80211 operations for querying a table with proxied mesh paths. Signed-off-by:
-
Fabian Frederick authored
Correct the kernel-doc, the parameter is called "blocked" not "state". Signed-off-by:
-
Luciano Coelho authored
The chandef of the channel context a vif is using may be different than the chandef of the vif itself. For instance, the bandwidth used by the vif may be narrower than the one configured in the channel context. To avoid confusion, return the vif's chandef in ieee80211_cfg_get_channel() instead of the chandef of the channel context. Signed-off-by:
-
Karl Beldan authored
Commit 5935839a ("mac80211: improve minstrel_ht rate sorting by throughput & probability") replaced the constant 8 with MCS_GROUP_RATES when getting the number of streams of an HT MCS. See commit 7a5e3fa2 ("mac80211: minstrel_ht: replace some occurences of MCS_GROUP_RATES"). Fixes: 5935839a ("mac80211: improve minstrel_ht rate sorting by throughput & probability") Signed-off-by:
-
Liad Kaufman authored
Forgot to add an entry to the struct description of sta_info. Signed-off-by:
Liad Kaufman <liad.kaufman@intel.com> Reviewed-by:
-
Ben Greear authored
Trivial cleanups. Signed-off-by:
Ben Greear <greearb@candelatech.com> Signed-off-by:
-
Ben Greear authored
If the packet can't be delivered to userspace (at all or quickly enough) then it can leak - fix that. Signed-off-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-nextLinus Torvalds authored
Pull networking updates from David Miller: "Most notable changes in here: 1) By far the biggest accomplishment, thanks to a large range of contributors, is the addition of multi-send for transmit. This is the result of discussions back in Chicago, and the hard work of several individuals. Now, when the ->ndo_start_xmit() method of a driver sees skb->xmit_more as true, it can choose to defer the doorbell telling the driver to start processing the new TX queue entires. skb->xmit_more means that the generic networking is guaranteed to call the driver immediately with another SKB to send. There is logic added to the qdisc layer to dequeue multiple packets at a time, and the handling mis-predicted offloads in software is now done with no locks held. Finally, pktgen is extended to have a "burst" parameter that can be used to test a multi-send implementation. Several drivers have xmit_more support: i40e, igb, ixgbe, mlx4, virtio_net Adding support is almost trivial, so export more drivers to support this optimization soon. I want to thank, in no particular or implied order, Jesper Dangaard Brouer, Eric Dumazet, Alexander Duyck, Tom Herbert, Jamal Hadi Salim, John Fastabend, Florian Westphal, Daniel Borkmann, David Tat, Hannes Frederic Sowa, and Rusty Russell. 2) PTP and timestamping support in bnx2x, from Michal Kalderon. 3) Allow adjusting the rx_copybreak threshold for a driver via ethtool, and add rx_copybreak support to enic driver. From Govindarajulu Varadarajan. 4) Significant enhancements to the generic PHY layer and the bcm7xxx driver in particular (EEE support, auto power down, etc.) from Florian Fainelli. 5) Allow raw buffers to be used for flow dissection, allowing drivers to determine the optimal "linear pull" size for devices that DMA into pools of pages. The objective is to get exactly the necessary amount of headers into the linear SKB area pre-pulled, but no more. The new interface drivers use is eth_get_headlen(). From WANG Cong, with driver conversions (several had their own by-hand duplicated implementations) by Alexander Duyck and Eric Dumazet. 6) Support checksumming more smoothly and efficiently for encapsulations, and add "foo over UDP" facility. From Tom Herbert. 7) Add Broadcom SF2 switch driver to DSA layer, from Florian Fainelli. 8) eBPF now can load programs via a system call and has an extensive testsuite. Alexei Starovoitov and Daniel Borkmann. 9) Major overhaul of the packet scheduler to use RCU in several major areas such as the classifiers and rate estimators. From John Fastabend. 10) Add driver for Intel FM10000 Ethernet Switch, from Alexander Duyck. 11) Rearrange TCP_SKB_CB() to reduce cache line misses, from Eric Dumazet. 12) Add Datacenter TCP congestion control algorithm support, From Florian Westphal. 13) Reorganize sk_buff so that __copy_skb_header() is significantly faster. From Eric Dumazet" * git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1558 commits) netlabel: directly return netlbl_unlabel_genl_init() net: add netdev_txq_bql_{enqueue, complete}_prefetchw() helpers net: description of dma_cookie cause make xmldocs warning cxgb4: clean up a type issue cxgb4: potential shift wrapping bug i40e: skb->xmit_more support net: fs_enet: Add NAPI TX net: fs_enet: Remove non NAPI RX r8169:add support for RTL8168EP net_sched: copy exts->type in tcf_exts_change() wimax: convert printk to pr_foo() af_unix: remove 0 assignment on static ipv6: Do not warn for informational ICMP messages, regardless of type. Update Intel Ethernet Driver maintainers list bridge: Save frag_max_size between PRE_ROUTING and POST_ROUTING tipc: fix bug in multicast congestion handling net: better IFF_XMIT_DST_RELEASE support net/mlx4_en: remove NETDEV_TX_BUSY 3c59x: fix bad split of cpu_to_le32(pci_map_single()) net: bcmgenet: fix Tx ring priority programming ...
-
