NetBSD/sys/net80211/ieee80211_node.h

380 lines
15 KiB
C
Raw Normal View History

/* $NetBSD: ieee80211_node.h,v 1.24 2011/10/07 16:51:45 dyoung Exp $ */
/*-
* Copyright (c) 2001 Atsushi Onoe
* Copyright (c) 2002-2005 Sam Leffler, Errno Consulting
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. The name of the author may not be used to endorse or promote products
* derived from this software without specific prior written permission.
*
* Alternatively, this software may be distributed under the terms of the
* GNU General Public License ("GPL") version 2 as published by the Free
* Software Foundation.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
* IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
* IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
* NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
* THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
Resolve conflicts and adapt to NetBSD. Thanks to dyoung@, scw@, and perry@ for help testing. 2005-08-30 15:27 avatar Properly set ic_curchan before calling back to device driver to do channel switching(ifconfig devX channel Y). This fix should make channel changing works again in monitor mode. Submitted by: sam X-MFC-With: other ic_curchan changes 2005-08-13 18:50 sam revert 1.64: we cannot use the channel characteristics to decide when to do 11g erp sta accounting because b/g channels show up as false positives when operating in 11b. Noticed by: Michal Mertl 2005-08-13 18:31 sam Extend acl support to pass ioctl requests through and use this to add support for getting the current policy setting and collecting the list of mac addresses in the acl table. Submitted by: Michal Mertl (original version) MFC after: 2 weeks 2005-08-10 18:42 sam Don't use ic_curmode to decide when to do 11g station accounting, use the station channel properties. Fixes assert failure/bogus operation when an ap is operating in 11a and has associated stations then switches to 11g. Noticed by: Michal Mertl Reviewed by: avatar MFC after: 2 weeks 2005-08-10 17:22 sam Clarify/fix handling of the current channel: o add ic_curchan and use it uniformly for specifying the current channel instead of overloading ic->ic_bss->ni_chan (or in some drivers ic_ibss_chan) o add ieee80211_scanparams structure to encapsulate scanning-related state captured for rx frames o move rx beacon+probe response frame handling into separate routines o change beacon+probe response handling to treat the scan table more like a scan cache--look for an existing entry before adding a new one; this combined with ic_curchan use corrects handling of stations that were previously found at a different channel o move adhoc neighbor discovery by beacon+probe response frames to a new ieee80211_add_neighbor routine Reviewed by: avatar Tested by: avatar, Michal Mertl MFC after: 2 weeks 2005-08-09 11:19 rwatson Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days 2005-08-08 19:46 sam Split crypto tx+rx key indices and add a key index -> node mapping table: Crypto changes: o change driver/net80211 key_alloc api to return tx+rx key indices; a driver can leave the rx key index set to IEEE80211_KEYIX_NONE or set it to be the same as the tx key index (the former disables use of the key index in building the keyix->node mapping table and is the default setup for naive drivers by null_key_alloc) o add cs_max_keyid to crypto state to specify the max h/w key index a driver will return; this is used to allocate the key index mapping table and to bounds check table loookups o while here introduce ieee80211_keyix (finally) for the type of a h/w key index o change crypto notifiers for rx failures to pass the rx key index up as appropriate (michael failure, replay, etc.) Node table changes: o optionally allocate a h/w key index to node mapping table for the station table using the max key index setting supplied by drivers (note the scan table does not get a map) o defer node table allocation to lateattach so the driver has a chance to set the max key id to size the key index map o while here also defer the aid bitmap allocation o add new ieee80211_find_rxnode_withkey api to find a sta/node entry on frame receive with an optional h/w key index to use in checking mapping table; also updates the map if it does a hash lookup and the found node has a rx key index set in the unicast key; note this work is separated from the old ieee80211_find_rxnode call so drivers do not need to be aware of the new mechanism o move some node table manipulation under the node table lock to close a race on node delete o add ieee80211_node_delucastkey to do the dirty work of deleting unicast key state for a node (deletes any key and handles key map references) Ath driver: o nuke private sc_keyixmap mechansim in favor of net80211 support o update key alloc api These changes close several race conditions for the ath driver operating in ap mode. Other drivers should see no change. Station mode operation for ath no longer uses the key index map but performance tests show no noticeable change and this will be fixed when the scan table is eliminated with the new scanning support. Tested by: Michal Mertl, avatar, others Reviewed by: avatar, others MFC after: 2 weeks 2005-08-08 06:49 sam use ieee80211_iterate_nodes to retrieve station data; the previous code walked the list w/o locking MFC after: 1 week 2005-08-08 04:30 sam Cleanup beacon/listen interval handling: o separate configured beacon interval from listen interval; this avoids potential use of one value for the other (e.g. setting powersavesleep to 0 clobbers the beacon interval used in hostap or ibss mode) o bounds check the beacon interval received in probe response and beacon frames and drop frames with bogus settings; not clear if we should instead clamp the value as any alteration would result in mismatched sta+ap configuration and probably be more confusing (don't want to log to the console but perhaps ok with rate limiting) o while here up max beacon interval to reflect WiFi standard Noticed by: Martin <nakal@nurfuerspam.de> MFC after: 1 week 2005-08-06 05:57 sam fix debug msg typo MFC after: 3 days 2005-08-06 05:56 sam Fix handling of frames sent prior to a station being authorized when operating in ap mode. Previously we allocated a node from the station table, sent the frame (using the node), then released the reference that "held the frame in the table". But while the frame was in flight the node might be reclaimed which could lead to problems. The solution is to add an ieee80211_tmp_node routine that crafts a node that does exist in a table and so isn't ever reclaimed; it exists only so long as the associated frame is in flight. MFC after: 5 days 2005-07-31 07:12 sam close a race between reclaiming a node when a station is inactive and sending the null data frame used to probe inactive stations MFC after: 5 days 2005-07-27 05:41 sam when bridging internally bypass the bss node as traffic to it must follow the normal input path Submitted by: Michal Mertl MFC after: 5 days 2005-07-27 03:53 sam bandaid ni_fails handling so ap's with association failures are reconsidered after a bit; a proper fix involves more changes to the scanning infrastructure Reviewed by: avatar, David Young MFC after: 5 days 2005-07-23 01:16 sam the AREF flag is only meaningful in ap mode; adhoc neighbors now are timed out of the sta/neighbor table 2005-07-23 00:25 sam o move inactivity-related debug msgs under IEEE80211_MSG_INACT o probe inactive neighbors in adhoc mode (they don't have an association id so previously were being timed out) MFC after: 3 days 2005-07-22 22:11 sam split xmit of probe request frame out into a separate routine that takes explicit parameters; this will be needed when scanning is decoupled from the state machine to do bg scanning MFC after: 3 days 2005-07-22 21:48 sam split 802.11 frame xmit setup code into ieee80211_send_setup MFC after: 3 days 2005-07-22 18:57 sam simplify ic_newassoc callback MFC after: 3 days 2005-07-22 18:54 sam simplify ieee80211_ibss_merge api MFC after: 3 days 2005-07-22 18:50 sam add stats we know we'll need soon and some spare fields for future expansion MFC after: 3 days 2005-07-22 18:45 sam simplify tim callback api MFC after: 3 days 2005-07-22 18:42 sam don't include 802.3 header in min frame length calculation as it may not be present for a frag; fixes problem with small (fragmented) frames being dropped Obtained from: Atheros MFC after: 3 days 2005-07-22 18:36 sam simplify ieee80211_node_authorize and ieee80211_node_unauthorize api's MFC after: 3 days 2005-07-22 18:31 sam simplifiy ieee80211_send_nulldata api MFC after: 3 days 2005-07-22 18:29 sam simplify rate set api's by removing ic parameter (implicit in node reference) MFC after: 3 days 2005-07-22 18:21 sam reject association requests with a wpa/rsn ie when wpa/rsn is not configured on the ap; previously we either ignored the ie or (possibly) failed an assertion Obtained from: Atheros MFC after: 3 days 2005-07-22 18:16 sam missed one in last commit; add device name to discard msgs 2005-07-22 18:13 sam include device name in discard msgs 2005-07-22 18:12 sam add diag msgs for frames discarded because the direction field is wrong 2005-07-22 18:08 sam split data frame delivery out to a new function ieee80211_deliver_data 2005-07-22 18:00 sam o add IEEE80211_IOC_FRAGTHRESHOLD for getting+setting the tx fragmentation threshold o fix bounds checking on IEEE80211_IOC_RTSTHRESHOLD MFC after: 3 days 2005-07-22 17:55 sam o add IEEE80211_FRAG_DEFAULT o move default settings for RTS and frag thresholds to ieee80211_var.h 2005-07-22 17:50 sam diff reduction against p4: define IEEE80211_FIXED_RATE_NONE and use it instead of -1 2005-07-22 17:37 sam add flags missed in last merge 2005-07-22 17:36 sam Diff reduction against p4: o add ic_flags_ext for eventual extention of ic_flags o define/reserve flag+capabilities bits for superg, bg scan, and roaming support o refactor debug msg macros MFC after: 3 days 2005-07-22 06:17 sam send a response when an auth request is denied due to an acl; might be better to silently ignore the frame but this way we give stations a chance of figuring out what's wrong 2005-07-22 06:15 sam remove excess whitespace 2005-07-22 05:55 sam use IF_HANDOFF when bridging frames internally so if_start gets called; fixes communication between associated sta's MFC after: 3 days 2005-07-11 04:06 sam Handle encrypt of arbitarily fragmented mbuf chains: previously we bailed if we couldn't collect the 16-bytes of data required for an aes block cipher in 2 mbufs; now we deal with it. While here make space accounting signed so a sanity check does the right thing for malformed mbuf chains. Approved by: re (scottl) 2005-07-11 04:00 sam nuke assert that duplicates real check Reviewed by: avatar Approved by: re (scottl)
2005-11-18 19:40:08 +03:00
* $FreeBSD: src/sys/net80211/ieee80211_node.h,v 1.22 2005/08/10 16:22:29 sam Exp $
*/
#ifndef _NET80211_IEEE80211_NODE_H_
#define _NET80211_IEEE80211_NODE_H_
#include <sys/atomic.h>
#include <net80211/ieee80211_netbsd.h>
#include <net80211/ieee80211_ioctl.h> /* for ieee80211_nodestats */
#ifdef _KERNEL
/*
* Each ieee80211com instance has a single timer that fires once a
* second. This is used to initiate various work depending on the
* state of the instance: scanning (passive or active), ``transition''
* (waiting for a response to a management frame when operating
* as a station), and node inactivity processing (when operating
* as an AP). For inactivity processing each node has a timeout
* set in it's ni_inact field that is decremented on each timeout
* and the node is reclaimed when the counter goes to zero. We
* use different inactivity timeout values depending on whether
* the node is associated and authorized (either by 802.1x or
* open/shared key authentication) or associated but yet to be
* authorized. The latter timeout is shorter to more aggressively
* reclaim nodes that leave part way through the 802.1x exchange.
*/
#define IEEE80211_INACT_WAIT 15 /* inactivity interval (secs) */
#define IEEE80211_INACT_INIT (30/IEEE80211_INACT_WAIT) /* initial */
#define IEEE80211_INACT_AUTH (180/IEEE80211_INACT_WAIT) /* associated but not authorized */
#define IEEE80211_INACT_RUN (300/IEEE80211_INACT_WAIT) /* authorized */
#define IEEE80211_INACT_PROBE (30/IEEE80211_INACT_WAIT) /* probe */
#define IEEE80211_INACT_SCAN (300/IEEE80211_INACT_WAIT) /* scanned */
#define IEEE80211_TRANS_WAIT 5 /* mgt frame tx timer (secs) */
#define IEEE80211_NODE_HASHSIZE 32
/* simple hash is enough for variation of macaddr */
#define IEEE80211_NODE_HASH(addr) \
(((const u_int8_t *)(addr))[IEEE80211_ADDR_LEN - 1] % \
IEEE80211_NODE_HASHSIZE)
struct ieee80211_rsnparms {
u_int8_t rsn_mcastcipher; /* mcast/group cipher */
u_int8_t rsn_mcastkeylen; /* mcast key length */
u_int8_t rsn_ucastcipherset; /* unicast cipher set */
u_int8_t rsn_ucastcipher; /* selected unicast cipher */
u_int8_t rsn_ucastkeylen; /* unicast key length */
u_int8_t rsn_keymgmtset; /* key mangement algorithms */
u_int8_t rsn_keymgmt; /* selected key mgmt algo */
u_int16_t rsn_caps; /* capabilities */
};
struct ieee80211_node_table;
struct ieee80211com;
/*
* Node specific information. Note that drivers are expected
* to derive from this structure to add device-specific per-node
* state. This is done by overriding the ic_node_* methods in
* the ieee80211com structure.
*/
struct ieee80211_node {
struct ieee80211com *ni_ic;
struct ieee80211_node_table *ni_table;
TAILQ_ENTRY(ieee80211_node) ni_list;
LIST_ENTRY(ieee80211_node) ni_hash;
u_int ni_refcnt;
u_int ni_scangen; /* gen# for timeout scan */
u_int8_t ni_authmode; /* authentication algorithm */
u_int16_t ni_flags; /* special-purpose state */
#define IEEE80211_NODE_AUTH 0x0001 /* authorized for data */
#define IEEE80211_NODE_QOS 0x0002 /* QoS enabled */
#define IEEE80211_NODE_ERP 0x0004 /* ERP enabled */
/* NB: this must have the same value as IEEE80211_FC1_PWR_MGT */
#define IEEE80211_NODE_PWR_MGT 0x0010 /* power save mode enabled */
2005-07-27 02:52:48 +04:00
#define IEEE80211_NODE_AREF 0x0020 /* authentication ref held */
u_int16_t ni_associd; /* assoc response */
u_int16_t ni_txpower; /* current transmit power */
u_int16_t ni_vlan; /* vlan tag */
u_int32_t *ni_challenge; /* shared-key challenge */
u_int8_t *ni_wpa_ie; /* captured WPA/RSN ie */
u_int8_t *ni_wme_ie; /* captured WME ie */
u_int16_t ni_txseqs[17]; /* tx seq per-tid */
u_int16_t ni_rxseqs[17]; /* rx seq previous per-tid*/
u_int32_t ni_rxfragstamp; /* time stamp of last rx frag */
struct mbuf *ni_rxfrag[3]; /* rx frag reassembly */
struct ieee80211_rsnparms ni_rsn; /* RSN/WPA parameters */
struct ieee80211_key ni_ucastkey; /* unicast key */
/* hardware */
u_int32_t ni_rstamp; /* recv timestamp */
u_int8_t ni_rssi; /* recv ssi */
/* header */
u_int8_t ni_macaddr[IEEE80211_ADDR_LEN];
u_int8_t ni_bssid[IEEE80211_ADDR_LEN];
/* beacon, probe response */
union {
u_int8_t data[8];
u_int64_t tsf;
} ni_tstamp; /* from last rcv'd beacon */
u_int16_t ni_intval; /* beacon interval */
u_int16_t ni_capinfo; /* capabilities */
u_int8_t ni_esslen;
u_int8_t ni_essid[IEEE80211_NWID_LEN];
struct ieee80211_rateset ni_rates; /* negotiated rate set */
Resolve conflicts and adapt to NetBSD. Thanks to dyoung@, scw@, and perry@ for help testing. 2005-08-30 15:27 avatar Properly set ic_curchan before calling back to device driver to do channel switching(ifconfig devX channel Y). This fix should make channel changing works again in monitor mode. Submitted by: sam X-MFC-With: other ic_curchan changes 2005-08-13 18:50 sam revert 1.64: we cannot use the channel characteristics to decide when to do 11g erp sta accounting because b/g channels show up as false positives when operating in 11b. Noticed by: Michal Mertl 2005-08-13 18:31 sam Extend acl support to pass ioctl requests through and use this to add support for getting the current policy setting and collecting the list of mac addresses in the acl table. Submitted by: Michal Mertl (original version) MFC after: 2 weeks 2005-08-10 18:42 sam Don't use ic_curmode to decide when to do 11g station accounting, use the station channel properties. Fixes assert failure/bogus operation when an ap is operating in 11a and has associated stations then switches to 11g. Noticed by: Michal Mertl Reviewed by: avatar MFC after: 2 weeks 2005-08-10 17:22 sam Clarify/fix handling of the current channel: o add ic_curchan and use it uniformly for specifying the current channel instead of overloading ic->ic_bss->ni_chan (or in some drivers ic_ibss_chan) o add ieee80211_scanparams structure to encapsulate scanning-related state captured for rx frames o move rx beacon+probe response frame handling into separate routines o change beacon+probe response handling to treat the scan table more like a scan cache--look for an existing entry before adding a new one; this combined with ic_curchan use corrects handling of stations that were previously found at a different channel o move adhoc neighbor discovery by beacon+probe response frames to a new ieee80211_add_neighbor routine Reviewed by: avatar Tested by: avatar, Michal Mertl MFC after: 2 weeks 2005-08-09 11:19 rwatson Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days 2005-08-08 19:46 sam Split crypto tx+rx key indices and add a key index -> node mapping table: Crypto changes: o change driver/net80211 key_alloc api to return tx+rx key indices; a driver can leave the rx key index set to IEEE80211_KEYIX_NONE or set it to be the same as the tx key index (the former disables use of the key index in building the keyix->node mapping table and is the default setup for naive drivers by null_key_alloc) o add cs_max_keyid to crypto state to specify the max h/w key index a driver will return; this is used to allocate the key index mapping table and to bounds check table loookups o while here introduce ieee80211_keyix (finally) for the type of a h/w key index o change crypto notifiers for rx failures to pass the rx key index up as appropriate (michael failure, replay, etc.) Node table changes: o optionally allocate a h/w key index to node mapping table for the station table using the max key index setting supplied by drivers (note the scan table does not get a map) o defer node table allocation to lateattach so the driver has a chance to set the max key id to size the key index map o while here also defer the aid bitmap allocation o add new ieee80211_find_rxnode_withkey api to find a sta/node entry on frame receive with an optional h/w key index to use in checking mapping table; also updates the map if it does a hash lookup and the found node has a rx key index set in the unicast key; note this work is separated from the old ieee80211_find_rxnode call so drivers do not need to be aware of the new mechanism o move some node table manipulation under the node table lock to close a race on node delete o add ieee80211_node_delucastkey to do the dirty work of deleting unicast key state for a node (deletes any key and handles key map references) Ath driver: o nuke private sc_keyixmap mechansim in favor of net80211 support o update key alloc api These changes close several race conditions for the ath driver operating in ap mode. Other drivers should see no change. Station mode operation for ath no longer uses the key index map but performance tests show no noticeable change and this will be fixed when the scan table is eliminated with the new scanning support. Tested by: Michal Mertl, avatar, others Reviewed by: avatar, others MFC after: 2 weeks 2005-08-08 06:49 sam use ieee80211_iterate_nodes to retrieve station data; the previous code walked the list w/o locking MFC after: 1 week 2005-08-08 04:30 sam Cleanup beacon/listen interval handling: o separate configured beacon interval from listen interval; this avoids potential use of one value for the other (e.g. setting powersavesleep to 0 clobbers the beacon interval used in hostap or ibss mode) o bounds check the beacon interval received in probe response and beacon frames and drop frames with bogus settings; not clear if we should instead clamp the value as any alteration would result in mismatched sta+ap configuration and probably be more confusing (don't want to log to the console but perhaps ok with rate limiting) o while here up max beacon interval to reflect WiFi standard Noticed by: Martin <nakal@nurfuerspam.de> MFC after: 1 week 2005-08-06 05:57 sam fix debug msg typo MFC after: 3 days 2005-08-06 05:56 sam Fix handling of frames sent prior to a station being authorized when operating in ap mode. Previously we allocated a node from the station table, sent the frame (using the node), then released the reference that "held the frame in the table". But while the frame was in flight the node might be reclaimed which could lead to problems. The solution is to add an ieee80211_tmp_node routine that crafts a node that does exist in a table and so isn't ever reclaimed; it exists only so long as the associated frame is in flight. MFC after: 5 days 2005-07-31 07:12 sam close a race between reclaiming a node when a station is inactive and sending the null data frame used to probe inactive stations MFC after: 5 days 2005-07-27 05:41 sam when bridging internally bypass the bss node as traffic to it must follow the normal input path Submitted by: Michal Mertl MFC after: 5 days 2005-07-27 03:53 sam bandaid ni_fails handling so ap's with association failures are reconsidered after a bit; a proper fix involves more changes to the scanning infrastructure Reviewed by: avatar, David Young MFC after: 5 days 2005-07-23 01:16 sam the AREF flag is only meaningful in ap mode; adhoc neighbors now are timed out of the sta/neighbor table 2005-07-23 00:25 sam o move inactivity-related debug msgs under IEEE80211_MSG_INACT o probe inactive neighbors in adhoc mode (they don't have an association id so previously were being timed out) MFC after: 3 days 2005-07-22 22:11 sam split xmit of probe request frame out into a separate routine that takes explicit parameters; this will be needed when scanning is decoupled from the state machine to do bg scanning MFC after: 3 days 2005-07-22 21:48 sam split 802.11 frame xmit setup code into ieee80211_send_setup MFC after: 3 days 2005-07-22 18:57 sam simplify ic_newassoc callback MFC after: 3 days 2005-07-22 18:54 sam simplify ieee80211_ibss_merge api MFC after: 3 days 2005-07-22 18:50 sam add stats we know we'll need soon and some spare fields for future expansion MFC after: 3 days 2005-07-22 18:45 sam simplify tim callback api MFC after: 3 days 2005-07-22 18:42 sam don't include 802.3 header in min frame length calculation as it may not be present for a frag; fixes problem with small (fragmented) frames being dropped Obtained from: Atheros MFC after: 3 days 2005-07-22 18:36 sam simplify ieee80211_node_authorize and ieee80211_node_unauthorize api's MFC after: 3 days 2005-07-22 18:31 sam simplifiy ieee80211_send_nulldata api MFC after: 3 days 2005-07-22 18:29 sam simplify rate set api's by removing ic parameter (implicit in node reference) MFC after: 3 days 2005-07-22 18:21 sam reject association requests with a wpa/rsn ie when wpa/rsn is not configured on the ap; previously we either ignored the ie or (possibly) failed an assertion Obtained from: Atheros MFC after: 3 days 2005-07-22 18:16 sam missed one in last commit; add device name to discard msgs 2005-07-22 18:13 sam include device name in discard msgs 2005-07-22 18:12 sam add diag msgs for frames discarded because the direction field is wrong 2005-07-22 18:08 sam split data frame delivery out to a new function ieee80211_deliver_data 2005-07-22 18:00 sam o add IEEE80211_IOC_FRAGTHRESHOLD for getting+setting the tx fragmentation threshold o fix bounds checking on IEEE80211_IOC_RTSTHRESHOLD MFC after: 3 days 2005-07-22 17:55 sam o add IEEE80211_FRAG_DEFAULT o move default settings for RTS and frag thresholds to ieee80211_var.h 2005-07-22 17:50 sam diff reduction against p4: define IEEE80211_FIXED_RATE_NONE and use it instead of -1 2005-07-22 17:37 sam add flags missed in last merge 2005-07-22 17:36 sam Diff reduction against p4: o add ic_flags_ext for eventual extention of ic_flags o define/reserve flag+capabilities bits for superg, bg scan, and roaming support o refactor debug msg macros MFC after: 3 days 2005-07-22 06:17 sam send a response when an auth request is denied due to an acl; might be better to silently ignore the frame but this way we give stations a chance of figuring out what's wrong 2005-07-22 06:15 sam remove excess whitespace 2005-07-22 05:55 sam use IF_HANDOFF when bridging frames internally so if_start gets called; fixes communication between associated sta's MFC after: 3 days 2005-07-11 04:06 sam Handle encrypt of arbitarily fragmented mbuf chains: previously we bailed if we couldn't collect the 16-bytes of data required for an aes block cipher in 2 mbufs; now we deal with it. While here make space accounting signed so a sanity check does the right thing for malformed mbuf chains. Approved by: re (scottl) 2005-07-11 04:00 sam nuke assert that duplicates real check Reviewed by: avatar Approved by: re (scottl)
2005-11-18 19:40:08 +03:00
struct ieee80211_channel *ni_chan; /* XXX multiple uses */
u_int16_t ni_fhdwell; /* FH only */
u_int8_t ni_fhindex; /* FH only */
u_int8_t ni_erp; /* ERP from beacon/probe resp */
u_int16_t ni_timoff; /* byte offset to TIM ie */
2005-07-27 02:52:48 +04:00
u_int8_t ni_dtim_period; /* DTIM period */
u_int8_t ni_dtim_count; /* DTIM count for last bcn */
/* others */
int ni_fails; /* failure count to associate */
short ni_inact; /* inactivity mark count */
short ni_inact_reload;/* inactivity reload value */
int ni_txrate; /* index to ni_rates[] */
struct ifqueue ni_savedq; /* ps-poll queue */
struct ieee80211_nodestats ni_stats; /* per-node statistics */
};
MALLOC_DECLARE(M_80211_NODE);
#define IEEE80211_NODE_AID(ni) IEEE80211_AID(ni->ni_associd)
#define IEEE80211_NODE_STAT(ni,stat) (ni->ni_stats.ns_##stat++)
#define IEEE80211_NODE_STAT_ADD(ni,stat,v) (ni->ni_stats.ns_##stat += v)
#define IEEE80211_NODE_STAT_SET(ni,stat,v) (ni->ni_stats.ns_##stat = v)
struct ieee80211com;
void ieee80211_node_attach(struct ieee80211com *);
void ieee80211_node_lateattach(struct ieee80211com *);
void ieee80211_node_detach(struct ieee80211com *);
2004-05-01 02:57:32 +04:00
static __inline int
ieee80211_node_is_authorized(const struct ieee80211_node *ni)
{
return (ni->ni_flags & IEEE80211_NODE_AUTH);
}
Resolve conflicts and adapt to NetBSD. Thanks to dyoung@, scw@, and perry@ for help testing. 2005-08-30 15:27 avatar Properly set ic_curchan before calling back to device driver to do channel switching(ifconfig devX channel Y). This fix should make channel changing works again in monitor mode. Submitted by: sam X-MFC-With: other ic_curchan changes 2005-08-13 18:50 sam revert 1.64: we cannot use the channel characteristics to decide when to do 11g erp sta accounting because b/g channels show up as false positives when operating in 11b. Noticed by: Michal Mertl 2005-08-13 18:31 sam Extend acl support to pass ioctl requests through and use this to add support for getting the current policy setting and collecting the list of mac addresses in the acl table. Submitted by: Michal Mertl (original version) MFC after: 2 weeks 2005-08-10 18:42 sam Don't use ic_curmode to decide when to do 11g station accounting, use the station channel properties. Fixes assert failure/bogus operation when an ap is operating in 11a and has associated stations then switches to 11g. Noticed by: Michal Mertl Reviewed by: avatar MFC after: 2 weeks 2005-08-10 17:22 sam Clarify/fix handling of the current channel: o add ic_curchan and use it uniformly for specifying the current channel instead of overloading ic->ic_bss->ni_chan (or in some drivers ic_ibss_chan) o add ieee80211_scanparams structure to encapsulate scanning-related state captured for rx frames o move rx beacon+probe response frame handling into separate routines o change beacon+probe response handling to treat the scan table more like a scan cache--look for an existing entry before adding a new one; this combined with ic_curchan use corrects handling of stations that were previously found at a different channel o move adhoc neighbor discovery by beacon+probe response frames to a new ieee80211_add_neighbor routine Reviewed by: avatar Tested by: avatar, Michal Mertl MFC after: 2 weeks 2005-08-09 11:19 rwatson Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days 2005-08-08 19:46 sam Split crypto tx+rx key indices and add a key index -> node mapping table: Crypto changes: o change driver/net80211 key_alloc api to return tx+rx key indices; a driver can leave the rx key index set to IEEE80211_KEYIX_NONE or set it to be the same as the tx key index (the former disables use of the key index in building the keyix->node mapping table and is the default setup for naive drivers by null_key_alloc) o add cs_max_keyid to crypto state to specify the max h/w key index a driver will return; this is used to allocate the key index mapping table and to bounds check table loookups o while here introduce ieee80211_keyix (finally) for the type of a h/w key index o change crypto notifiers for rx failures to pass the rx key index up as appropriate (michael failure, replay, etc.) Node table changes: o optionally allocate a h/w key index to node mapping table for the station table using the max key index setting supplied by drivers (note the scan table does not get a map) o defer node table allocation to lateattach so the driver has a chance to set the max key id to size the key index map o while here also defer the aid bitmap allocation o add new ieee80211_find_rxnode_withkey api to find a sta/node entry on frame receive with an optional h/w key index to use in checking mapping table; also updates the map if it does a hash lookup and the found node has a rx key index set in the unicast key; note this work is separated from the old ieee80211_find_rxnode call so drivers do not need to be aware of the new mechanism o move some node table manipulation under the node table lock to close a race on node delete o add ieee80211_node_delucastkey to do the dirty work of deleting unicast key state for a node (deletes any key and handles key map references) Ath driver: o nuke private sc_keyixmap mechansim in favor of net80211 support o update key alloc api These changes close several race conditions for the ath driver operating in ap mode. Other drivers should see no change. Station mode operation for ath no longer uses the key index map but performance tests show no noticeable change and this will be fixed when the scan table is eliminated with the new scanning support. Tested by: Michal Mertl, avatar, others Reviewed by: avatar, others MFC after: 2 weeks 2005-08-08 06:49 sam use ieee80211_iterate_nodes to retrieve station data; the previous code walked the list w/o locking MFC after: 1 week 2005-08-08 04:30 sam Cleanup beacon/listen interval handling: o separate configured beacon interval from listen interval; this avoids potential use of one value for the other (e.g. setting powersavesleep to 0 clobbers the beacon interval used in hostap or ibss mode) o bounds check the beacon interval received in probe response and beacon frames and drop frames with bogus settings; not clear if we should instead clamp the value as any alteration would result in mismatched sta+ap configuration and probably be more confusing (don't want to log to the console but perhaps ok with rate limiting) o while here up max beacon interval to reflect WiFi standard Noticed by: Martin <nakal@nurfuerspam.de> MFC after: 1 week 2005-08-06 05:57 sam fix debug msg typo MFC after: 3 days 2005-08-06 05:56 sam Fix handling of frames sent prior to a station being authorized when operating in ap mode. Previously we allocated a node from the station table, sent the frame (using the node), then released the reference that "held the frame in the table". But while the frame was in flight the node might be reclaimed which could lead to problems. The solution is to add an ieee80211_tmp_node routine that crafts a node that does exist in a table and so isn't ever reclaimed; it exists only so long as the associated frame is in flight. MFC after: 5 days 2005-07-31 07:12 sam close a race between reclaiming a node when a station is inactive and sending the null data frame used to probe inactive stations MFC after: 5 days 2005-07-27 05:41 sam when bridging internally bypass the bss node as traffic to it must follow the normal input path Submitted by: Michal Mertl MFC after: 5 days 2005-07-27 03:53 sam bandaid ni_fails handling so ap's with association failures are reconsidered after a bit; a proper fix involves more changes to the scanning infrastructure Reviewed by: avatar, David Young MFC after: 5 days 2005-07-23 01:16 sam the AREF flag is only meaningful in ap mode; adhoc neighbors now are timed out of the sta/neighbor table 2005-07-23 00:25 sam o move inactivity-related debug msgs under IEEE80211_MSG_INACT o probe inactive neighbors in adhoc mode (they don't have an association id so previously were being timed out) MFC after: 3 days 2005-07-22 22:11 sam split xmit of probe request frame out into a separate routine that takes explicit parameters; this will be needed when scanning is decoupled from the state machine to do bg scanning MFC after: 3 days 2005-07-22 21:48 sam split 802.11 frame xmit setup code into ieee80211_send_setup MFC after: 3 days 2005-07-22 18:57 sam simplify ic_newassoc callback MFC after: 3 days 2005-07-22 18:54 sam simplify ieee80211_ibss_merge api MFC after: 3 days 2005-07-22 18:50 sam add stats we know we'll need soon and some spare fields for future expansion MFC after: 3 days 2005-07-22 18:45 sam simplify tim callback api MFC after: 3 days 2005-07-22 18:42 sam don't include 802.3 header in min frame length calculation as it may not be present for a frag; fixes problem with small (fragmented) frames being dropped Obtained from: Atheros MFC after: 3 days 2005-07-22 18:36 sam simplify ieee80211_node_authorize and ieee80211_node_unauthorize api's MFC after: 3 days 2005-07-22 18:31 sam simplifiy ieee80211_send_nulldata api MFC after: 3 days 2005-07-22 18:29 sam simplify rate set api's by removing ic parameter (implicit in node reference) MFC after: 3 days 2005-07-22 18:21 sam reject association requests with a wpa/rsn ie when wpa/rsn is not configured on the ap; previously we either ignored the ie or (possibly) failed an assertion Obtained from: Atheros MFC after: 3 days 2005-07-22 18:16 sam missed one in last commit; add device name to discard msgs 2005-07-22 18:13 sam include device name in discard msgs 2005-07-22 18:12 sam add diag msgs for frames discarded because the direction field is wrong 2005-07-22 18:08 sam split data frame delivery out to a new function ieee80211_deliver_data 2005-07-22 18:00 sam o add IEEE80211_IOC_FRAGTHRESHOLD for getting+setting the tx fragmentation threshold o fix bounds checking on IEEE80211_IOC_RTSTHRESHOLD MFC after: 3 days 2005-07-22 17:55 sam o add IEEE80211_FRAG_DEFAULT o move default settings for RTS and frag thresholds to ieee80211_var.h 2005-07-22 17:50 sam diff reduction against p4: define IEEE80211_FIXED_RATE_NONE and use it instead of -1 2005-07-22 17:37 sam add flags missed in last merge 2005-07-22 17:36 sam Diff reduction against p4: o add ic_flags_ext for eventual extention of ic_flags o define/reserve flag+capabilities bits for superg, bg scan, and roaming support o refactor debug msg macros MFC after: 3 days 2005-07-22 06:17 sam send a response when an auth request is denied due to an acl; might be better to silently ignore the frame but this way we give stations a chance of figuring out what's wrong 2005-07-22 06:15 sam remove excess whitespace 2005-07-22 05:55 sam use IF_HANDOFF when bridging frames internally so if_start gets called; fixes communication between associated sta's MFC after: 3 days 2005-07-11 04:06 sam Handle encrypt of arbitarily fragmented mbuf chains: previously we bailed if we couldn't collect the 16-bytes of data required for an aes block cipher in 2 mbufs; now we deal with it. While here make space accounting signed so a sanity check does the right thing for malformed mbuf chains. Approved by: re (scottl) 2005-07-11 04:00 sam nuke assert that duplicates real check Reviewed by: avatar Approved by: re (scottl)
2005-11-18 19:40:08 +03:00
void ieee80211_node_authorize(struct ieee80211_node *);
void ieee80211_node_unauthorize(struct ieee80211_node *);
void ieee80211_begin_scan(struct ieee80211com *, int);
int ieee80211_next_scan(struct ieee80211com *);
void ieee80211_probe_curchan(struct ieee80211com *, int);
void ieee80211_create_ibss(struct ieee80211com*, struct ieee80211_channel *);
void ieee80211_reset_bss(struct ieee80211com *);
void ieee80211_cancel_scan(struct ieee80211com *);
void ieee80211_end_scan(struct ieee80211com *);
Resolve conflicts and adapt to NetBSD. Thanks to dyoung@, scw@, and perry@ for help testing. 2005-08-30 15:27 avatar Properly set ic_curchan before calling back to device driver to do channel switching(ifconfig devX channel Y). This fix should make channel changing works again in monitor mode. Submitted by: sam X-MFC-With: other ic_curchan changes 2005-08-13 18:50 sam revert 1.64: we cannot use the channel characteristics to decide when to do 11g erp sta accounting because b/g channels show up as false positives when operating in 11b. Noticed by: Michal Mertl 2005-08-13 18:31 sam Extend acl support to pass ioctl requests through and use this to add support for getting the current policy setting and collecting the list of mac addresses in the acl table. Submitted by: Michal Mertl (original version) MFC after: 2 weeks 2005-08-10 18:42 sam Don't use ic_curmode to decide when to do 11g station accounting, use the station channel properties. Fixes assert failure/bogus operation when an ap is operating in 11a and has associated stations then switches to 11g. Noticed by: Michal Mertl Reviewed by: avatar MFC after: 2 weeks 2005-08-10 17:22 sam Clarify/fix handling of the current channel: o add ic_curchan and use it uniformly for specifying the current channel instead of overloading ic->ic_bss->ni_chan (or in some drivers ic_ibss_chan) o add ieee80211_scanparams structure to encapsulate scanning-related state captured for rx frames o move rx beacon+probe response frame handling into separate routines o change beacon+probe response handling to treat the scan table more like a scan cache--look for an existing entry before adding a new one; this combined with ic_curchan use corrects handling of stations that were previously found at a different channel o move adhoc neighbor discovery by beacon+probe response frames to a new ieee80211_add_neighbor routine Reviewed by: avatar Tested by: avatar, Michal Mertl MFC after: 2 weeks 2005-08-09 11:19 rwatson Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days 2005-08-08 19:46 sam Split crypto tx+rx key indices and add a key index -> node mapping table: Crypto changes: o change driver/net80211 key_alloc api to return tx+rx key indices; a driver can leave the rx key index set to IEEE80211_KEYIX_NONE or set it to be the same as the tx key index (the former disables use of the key index in building the keyix->node mapping table and is the default setup for naive drivers by null_key_alloc) o add cs_max_keyid to crypto state to specify the max h/w key index a driver will return; this is used to allocate the key index mapping table and to bounds check table loookups o while here introduce ieee80211_keyix (finally) for the type of a h/w key index o change crypto notifiers for rx failures to pass the rx key index up as appropriate (michael failure, replay, etc.) Node table changes: o optionally allocate a h/w key index to node mapping table for the station table using the max key index setting supplied by drivers (note the scan table does not get a map) o defer node table allocation to lateattach so the driver has a chance to set the max key id to size the key index map o while here also defer the aid bitmap allocation o add new ieee80211_find_rxnode_withkey api to find a sta/node entry on frame receive with an optional h/w key index to use in checking mapping table; also updates the map if it does a hash lookup and the found node has a rx key index set in the unicast key; note this work is separated from the old ieee80211_find_rxnode call so drivers do not need to be aware of the new mechanism o move some node table manipulation under the node table lock to close a race on node delete o add ieee80211_node_delucastkey to do the dirty work of deleting unicast key state for a node (deletes any key and handles key map references) Ath driver: o nuke private sc_keyixmap mechansim in favor of net80211 support o update key alloc api These changes close several race conditions for the ath driver operating in ap mode. Other drivers should see no change. Station mode operation for ath no longer uses the key index map but performance tests show no noticeable change and this will be fixed when the scan table is eliminated with the new scanning support. Tested by: Michal Mertl, avatar, others Reviewed by: avatar, others MFC after: 2 weeks 2005-08-08 06:49 sam use ieee80211_iterate_nodes to retrieve station data; the previous code walked the list w/o locking MFC after: 1 week 2005-08-08 04:30 sam Cleanup beacon/listen interval handling: o separate configured beacon interval from listen interval; this avoids potential use of one value for the other (e.g. setting powersavesleep to 0 clobbers the beacon interval used in hostap or ibss mode) o bounds check the beacon interval received in probe response and beacon frames and drop frames with bogus settings; not clear if we should instead clamp the value as any alteration would result in mismatched sta+ap configuration and probably be more confusing (don't want to log to the console but perhaps ok with rate limiting) o while here up max beacon interval to reflect WiFi standard Noticed by: Martin <nakal@nurfuerspam.de> MFC after: 1 week 2005-08-06 05:57 sam fix debug msg typo MFC after: 3 days 2005-08-06 05:56 sam Fix handling of frames sent prior to a station being authorized when operating in ap mode. Previously we allocated a node from the station table, sent the frame (using the node), then released the reference that "held the frame in the table". But while the frame was in flight the node might be reclaimed which could lead to problems. The solution is to add an ieee80211_tmp_node routine that crafts a node that does exist in a table and so isn't ever reclaimed; it exists only so long as the associated frame is in flight. MFC after: 5 days 2005-07-31 07:12 sam close a race between reclaiming a node when a station is inactive and sending the null data frame used to probe inactive stations MFC after: 5 days 2005-07-27 05:41 sam when bridging internally bypass the bss node as traffic to it must follow the normal input path Submitted by: Michal Mertl MFC after: 5 days 2005-07-27 03:53 sam bandaid ni_fails handling so ap's with association failures are reconsidered after a bit; a proper fix involves more changes to the scanning infrastructure Reviewed by: avatar, David Young MFC after: 5 days 2005-07-23 01:16 sam the AREF flag is only meaningful in ap mode; adhoc neighbors now are timed out of the sta/neighbor table 2005-07-23 00:25 sam o move inactivity-related debug msgs under IEEE80211_MSG_INACT o probe inactive neighbors in adhoc mode (they don't have an association id so previously were being timed out) MFC after: 3 days 2005-07-22 22:11 sam split xmit of probe request frame out into a separate routine that takes explicit parameters; this will be needed when scanning is decoupled from the state machine to do bg scanning MFC after: 3 days 2005-07-22 21:48 sam split 802.11 frame xmit setup code into ieee80211_send_setup MFC after: 3 days 2005-07-22 18:57 sam simplify ic_newassoc callback MFC after: 3 days 2005-07-22 18:54 sam simplify ieee80211_ibss_merge api MFC after: 3 days 2005-07-22 18:50 sam add stats we know we'll need soon and some spare fields for future expansion MFC after: 3 days 2005-07-22 18:45 sam simplify tim callback api MFC after: 3 days 2005-07-22 18:42 sam don't include 802.3 header in min frame length calculation as it may not be present for a frag; fixes problem with small (fragmented) frames being dropped Obtained from: Atheros MFC after: 3 days 2005-07-22 18:36 sam simplify ieee80211_node_authorize and ieee80211_node_unauthorize api's MFC after: 3 days 2005-07-22 18:31 sam simplifiy ieee80211_send_nulldata api MFC after: 3 days 2005-07-22 18:29 sam simplify rate set api's by removing ic parameter (implicit in node reference) MFC after: 3 days 2005-07-22 18:21 sam reject association requests with a wpa/rsn ie when wpa/rsn is not configured on the ap; previously we either ignored the ie or (possibly) failed an assertion Obtained from: Atheros MFC after: 3 days 2005-07-22 18:16 sam missed one in last commit; add device name to discard msgs 2005-07-22 18:13 sam include device name in discard msgs 2005-07-22 18:12 sam add diag msgs for frames discarded because the direction field is wrong 2005-07-22 18:08 sam split data frame delivery out to a new function ieee80211_deliver_data 2005-07-22 18:00 sam o add IEEE80211_IOC_FRAGTHRESHOLD for getting+setting the tx fragmentation threshold o fix bounds checking on IEEE80211_IOC_RTSTHRESHOLD MFC after: 3 days 2005-07-22 17:55 sam o add IEEE80211_FRAG_DEFAULT o move default settings for RTS and frag thresholds to ieee80211_var.h 2005-07-22 17:50 sam diff reduction against p4: define IEEE80211_FIXED_RATE_NONE and use it instead of -1 2005-07-22 17:37 sam add flags missed in last merge 2005-07-22 17:36 sam Diff reduction against p4: o add ic_flags_ext for eventual extention of ic_flags o define/reserve flag+capabilities bits for superg, bg scan, and roaming support o refactor debug msg macros MFC after: 3 days 2005-07-22 06:17 sam send a response when an auth request is denied due to an acl; might be better to silently ignore the frame but this way we give stations a chance of figuring out what's wrong 2005-07-22 06:15 sam remove excess whitespace 2005-07-22 05:55 sam use IF_HANDOFF when bridging frames internally so if_start gets called; fixes communication between associated sta's MFC after: 3 days 2005-07-11 04:06 sam Handle encrypt of arbitarily fragmented mbuf chains: previously we bailed if we couldn't collect the 16-bytes of data required for an aes block cipher in 2 mbufs; now we deal with it. While here make space accounting signed so a sanity check does the right thing for malformed mbuf chains. Approved by: re (scottl) 2005-07-11 04:00 sam nuke assert that duplicates real check Reviewed by: avatar Approved by: re (scottl)
2005-11-18 19:40:08 +03:00
int ieee80211_ibss_merge(struct ieee80211_node *);
int ieee80211_sta_join(struct ieee80211com *, struct ieee80211_node *);
void ieee80211_sta_leave(struct ieee80211com *, struct ieee80211_node *);
/*
* Table of ieee80211_node instances. Each ieee80211com
* has at least one for holding the scan candidates.
* When operating as an access point or in ibss mode there
* is a second table for associated stations or neighbors.
*/
struct ieee80211_node_table {
struct ieee80211com *nt_ic; /* back reference */
ieee80211_node_lock_t nt_nodelock; /* on node table */
TAILQ_HEAD(, ieee80211_node) nt_node; /* information of all nodes */
LIST_HEAD(, ieee80211_node) nt_hash[IEEE80211_NODE_HASHSIZE];
const char *nt_name; /* for debugging */
ieee80211_scan_lock_t nt_scanlock; /* on nt_scangen */
u_int nt_scangen; /* gen# for timeout scan */
int nt_inact_timer; /* inactivity timer */
int nt_inact_init; /* initial node inact setting */
Resolve conflicts and adapt to NetBSD. Thanks to dyoung@, scw@, and perry@ for help testing. 2005-08-30 15:27 avatar Properly set ic_curchan before calling back to device driver to do channel switching(ifconfig devX channel Y). This fix should make channel changing works again in monitor mode. Submitted by: sam X-MFC-With: other ic_curchan changes 2005-08-13 18:50 sam revert 1.64: we cannot use the channel characteristics to decide when to do 11g erp sta accounting because b/g channels show up as false positives when operating in 11b. Noticed by: Michal Mertl 2005-08-13 18:31 sam Extend acl support to pass ioctl requests through and use this to add support for getting the current policy setting and collecting the list of mac addresses in the acl table. Submitted by: Michal Mertl (original version) MFC after: 2 weeks 2005-08-10 18:42 sam Don't use ic_curmode to decide when to do 11g station accounting, use the station channel properties. Fixes assert failure/bogus operation when an ap is operating in 11a and has associated stations then switches to 11g. Noticed by: Michal Mertl Reviewed by: avatar MFC after: 2 weeks 2005-08-10 17:22 sam Clarify/fix handling of the current channel: o add ic_curchan and use it uniformly for specifying the current channel instead of overloading ic->ic_bss->ni_chan (or in some drivers ic_ibss_chan) o add ieee80211_scanparams structure to encapsulate scanning-related state captured for rx frames o move rx beacon+probe response frame handling into separate routines o change beacon+probe response handling to treat the scan table more like a scan cache--look for an existing entry before adding a new one; this combined with ic_curchan use corrects handling of stations that were previously found at a different channel o move adhoc neighbor discovery by beacon+probe response frames to a new ieee80211_add_neighbor routine Reviewed by: avatar Tested by: avatar, Michal Mertl MFC after: 2 weeks 2005-08-09 11:19 rwatson Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days 2005-08-08 19:46 sam Split crypto tx+rx key indices and add a key index -> node mapping table: Crypto changes: o change driver/net80211 key_alloc api to return tx+rx key indices; a driver can leave the rx key index set to IEEE80211_KEYIX_NONE or set it to be the same as the tx key index (the former disables use of the key index in building the keyix->node mapping table and is the default setup for naive drivers by null_key_alloc) o add cs_max_keyid to crypto state to specify the max h/w key index a driver will return; this is used to allocate the key index mapping table and to bounds check table loookups o while here introduce ieee80211_keyix (finally) for the type of a h/w key index o change crypto notifiers for rx failures to pass the rx key index up as appropriate (michael failure, replay, etc.) Node table changes: o optionally allocate a h/w key index to node mapping table for the station table using the max key index setting supplied by drivers (note the scan table does not get a map) o defer node table allocation to lateattach so the driver has a chance to set the max key id to size the key index map o while here also defer the aid bitmap allocation o add new ieee80211_find_rxnode_withkey api to find a sta/node entry on frame receive with an optional h/w key index to use in checking mapping table; also updates the map if it does a hash lookup and the found node has a rx key index set in the unicast key; note this work is separated from the old ieee80211_find_rxnode call so drivers do not need to be aware of the new mechanism o move some node table manipulation under the node table lock to close a race on node delete o add ieee80211_node_delucastkey to do the dirty work of deleting unicast key state for a node (deletes any key and handles key map references) Ath driver: o nuke private sc_keyixmap mechansim in favor of net80211 support o update key alloc api These changes close several race conditions for the ath driver operating in ap mode. Other drivers should see no change. Station mode operation for ath no longer uses the key index map but performance tests show no noticeable change and this will be fixed when the scan table is eliminated with the new scanning support. Tested by: Michal Mertl, avatar, others Reviewed by: avatar, others MFC after: 2 weeks 2005-08-08 06:49 sam use ieee80211_iterate_nodes to retrieve station data; the previous code walked the list w/o locking MFC after: 1 week 2005-08-08 04:30 sam Cleanup beacon/listen interval handling: o separate configured beacon interval from listen interval; this avoids potential use of one value for the other (e.g. setting powersavesleep to 0 clobbers the beacon interval used in hostap or ibss mode) o bounds check the beacon interval received in probe response and beacon frames and drop frames with bogus settings; not clear if we should instead clamp the value as any alteration would result in mismatched sta+ap configuration and probably be more confusing (don't want to log to the console but perhaps ok with rate limiting) o while here up max beacon interval to reflect WiFi standard Noticed by: Martin <nakal@nurfuerspam.de> MFC after: 1 week 2005-08-06 05:57 sam fix debug msg typo MFC after: 3 days 2005-08-06 05:56 sam Fix handling of frames sent prior to a station being authorized when operating in ap mode. Previously we allocated a node from the station table, sent the frame (using the node), then released the reference that "held the frame in the table". But while the frame was in flight the node might be reclaimed which could lead to problems. The solution is to add an ieee80211_tmp_node routine that crafts a node that does exist in a table and so isn't ever reclaimed; it exists only so long as the associated frame is in flight. MFC after: 5 days 2005-07-31 07:12 sam close a race between reclaiming a node when a station is inactive and sending the null data frame used to probe inactive stations MFC after: 5 days 2005-07-27 05:41 sam when bridging internally bypass the bss node as traffic to it must follow the normal input path Submitted by: Michal Mertl MFC after: 5 days 2005-07-27 03:53 sam bandaid ni_fails handling so ap's with association failures are reconsidered after a bit; a proper fix involves more changes to the scanning infrastructure Reviewed by: avatar, David Young MFC after: 5 days 2005-07-23 01:16 sam the AREF flag is only meaningful in ap mode; adhoc neighbors now are timed out of the sta/neighbor table 2005-07-23 00:25 sam o move inactivity-related debug msgs under IEEE80211_MSG_INACT o probe inactive neighbors in adhoc mode (they don't have an association id so previously were being timed out) MFC after: 3 days 2005-07-22 22:11 sam split xmit of probe request frame out into a separate routine that takes explicit parameters; this will be needed when scanning is decoupled from the state machine to do bg scanning MFC after: 3 days 2005-07-22 21:48 sam split 802.11 frame xmit setup code into ieee80211_send_setup MFC after: 3 days 2005-07-22 18:57 sam simplify ic_newassoc callback MFC after: 3 days 2005-07-22 18:54 sam simplify ieee80211_ibss_merge api MFC after: 3 days 2005-07-22 18:50 sam add stats we know we'll need soon and some spare fields for future expansion MFC after: 3 days 2005-07-22 18:45 sam simplify tim callback api MFC after: 3 days 2005-07-22 18:42 sam don't include 802.3 header in min frame length calculation as it may not be present for a frag; fixes problem with small (fragmented) frames being dropped Obtained from: Atheros MFC after: 3 days 2005-07-22 18:36 sam simplify ieee80211_node_authorize and ieee80211_node_unauthorize api's MFC after: 3 days 2005-07-22 18:31 sam simplifiy ieee80211_send_nulldata api MFC after: 3 days 2005-07-22 18:29 sam simplify rate set api's by removing ic parameter (implicit in node reference) MFC after: 3 days 2005-07-22 18:21 sam reject association requests with a wpa/rsn ie when wpa/rsn is not configured on the ap; previously we either ignored the ie or (possibly) failed an assertion Obtained from: Atheros MFC after: 3 days 2005-07-22 18:16 sam missed one in last commit; add device name to discard msgs 2005-07-22 18:13 sam include device name in discard msgs 2005-07-22 18:12 sam add diag msgs for frames discarded because the direction field is wrong 2005-07-22 18:08 sam split data frame delivery out to a new function ieee80211_deliver_data 2005-07-22 18:00 sam o add IEEE80211_IOC_FRAGTHRESHOLD for getting+setting the tx fragmentation threshold o fix bounds checking on IEEE80211_IOC_RTSTHRESHOLD MFC after: 3 days 2005-07-22 17:55 sam o add IEEE80211_FRAG_DEFAULT o move default settings for RTS and frag thresholds to ieee80211_var.h 2005-07-22 17:50 sam diff reduction against p4: define IEEE80211_FIXED_RATE_NONE and use it instead of -1 2005-07-22 17:37 sam add flags missed in last merge 2005-07-22 17:36 sam Diff reduction against p4: o add ic_flags_ext for eventual extention of ic_flags o define/reserve flag+capabilities bits for superg, bg scan, and roaming support o refactor debug msg macros MFC after: 3 days 2005-07-22 06:17 sam send a response when an auth request is denied due to an acl; might be better to silently ignore the frame but this way we give stations a chance of figuring out what's wrong 2005-07-22 06:15 sam remove excess whitespace 2005-07-22 05:55 sam use IF_HANDOFF when bridging frames internally so if_start gets called; fixes communication between associated sta's MFC after: 3 days 2005-07-11 04:06 sam Handle encrypt of arbitarily fragmented mbuf chains: previously we bailed if we couldn't collect the 16-bytes of data required for an aes block cipher in 2 mbufs; now we deal with it. While here make space accounting signed so a sanity check does the right thing for malformed mbuf chains. Approved by: re (scottl) 2005-07-11 04:00 sam nuke assert that duplicates real check Reviewed by: avatar Approved by: re (scottl)
2005-11-18 19:40:08 +03:00
struct ieee80211_node **nt_keyixmap; /* key ix -> node map */
int nt_keyixmax; /* keyixmap size */
void (*nt_timeout)(struct ieee80211_node_table *);
};
void ieee80211_node_table_reset(struct ieee80211_node_table *);
struct ieee80211_node *ieee80211_alloc_node(
struct ieee80211_node_table *, const u_int8_t *);
Resolve conflicts and adapt to NetBSD. Thanks to dyoung@, scw@, and perry@ for help testing. 2005-08-30 15:27 avatar Properly set ic_curchan before calling back to device driver to do channel switching(ifconfig devX channel Y). This fix should make channel changing works again in monitor mode. Submitted by: sam X-MFC-With: other ic_curchan changes 2005-08-13 18:50 sam revert 1.64: we cannot use the channel characteristics to decide when to do 11g erp sta accounting because b/g channels show up as false positives when operating in 11b. Noticed by: Michal Mertl 2005-08-13 18:31 sam Extend acl support to pass ioctl requests through and use this to add support for getting the current policy setting and collecting the list of mac addresses in the acl table. Submitted by: Michal Mertl (original version) MFC after: 2 weeks 2005-08-10 18:42 sam Don't use ic_curmode to decide when to do 11g station accounting, use the station channel properties. Fixes assert failure/bogus operation when an ap is operating in 11a and has associated stations then switches to 11g. Noticed by: Michal Mertl Reviewed by: avatar MFC after: 2 weeks 2005-08-10 17:22 sam Clarify/fix handling of the current channel: o add ic_curchan and use it uniformly for specifying the current channel instead of overloading ic->ic_bss->ni_chan (or in some drivers ic_ibss_chan) o add ieee80211_scanparams structure to encapsulate scanning-related state captured for rx frames o move rx beacon+probe response frame handling into separate routines o change beacon+probe response handling to treat the scan table more like a scan cache--look for an existing entry before adding a new one; this combined with ic_curchan use corrects handling of stations that were previously found at a different channel o move adhoc neighbor discovery by beacon+probe response frames to a new ieee80211_add_neighbor routine Reviewed by: avatar Tested by: avatar, Michal Mertl MFC after: 2 weeks 2005-08-09 11:19 rwatson Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days 2005-08-08 19:46 sam Split crypto tx+rx key indices and add a key index -> node mapping table: Crypto changes: o change driver/net80211 key_alloc api to return tx+rx key indices; a driver can leave the rx key index set to IEEE80211_KEYIX_NONE or set it to be the same as the tx key index (the former disables use of the key index in building the keyix->node mapping table and is the default setup for naive drivers by null_key_alloc) o add cs_max_keyid to crypto state to specify the max h/w key index a driver will return; this is used to allocate the key index mapping table and to bounds check table loookups o while here introduce ieee80211_keyix (finally) for the type of a h/w key index o change crypto notifiers for rx failures to pass the rx key index up as appropriate (michael failure, replay, etc.) Node table changes: o optionally allocate a h/w key index to node mapping table for the station table using the max key index setting supplied by drivers (note the scan table does not get a map) o defer node table allocation to lateattach so the driver has a chance to set the max key id to size the key index map o while here also defer the aid bitmap allocation o add new ieee80211_find_rxnode_withkey api to find a sta/node entry on frame receive with an optional h/w key index to use in checking mapping table; also updates the map if it does a hash lookup and the found node has a rx key index set in the unicast key; note this work is separated from the old ieee80211_find_rxnode call so drivers do not need to be aware of the new mechanism o move some node table manipulation under the node table lock to close a race on node delete o add ieee80211_node_delucastkey to do the dirty work of deleting unicast key state for a node (deletes any key and handles key map references) Ath driver: o nuke private sc_keyixmap mechansim in favor of net80211 support o update key alloc api These changes close several race conditions for the ath driver operating in ap mode. Other drivers should see no change. Station mode operation for ath no longer uses the key index map but performance tests show no noticeable change and this will be fixed when the scan table is eliminated with the new scanning support. Tested by: Michal Mertl, avatar, others Reviewed by: avatar, others MFC after: 2 weeks 2005-08-08 06:49 sam use ieee80211_iterate_nodes to retrieve station data; the previous code walked the list w/o locking MFC after: 1 week 2005-08-08 04:30 sam Cleanup beacon/listen interval handling: o separate configured beacon interval from listen interval; this avoids potential use of one value for the other (e.g. setting powersavesleep to 0 clobbers the beacon interval used in hostap or ibss mode) o bounds check the beacon interval received in probe response and beacon frames and drop frames with bogus settings; not clear if we should instead clamp the value as any alteration would result in mismatched sta+ap configuration and probably be more confusing (don't want to log to the console but perhaps ok with rate limiting) o while here up max beacon interval to reflect WiFi standard Noticed by: Martin <nakal@nurfuerspam.de> MFC after: 1 week 2005-08-06 05:57 sam fix debug msg typo MFC after: 3 days 2005-08-06 05:56 sam Fix handling of frames sent prior to a station being authorized when operating in ap mode. Previously we allocated a node from the station table, sent the frame (using the node), then released the reference that "held the frame in the table". But while the frame was in flight the node might be reclaimed which could lead to problems. The solution is to add an ieee80211_tmp_node routine that crafts a node that does exist in a table and so isn't ever reclaimed; it exists only so long as the associated frame is in flight. MFC after: 5 days 2005-07-31 07:12 sam close a race between reclaiming a node when a station is inactive and sending the null data frame used to probe inactive stations MFC after: 5 days 2005-07-27 05:41 sam when bridging internally bypass the bss node as traffic to it must follow the normal input path Submitted by: Michal Mertl MFC after: 5 days 2005-07-27 03:53 sam bandaid ni_fails handling so ap's with association failures are reconsidered after a bit; a proper fix involves more changes to the scanning infrastructure Reviewed by: avatar, David Young MFC after: 5 days 2005-07-23 01:16 sam the AREF flag is only meaningful in ap mode; adhoc neighbors now are timed out of the sta/neighbor table 2005-07-23 00:25 sam o move inactivity-related debug msgs under IEEE80211_MSG_INACT o probe inactive neighbors in adhoc mode (they don't have an association id so previously were being timed out) MFC after: 3 days 2005-07-22 22:11 sam split xmit of probe request frame out into a separate routine that takes explicit parameters; this will be needed when scanning is decoupled from the state machine to do bg scanning MFC after: 3 days 2005-07-22 21:48 sam split 802.11 frame xmit setup code into ieee80211_send_setup MFC after: 3 days 2005-07-22 18:57 sam simplify ic_newassoc callback MFC after: 3 days 2005-07-22 18:54 sam simplify ieee80211_ibss_merge api MFC after: 3 days 2005-07-22 18:50 sam add stats we know we'll need soon and some spare fields for future expansion MFC after: 3 days 2005-07-22 18:45 sam simplify tim callback api MFC after: 3 days 2005-07-22 18:42 sam don't include 802.3 header in min frame length calculation as it may not be present for a frag; fixes problem with small (fragmented) frames being dropped Obtained from: Atheros MFC after: 3 days 2005-07-22 18:36 sam simplify ieee80211_node_authorize and ieee80211_node_unauthorize api's MFC after: 3 days 2005-07-22 18:31 sam simplifiy ieee80211_send_nulldata api MFC after: 3 days 2005-07-22 18:29 sam simplify rate set api's by removing ic parameter (implicit in node reference) MFC after: 3 days 2005-07-22 18:21 sam reject association requests with a wpa/rsn ie when wpa/rsn is not configured on the ap; previously we either ignored the ie or (possibly) failed an assertion Obtained from: Atheros MFC after: 3 days 2005-07-22 18:16 sam missed one in last commit; add device name to discard msgs 2005-07-22 18:13 sam include device name in discard msgs 2005-07-22 18:12 sam add diag msgs for frames discarded because the direction field is wrong 2005-07-22 18:08 sam split data frame delivery out to a new function ieee80211_deliver_data 2005-07-22 18:00 sam o add IEEE80211_IOC_FRAGTHRESHOLD for getting+setting the tx fragmentation threshold o fix bounds checking on IEEE80211_IOC_RTSTHRESHOLD MFC after: 3 days 2005-07-22 17:55 sam o add IEEE80211_FRAG_DEFAULT o move default settings for RTS and frag thresholds to ieee80211_var.h 2005-07-22 17:50 sam diff reduction against p4: define IEEE80211_FIXED_RATE_NONE and use it instead of -1 2005-07-22 17:37 sam add flags missed in last merge 2005-07-22 17:36 sam Diff reduction against p4: o add ic_flags_ext for eventual extention of ic_flags o define/reserve flag+capabilities bits for superg, bg scan, and roaming support o refactor debug msg macros MFC after: 3 days 2005-07-22 06:17 sam send a response when an auth request is denied due to an acl; might be better to silently ignore the frame but this way we give stations a chance of figuring out what's wrong 2005-07-22 06:15 sam remove excess whitespace 2005-07-22 05:55 sam use IF_HANDOFF when bridging frames internally so if_start gets called; fixes communication between associated sta's MFC after: 3 days 2005-07-11 04:06 sam Handle encrypt of arbitarily fragmented mbuf chains: previously we bailed if we couldn't collect the 16-bytes of data required for an aes block cipher in 2 mbufs; now we deal with it. While here make space accounting signed so a sanity check does the right thing for malformed mbuf chains. Approved by: re (scottl) 2005-07-11 04:00 sam nuke assert that duplicates real check Reviewed by: avatar Approved by: re (scottl)
2005-11-18 19:40:08 +03:00
struct ieee80211_node *ieee80211_tmp_node(struct ieee80211com *,
const u_int8_t *macaddr);
struct ieee80211_node *ieee80211_dup_bss(struct ieee80211_node_table *,
const u_int8_t *);
#ifdef IEEE80211_DEBUG_REFCNT
void ieee80211_free_node_debug(struct ieee80211_node *,
const char *func, int line);
struct ieee80211_node *ieee80211_find_node_debug(
struct ieee80211_node_table *, const u_int8_t *,
const char *func, int line);
struct ieee80211_node * ieee80211_find_rxnode_debug(
struct ieee80211com *, const struct ieee80211_frame_min *,
const char *func, int line);
Resolve conflicts and adapt to NetBSD. Thanks to dyoung@, scw@, and perry@ for help testing. 2005-08-30 15:27 avatar Properly set ic_curchan before calling back to device driver to do channel switching(ifconfig devX channel Y). This fix should make channel changing works again in monitor mode. Submitted by: sam X-MFC-With: other ic_curchan changes 2005-08-13 18:50 sam revert 1.64: we cannot use the channel characteristics to decide when to do 11g erp sta accounting because b/g channels show up as false positives when operating in 11b. Noticed by: Michal Mertl 2005-08-13 18:31 sam Extend acl support to pass ioctl requests through and use this to add support for getting the current policy setting and collecting the list of mac addresses in the acl table. Submitted by: Michal Mertl (original version) MFC after: 2 weeks 2005-08-10 18:42 sam Don't use ic_curmode to decide when to do 11g station accounting, use the station channel properties. Fixes assert failure/bogus operation when an ap is operating in 11a and has associated stations then switches to 11g. Noticed by: Michal Mertl Reviewed by: avatar MFC after: 2 weeks 2005-08-10 17:22 sam Clarify/fix handling of the current channel: o add ic_curchan and use it uniformly for specifying the current channel instead of overloading ic->ic_bss->ni_chan (or in some drivers ic_ibss_chan) o add ieee80211_scanparams structure to encapsulate scanning-related state captured for rx frames o move rx beacon+probe response frame handling into separate routines o change beacon+probe response handling to treat the scan table more like a scan cache--look for an existing entry before adding a new one; this combined with ic_curchan use corrects handling of stations that were previously found at a different channel o move adhoc neighbor discovery by beacon+probe response frames to a new ieee80211_add_neighbor routine Reviewed by: avatar Tested by: avatar, Michal Mertl MFC after: 2 weeks 2005-08-09 11:19 rwatson Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days 2005-08-08 19:46 sam Split crypto tx+rx key indices and add a key index -> node mapping table: Crypto changes: o change driver/net80211 key_alloc api to return tx+rx key indices; a driver can leave the rx key index set to IEEE80211_KEYIX_NONE or set it to be the same as the tx key index (the former disables use of the key index in building the keyix->node mapping table and is the default setup for naive drivers by null_key_alloc) o add cs_max_keyid to crypto state to specify the max h/w key index a driver will return; this is used to allocate the key index mapping table and to bounds check table loookups o while here introduce ieee80211_keyix (finally) for the type of a h/w key index o change crypto notifiers for rx failures to pass the rx key index up as appropriate (michael failure, replay, etc.) Node table changes: o optionally allocate a h/w key index to node mapping table for the station table using the max key index setting supplied by drivers (note the scan table does not get a map) o defer node table allocation to lateattach so the driver has a chance to set the max key id to size the key index map o while here also defer the aid bitmap allocation o add new ieee80211_find_rxnode_withkey api to find a sta/node entry on frame receive with an optional h/w key index to use in checking mapping table; also updates the map if it does a hash lookup and the found node has a rx key index set in the unicast key; note this work is separated from the old ieee80211_find_rxnode call so drivers do not need to be aware of the new mechanism o move some node table manipulation under the node table lock to close a race on node delete o add ieee80211_node_delucastkey to do the dirty work of deleting unicast key state for a node (deletes any key and handles key map references) Ath driver: o nuke private sc_keyixmap mechansim in favor of net80211 support o update key alloc api These changes close several race conditions for the ath driver operating in ap mode. Other drivers should see no change. Station mode operation for ath no longer uses the key index map but performance tests show no noticeable change and this will be fixed when the scan table is eliminated with the new scanning support. Tested by: Michal Mertl, avatar, others Reviewed by: avatar, others MFC after: 2 weeks 2005-08-08 06:49 sam use ieee80211_iterate_nodes to retrieve station data; the previous code walked the list w/o locking MFC after: 1 week 2005-08-08 04:30 sam Cleanup beacon/listen interval handling: o separate configured beacon interval from listen interval; this avoids potential use of one value for the other (e.g. setting powersavesleep to 0 clobbers the beacon interval used in hostap or ibss mode) o bounds check the beacon interval received in probe response and beacon frames and drop frames with bogus settings; not clear if we should instead clamp the value as any alteration would result in mismatched sta+ap configuration and probably be more confusing (don't want to log to the console but perhaps ok with rate limiting) o while here up max beacon interval to reflect WiFi standard Noticed by: Martin <nakal@nurfuerspam.de> MFC after: 1 week 2005-08-06 05:57 sam fix debug msg typo MFC after: 3 days 2005-08-06 05:56 sam Fix handling of frames sent prior to a station being authorized when operating in ap mode. Previously we allocated a node from the station table, sent the frame (using the node), then released the reference that "held the frame in the table". But while the frame was in flight the node might be reclaimed which could lead to problems. The solution is to add an ieee80211_tmp_node routine that crafts a node that does exist in a table and so isn't ever reclaimed; it exists only so long as the associated frame is in flight. MFC after: 5 days 2005-07-31 07:12 sam close a race between reclaiming a node when a station is inactive and sending the null data frame used to probe inactive stations MFC after: 5 days 2005-07-27 05:41 sam when bridging internally bypass the bss node as traffic to it must follow the normal input path Submitted by: Michal Mertl MFC after: 5 days 2005-07-27 03:53 sam bandaid ni_fails handling so ap's with association failures are reconsidered after a bit; a proper fix involves more changes to the scanning infrastructure Reviewed by: avatar, David Young MFC after: 5 days 2005-07-23 01:16 sam the AREF flag is only meaningful in ap mode; adhoc neighbors now are timed out of the sta/neighbor table 2005-07-23 00:25 sam o move inactivity-related debug msgs under IEEE80211_MSG_INACT o probe inactive neighbors in adhoc mode (they don't have an association id so previously were being timed out) MFC after: 3 days 2005-07-22 22:11 sam split xmit of probe request frame out into a separate routine that takes explicit parameters; this will be needed when scanning is decoupled from the state machine to do bg scanning MFC after: 3 days 2005-07-22 21:48 sam split 802.11 frame xmit setup code into ieee80211_send_setup MFC after: 3 days 2005-07-22 18:57 sam simplify ic_newassoc callback MFC after: 3 days 2005-07-22 18:54 sam simplify ieee80211_ibss_merge api MFC after: 3 days 2005-07-22 18:50 sam add stats we know we'll need soon and some spare fields for future expansion MFC after: 3 days 2005-07-22 18:45 sam simplify tim callback api MFC after: 3 days 2005-07-22 18:42 sam don't include 802.3 header in min frame length calculation as it may not be present for a frag; fixes problem with small (fragmented) frames being dropped Obtained from: Atheros MFC after: 3 days 2005-07-22 18:36 sam simplify ieee80211_node_authorize and ieee80211_node_unauthorize api's MFC after: 3 days 2005-07-22 18:31 sam simplifiy ieee80211_send_nulldata api MFC after: 3 days 2005-07-22 18:29 sam simplify rate set api's by removing ic parameter (implicit in node reference) MFC after: 3 days 2005-07-22 18:21 sam reject association requests with a wpa/rsn ie when wpa/rsn is not configured on the ap; previously we either ignored the ie or (possibly) failed an assertion Obtained from: Atheros MFC after: 3 days 2005-07-22 18:16 sam missed one in last commit; add device name to discard msgs 2005-07-22 18:13 sam include device name in discard msgs 2005-07-22 18:12 sam add diag msgs for frames discarded because the direction field is wrong 2005-07-22 18:08 sam split data frame delivery out to a new function ieee80211_deliver_data 2005-07-22 18:00 sam o add IEEE80211_IOC_FRAGTHRESHOLD for getting+setting the tx fragmentation threshold o fix bounds checking on IEEE80211_IOC_RTSTHRESHOLD MFC after: 3 days 2005-07-22 17:55 sam o add IEEE80211_FRAG_DEFAULT o move default settings for RTS and frag thresholds to ieee80211_var.h 2005-07-22 17:50 sam diff reduction against p4: define IEEE80211_FIXED_RATE_NONE and use it instead of -1 2005-07-22 17:37 sam add flags missed in last merge 2005-07-22 17:36 sam Diff reduction against p4: o add ic_flags_ext for eventual extention of ic_flags o define/reserve flag+capabilities bits for superg, bg scan, and roaming support o refactor debug msg macros MFC after: 3 days 2005-07-22 06:17 sam send a response when an auth request is denied due to an acl; might be better to silently ignore the frame but this way we give stations a chance of figuring out what's wrong 2005-07-22 06:15 sam remove excess whitespace 2005-07-22 05:55 sam use IF_HANDOFF when bridging frames internally so if_start gets called; fixes communication between associated sta's MFC after: 3 days 2005-07-11 04:06 sam Handle encrypt of arbitarily fragmented mbuf chains: previously we bailed if we couldn't collect the 16-bytes of data required for an aes block cipher in 2 mbufs; now we deal with it. While here make space accounting signed so a sanity check does the right thing for malformed mbuf chains. Approved by: re (scottl) 2005-07-11 04:00 sam nuke assert that duplicates real check Reviewed by: avatar Approved by: re (scottl)
2005-11-18 19:40:08 +03:00
struct ieee80211_node * ieee80211_find_rxnode_withkey_debug(
struct ieee80211com *,
const struct ieee80211_frame_min *, u_int16_t keyix,
const char *func, int line);
struct ieee80211_node *ieee80211_find_txnode_debug(
struct ieee80211com *, const u_int8_t *,
const char *func, int line);
struct ieee80211_node *ieee80211_find_node_with_channel_debug(
struct ieee80211_node_table *, const u_int8_t *macaddr,
struct ieee80211_channel *, const char *func, int line);
struct ieee80211_node *ieee80211_find_node_with_ssid_debug(
struct ieee80211_node_table *, const u_int8_t *macaddr,
u_int ssidlen, const u_int8_t *ssid,
const char *func, int line);
#define ieee80211_free_node(ni) \
ieee80211_free_node_debug(ni, __func__, __LINE__)
#define ieee80211_find_node(nt, mac) \
ieee80211_find_node_debug(nt, mac, __func__, __LINE__)
#define ieee80211_find_rxnode(nt, wh) \
ieee80211_find_rxnode_debug(nt, wh, __func__, __LINE__)
Resolve conflicts and adapt to NetBSD. Thanks to dyoung@, scw@, and perry@ for help testing. 2005-08-30 15:27 avatar Properly set ic_curchan before calling back to device driver to do channel switching(ifconfig devX channel Y). This fix should make channel changing works again in monitor mode. Submitted by: sam X-MFC-With: other ic_curchan changes 2005-08-13 18:50 sam revert 1.64: we cannot use the channel characteristics to decide when to do 11g erp sta accounting because b/g channels show up as false positives when operating in 11b. Noticed by: Michal Mertl 2005-08-13 18:31 sam Extend acl support to pass ioctl requests through and use this to add support for getting the current policy setting and collecting the list of mac addresses in the acl table. Submitted by: Michal Mertl (original version) MFC after: 2 weeks 2005-08-10 18:42 sam Don't use ic_curmode to decide when to do 11g station accounting, use the station channel properties. Fixes assert failure/bogus operation when an ap is operating in 11a and has associated stations then switches to 11g. Noticed by: Michal Mertl Reviewed by: avatar MFC after: 2 weeks 2005-08-10 17:22 sam Clarify/fix handling of the current channel: o add ic_curchan and use it uniformly for specifying the current channel instead of overloading ic->ic_bss->ni_chan (or in some drivers ic_ibss_chan) o add ieee80211_scanparams structure to encapsulate scanning-related state captured for rx frames o move rx beacon+probe response frame handling into separate routines o change beacon+probe response handling to treat the scan table more like a scan cache--look for an existing entry before adding a new one; this combined with ic_curchan use corrects handling of stations that were previously found at a different channel o move adhoc neighbor discovery by beacon+probe response frames to a new ieee80211_add_neighbor routine Reviewed by: avatar Tested by: avatar, Michal Mertl MFC after: 2 weeks 2005-08-09 11:19 rwatson Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days 2005-08-08 19:46 sam Split crypto tx+rx key indices and add a key index -> node mapping table: Crypto changes: o change driver/net80211 key_alloc api to return tx+rx key indices; a driver can leave the rx key index set to IEEE80211_KEYIX_NONE or set it to be the same as the tx key index (the former disables use of the key index in building the keyix->node mapping table and is the default setup for naive drivers by null_key_alloc) o add cs_max_keyid to crypto state to specify the max h/w key index a driver will return; this is used to allocate the key index mapping table and to bounds check table loookups o while here introduce ieee80211_keyix (finally) for the type of a h/w key index o change crypto notifiers for rx failures to pass the rx key index up as appropriate (michael failure, replay, etc.) Node table changes: o optionally allocate a h/w key index to node mapping table for the station table using the max key index setting supplied by drivers (note the scan table does not get a map) o defer node table allocation to lateattach so the driver has a chance to set the max key id to size the key index map o while here also defer the aid bitmap allocation o add new ieee80211_find_rxnode_withkey api to find a sta/node entry on frame receive with an optional h/w key index to use in checking mapping table; also updates the map if it does a hash lookup and the found node has a rx key index set in the unicast key; note this work is separated from the old ieee80211_find_rxnode call so drivers do not need to be aware of the new mechanism o move some node table manipulation under the node table lock to close a race on node delete o add ieee80211_node_delucastkey to do the dirty work of deleting unicast key state for a node (deletes any key and handles key map references) Ath driver: o nuke private sc_keyixmap mechansim in favor of net80211 support o update key alloc api These changes close several race conditions for the ath driver operating in ap mode. Other drivers should see no change. Station mode operation for ath no longer uses the key index map but performance tests show no noticeable change and this will be fixed when the scan table is eliminated with the new scanning support. Tested by: Michal Mertl, avatar, others Reviewed by: avatar, others MFC after: 2 weeks 2005-08-08 06:49 sam use ieee80211_iterate_nodes to retrieve station data; the previous code walked the list w/o locking MFC after: 1 week 2005-08-08 04:30 sam Cleanup beacon/listen interval handling: o separate configured beacon interval from listen interval; this avoids potential use of one value for the other (e.g. setting powersavesleep to 0 clobbers the beacon interval used in hostap or ibss mode) o bounds check the beacon interval received in probe response and beacon frames and drop frames with bogus settings; not clear if we should instead clamp the value as any alteration would result in mismatched sta+ap configuration and probably be more confusing (don't want to log to the console but perhaps ok with rate limiting) o while here up max beacon interval to reflect WiFi standard Noticed by: Martin <nakal@nurfuerspam.de> MFC after: 1 week 2005-08-06 05:57 sam fix debug msg typo MFC after: 3 days 2005-08-06 05:56 sam Fix handling of frames sent prior to a station being authorized when operating in ap mode. Previously we allocated a node from the station table, sent the frame (using the node), then released the reference that "held the frame in the table". But while the frame was in flight the node might be reclaimed which could lead to problems. The solution is to add an ieee80211_tmp_node routine that crafts a node that does exist in a table and so isn't ever reclaimed; it exists only so long as the associated frame is in flight. MFC after: 5 days 2005-07-31 07:12 sam close a race between reclaiming a node when a station is inactive and sending the null data frame used to probe inactive stations MFC after: 5 days 2005-07-27 05:41 sam when bridging internally bypass the bss node as traffic to it must follow the normal input path Submitted by: Michal Mertl MFC after: 5 days 2005-07-27 03:53 sam bandaid ni_fails handling so ap's with association failures are reconsidered after a bit; a proper fix involves more changes to the scanning infrastructure Reviewed by: avatar, David Young MFC after: 5 days 2005-07-23 01:16 sam the AREF flag is only meaningful in ap mode; adhoc neighbors now are timed out of the sta/neighbor table 2005-07-23 00:25 sam o move inactivity-related debug msgs under IEEE80211_MSG_INACT o probe inactive neighbors in adhoc mode (they don't have an association id so previously were being timed out) MFC after: 3 days 2005-07-22 22:11 sam split xmit of probe request frame out into a separate routine that takes explicit parameters; this will be needed when scanning is decoupled from the state machine to do bg scanning MFC after: 3 days 2005-07-22 21:48 sam split 802.11 frame xmit setup code into ieee80211_send_setup MFC after: 3 days 2005-07-22 18:57 sam simplify ic_newassoc callback MFC after: 3 days 2005-07-22 18:54 sam simplify ieee80211_ibss_merge api MFC after: 3 days 2005-07-22 18:50 sam add stats we know we'll need soon and some spare fields for future expansion MFC after: 3 days 2005-07-22 18:45 sam simplify tim callback api MFC after: 3 days 2005-07-22 18:42 sam don't include 802.3 header in min frame length calculation as it may not be present for a frag; fixes problem with small (fragmented) frames being dropped Obtained from: Atheros MFC after: 3 days 2005-07-22 18:36 sam simplify ieee80211_node_authorize and ieee80211_node_unauthorize api's MFC after: 3 days 2005-07-22 18:31 sam simplifiy ieee80211_send_nulldata api MFC after: 3 days 2005-07-22 18:29 sam simplify rate set api's by removing ic parameter (implicit in node reference) MFC after: 3 days 2005-07-22 18:21 sam reject association requests with a wpa/rsn ie when wpa/rsn is not configured on the ap; previously we either ignored the ie or (possibly) failed an assertion Obtained from: Atheros MFC after: 3 days 2005-07-22 18:16 sam missed one in last commit; add device name to discard msgs 2005-07-22 18:13 sam include device name in discard msgs 2005-07-22 18:12 sam add diag msgs for frames discarded because the direction field is wrong 2005-07-22 18:08 sam split data frame delivery out to a new function ieee80211_deliver_data 2005-07-22 18:00 sam o add IEEE80211_IOC_FRAGTHRESHOLD for getting+setting the tx fragmentation threshold o fix bounds checking on IEEE80211_IOC_RTSTHRESHOLD MFC after: 3 days 2005-07-22 17:55 sam o add IEEE80211_FRAG_DEFAULT o move default settings for RTS and frag thresholds to ieee80211_var.h 2005-07-22 17:50 sam diff reduction against p4: define IEEE80211_FIXED_RATE_NONE and use it instead of -1 2005-07-22 17:37 sam add flags missed in last merge 2005-07-22 17:36 sam Diff reduction against p4: o add ic_flags_ext for eventual extention of ic_flags o define/reserve flag+capabilities bits for superg, bg scan, and roaming support o refactor debug msg macros MFC after: 3 days 2005-07-22 06:17 sam send a response when an auth request is denied due to an acl; might be better to silently ignore the frame but this way we give stations a chance of figuring out what's wrong 2005-07-22 06:15 sam remove excess whitespace 2005-07-22 05:55 sam use IF_HANDOFF when bridging frames internally so if_start gets called; fixes communication between associated sta's MFC after: 3 days 2005-07-11 04:06 sam Handle encrypt of arbitarily fragmented mbuf chains: previously we bailed if we couldn't collect the 16-bytes of data required for an aes block cipher in 2 mbufs; now we deal with it. While here make space accounting signed so a sanity check does the right thing for malformed mbuf chains. Approved by: re (scottl) 2005-07-11 04:00 sam nuke assert that duplicates real check Reviewed by: avatar Approved by: re (scottl)
2005-11-18 19:40:08 +03:00
#define ieee80211_find_rxnode_withkey(nt, wh, keyix) \
ieee80211_find_rxnode_withkey_debug(nt, wh, keyix, __func__, __LINE__)
#define ieee80211_find_txnode(nt, mac) \
ieee80211_find_txnode_debug(nt, mac, __func__, __LINE__)
#define ieee80211_find_node_with_channel(nt, mac, c) \
ieee80211_find_node_with_channel_debug(nt, mac, c, __func__, __LINE__)
#define ieee80211_find_node_with_ssid(nt, mac, sl, ss) \
ieee80211_find_node_with_ssid_debug(nt, mac, sl, ss, __func__, __LINE__)
#else
void ieee80211_free_node(struct ieee80211_node *);
struct ieee80211_node *ieee80211_find_node(
struct ieee80211_node_table *, const u_int8_t *);
struct ieee80211_node * ieee80211_find_rxnode(
struct ieee80211com *, const struct ieee80211_frame_min *);
Resolve conflicts and adapt to NetBSD. Thanks to dyoung@, scw@, and perry@ for help testing. 2005-08-30 15:27 avatar Properly set ic_curchan before calling back to device driver to do channel switching(ifconfig devX channel Y). This fix should make channel changing works again in monitor mode. Submitted by: sam X-MFC-With: other ic_curchan changes 2005-08-13 18:50 sam revert 1.64: we cannot use the channel characteristics to decide when to do 11g erp sta accounting because b/g channels show up as false positives when operating in 11b. Noticed by: Michal Mertl 2005-08-13 18:31 sam Extend acl support to pass ioctl requests through and use this to add support for getting the current policy setting and collecting the list of mac addresses in the acl table. Submitted by: Michal Mertl (original version) MFC after: 2 weeks 2005-08-10 18:42 sam Don't use ic_curmode to decide when to do 11g station accounting, use the station channel properties. Fixes assert failure/bogus operation when an ap is operating in 11a and has associated stations then switches to 11g. Noticed by: Michal Mertl Reviewed by: avatar MFC after: 2 weeks 2005-08-10 17:22 sam Clarify/fix handling of the current channel: o add ic_curchan and use it uniformly for specifying the current channel instead of overloading ic->ic_bss->ni_chan (or in some drivers ic_ibss_chan) o add ieee80211_scanparams structure to encapsulate scanning-related state captured for rx frames o move rx beacon+probe response frame handling into separate routines o change beacon+probe response handling to treat the scan table more like a scan cache--look for an existing entry before adding a new one; this combined with ic_curchan use corrects handling of stations that were previously found at a different channel o move adhoc neighbor discovery by beacon+probe response frames to a new ieee80211_add_neighbor routine Reviewed by: avatar Tested by: avatar, Michal Mertl MFC after: 2 weeks 2005-08-09 11:19 rwatson Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days 2005-08-08 19:46 sam Split crypto tx+rx key indices and add a key index -> node mapping table: Crypto changes: o change driver/net80211 key_alloc api to return tx+rx key indices; a driver can leave the rx key index set to IEEE80211_KEYIX_NONE or set it to be the same as the tx key index (the former disables use of the key index in building the keyix->node mapping table and is the default setup for naive drivers by null_key_alloc) o add cs_max_keyid to crypto state to specify the max h/w key index a driver will return; this is used to allocate the key index mapping table and to bounds check table loookups o while here introduce ieee80211_keyix (finally) for the type of a h/w key index o change crypto notifiers for rx failures to pass the rx key index up as appropriate (michael failure, replay, etc.) Node table changes: o optionally allocate a h/w key index to node mapping table for the station table using the max key index setting supplied by drivers (note the scan table does not get a map) o defer node table allocation to lateattach so the driver has a chance to set the max key id to size the key index map o while here also defer the aid bitmap allocation o add new ieee80211_find_rxnode_withkey api to find a sta/node entry on frame receive with an optional h/w key index to use in checking mapping table; also updates the map if it does a hash lookup and the found node has a rx key index set in the unicast key; note this work is separated from the old ieee80211_find_rxnode call so drivers do not need to be aware of the new mechanism o move some node table manipulation under the node table lock to close a race on node delete o add ieee80211_node_delucastkey to do the dirty work of deleting unicast key state for a node (deletes any key and handles key map references) Ath driver: o nuke private sc_keyixmap mechansim in favor of net80211 support o update key alloc api These changes close several race conditions for the ath driver operating in ap mode. Other drivers should see no change. Station mode operation for ath no longer uses the key index map but performance tests show no noticeable change and this will be fixed when the scan table is eliminated with the new scanning support. Tested by: Michal Mertl, avatar, others Reviewed by: avatar, others MFC after: 2 weeks 2005-08-08 06:49 sam use ieee80211_iterate_nodes to retrieve station data; the previous code walked the list w/o locking MFC after: 1 week 2005-08-08 04:30 sam Cleanup beacon/listen interval handling: o separate configured beacon interval from listen interval; this avoids potential use of one value for the other (e.g. setting powersavesleep to 0 clobbers the beacon interval used in hostap or ibss mode) o bounds check the beacon interval received in probe response and beacon frames and drop frames with bogus settings; not clear if we should instead clamp the value as any alteration would result in mismatched sta+ap configuration and probably be more confusing (don't want to log to the console but perhaps ok with rate limiting) o while here up max beacon interval to reflect WiFi standard Noticed by: Martin <nakal@nurfuerspam.de> MFC after: 1 week 2005-08-06 05:57 sam fix debug msg typo MFC after: 3 days 2005-08-06 05:56 sam Fix handling of frames sent prior to a station being authorized when operating in ap mode. Previously we allocated a node from the station table, sent the frame (using the node), then released the reference that "held the frame in the table". But while the frame was in flight the node might be reclaimed which could lead to problems. The solution is to add an ieee80211_tmp_node routine that crafts a node that does exist in a table and so isn't ever reclaimed; it exists only so long as the associated frame is in flight. MFC after: 5 days 2005-07-31 07:12 sam close a race between reclaiming a node when a station is inactive and sending the null data frame used to probe inactive stations MFC after: 5 days 2005-07-27 05:41 sam when bridging internally bypass the bss node as traffic to it must follow the normal input path Submitted by: Michal Mertl MFC after: 5 days 2005-07-27 03:53 sam bandaid ni_fails handling so ap's with association failures are reconsidered after a bit; a proper fix involves more changes to the scanning infrastructure Reviewed by: avatar, David Young MFC after: 5 days 2005-07-23 01:16 sam the AREF flag is only meaningful in ap mode; adhoc neighbors now are timed out of the sta/neighbor table 2005-07-23 00:25 sam o move inactivity-related debug msgs under IEEE80211_MSG_INACT o probe inactive neighbors in adhoc mode (they don't have an association id so previously were being timed out) MFC after: 3 days 2005-07-22 22:11 sam split xmit of probe request frame out into a separate routine that takes explicit parameters; this will be needed when scanning is decoupled from the state machine to do bg scanning MFC after: 3 days 2005-07-22 21:48 sam split 802.11 frame xmit setup code into ieee80211_send_setup MFC after: 3 days 2005-07-22 18:57 sam simplify ic_newassoc callback MFC after: 3 days 2005-07-22 18:54 sam simplify ieee80211_ibss_merge api MFC after: 3 days 2005-07-22 18:50 sam add stats we know we'll need soon and some spare fields for future expansion MFC after: 3 days 2005-07-22 18:45 sam simplify tim callback api MFC after: 3 days 2005-07-22 18:42 sam don't include 802.3 header in min frame length calculation as it may not be present for a frag; fixes problem with small (fragmented) frames being dropped Obtained from: Atheros MFC after: 3 days 2005-07-22 18:36 sam simplify ieee80211_node_authorize and ieee80211_node_unauthorize api's MFC after: 3 days 2005-07-22 18:31 sam simplifiy ieee80211_send_nulldata api MFC after: 3 days 2005-07-22 18:29 sam simplify rate set api's by removing ic parameter (implicit in node reference) MFC after: 3 days 2005-07-22 18:21 sam reject association requests with a wpa/rsn ie when wpa/rsn is not configured on the ap; previously we either ignored the ie or (possibly) failed an assertion Obtained from: Atheros MFC after: 3 days 2005-07-22 18:16 sam missed one in last commit; add device name to discard msgs 2005-07-22 18:13 sam include device name in discard msgs 2005-07-22 18:12 sam add diag msgs for frames discarded because the direction field is wrong 2005-07-22 18:08 sam split data frame delivery out to a new function ieee80211_deliver_data 2005-07-22 18:00 sam o add IEEE80211_IOC_FRAGTHRESHOLD for getting+setting the tx fragmentation threshold o fix bounds checking on IEEE80211_IOC_RTSTHRESHOLD MFC after: 3 days 2005-07-22 17:55 sam o add IEEE80211_FRAG_DEFAULT o move default settings for RTS and frag thresholds to ieee80211_var.h 2005-07-22 17:50 sam diff reduction against p4: define IEEE80211_FIXED_RATE_NONE and use it instead of -1 2005-07-22 17:37 sam add flags missed in last merge 2005-07-22 17:36 sam Diff reduction against p4: o add ic_flags_ext for eventual extention of ic_flags o define/reserve flag+capabilities bits for superg, bg scan, and roaming support o refactor debug msg macros MFC after: 3 days 2005-07-22 06:17 sam send a response when an auth request is denied due to an acl; might be better to silently ignore the frame but this way we give stations a chance of figuring out what's wrong 2005-07-22 06:15 sam remove excess whitespace 2005-07-22 05:55 sam use IF_HANDOFF when bridging frames internally so if_start gets called; fixes communication between associated sta's MFC after: 3 days 2005-07-11 04:06 sam Handle encrypt of arbitarily fragmented mbuf chains: previously we bailed if we couldn't collect the 16-bytes of data required for an aes block cipher in 2 mbufs; now we deal with it. While here make space accounting signed so a sanity check does the right thing for malformed mbuf chains. Approved by: re (scottl) 2005-07-11 04:00 sam nuke assert that duplicates real check Reviewed by: avatar Approved by: re (scottl)
2005-11-18 19:40:08 +03:00
struct ieee80211_node * ieee80211_find_rxnode_withkey(struct ieee80211com *,
const struct ieee80211_frame_min *, u_int16_t keyix);
struct ieee80211_node *ieee80211_find_txnode(
struct ieee80211com *, const u_int8_t *);
struct ieee80211_node *ieee80211_find_node_with_channel(
struct ieee80211_node_table *, const u_int8_t *macaddr,
struct ieee80211_channel *);
struct ieee80211_node *ieee80211_find_node_with_ssid(
struct ieee80211_node_table *, const u_int8_t *macaddr,
u_int ssidlen, const u_int8_t *ssid);
#endif
Resolve conflicts and adapt to NetBSD. Thanks to dyoung@, scw@, and perry@ for help testing. 2005-08-30 15:27 avatar Properly set ic_curchan before calling back to device driver to do channel switching(ifconfig devX channel Y). This fix should make channel changing works again in monitor mode. Submitted by: sam X-MFC-With: other ic_curchan changes 2005-08-13 18:50 sam revert 1.64: we cannot use the channel characteristics to decide when to do 11g erp sta accounting because b/g channels show up as false positives when operating in 11b. Noticed by: Michal Mertl 2005-08-13 18:31 sam Extend acl support to pass ioctl requests through and use this to add support for getting the current policy setting and collecting the list of mac addresses in the acl table. Submitted by: Michal Mertl (original version) MFC after: 2 weeks 2005-08-10 18:42 sam Don't use ic_curmode to decide when to do 11g station accounting, use the station channel properties. Fixes assert failure/bogus operation when an ap is operating in 11a and has associated stations then switches to 11g. Noticed by: Michal Mertl Reviewed by: avatar MFC after: 2 weeks 2005-08-10 17:22 sam Clarify/fix handling of the current channel: o add ic_curchan and use it uniformly for specifying the current channel instead of overloading ic->ic_bss->ni_chan (or in some drivers ic_ibss_chan) o add ieee80211_scanparams structure to encapsulate scanning-related state captured for rx frames o move rx beacon+probe response frame handling into separate routines o change beacon+probe response handling to treat the scan table more like a scan cache--look for an existing entry before adding a new one; this combined with ic_curchan use corrects handling of stations that were previously found at a different channel o move adhoc neighbor discovery by beacon+probe response frames to a new ieee80211_add_neighbor routine Reviewed by: avatar Tested by: avatar, Michal Mertl MFC after: 2 weeks 2005-08-09 11:19 rwatson Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days 2005-08-08 19:46 sam Split crypto tx+rx key indices and add a key index -> node mapping table: Crypto changes: o change driver/net80211 key_alloc api to return tx+rx key indices; a driver can leave the rx key index set to IEEE80211_KEYIX_NONE or set it to be the same as the tx key index (the former disables use of the key index in building the keyix->node mapping table and is the default setup for naive drivers by null_key_alloc) o add cs_max_keyid to crypto state to specify the max h/w key index a driver will return; this is used to allocate the key index mapping table and to bounds check table loookups o while here introduce ieee80211_keyix (finally) for the type of a h/w key index o change crypto notifiers for rx failures to pass the rx key index up as appropriate (michael failure, replay, etc.) Node table changes: o optionally allocate a h/w key index to node mapping table for the station table using the max key index setting supplied by drivers (note the scan table does not get a map) o defer node table allocation to lateattach so the driver has a chance to set the max key id to size the key index map o while here also defer the aid bitmap allocation o add new ieee80211_find_rxnode_withkey api to find a sta/node entry on frame receive with an optional h/w key index to use in checking mapping table; also updates the map if it does a hash lookup and the found node has a rx key index set in the unicast key; note this work is separated from the old ieee80211_find_rxnode call so drivers do not need to be aware of the new mechanism o move some node table manipulation under the node table lock to close a race on node delete o add ieee80211_node_delucastkey to do the dirty work of deleting unicast key state for a node (deletes any key and handles key map references) Ath driver: o nuke private sc_keyixmap mechansim in favor of net80211 support o update key alloc api These changes close several race conditions for the ath driver operating in ap mode. Other drivers should see no change. Station mode operation for ath no longer uses the key index map but performance tests show no noticeable change and this will be fixed when the scan table is eliminated with the new scanning support. Tested by: Michal Mertl, avatar, others Reviewed by: avatar, others MFC after: 2 weeks 2005-08-08 06:49 sam use ieee80211_iterate_nodes to retrieve station data; the previous code walked the list w/o locking MFC after: 1 week 2005-08-08 04:30 sam Cleanup beacon/listen interval handling: o separate configured beacon interval from listen interval; this avoids potential use of one value for the other (e.g. setting powersavesleep to 0 clobbers the beacon interval used in hostap or ibss mode) o bounds check the beacon interval received in probe response and beacon frames and drop frames with bogus settings; not clear if we should instead clamp the value as any alteration would result in mismatched sta+ap configuration and probably be more confusing (don't want to log to the console but perhaps ok with rate limiting) o while here up max beacon interval to reflect WiFi standard Noticed by: Martin <nakal@nurfuerspam.de> MFC after: 1 week 2005-08-06 05:57 sam fix debug msg typo MFC after: 3 days 2005-08-06 05:56 sam Fix handling of frames sent prior to a station being authorized when operating in ap mode. Previously we allocated a node from the station table, sent the frame (using the node), then released the reference that "held the frame in the table". But while the frame was in flight the node might be reclaimed which could lead to problems. The solution is to add an ieee80211_tmp_node routine that crafts a node that does exist in a table and so isn't ever reclaimed; it exists only so long as the associated frame is in flight. MFC after: 5 days 2005-07-31 07:12 sam close a race between reclaiming a node when a station is inactive and sending the null data frame used to probe inactive stations MFC after: 5 days 2005-07-27 05:41 sam when bridging internally bypass the bss node as traffic to it must follow the normal input path Submitted by: Michal Mertl MFC after: 5 days 2005-07-27 03:53 sam bandaid ni_fails handling so ap's with association failures are reconsidered after a bit; a proper fix involves more changes to the scanning infrastructure Reviewed by: avatar, David Young MFC after: 5 days 2005-07-23 01:16 sam the AREF flag is only meaningful in ap mode; adhoc neighbors now are timed out of the sta/neighbor table 2005-07-23 00:25 sam o move inactivity-related debug msgs under IEEE80211_MSG_INACT o probe inactive neighbors in adhoc mode (they don't have an association id so previously were being timed out) MFC after: 3 days 2005-07-22 22:11 sam split xmit of probe request frame out into a separate routine that takes explicit parameters; this will be needed when scanning is decoupled from the state machine to do bg scanning MFC after: 3 days 2005-07-22 21:48 sam split 802.11 frame xmit setup code into ieee80211_send_setup MFC after: 3 days 2005-07-22 18:57 sam simplify ic_newassoc callback MFC after: 3 days 2005-07-22 18:54 sam simplify ieee80211_ibss_merge api MFC after: 3 days 2005-07-22 18:50 sam add stats we know we'll need soon and some spare fields for future expansion MFC after: 3 days 2005-07-22 18:45 sam simplify tim callback api MFC after: 3 days 2005-07-22 18:42 sam don't include 802.3 header in min frame length calculation as it may not be present for a frag; fixes problem with small (fragmented) frames being dropped Obtained from: Atheros MFC after: 3 days 2005-07-22 18:36 sam simplify ieee80211_node_authorize and ieee80211_node_unauthorize api's MFC after: 3 days 2005-07-22 18:31 sam simplifiy ieee80211_send_nulldata api MFC after: 3 days 2005-07-22 18:29 sam simplify rate set api's by removing ic parameter (implicit in node reference) MFC after: 3 days 2005-07-22 18:21 sam reject association requests with a wpa/rsn ie when wpa/rsn is not configured on the ap; previously we either ignored the ie or (possibly) failed an assertion Obtained from: Atheros MFC after: 3 days 2005-07-22 18:16 sam missed one in last commit; add device name to discard msgs 2005-07-22 18:13 sam include device name in discard msgs 2005-07-22 18:12 sam add diag msgs for frames discarded because the direction field is wrong 2005-07-22 18:08 sam split data frame delivery out to a new function ieee80211_deliver_data 2005-07-22 18:00 sam o add IEEE80211_IOC_FRAGTHRESHOLD for getting+setting the tx fragmentation threshold o fix bounds checking on IEEE80211_IOC_RTSTHRESHOLD MFC after: 3 days 2005-07-22 17:55 sam o add IEEE80211_FRAG_DEFAULT o move default settings for RTS and frag thresholds to ieee80211_var.h 2005-07-22 17:50 sam diff reduction against p4: define IEEE80211_FIXED_RATE_NONE and use it instead of -1 2005-07-22 17:37 sam add flags missed in last merge 2005-07-22 17:36 sam Diff reduction against p4: o add ic_flags_ext for eventual extention of ic_flags o define/reserve flag+capabilities bits for superg, bg scan, and roaming support o refactor debug msg macros MFC after: 3 days 2005-07-22 06:17 sam send a response when an auth request is denied due to an acl; might be better to silently ignore the frame but this way we give stations a chance of figuring out what's wrong 2005-07-22 06:15 sam remove excess whitespace 2005-07-22 05:55 sam use IF_HANDOFF when bridging frames internally so if_start gets called; fixes communication between associated sta's MFC after: 3 days 2005-07-11 04:06 sam Handle encrypt of arbitarily fragmented mbuf chains: previously we bailed if we couldn't collect the 16-bytes of data required for an aes block cipher in 2 mbufs; now we deal with it. While here make space accounting signed so a sanity check does the right thing for malformed mbuf chains. Approved by: re (scottl) 2005-07-11 04:00 sam nuke assert that duplicates real check Reviewed by: avatar Approved by: re (scottl)
2005-11-18 19:40:08 +03:00
int ieee80211_node_delucastkey(struct ieee80211_node *);
struct ieee80211_node *ieee80211_refine_node_for_beacon(
struct ieee80211com *, struct ieee80211_node *,
struct ieee80211_channel *, const u_int8_t *ssid);
typedef void ieee80211_iter_func(void *, struct ieee80211_node *);
void ieee80211_iterate_nodes(struct ieee80211_node_table *,
ieee80211_iter_func *, void *);
void ieee80211_dump_node(struct ieee80211_node_table *,
struct ieee80211_node *);
void ieee80211_dump_nodes(struct ieee80211_node_table *);
struct ieee80211_node *ieee80211_fakeup_adhoc_node(
struct ieee80211_node_table *, const u_int8_t macaddr[]);
void ieee80211_node_join(struct ieee80211com *, struct ieee80211_node *,int);
void ieee80211_node_leave(struct ieee80211com *, struct ieee80211_node *);
u_int8_t ieee80211_getrssi(struct ieee80211com *ic);
Resolve conflicts and adapt to NetBSD. Thanks to dyoung@, scw@, and perry@ for help testing. 2005-08-30 15:27 avatar Properly set ic_curchan before calling back to device driver to do channel switching(ifconfig devX channel Y). This fix should make channel changing works again in monitor mode. Submitted by: sam X-MFC-With: other ic_curchan changes 2005-08-13 18:50 sam revert 1.64: we cannot use the channel characteristics to decide when to do 11g erp sta accounting because b/g channels show up as false positives when operating in 11b. Noticed by: Michal Mertl 2005-08-13 18:31 sam Extend acl support to pass ioctl requests through and use this to add support for getting the current policy setting and collecting the list of mac addresses in the acl table. Submitted by: Michal Mertl (original version) MFC after: 2 weeks 2005-08-10 18:42 sam Don't use ic_curmode to decide when to do 11g station accounting, use the station channel properties. Fixes assert failure/bogus operation when an ap is operating in 11a and has associated stations then switches to 11g. Noticed by: Michal Mertl Reviewed by: avatar MFC after: 2 weeks 2005-08-10 17:22 sam Clarify/fix handling of the current channel: o add ic_curchan and use it uniformly for specifying the current channel instead of overloading ic->ic_bss->ni_chan (or in some drivers ic_ibss_chan) o add ieee80211_scanparams structure to encapsulate scanning-related state captured for rx frames o move rx beacon+probe response frame handling into separate routines o change beacon+probe response handling to treat the scan table more like a scan cache--look for an existing entry before adding a new one; this combined with ic_curchan use corrects handling of stations that were previously found at a different channel o move adhoc neighbor discovery by beacon+probe response frames to a new ieee80211_add_neighbor routine Reviewed by: avatar Tested by: avatar, Michal Mertl MFC after: 2 weeks 2005-08-09 11:19 rwatson Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days 2005-08-08 19:46 sam Split crypto tx+rx key indices and add a key index -> node mapping table: Crypto changes: o change driver/net80211 key_alloc api to return tx+rx key indices; a driver can leave the rx key index set to IEEE80211_KEYIX_NONE or set it to be the same as the tx key index (the former disables use of the key index in building the keyix->node mapping table and is the default setup for naive drivers by null_key_alloc) o add cs_max_keyid to crypto state to specify the max h/w key index a driver will return; this is used to allocate the key index mapping table and to bounds check table loookups o while here introduce ieee80211_keyix (finally) for the type of a h/w key index o change crypto notifiers for rx failures to pass the rx key index up as appropriate (michael failure, replay, etc.) Node table changes: o optionally allocate a h/w key index to node mapping table for the station table using the max key index setting supplied by drivers (note the scan table does not get a map) o defer node table allocation to lateattach so the driver has a chance to set the max key id to size the key index map o while here also defer the aid bitmap allocation o add new ieee80211_find_rxnode_withkey api to find a sta/node entry on frame receive with an optional h/w key index to use in checking mapping table; also updates the map if it does a hash lookup and the found node has a rx key index set in the unicast key; note this work is separated from the old ieee80211_find_rxnode call so drivers do not need to be aware of the new mechanism o move some node table manipulation under the node table lock to close a race on node delete o add ieee80211_node_delucastkey to do the dirty work of deleting unicast key state for a node (deletes any key and handles key map references) Ath driver: o nuke private sc_keyixmap mechansim in favor of net80211 support o update key alloc api These changes close several race conditions for the ath driver operating in ap mode. Other drivers should see no change. Station mode operation for ath no longer uses the key index map but performance tests show no noticeable change and this will be fixed when the scan table is eliminated with the new scanning support. Tested by: Michal Mertl, avatar, others Reviewed by: avatar, others MFC after: 2 weeks 2005-08-08 06:49 sam use ieee80211_iterate_nodes to retrieve station data; the previous code walked the list w/o locking MFC after: 1 week 2005-08-08 04:30 sam Cleanup beacon/listen interval handling: o separate configured beacon interval from listen interval; this avoids potential use of one value for the other (e.g. setting powersavesleep to 0 clobbers the beacon interval used in hostap or ibss mode) o bounds check the beacon interval received in probe response and beacon frames and drop frames with bogus settings; not clear if we should instead clamp the value as any alteration would result in mismatched sta+ap configuration and probably be more confusing (don't want to log to the console but perhaps ok with rate limiting) o while here up max beacon interval to reflect WiFi standard Noticed by: Martin <nakal@nurfuerspam.de> MFC after: 1 week 2005-08-06 05:57 sam fix debug msg typo MFC after: 3 days 2005-08-06 05:56 sam Fix handling of frames sent prior to a station being authorized when operating in ap mode. Previously we allocated a node from the station table, sent the frame (using the node), then released the reference that "held the frame in the table". But while the frame was in flight the node might be reclaimed which could lead to problems. The solution is to add an ieee80211_tmp_node routine that crafts a node that does exist in a table and so isn't ever reclaimed; it exists only so long as the associated frame is in flight. MFC after: 5 days 2005-07-31 07:12 sam close a race between reclaiming a node when a station is inactive and sending the null data frame used to probe inactive stations MFC after: 5 days 2005-07-27 05:41 sam when bridging internally bypass the bss node as traffic to it must follow the normal input path Submitted by: Michal Mertl MFC after: 5 days 2005-07-27 03:53 sam bandaid ni_fails handling so ap's with association failures are reconsidered after a bit; a proper fix involves more changes to the scanning infrastructure Reviewed by: avatar, David Young MFC after: 5 days 2005-07-23 01:16 sam the AREF flag is only meaningful in ap mode; adhoc neighbors now are timed out of the sta/neighbor table 2005-07-23 00:25 sam o move inactivity-related debug msgs under IEEE80211_MSG_INACT o probe inactive neighbors in adhoc mode (they don't have an association id so previously were being timed out) MFC after: 3 days 2005-07-22 22:11 sam split xmit of probe request frame out into a separate routine that takes explicit parameters; this will be needed when scanning is decoupled from the state machine to do bg scanning MFC after: 3 days 2005-07-22 21:48 sam split 802.11 frame xmit setup code into ieee80211_send_setup MFC after: 3 days 2005-07-22 18:57 sam simplify ic_newassoc callback MFC after: 3 days 2005-07-22 18:54 sam simplify ieee80211_ibss_merge api MFC after: 3 days 2005-07-22 18:50 sam add stats we know we'll need soon and some spare fields for future expansion MFC after: 3 days 2005-07-22 18:45 sam simplify tim callback api MFC after: 3 days 2005-07-22 18:42 sam don't include 802.3 header in min frame length calculation as it may not be present for a frag; fixes problem with small (fragmented) frames being dropped Obtained from: Atheros MFC after: 3 days 2005-07-22 18:36 sam simplify ieee80211_node_authorize and ieee80211_node_unauthorize api's MFC after: 3 days 2005-07-22 18:31 sam simplifiy ieee80211_send_nulldata api MFC after: 3 days 2005-07-22 18:29 sam simplify rate set api's by removing ic parameter (implicit in node reference) MFC after: 3 days 2005-07-22 18:21 sam reject association requests with a wpa/rsn ie when wpa/rsn is not configured on the ap; previously we either ignored the ie or (possibly) failed an assertion Obtained from: Atheros MFC after: 3 days 2005-07-22 18:16 sam missed one in last commit; add device name to discard msgs 2005-07-22 18:13 sam include device name in discard msgs 2005-07-22 18:12 sam add diag msgs for frames discarded because the direction field is wrong 2005-07-22 18:08 sam split data frame delivery out to a new function ieee80211_deliver_data 2005-07-22 18:00 sam o add IEEE80211_IOC_FRAGTHRESHOLD for getting+setting the tx fragmentation threshold o fix bounds checking on IEEE80211_IOC_RTSTHRESHOLD MFC after: 3 days 2005-07-22 17:55 sam o add IEEE80211_FRAG_DEFAULT o move default settings for RTS and frag thresholds to ieee80211_var.h 2005-07-22 17:50 sam diff reduction against p4: define IEEE80211_FIXED_RATE_NONE and use it instead of -1 2005-07-22 17:37 sam add flags missed in last merge 2005-07-22 17:36 sam Diff reduction against p4: o add ic_flags_ext for eventual extention of ic_flags o define/reserve flag+capabilities bits for superg, bg scan, and roaming support o refactor debug msg macros MFC after: 3 days 2005-07-22 06:17 sam send a response when an auth request is denied due to an acl; might be better to silently ignore the frame but this way we give stations a chance of figuring out what's wrong 2005-07-22 06:15 sam remove excess whitespace 2005-07-22 05:55 sam use IF_HANDOFF when bridging frames internally so if_start gets called; fixes communication between associated sta's MFC after: 3 days 2005-07-11 04:06 sam Handle encrypt of arbitarily fragmented mbuf chains: previously we bailed if we couldn't collect the 16-bytes of data required for an aes block cipher in 2 mbufs; now we deal with it. While here make space accounting signed so a sanity check does the right thing for malformed mbuf chains. Approved by: re (scottl) 2005-07-11 04:00 sam nuke assert that duplicates real check Reviewed by: avatar Approved by: re (scottl)
2005-11-18 19:40:08 +03:00
/*
* Parameters supplied when adding/updating an entry in a
* scan cache. Pointer variables should be set to NULL
* if no data is available. Pointer references can be to
* local data; any information that is saved will be copied.
* All multi-byte values must be in host byte order.
*/
struct ieee80211_scanparams {
u_int16_t capinfo; /* 802.11 capabilities */
u_int16_t fhdwell; /* FHSS dwell interval */
u_int8_t chan; /* */
u_int8_t bchan;
u_int8_t fhindex;
u_int8_t erp;
u_int16_t bintval;
u_int8_t timoff;
u_int8_t *tim;
u_int8_t *tstamp;
u_int8_t *country;
u_int8_t *ssid;
u_int8_t *rates;
u_int8_t *xrates;
u_int8_t *wpa;
u_int8_t *wme;
};
/*
* Node reference counting definitions.
*
* ieee80211_node_initref initialize the reference count to 1
* ieee80211_node_incref add a reference
* ieee80211_node_decref remove a reference
* ieee80211_node_dectestref remove a reference and return 1 if this
* is the last reference, otherwise 0
* ieee80211_node_refcnt reference count for printing (only)
*/
static inline void
ieee80211_node_initref(struct ieee80211_node *ni)
{
ni->ni_refcnt = 1;
}
static inline void
ieee80211_node_incref(struct ieee80211_node *ni)
{
atomic_inc_uint(&ni->ni_refcnt);
}
static inline void
ieee80211_node_decref(struct ieee80211_node *ni)
{
atomic_dec_uint(&ni->ni_refcnt);
}
int ieee80211_node_dectestref(struct ieee80211_node *ni);
static inline unsigned int
ieee80211_node_refcnt(const struct ieee80211_node *ni)
{
return ni->ni_refcnt;
}
static __inline struct ieee80211_node *
ieee80211_ref_node(struct ieee80211_node *ni)
{
ieee80211_node_incref(ni);
return ni;
}
static __inline void
ieee80211_unref_node(struct ieee80211_node **ni)
{
ieee80211_node_decref(*ni);
*ni = NULL; /* guard against use */
}
Resolve conflicts and adapt to NetBSD. Thanks to dyoung@, scw@, and perry@ for help testing. 2005-08-30 15:27 avatar Properly set ic_curchan before calling back to device driver to do channel switching(ifconfig devX channel Y). This fix should make channel changing works again in monitor mode. Submitted by: sam X-MFC-With: other ic_curchan changes 2005-08-13 18:50 sam revert 1.64: we cannot use the channel characteristics to decide when to do 11g erp sta accounting because b/g channels show up as false positives when operating in 11b. Noticed by: Michal Mertl 2005-08-13 18:31 sam Extend acl support to pass ioctl requests through and use this to add support for getting the current policy setting and collecting the list of mac addresses in the acl table. Submitted by: Michal Mertl (original version) MFC after: 2 weeks 2005-08-10 18:42 sam Don't use ic_curmode to decide when to do 11g station accounting, use the station channel properties. Fixes assert failure/bogus operation when an ap is operating in 11a and has associated stations then switches to 11g. Noticed by: Michal Mertl Reviewed by: avatar MFC after: 2 weeks 2005-08-10 17:22 sam Clarify/fix handling of the current channel: o add ic_curchan and use it uniformly for specifying the current channel instead of overloading ic->ic_bss->ni_chan (or in some drivers ic_ibss_chan) o add ieee80211_scanparams structure to encapsulate scanning-related state captured for rx frames o move rx beacon+probe response frame handling into separate routines o change beacon+probe response handling to treat the scan table more like a scan cache--look for an existing entry before adding a new one; this combined with ic_curchan use corrects handling of stations that were previously found at a different channel o move adhoc neighbor discovery by beacon+probe response frames to a new ieee80211_add_neighbor routine Reviewed by: avatar Tested by: avatar, Michal Mertl MFC after: 2 weeks 2005-08-09 11:19 rwatson Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days 2005-08-08 19:46 sam Split crypto tx+rx key indices and add a key index -> node mapping table: Crypto changes: o change driver/net80211 key_alloc api to return tx+rx key indices; a driver can leave the rx key index set to IEEE80211_KEYIX_NONE or set it to be the same as the tx key index (the former disables use of the key index in building the keyix->node mapping table and is the default setup for naive drivers by null_key_alloc) o add cs_max_keyid to crypto state to specify the max h/w key index a driver will return; this is used to allocate the key index mapping table and to bounds check table loookups o while here introduce ieee80211_keyix (finally) for the type of a h/w key index o change crypto notifiers for rx failures to pass the rx key index up as appropriate (michael failure, replay, etc.) Node table changes: o optionally allocate a h/w key index to node mapping table for the station table using the max key index setting supplied by drivers (note the scan table does not get a map) o defer node table allocation to lateattach so the driver has a chance to set the max key id to size the key index map o while here also defer the aid bitmap allocation o add new ieee80211_find_rxnode_withkey api to find a sta/node entry on frame receive with an optional h/w key index to use in checking mapping table; also updates the map if it does a hash lookup and the found node has a rx key index set in the unicast key; note this work is separated from the old ieee80211_find_rxnode call so drivers do not need to be aware of the new mechanism o move some node table manipulation under the node table lock to close a race on node delete o add ieee80211_node_delucastkey to do the dirty work of deleting unicast key state for a node (deletes any key and handles key map references) Ath driver: o nuke private sc_keyixmap mechansim in favor of net80211 support o update key alloc api These changes close several race conditions for the ath driver operating in ap mode. Other drivers should see no change. Station mode operation for ath no longer uses the key index map but performance tests show no noticeable change and this will be fixed when the scan table is eliminated with the new scanning support. Tested by: Michal Mertl, avatar, others Reviewed by: avatar, others MFC after: 2 weeks 2005-08-08 06:49 sam use ieee80211_iterate_nodes to retrieve station data; the previous code walked the list w/o locking MFC after: 1 week 2005-08-08 04:30 sam Cleanup beacon/listen interval handling: o separate configured beacon interval from listen interval; this avoids potential use of one value for the other (e.g. setting powersavesleep to 0 clobbers the beacon interval used in hostap or ibss mode) o bounds check the beacon interval received in probe response and beacon frames and drop frames with bogus settings; not clear if we should instead clamp the value as any alteration would result in mismatched sta+ap configuration and probably be more confusing (don't want to log to the console but perhaps ok with rate limiting) o while here up max beacon interval to reflect WiFi standard Noticed by: Martin <nakal@nurfuerspam.de> MFC after: 1 week 2005-08-06 05:57 sam fix debug msg typo MFC after: 3 days 2005-08-06 05:56 sam Fix handling of frames sent prior to a station being authorized when operating in ap mode. Previously we allocated a node from the station table, sent the frame (using the node), then released the reference that "held the frame in the table". But while the frame was in flight the node might be reclaimed which could lead to problems. The solution is to add an ieee80211_tmp_node routine that crafts a node that does exist in a table and so isn't ever reclaimed; it exists only so long as the associated frame is in flight. MFC after: 5 days 2005-07-31 07:12 sam close a race between reclaiming a node when a station is inactive and sending the null data frame used to probe inactive stations MFC after: 5 days 2005-07-27 05:41 sam when bridging internally bypass the bss node as traffic to it must follow the normal input path Submitted by: Michal Mertl MFC after: 5 days 2005-07-27 03:53 sam bandaid ni_fails handling so ap's with association failures are reconsidered after a bit; a proper fix involves more changes to the scanning infrastructure Reviewed by: avatar, David Young MFC after: 5 days 2005-07-23 01:16 sam the AREF flag is only meaningful in ap mode; adhoc neighbors now are timed out of the sta/neighbor table 2005-07-23 00:25 sam o move inactivity-related debug msgs under IEEE80211_MSG_INACT o probe inactive neighbors in adhoc mode (they don't have an association id so previously were being timed out) MFC after: 3 days 2005-07-22 22:11 sam split xmit of probe request frame out into a separate routine that takes explicit parameters; this will be needed when scanning is decoupled from the state machine to do bg scanning MFC after: 3 days 2005-07-22 21:48 sam split 802.11 frame xmit setup code into ieee80211_send_setup MFC after: 3 days 2005-07-22 18:57 sam simplify ic_newassoc callback MFC after: 3 days 2005-07-22 18:54 sam simplify ieee80211_ibss_merge api MFC after: 3 days 2005-07-22 18:50 sam add stats we know we'll need soon and some spare fields for future expansion MFC after: 3 days 2005-07-22 18:45 sam simplify tim callback api MFC after: 3 days 2005-07-22 18:42 sam don't include 802.3 header in min frame length calculation as it may not be present for a frag; fixes problem with small (fragmented) frames being dropped Obtained from: Atheros MFC after: 3 days 2005-07-22 18:36 sam simplify ieee80211_node_authorize and ieee80211_node_unauthorize api's MFC after: 3 days 2005-07-22 18:31 sam simplifiy ieee80211_send_nulldata api MFC after: 3 days 2005-07-22 18:29 sam simplify rate set api's by removing ic parameter (implicit in node reference) MFC after: 3 days 2005-07-22 18:21 sam reject association requests with a wpa/rsn ie when wpa/rsn is not configured on the ap; previously we either ignored the ie or (possibly) failed an assertion Obtained from: Atheros MFC after: 3 days 2005-07-22 18:16 sam missed one in last commit; add device name to discard msgs 2005-07-22 18:13 sam include device name in discard msgs 2005-07-22 18:12 sam add diag msgs for frames discarded because the direction field is wrong 2005-07-22 18:08 sam split data frame delivery out to a new function ieee80211_deliver_data 2005-07-22 18:00 sam o add IEEE80211_IOC_FRAGTHRESHOLD for getting+setting the tx fragmentation threshold o fix bounds checking on IEEE80211_IOC_RTSTHRESHOLD MFC after: 3 days 2005-07-22 17:55 sam o add IEEE80211_FRAG_DEFAULT o move default settings for RTS and frag thresholds to ieee80211_var.h 2005-07-22 17:50 sam diff reduction against p4: define IEEE80211_FIXED_RATE_NONE and use it instead of -1 2005-07-22 17:37 sam add flags missed in last merge 2005-07-22 17:36 sam Diff reduction against p4: o add ic_flags_ext for eventual extention of ic_flags o define/reserve flag+capabilities bits for superg, bg scan, and roaming support o refactor debug msg macros MFC after: 3 days 2005-07-22 06:17 sam send a response when an auth request is denied due to an acl; might be better to silently ignore the frame but this way we give stations a chance of figuring out what's wrong 2005-07-22 06:15 sam remove excess whitespace 2005-07-22 05:55 sam use IF_HANDOFF when bridging frames internally so if_start gets called; fixes communication between associated sta's MFC after: 3 days 2005-07-11 04:06 sam Handle encrypt of arbitarily fragmented mbuf chains: previously we bailed if we couldn't collect the 16-bytes of data required for an aes block cipher in 2 mbufs; now we deal with it. While here make space accounting signed so a sanity check does the right thing for malformed mbuf chains. Approved by: re (scottl) 2005-07-11 04:00 sam nuke assert that duplicates real check Reviewed by: avatar Approved by: re (scottl)
2005-11-18 19:40:08 +03:00
void ieee80211_add_scan(struct ieee80211com *,
const struct ieee80211_scanparams *,
const struct ieee80211_frame *,
int subtype, int rssi, int rstamp);
void ieee80211_init_neighbor(struct ieee80211com *, struct ieee80211_node *,
const struct ieee80211_frame *,
const struct ieee80211_scanparams *, int);
Resolve conflicts and adapt to NetBSD. Thanks to dyoung@, scw@, and perry@ for help testing. 2005-08-30 15:27 avatar Properly set ic_curchan before calling back to device driver to do channel switching(ifconfig devX channel Y). This fix should make channel changing works again in monitor mode. Submitted by: sam X-MFC-With: other ic_curchan changes 2005-08-13 18:50 sam revert 1.64: we cannot use the channel characteristics to decide when to do 11g erp sta accounting because b/g channels show up as false positives when operating in 11b. Noticed by: Michal Mertl 2005-08-13 18:31 sam Extend acl support to pass ioctl requests through and use this to add support for getting the current policy setting and collecting the list of mac addresses in the acl table. Submitted by: Michal Mertl (original version) MFC after: 2 weeks 2005-08-10 18:42 sam Don't use ic_curmode to decide when to do 11g station accounting, use the station channel properties. Fixes assert failure/bogus operation when an ap is operating in 11a and has associated stations then switches to 11g. Noticed by: Michal Mertl Reviewed by: avatar MFC after: 2 weeks 2005-08-10 17:22 sam Clarify/fix handling of the current channel: o add ic_curchan and use it uniformly for specifying the current channel instead of overloading ic->ic_bss->ni_chan (or in some drivers ic_ibss_chan) o add ieee80211_scanparams structure to encapsulate scanning-related state captured for rx frames o move rx beacon+probe response frame handling into separate routines o change beacon+probe response handling to treat the scan table more like a scan cache--look for an existing entry before adding a new one; this combined with ic_curchan use corrects handling of stations that were previously found at a different channel o move adhoc neighbor discovery by beacon+probe response frames to a new ieee80211_add_neighbor routine Reviewed by: avatar Tested by: avatar, Michal Mertl MFC after: 2 weeks 2005-08-09 11:19 rwatson Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days 2005-08-08 19:46 sam Split crypto tx+rx key indices and add a key index -> node mapping table: Crypto changes: o change driver/net80211 key_alloc api to return tx+rx key indices; a driver can leave the rx key index set to IEEE80211_KEYIX_NONE or set it to be the same as the tx key index (the former disables use of the key index in building the keyix->node mapping table and is the default setup for naive drivers by null_key_alloc) o add cs_max_keyid to crypto state to specify the max h/w key index a driver will return; this is used to allocate the key index mapping table and to bounds check table loookups o while here introduce ieee80211_keyix (finally) for the type of a h/w key index o change crypto notifiers for rx failures to pass the rx key index up as appropriate (michael failure, replay, etc.) Node table changes: o optionally allocate a h/w key index to node mapping table for the station table using the max key index setting supplied by drivers (note the scan table does not get a map) o defer node table allocation to lateattach so the driver has a chance to set the max key id to size the key index map o while here also defer the aid bitmap allocation o add new ieee80211_find_rxnode_withkey api to find a sta/node entry on frame receive with an optional h/w key index to use in checking mapping table; also updates the map if it does a hash lookup and the found node has a rx key index set in the unicast key; note this work is separated from the old ieee80211_find_rxnode call so drivers do not need to be aware of the new mechanism o move some node table manipulation under the node table lock to close a race on node delete o add ieee80211_node_delucastkey to do the dirty work of deleting unicast key state for a node (deletes any key and handles key map references) Ath driver: o nuke private sc_keyixmap mechansim in favor of net80211 support o update key alloc api These changes close several race conditions for the ath driver operating in ap mode. Other drivers should see no change. Station mode operation for ath no longer uses the key index map but performance tests show no noticeable change and this will be fixed when the scan table is eliminated with the new scanning support. Tested by: Michal Mertl, avatar, others Reviewed by: avatar, others MFC after: 2 weeks 2005-08-08 06:49 sam use ieee80211_iterate_nodes to retrieve station data; the previous code walked the list w/o locking MFC after: 1 week 2005-08-08 04:30 sam Cleanup beacon/listen interval handling: o separate configured beacon interval from listen interval; this avoids potential use of one value for the other (e.g. setting powersavesleep to 0 clobbers the beacon interval used in hostap or ibss mode) o bounds check the beacon interval received in probe response and beacon frames and drop frames with bogus settings; not clear if we should instead clamp the value as any alteration would result in mismatched sta+ap configuration and probably be more confusing (don't want to log to the console but perhaps ok with rate limiting) o while here up max beacon interval to reflect WiFi standard Noticed by: Martin <nakal@nurfuerspam.de> MFC after: 1 week 2005-08-06 05:57 sam fix debug msg typo MFC after: 3 days 2005-08-06 05:56 sam Fix handling of frames sent prior to a station being authorized when operating in ap mode. Previously we allocated a node from the station table, sent the frame (using the node), then released the reference that "held the frame in the table". But while the frame was in flight the node might be reclaimed which could lead to problems. The solution is to add an ieee80211_tmp_node routine that crafts a node that does exist in a table and so isn't ever reclaimed; it exists only so long as the associated frame is in flight. MFC after: 5 days 2005-07-31 07:12 sam close a race between reclaiming a node when a station is inactive and sending the null data frame used to probe inactive stations MFC after: 5 days 2005-07-27 05:41 sam when bridging internally bypass the bss node as traffic to it must follow the normal input path Submitted by: Michal Mertl MFC after: 5 days 2005-07-27 03:53 sam bandaid ni_fails handling so ap's with association failures are reconsidered after a bit; a proper fix involves more changes to the scanning infrastructure Reviewed by: avatar, David Young MFC after: 5 days 2005-07-23 01:16 sam the AREF flag is only meaningful in ap mode; adhoc neighbors now are timed out of the sta/neighbor table 2005-07-23 00:25 sam o move inactivity-related debug msgs under IEEE80211_MSG_INACT o probe inactive neighbors in adhoc mode (they don't have an association id so previously were being timed out) MFC after: 3 days 2005-07-22 22:11 sam split xmit of probe request frame out into a separate routine that takes explicit parameters; this will be needed when scanning is decoupled from the state machine to do bg scanning MFC after: 3 days 2005-07-22 21:48 sam split 802.11 frame xmit setup code into ieee80211_send_setup MFC after: 3 days 2005-07-22 18:57 sam simplify ic_newassoc callback MFC after: 3 days 2005-07-22 18:54 sam simplify ieee80211_ibss_merge api MFC after: 3 days 2005-07-22 18:50 sam add stats we know we'll need soon and some spare fields for future expansion MFC after: 3 days 2005-07-22 18:45 sam simplify tim callback api MFC after: 3 days 2005-07-22 18:42 sam don't include 802.3 header in min frame length calculation as it may not be present for a frag; fixes problem with small (fragmented) frames being dropped Obtained from: Atheros MFC after: 3 days 2005-07-22 18:36 sam simplify ieee80211_node_authorize and ieee80211_node_unauthorize api's MFC after: 3 days 2005-07-22 18:31 sam simplifiy ieee80211_send_nulldata api MFC after: 3 days 2005-07-22 18:29 sam simplify rate set api's by removing ic parameter (implicit in node reference) MFC after: 3 days 2005-07-22 18:21 sam reject association requests with a wpa/rsn ie when wpa/rsn is not configured on the ap; previously we either ignored the ie or (possibly) failed an assertion Obtained from: Atheros MFC after: 3 days 2005-07-22 18:16 sam missed one in last commit; add device name to discard msgs 2005-07-22 18:13 sam include device name in discard msgs 2005-07-22 18:12 sam add diag msgs for frames discarded because the direction field is wrong 2005-07-22 18:08 sam split data frame delivery out to a new function ieee80211_deliver_data 2005-07-22 18:00 sam o add IEEE80211_IOC_FRAGTHRESHOLD for getting+setting the tx fragmentation threshold o fix bounds checking on IEEE80211_IOC_RTSTHRESHOLD MFC after: 3 days 2005-07-22 17:55 sam o add IEEE80211_FRAG_DEFAULT o move default settings for RTS and frag thresholds to ieee80211_var.h 2005-07-22 17:50 sam diff reduction against p4: define IEEE80211_FIXED_RATE_NONE and use it instead of -1 2005-07-22 17:37 sam add flags missed in last merge 2005-07-22 17:36 sam Diff reduction against p4: o add ic_flags_ext for eventual extention of ic_flags o define/reserve flag+capabilities bits for superg, bg scan, and roaming support o refactor debug msg macros MFC after: 3 days 2005-07-22 06:17 sam send a response when an auth request is denied due to an acl; might be better to silently ignore the frame but this way we give stations a chance of figuring out what's wrong 2005-07-22 06:15 sam remove excess whitespace 2005-07-22 05:55 sam use IF_HANDOFF when bridging frames internally so if_start gets called; fixes communication between associated sta's MFC after: 3 days 2005-07-11 04:06 sam Handle encrypt of arbitarily fragmented mbuf chains: previously we bailed if we couldn't collect the 16-bytes of data required for an aes block cipher in 2 mbufs; now we deal with it. While here make space accounting signed so a sanity check does the right thing for malformed mbuf chains. Approved by: re (scottl) 2005-07-11 04:00 sam nuke assert that duplicates real check Reviewed by: avatar Approved by: re (scottl)
2005-11-18 19:40:08 +03:00
struct ieee80211_node *ieee80211_add_neighbor(struct ieee80211com *,
const struct ieee80211_frame *,
const struct ieee80211_scanparams *);
#endif /* _KERNEL */
#endif /* !_NET80211_IEEE80211_NODE_H_ */