NetBSD/sys/net80211/ieee80211_netbsd.c

714 lines
20 KiB
C
Raw Normal View History

/* $NetBSD: ieee80211_netbsd.c,v 1.26 2014/04/07 00:07:40 pooka Exp $ */
/*-
* Copyright (c) 2003-2005 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.
*
* 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.
*/
#include <sys/cdefs.h>
#ifdef __FreeBSD__
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
__FBSDID("$FreeBSD: src/sys/net80211/ieee80211_freebsd.c,v 1.8 2005/08/08 18:46:35 sam Exp $");
#else
__KERNEL_RCSID(0, "$NetBSD: ieee80211_netbsd.c,v 1.26 2014/04/07 00:07:40 pooka Exp $");
#endif
/*
* IEEE 802.11 support (NetBSD-specific code)
*/
#include <sys/param.h>
#include <sys/kernel.h>
#include <sys/systm.h>
#include <sys/mbuf.h>
#include <sys/proc.h>
#include <sys/sysctl.h>
#include <sys/once.h>
#include <sys/socket.h>
First step of random number subsystem rework described in <20111022023242.BA26F14A158@mail.netbsd.org>. This change includes the following: An initial cleanup and minor reorganization of the entropy pool code in sys/dev/rnd.c and sys/dev/rndpool.c. Several bugs are fixed. Some effort is made to accumulate entropy more quickly at boot time. A generic interface, "rndsink", is added, for stream generators to request that they be re-keyed with good quality entropy from the pool as soon as it is available. The arc4random()/arc4randbytes() implementation in libkern is adjusted to use the rndsink interface for rekeying, which helps address the problem of low-quality keys at boot time. An implementation of the FIPS 140-2 statistical tests for random number generator quality is provided (libkern/rngtest.c). This is based on Greg Rose's implementation from Qualcomm. A new random stream generator, nist_ctr_drbg, is provided. It is based on an implementation of the NIST SP800-90 CTR_DRBG by Henric Jungheim. This generator users AES in a modified counter mode to generate a backtracking-resistant random stream. An abstraction layer, "cprng", is provided for in-kernel consumers of randomness. The arc4random/arc4randbytes API is deprecated for in-kernel use. It is replaced by "cprng_strong". The current cprng_fast implementation wraps the existing arc4random implementation. The current cprng_strong implementation wraps the new CTR_DRBG implementation. Both interfaces are rekeyed from the entropy pool automatically at intervals justifiable from best current cryptographic practice. In some quick tests, cprng_fast() is about the same speed as the old arc4randbytes(), and cprng_strong() is about 20% faster than rnd_extract_data(). Performance is expected to improve. The AES code in src/crypto/rijndael is no longer an optional kernel component, as it is required by cprng_strong, which is not an optional kernel component. The entropy pool output is subjected to the rngtest tests at startup time; if it fails, the system will reboot. There is approximately a 3/10000 chance of a false positive from these tests. Entropy pool _input_ from hardware random numbers is subjected to the rngtest tests at attach time, as well as the FIPS continuous-output test, to detect bad or stuck hardware RNGs; if any are detected, they are detached, but the system continues to run. A problem with rndctl(8) is fixed -- datastructures with pointers in arrays are no longer passed to userspace (this was not a security problem, but rather a major issue for compat32). A new kernel will require a new rndctl. The sysctl kern.arandom() and kern.urandom() nodes are hooked up to the new generators, but the /dev/*random pseudodevices are not, yet. Manual pages for the new kernel interfaces are forthcoming.
2011-11-20 02:51:18 +04:00
#include <sys/cprng.h>
#include <net/if.h>
#include <net/if_media.h>
#include <net/if_ether.h>
#include <net/route.h>
#include <net80211/ieee80211_netbsd.h>
#include <net80211/ieee80211_var.h>
#include <net80211/ieee80211_sysctl.h>
#define LOGICALLY_EQUAL(x, y) (!(x) == !(y))
static void ieee80211_sysctl_fill_node(struct ieee80211_node *,
struct ieee80211_node_sysctl *, int, const struct ieee80211_channel *,
uint32_t);
static struct ieee80211_node *ieee80211_node_walknext(
struct ieee80211_node_walk *);
static struct ieee80211_node *ieee80211_node_walkfirst(
struct ieee80211_node_walk *, u_short);
static int ieee80211_sysctl_node(SYSCTLFN_ARGS);
static void ieee80211_sysctl_setup(void);
#ifdef IEEE80211_DEBUG
int ieee80211_debug = 0;
#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
typedef void (*ieee80211_setup_func)(void);
__link_set_decl(ieee80211_funcs, ieee80211_setup_func);
static int
ieee80211_init0(void)
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
{
ieee80211_setup_func * const *ieee80211_setup, f;
ieee80211_sysctl_setup();
if (max_linkhdr < ALIGN(sizeof(struct ieee80211_qosframe_addr4))) {
max_linkhdr = ALIGN(sizeof(struct ieee80211_qosframe_addr4));
}
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
__link_set_foreach(ieee80211_setup, ieee80211_funcs) {
f = (void*)*ieee80211_setup;
(*f)();
}
return 0;
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_init(void)
{
static ONCE_DECL(ieee80211_init_once);
RUN_ONCE(&ieee80211_init_once, ieee80211_init0);
}
static int
ieee80211_sysctl_inact(SYSCTLFN_ARGS)
{
int error, t;
struct sysctlnode node;
node = *rnode;
/* sysctl_lookup copies the product from t. Then, it
* copies the new value onto t.
*/
t = *(int*)rnode->sysctl_data * IEEE80211_INACT_WAIT;
node.sysctl_data = &t;
error = sysctl_lookup(SYSCTLFN_CALL(&node));
if (error || newp == NULL)
return (error);
/* The new value was in seconds. Convert to inactivity-wait
* intervals. There are IEEE80211_INACT_WAIT seconds per
* interval.
*/
*(int*)rnode->sysctl_data = t / IEEE80211_INACT_WAIT;
return (0);
}
static int
ieee80211_sysctl_parent(SYSCTLFN_ARGS)
{
struct ieee80211com *ic;
char pname[IFNAMSIZ];
struct sysctlnode node;
node = *rnode;
ic = node.sysctl_data;
strncpy(pname, ic->ic_ifp->if_xname, IFNAMSIZ);
node.sysctl_data = pname;
return sysctl_lookup(SYSCTLFN_CALL(&node));
}
/*
* Create or get top of sysctl tree net.link.ieee80211.
*/
static const struct sysctlnode *
ieee80211_sysctl_treetop(struct sysctllog **log)
{
int rc;
const struct sysctlnode *rnode;
if ((rc = sysctl_createv(log, 0, NULL, &rnode,
CTLFLAG_PERMANENT, CTLTYPE_NODE, "link",
"link-layer statistics and controls",
NULL, 0, NULL, 0, CTL_NET, PF_LINK, CTL_EOL)) != 0)
goto err;
if ((rc = sysctl_createv(log, 0, &rnode, &rnode,
CTLFLAG_PERMANENT, CTLTYPE_NODE, "ieee80211",
"IEEE 802.11 WLAN statistics and controls",
NULL, 0, NULL, 0, CTL_CREATE, CTL_EOL)) != 0)
goto err;
return rnode;
err:
printf("%s: sysctl_createv failed, rc = %d\n", __func__, rc);
return NULL;
}
void
ieee80211_sysctl_attach(struct ieee80211com *ic)
{
int rc;
const struct sysctlnode *cnode, *rnode;
char num[sizeof("vap") + 14]; /* sufficient for 32 bits */
if ((rnode = ieee80211_sysctl_treetop(NULL)) == NULL)
return;
snprintf(num, sizeof(num), "vap%u", ic->ic_vap);
if ((rc = sysctl_createv(&ic->ic_sysctllog, 0, &rnode, &rnode,
CTLFLAG_PERMANENT, CTLTYPE_NODE, num, SYSCTL_DESCR("virtual AP"),
NULL, 0, NULL, 0, CTL_CREATE, CTL_EOL)) != 0)
goto err;
/* control debugging printfs */
if ((rc = sysctl_createv(&ic->ic_sysctllog, 0, &rnode, &cnode,
CTLFLAG_PERMANENT|CTLFLAG_READONLY, CTLTYPE_STRING,
"parent", SYSCTL_DESCR("parent device"),
ieee80211_sysctl_parent, 0, (void *)ic, IFNAMSIZ, CTL_CREATE,
CTL_EOL)) != 0)
goto err;
#ifdef IEEE80211_DEBUG
/* control debugging printfs */
if ((rc = sysctl_createv(&ic->ic_sysctllog, 0, &rnode, &cnode,
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, CTLTYPE_INT,
"debug", SYSCTL_DESCR("control debugging printfs"),
NULL, ieee80211_debug, &ic->ic_debug, 0,
CTL_CREATE, CTL_EOL)) != 0)
goto err;
#endif
/* XXX inherit from tunables */
if ((rc = sysctl_createv(&ic->ic_sysctllog, 0, &rnode, &cnode,
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, CTLTYPE_INT,
"inact_run", SYSCTL_DESCR("station inactivity timeout (sec)"),
ieee80211_sysctl_inact, 0, &ic->ic_inact_run, 0,
CTL_CREATE, CTL_EOL)) != 0)
goto err;
if ((rc = sysctl_createv(&ic->ic_sysctllog, 0, &rnode, &cnode,
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, CTLTYPE_INT,
"inact_probe",
SYSCTL_DESCR("station inactivity probe timeout (sec)"),
ieee80211_sysctl_inact, 0, &ic->ic_inact_probe, 0,
CTL_CREATE, CTL_EOL)) != 0)
goto err;
if ((rc = sysctl_createv(&ic->ic_sysctllog, 0, &rnode, &cnode,
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, CTLTYPE_INT,
"inact_auth",
SYSCTL_DESCR("station authentication timeout (sec)"),
ieee80211_sysctl_inact, 0, &ic->ic_inact_auth, 0,
CTL_CREATE, CTL_EOL)) != 0)
goto err;
if ((rc = sysctl_createv(&ic->ic_sysctllog, 0, &rnode, &cnode,
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, CTLTYPE_INT,
"inact_init",
SYSCTL_DESCR("station initial state timeout (sec)"),
ieee80211_sysctl_inact, 0, &ic->ic_inact_init, 0,
CTL_CREATE, CTL_EOL)) != 0)
goto err;
if ((rc = sysctl_createv(&ic->ic_sysctllog, 0, &rnode, &cnode,
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, CTLTYPE_INT,
"driver_caps", SYSCTL_DESCR("driver capabilities"),
NULL, 0, &ic->ic_caps, 0, CTL_CREATE, CTL_EOL)) != 0)
goto err;
if ((rc = sysctl_createv(&ic->ic_sysctllog, 0, &rnode, &cnode,
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, CTLTYPE_INT,
"bmiss_max", SYSCTL_DESCR("consecutive beacon misses before scanning"),
NULL, 0, &ic->ic_bmiss_max, 0, CTL_CREATE, CTL_EOL)) != 0)
goto err;
return;
err:
printf("%s: sysctl_createv failed, rc = %d\n", __func__, rc);
}
void
ieee80211_sysctl_detach(struct ieee80211com *ic)
{
sysctl_teardown(&ic->ic_sysctllog);
}
/*
* Pointers for testing:
*
* If there are no interfaces, or else no 802.11 interfaces,
* ieee80211_node_walkfirst must return NULL.
*
* If there is any single 802.11 interface, ieee80211_node_walkfirst
* must not return NULL.
*/
static struct ieee80211_node *
ieee80211_node_walkfirst(struct ieee80211_node_walk *nw, u_short if_index)
{
(void)memset(nw, 0, sizeof(*nw));
nw->nw_ifindex = if_index;
LIST_FOREACH(nw->nw_ic, &ieee80211com_head, ic_list) {
if (if_index != 0 && nw->nw_ic->ic_ifp->if_index != if_index)
continue;
if (!TAILQ_EMPTY(&nw->nw_ic->ic_sta.nt_node))
nw->nw_nt = &nw->nw_ic->ic_sta;
else if (!TAILQ_EMPTY(&nw->nw_ic->ic_scan.nt_node))
nw->nw_nt = &nw->nw_ic->ic_scan;
else if (nw->nw_ic->ic_bss == NULL)
continue;
break;
}
if (nw->nw_ic == NULL)
return NULL;
if (nw->nw_nt == NULL)
nw->nw_ni = nw->nw_ic->ic_bss;
else
nw->nw_ni = TAILQ_FIRST(&nw->nw_nt->nt_node);
return nw->nw_ni;
}
static struct ieee80211_node *
ieee80211_node_walknext(struct ieee80211_node_walk *nw)
{
if (nw->nw_nt != NULL)
nw->nw_ni = TAILQ_NEXT(nw->nw_ni, ni_list);
else
nw->nw_ni = NULL;
while (nw->nw_ni == NULL) {
if (nw->nw_nt == &nw->nw_ic->ic_sta) {
nw->nw_nt = &nw->nw_ic->ic_scan;
nw->nw_ni = TAILQ_FIRST(&nw->nw_nt->nt_node);
continue;
} else if (nw->nw_nt == &nw->nw_ic->ic_scan) {
nw->nw_nt = NULL;
nw->nw_ni = nw->nw_ic->ic_bss;
continue;
}
KASSERT(nw->nw_nt == NULL);
if (nw->nw_ifindex != 0)
return NULL;
nw->nw_ic = LIST_NEXT(nw->nw_ic, ic_list);
if (nw->nw_ic == NULL)
return NULL;
nw->nw_nt = &nw->nw_ic->ic_sta;
nw->nw_ni = TAILQ_FIRST(&nw->nw_nt->nt_node);
}
return nw->nw_ni;
}
static void
ieee80211_sysctl_fill_node(struct ieee80211_node *ni,
struct ieee80211_node_sysctl *ns, int ifindex,
const struct ieee80211_channel *chan0, uint32_t flags)
{
ns->ns_ifindex = ifindex;
ns->ns_capinfo = ni->ni_capinfo;
ns->ns_flags = flags;
(void)memcpy(ns->ns_macaddr, ni->ni_macaddr, sizeof(ns->ns_macaddr));
(void)memcpy(ns->ns_bssid, ni->ni_bssid, sizeof(ns->ns_bssid));
if (ni->ni_chan != IEEE80211_CHAN_ANYC) {
ns->ns_freq = ni->ni_chan->ic_freq;
ns->ns_chanflags = ni->ni_chan->ic_flags;
ns->ns_chanidx = ni->ni_chan - chan0;
} else {
ns->ns_freq = ns->ns_chanflags = 0;
ns->ns_chanidx = 0;
}
ns->ns_rssi = ni->ni_rssi;
ns->ns_esslen = ni->ni_esslen;
(void)memcpy(ns->ns_essid, ni->ni_essid, sizeof(ns->ns_essid));
ns->ns_erp = ni->ni_erp;
ns->ns_associd = ni->ni_associd;
ns->ns_inact = ni->ni_inact * IEEE80211_INACT_WAIT;
ns->ns_rstamp = ni->ni_rstamp;
ns->ns_rates = ni->ni_rates;
ns->ns_txrate = ni->ni_txrate;
ns->ns_intval = ni->ni_intval;
(void)memcpy(ns->ns_tstamp, &ni->ni_tstamp, sizeof(ns->ns_tstamp));
ns->ns_txseq = ni->ni_txseqs[0];
ns->ns_rxseq = ni->ni_rxseqs[0];
ns->ns_fhdwell = ni->ni_fhdwell;
ns->ns_fhindex = ni->ni_fhindex;
ns->ns_fails = ni->ni_fails;
}
/* Between two examinations of the sysctl tree, I expect each
* interface to add no more than 5 nodes.
*/
#define IEEE80211_SYSCTL_NODE_GROWTH 5
static int
ieee80211_sysctl_node(SYSCTLFN_ARGS)
{
struct ieee80211_node_walk nw;
struct ieee80211_node *ni;
struct ieee80211_node_sysctl ns;
char *dp;
u_int cur_ifindex, ifcount, ifindex, last_ifindex, op, arg, hdr_type;
uint32_t flags;
size_t len, needed, eltsize, out_size;
int error, s, saw_bss = 0, nelt;
if (namelen == 1 && name[0] == CTL_QUERY)
return (sysctl_query(SYSCTLFN_CALL(rnode)));
if (namelen != IEEE80211_SYSCTL_NODENAMELEN)
return (EINVAL);
/* ifindex.op.arg.header-type.eltsize.nelt */
dp = oldp;
len = (oldp != NULL) ? *oldlenp : 0;
ifindex = name[IEEE80211_SYSCTL_NODENAME_IF];
op = name[IEEE80211_SYSCTL_NODENAME_OP];
arg = name[IEEE80211_SYSCTL_NODENAME_ARG];
hdr_type = name[IEEE80211_SYSCTL_NODENAME_TYPE];
eltsize = name[IEEE80211_SYSCTL_NODENAME_ELTSIZE];
nelt = name[IEEE80211_SYSCTL_NODENAME_ELTCOUNT];
out_size = MIN(sizeof(ns), eltsize);
if (op != IEEE80211_SYSCTL_OP_ALL || arg != 0 ||
hdr_type != IEEE80211_SYSCTL_T_NODE || eltsize < 1 || nelt < 0)
return (EINVAL);
error = 0;
needed = 0;
ifcount = 0;
last_ifindex = 0;
s = splnet();
for (ni = ieee80211_node_walkfirst(&nw, ifindex); ni != NULL;
ni = ieee80211_node_walknext(&nw)) {
struct ieee80211com *ic;
ic = nw.nw_ic;
cur_ifindex = ic->ic_ifp->if_index;
if (cur_ifindex != last_ifindex) {
saw_bss = 0;
ifcount++;
last_ifindex = cur_ifindex;
}
if (nelt <= 0)
continue;
if (saw_bss && ni == ic->ic_bss)
continue;
else if (ni == ic->ic_bss) {
saw_bss = 1;
flags = IEEE80211_NODE_SYSCTL_F_BSS;
} else
flags = 0;
if (ni->ni_table == &ic->ic_scan)
flags |= IEEE80211_NODE_SYSCTL_F_SCAN;
else if (ni->ni_table == &ic->ic_sta)
flags |= IEEE80211_NODE_SYSCTL_F_STA;
if (len >= eltsize) {
ieee80211_sysctl_fill_node(ni, &ns, cur_ifindex,
&ic->ic_channels[0], flags);
error = copyout(&ns, dp, out_size);
if (error)
goto cleanup;
dp += eltsize;
len -= eltsize;
}
needed += eltsize;
if (nelt != INT_MAX)
nelt--;
}
cleanup:
splx(s);
*oldlenp = needed;
if (oldp == NULL)
*oldlenp += ifcount * IEEE80211_SYSCTL_NODE_GROWTH * eltsize;
return (error);
}
/*
* Setup sysctl(3) MIB, net.ieee80211.*
*
* TBD condition CTLFLAG_PERMANENT on being a module or not
*/
static struct sysctllog *ieee80211_sysctllog;
static void
ieee80211_sysctl_setup(void)
{
int rc;
const struct sysctlnode *cnode, *rnode;
if ((rnode = ieee80211_sysctl_treetop(&ieee80211_sysctllog)) == NULL)
return;
if ((rc = sysctl_createv(&ieee80211_sysctllog, 0, &rnode, NULL,
CTLFLAG_PERMANENT, CTLTYPE_NODE, "nodes", "client/peer stations",
ieee80211_sysctl_node, 0, NULL, 0, CTL_CREATE, CTL_EOL)) != 0)
goto err;
#ifdef IEEE80211_DEBUG
/* control debugging printfs */
if ((rc = sysctl_createv(&ieee80211_sysctllog, 0, &rnode, &cnode,
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, CTLTYPE_INT,
"debug", SYSCTL_DESCR("control debugging printfs"),
NULL, 0, &ieee80211_debug, 0, CTL_CREATE, CTL_EOL)) != 0)
goto err;
#endif /* IEEE80211_DEBUG */
ieee80211_rssadapt_sysctl_setup(&ieee80211_sysctllog);
return;
err:
printf("%s: sysctl_createv failed (rc = %d)\n", __func__, rc);
}
int
ieee80211_node_dectestref(struct ieee80211_node *ni)
{
if (atomic_dec_uint_nv(&ni->ni_refcnt) == 0) {
atomic_inc_uint(&ni->ni_refcnt);
return 1;
} else
return 0;
}
void
ieee80211_drain_ifq(struct ifqueue *ifq)
{
struct ieee80211_node *ni;
struct mbuf *m;
for (;;) {
IF_DEQUEUE(ifq, m);
if (m == NULL)
break;
ni = (struct ieee80211_node *)m->m_pkthdr.rcvif;
KASSERT(ni != NULL);
ieee80211_free_node(ni);
m->m_pkthdr.rcvif = NULL;
m_freem(m);
}
}
void
if_printf(struct ifnet *ifp, const char *fmt, ...)
{
va_list ap;
va_start(ap, fmt);
printf("%s: ", ifp->if_xname);
vprintf(fmt, ap);
va_end(ap);
return;
}
/*
* Allocate and setup a management frame of the specified
* size. We return the mbuf and a pointer to the start
* of the contiguous data area that's been reserved based
* on the packet length. The data area is forced to 32-bit
* alignment and the buffer length to a multiple of 4 bytes.
* This is done mainly so beacon frames (that require this)
* can use this interface too.
*/
struct mbuf *
ieee80211_getmgtframe(u_int8_t **frm, u_int pktlen)
{
struct mbuf *m;
u_int len;
/*
* NB: we know the mbuf routines will align the data area
* so we don't need to do anything special.
*/
/* XXX 4-address frame? */
len = roundup(sizeof(struct ieee80211_frame) + pktlen, 4);
IASSERT(len <= MCLBYTES, ("802.11 mgt frame too large: %u", len));
if (len <= MHLEN) {
m = m_gethdr(M_NOWAIT, MT_HEADER);
/*
* Align the data in case additional headers are added.
* This should only happen when a WEP header is added
* which only happens for shared key authentication mgt
* frames which all fit in MHLEN.
*/
if (m != NULL)
MH_ALIGN(m, len);
} else
m = m_getcl(M_NOWAIT, MT_HEADER, M_PKTHDR);
if (m != NULL) {
m->m_data += sizeof(struct ieee80211_frame);
*frm = m->m_data;
IASSERT((uintptr_t)*frm % 4 == 0, ("bad beacon boundary"));
}
return m;
}
void
get_random_bytes(void *p, size_t n)
{
First step of random number subsystem rework described in <20111022023242.BA26F14A158@mail.netbsd.org>. This change includes the following: An initial cleanup and minor reorganization of the entropy pool code in sys/dev/rnd.c and sys/dev/rndpool.c. Several bugs are fixed. Some effort is made to accumulate entropy more quickly at boot time. A generic interface, "rndsink", is added, for stream generators to request that they be re-keyed with good quality entropy from the pool as soon as it is available. The arc4random()/arc4randbytes() implementation in libkern is adjusted to use the rndsink interface for rekeying, which helps address the problem of low-quality keys at boot time. An implementation of the FIPS 140-2 statistical tests for random number generator quality is provided (libkern/rngtest.c). This is based on Greg Rose's implementation from Qualcomm. A new random stream generator, nist_ctr_drbg, is provided. It is based on an implementation of the NIST SP800-90 CTR_DRBG by Henric Jungheim. This generator users AES in a modified counter mode to generate a backtracking-resistant random stream. An abstraction layer, "cprng", is provided for in-kernel consumers of randomness. The arc4random/arc4randbytes API is deprecated for in-kernel use. It is replaced by "cprng_strong". The current cprng_fast implementation wraps the existing arc4random implementation. The current cprng_strong implementation wraps the new CTR_DRBG implementation. Both interfaces are rekeyed from the entropy pool automatically at intervals justifiable from best current cryptographic practice. In some quick tests, cprng_fast() is about the same speed as the old arc4randbytes(), and cprng_strong() is about 20% faster than rnd_extract_data(). Performance is expected to improve. The AES code in src/crypto/rijndael is no longer an optional kernel component, as it is required by cprng_strong, which is not an optional kernel component. The entropy pool output is subjected to the rngtest tests at startup time; if it fails, the system will reboot. There is approximately a 3/10000 chance of a false positive from these tests. Entropy pool _input_ from hardware random numbers is subjected to the rngtest tests at attach time, as well as the FIPS continuous-output test, to detect bad or stuck hardware RNGs; if any are detected, they are detached, but the system continues to run. A problem with rndctl(8) is fixed -- datastructures with pointers in arrays are no longer passed to userspace (this was not a security problem, but rather a major issue for compat32). A new kernel will require a new rndctl. The sysctl kern.arandom() and kern.urandom() nodes are hooked up to the new generators, but the /dev/*random pseudodevices are not, yet. Manual pages for the new kernel interfaces are forthcoming.
2011-11-20 02:51:18 +04:00
cprng_fast(p, n);
}
void
ieee80211_notify_node_join(struct ieee80211com *ic, struct ieee80211_node *ni, int newassoc)
{
struct ifnet *ifp = ic->ic_ifp;
struct ieee80211_join_event iev;
2013-02-04 19:44:45 +04:00
IEEE80211_DPRINTF(ic, IEEE80211_MSG_NODE, "%snode %s join\n",
(ni == ic->ic_bss) ? "bss " : "",
ether_sprintf(ni->ni_macaddr));
memset(&iev, 0, sizeof(iev));
if (ni == ic->ic_bss) {
IEEE80211_ADDR_COPY(iev.iev_addr, ni->ni_bssid);
rt_ieee80211msg(ifp, newassoc ?
RTM_IEEE80211_ASSOC : RTM_IEEE80211_REASSOC,
&iev, sizeof(iev));
if_link_state_change(ifp, LINK_STATE_UP);
} else {
IEEE80211_ADDR_COPY(iev.iev_addr, ni->ni_macaddr);
rt_ieee80211msg(ifp, newassoc ?
RTM_IEEE80211_JOIN : RTM_IEEE80211_REJOIN,
&iev, sizeof(iev));
}
}
void
ieee80211_notify_node_leave(struct ieee80211com *ic, struct ieee80211_node *ni)
{
struct ifnet *ifp = ic->ic_ifp;
struct ieee80211_leave_event iev;
2013-02-04 19:44:45 +04:00
IEEE80211_DPRINTF(ic, IEEE80211_MSG_NODE, "%snode %s leave\n",
(ni == ic->ic_bss) ? "bss " : "",
ether_sprintf(ni->ni_macaddr));
if (ni == ic->ic_bss) {
rt_ieee80211msg(ifp, RTM_IEEE80211_DISASSOC, NULL, 0);
if_link_state_change(ifp, LINK_STATE_DOWN);
} else {
/* fire off wireless event station leaving */
memset(&iev, 0, sizeof(iev));
IEEE80211_ADDR_COPY(iev.iev_addr, ni->ni_macaddr);
rt_ieee80211msg(ifp, RTM_IEEE80211_LEAVE, &iev, sizeof(iev));
}
}
void
ieee80211_notify_scan_done(struct ieee80211com *ic)
{
struct ifnet *ifp = ic->ic_ifp;
IEEE80211_DPRINTF(ic, IEEE80211_MSG_SCAN,
2013-02-04 19:44:45 +04:00
"%s", "notify scan done\n");
/* dispatch wireless event indicating scan completed */
rt_ieee80211msg(ifp, RTM_IEEE80211_SCAN, NULL, 0);
}
void
ieee80211_notify_replay_failure(struct ieee80211com *ic,
const struct ieee80211_frame *wh, const struct ieee80211_key *k,
u_int64_t rsc)
{
struct ifnet *ifp = ic->ic_ifp;
IEEE80211_DPRINTF(ic, IEEE80211_MSG_CRYPTO,
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
"[%s] %s replay detected <rsc %ju, csc %ju, keyix %u rxkeyix %u>\n",
ether_sprintf(wh->i_addr2), k->wk_cipher->ic_name,
(intmax_t) rsc, (intmax_t) k->wk_keyrsc,
k->wk_keyix, k->wk_rxkeyix);
if (ifp != NULL) { /* NB: for cipher test modules */
struct ieee80211_replay_event iev;
IEEE80211_ADDR_COPY(iev.iev_dst, wh->i_addr1);
IEEE80211_ADDR_COPY(iev.iev_src, wh->i_addr2);
iev.iev_cipher = k->wk_cipher->ic_cipher;
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
if (k->wk_rxkeyix != IEEE80211_KEYIX_NONE)
iev.iev_keyix = k->wk_rxkeyix;
else
iev.iev_keyix = k->wk_keyix;
iev.iev_keyrsc = k->wk_keyrsc;
iev.iev_rsc = rsc;
rt_ieee80211msg(ifp, RTM_IEEE80211_REPLAY, &iev, sizeof(iev));
}
}
void
ieee80211_notify_michael_failure(struct ieee80211com *ic,
const struct ieee80211_frame *wh, u_int keyix)
{
struct ifnet *ifp = ic->ic_ifp;
IEEE80211_DPRINTF(ic, IEEE80211_MSG_CRYPTO,
"[%s] michael MIC verification failed <keyix %u>\n",
ether_sprintf(wh->i_addr2), keyix);
ic->ic_stats.is_rx_tkipmic++;
if (ifp != NULL) { /* NB: for cipher test modules */
struct ieee80211_michael_event iev;
IEEE80211_ADDR_COPY(iev.iev_dst, wh->i_addr1);
IEEE80211_ADDR_COPY(iev.iev_src, wh->i_addr2);
iev.iev_cipher = IEEE80211_CIPHER_TKIP;
iev.iev_keyix = keyix;
rt_ieee80211msg(ifp, RTM_IEEE80211_MICHAEL, &iev, sizeof(iev));
}
}
void
ieee80211_load_module(const char *modname)
{
#ifdef notyet
struct thread *td = curthread;
if (suser(td) == 0 && securelevel_gt(td->td_ucred, 0) == 0) {
mtx_lock(&Giant);
(void) linker_load_module(modname, NULL, NULL, NULL, NULL);
mtx_unlock(&Giant);
}
#else
printf("%s: load the %s module by hand for now.\n", __func__, modname);
#endif
}