freebsd-dev/sys/net80211/ieee80211_wds.c

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

806 lines
24 KiB
C
Raw Normal View History

/*-
* SPDX-License-Identifier: BSD-2-Clause-FreeBSD
*
* Copyright (c) 2007-2008 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.
*
* 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__
__FBSDID("$FreeBSD$");
#endif
/*
* IEEE 802.11 WDS mode support.
*/
#include "opt_inet.h"
#include "opt_wlan.h"
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/mbuf.h>
#include <sys/malloc.h>
#include <sys/kernel.h>
#include <sys/socket.h>
#include <sys/sockio.h>
#include <sys/endian.h>
#include <sys/errno.h>
#include <sys/proc.h>
#include <sys/sysctl.h>
#include <net/if.h>
#include <net/if_var.h>
#include <net/if_media.h>
#include <net/if_llc.h>
#include <net/if_private.h>
#include <net/ethernet.h>
#include <net/bpf.h>
#include <net80211/ieee80211_var.h>
#include <net80211/ieee80211_wds.h>
#include <net80211/ieee80211_input.h>
#ifdef IEEE80211_SUPPORT_SUPERG
#include <net80211/ieee80211_superg.h>
#endif
static void wds_vattach(struct ieee80211vap *);
static int wds_newstate(struct ieee80211vap *, enum ieee80211_state, int);
Begin plumbing ieee80211_rx_stats through the receive path. Smart NICs with firmware (eg wpi, iwn, the new atheros parts, the intel 7260 series, etc) support doing a lot of things in firmware. This includes but isn't limited to things like scanning, sending probe requests and receiving probe responses. However, net80211 doesn't know about any of this - it still drives the whole scan/probe infrastructure itself. In order to move towards suppoting smart NICs, the receive path needs to know about the channel/details for each received packet. In at least the iwn and 7260 firmware (and I believe wpi, but I haven't tried it yet) it will do the scanning, power-save and off-channel buffering for you - all you need to do is handle receiving beacons and probe responses on channels that aren't what you're currently on. However the whole receive path is peppered with ic->ic_curchan and manual scan/powersave handling. The beacon parsing code also checks ic->ic_curchan to determine if the received beacon is on the correct channel or not.[1] So: * add freq/ieee values to ieee80211_rx_stats; * change ieee80211_parse_beacon() to accept the 'current' channel as an argument; * modify the iv_input() and iv_recv_mgmt() methods to include the rx_stats; * add a new method - ieee80211_lookup_channel_rxstats() - that looks up a channel based on the contents of ieee80211_rx_stats; * if it exists, use it in the mgmt path to switch the current channel (which still defaults to ic->ic_curchan) over to something determined by rx_stats. This is enough to kick-start scan offload support in the Intel 7260 driver that Rui/I are working on. It also is a good start for scan offload support for a handful of existing NICs (wpi, iwn, some USB parts) and it'll very likely dramatically improve stability/performance there. It's not the whole thing - notably, we don't need to do powersave, we should not scan all channels, and we should leave probe request sending to the firmware and not do it ourselves. But, this allows for continued development on the above features whilst actually having a somewhat working NIC. TODO: * Finish tidying up how the net80211 input path works. Right now ieee80211_input / ieee80211_input_all act as the top-level that everything feeds into; it should change so the MIMO input routines are those and the legacy routines are phased out. * The band selection should be done by the driver, not by the net80211 layer. * ieee80211_lookup_channel_rxstats() only determines 11b or 11g channels for now - this is enough for scanning, but not 100% true in all cases. If we ever need to handle off-channel scan support for things like static-40MHz or static-80MHz, or turbo-G, or half/quarter rates, then we should extend this. [1] This is a side effect of frequency-hopping and CCK modes - you can receive beacons when you think you're on a different channel. In particular, CCK (which is used by the low 11b rates, eg beacons!) is decodable from adjacent channels - just at a low SNR. FH is a side effect of having the hardware/firmware do the frequency hopping - it may pick up beacons transmitted from other FH networks that are in a different phase of hopping frequencies.
2015-05-25 16:37:41 +00:00
static int wds_input(struct ieee80211_node *ni, struct mbuf *m,
const struct ieee80211_rx_stats *rxs, int, int);
static void wds_recv_mgmt(struct ieee80211_node *, struct mbuf *, int subtype,
const struct ieee80211_rx_stats *, int, int);
void
ieee80211_wds_attach(struct ieee80211com *ic)
{
ic->ic_vattach[IEEE80211_M_WDS] = wds_vattach;
}
void
ieee80211_wds_detach(struct ieee80211com *ic)
{
}
static void
wds_vdetach(struct ieee80211vap *vap)
{
if (vap->iv_bss != NULL) {
/* XXX locking? */
if (vap->iv_bss->ni_wdsvap == vap)
vap->iv_bss->ni_wdsvap = NULL;
}
}
static void
wds_vattach(struct ieee80211vap *vap)
{
vap->iv_newstate = wds_newstate;
vap->iv_input = wds_input;
vap->iv_recv_mgmt = wds_recv_mgmt;
vap->iv_opdetach = wds_vdetach;
}
static void
wds_flush(struct ieee80211_node *ni)
{
struct ieee80211com *ic = ni->ni_ic;
struct mbuf *m, *next;
int8_t rssi, nf;
m = ieee80211_ageq_remove(&ic->ic_stageq,
(void *)(uintptr_t) ieee80211_mac_hash(ic, ni->ni_macaddr));
if (m == NULL)
return;
IEEE80211_NOTE(ni->ni_vap, IEEE80211_MSG_WDS, ni,
"%s", "flush wds queue");
ic->ic_node_getsignal(ni, &rssi, &nf);
for (; m != NULL; m = next) {
next = m->m_nextpkt;
m->m_nextpkt = NULL;
ieee80211_input(ni, m, rssi, nf);
}
}
static int
ieee80211_create_wds(struct ieee80211vap *vap, struct ieee80211_channel *chan)
{
struct ieee80211com *ic = vap->iv_ic;
struct ieee80211_node_table *nt = &ic->ic_sta;
struct ieee80211_node *ni, *obss;
IEEE80211_DPRINTF(vap, IEEE80211_MSG_WDS,
"%s: creating link to %s on channel %u\n", __func__,
ether_sprintf(vap->iv_des_bssid), ieee80211_chan2ieee(ic, chan));
/* NB: vap create must specify the bssid for the link */
KASSERT(vap->iv_flags & IEEE80211_F_DESBSSID, ("no bssid"));
/* NB: we should only be called on RUN transition */
KASSERT(vap->iv_state == IEEE80211_S_RUN, ("!RUN state"));
if ((vap->iv_flags_ext & IEEE80211_FEXT_WDSLEGACY) == 0) {
/*
* Dynamic/non-legacy WDS. Reference the associated
* station specified by the desired bssid setup at vap
* create. Point ni_wdsvap at the WDS vap so 4-address
* frames received through the associated AP vap will
* be dispatched upward (e.g. to a bridge) as though
* they arrived on the WDS vap.
*/
IEEE80211_NODE_LOCK(nt);
obss = NULL;
ni = ieee80211_find_node_locked(&ic->ic_sta, vap->iv_des_bssid);
if (ni == NULL) {
/*
* Node went away before we could hookup. This
* should be ok; no traffic will flow and a leave
* event will be dispatched that should cause
* the vap to be destroyed.
*/
IEEE80211_DPRINTF(vap, IEEE80211_MSG_WDS,
"%s: station %s went away\n",
__func__, ether_sprintf(vap->iv_des_bssid));
/* XXX stat? */
} else if (ni->ni_wdsvap != NULL) {
/*
* Node already setup with a WDS vap; we cannot
* allow multiple references so disallow. If
* ni_wdsvap points at us that's ok; we should
* do nothing anyway.
*/
/* XXX printf instead? */
IEEE80211_DPRINTF(vap, IEEE80211_MSG_WDS,
"%s: station %s in use with %s\n",
__func__, ether_sprintf(vap->iv_des_bssid),
ni->ni_wdsvap->iv_ifp->if_xname);
/* XXX stat? */
} else {
/*
* Committed to new node, setup state.
*/
obss = vap->iv_update_bss(vap, ni);
ni->ni_wdsvap = vap;
}
IEEE80211_NODE_UNLOCK(nt);
if (obss != NULL) {
/* NB: deferred to avoid recursive lock */
ieee80211_free_node(obss);
}
} else {
/*
* Legacy WDS vap setup.
*/
/*
* The far end does not associate so we just create
* create a new node and install it as the vap's
* bss node. We must simulate an association and
* authorize the port for traffic to flow.
* XXX check if node already in sta table?
*/
ni = ieee80211_node_create_wds(vap, vap->iv_des_bssid, chan);
if (ni != NULL) {
obss = vap->iv_update_bss(vap, ieee80211_ref_node(ni));
ni->ni_flags |= IEEE80211_NODE_AREF;
if (obss != NULL)
ieee80211_free_node(obss);
/* give driver a chance to setup state like ni_txrate */
if (ic->ic_newassoc != NULL)
ic->ic_newassoc(ni, 1);
/* tell the authenticator about new station */
if (vap->iv_auth->ia_node_join != NULL)
vap->iv_auth->ia_node_join(ni);
if (ni->ni_authmode != IEEE80211_AUTH_8021X)
ieee80211_node_authorize(ni);
ieee80211_notify_node_join(ni, 1 /*newassoc*/);
/* XXX inject l2uf frame */
}
}
/*
* Flush any pending frames now that were setup.
*/
if (ni != NULL)
wds_flush(ni);
return (ni == NULL ? ENOENT : 0);
}
/*
* Propagate multicast frames of an ap vap to all DWDS links.
* The caller is assumed to have verified this frame is multicast.
*/
void
ieee80211_dwds_mcast(struct ieee80211vap *vap0, struct mbuf *m)
{
struct ieee80211com *ic = vap0->iv_ic;
const struct ether_header *eh = mtod(m, const struct ether_header *);
struct ieee80211_node *ni;
struct ieee80211vap *vap;
struct ifnet *ifp;
struct mbuf *mcopy;
int err;
KASSERT(ETHER_IS_MULTICAST(eh->ether_dhost),
("%s not mcast", ether_sprintf(eh->ether_dhost)));
/* XXX locking */
TAILQ_FOREACH(vap, &ic->ic_vaps, iv_next) {
/* only DWDS vaps are interesting */
if (vap->iv_opmode != IEEE80211_M_WDS ||
(vap->iv_flags_ext & IEEE80211_FEXT_WDSLEGACY))
continue;
/* if it came in this interface, don't send it back out */
ifp = vap->iv_ifp;
if (ifp == m->m_pkthdr.rcvif)
continue;
/*
* Duplicate the frame and send it.
*/
mcopy = m_copypacket(m, IEEE80211_M_NOWAIT);
if (mcopy == NULL) {
if_inc_counter(ifp, IFCOUNTER_OERRORS, 1);
/* XXX stat + msg */
continue;
}
ni = ieee80211_find_txnode(vap, eh->ether_dhost);
if (ni == NULL) {
/* NB: ieee80211_find_txnode does stat+msg */
if_inc_counter(ifp, IFCOUNTER_OERRORS, 1);
m_freem(mcopy);
continue;
}
/* calculate priority so drivers can find the tx queue */
if (ieee80211_classify(ni, mcopy)) {
IEEE80211_DISCARD_MAC(vap,
IEEE80211_MSG_OUTPUT | IEEE80211_MSG_WDS,
eh->ether_dhost, NULL,
"%s", "classification failure");
vap->iv_stats.is_tx_classify++;
if_inc_counter(ifp, IFCOUNTER_OERRORS, 1);
m_freem(mcopy);
ieee80211_free_node(ni);
continue;
}
BPF_MTAP(ifp, m); /* 802.3 tx */
/*
* Encapsulate the packet in prep for transmission.
*/
IEEE80211_TX_LOCK(ic);
mcopy = ieee80211_encap(vap, ni, mcopy);
if (mcopy == NULL) {
/* NB: stat+msg handled in ieee80211_encap */
IEEE80211_TX_UNLOCK(ic);
ieee80211_free_node(ni);
continue;
}
mcopy->m_flags |= M_MCAST;
Restructure mbuf send tags to provide stronger guarantees. - Perform ifp mismatch checks (to determine if a send tag is allocated for a different ifp than the one the packet is being output on), in ip_output() and ip6_output(). This avoids sending packets with send tags to ifnet drivers that don't support send tags. Since we are now checking for ifp mismatches before invoking if_output, we can now try to allocate a new tag before invoking if_output sending the original packet on the new tag if allocation succeeds. To avoid code duplication for the fragment and unfragmented cases, add ip_output_send() and ip6_output_send() as wrappers around if_output and nd6_output_ifp, respectively. All of the logic for setting send tags and dealing with send tag-related errors is done in these wrapper functions. For pseudo interfaces that wrap other network interfaces (vlan and lagg), wrapper send tags are now allocated so that ip*_output see the wrapper ifp as the ifp in the send tag. The if_transmit routines rewrite the send tags after performing an ifp mismatch check. If an ifp mismatch is detected, the transmit routines fail with EAGAIN. - To provide clearer life cycle management of send tags, especially in the presence of vlan and lagg wrapper tags, add a reference count to send tags managed via m_snd_tag_ref() and m_snd_tag_rele(). Provide a helper function (m_snd_tag_init()) for use by drivers supporting send tags. m_snd_tag_init() takes care of the if_ref on the ifp meaning that code alloating send tags via if_snd_tag_alloc no longer has to manage that manually. Similarly, m_snd_tag_rele drops the refcount on the ifp after invoking if_snd_tag_free when the last reference to a send tag is dropped. This also closes use after free races if there are pending packets in driver tx rings after the socket is closed (e.g. from tcpdrop). In order for m_free to work reliably, add a new CSUM_SND_TAG flag in csum_flags to indicate 'snd_tag' is set (rather than 'rcvif'). Drivers now also check this flag instead of checking snd_tag against NULL. This avoids false positive matches when a forwarded packet has a non-NULL rcvif that was treated as a send tag. - cxgbe was relying on snd_tag_free being called when the inp was detached so that it could kick the firmware to flush any pending work on the flow. This is because the driver doesn't require ACK messages from the firmware for every request, but instead does a kind of manual interrupt coalescing by only setting a flag to request a completion on a subset of requests. If all of the in-flight requests don't have the flag when the tag is detached from the inp, the flow might never return the credits. The current snd_tag_free command issues a flush command to force the credits to return. However, the credit return is what also frees the mbufs, and since those mbufs now hold references on the tag, this meant that snd_tag_free would never be called. To fix, explicitly drop the mbuf's reference on the snd tag when the mbuf is queued in the firmware work queue. This means that once the inp's reference on the tag goes away and all in-flight mbufs have been queued to the firmware, tag's refcount will drop to zero and snd_tag_free will kick in and send the flush request. Note that we need to avoid doing this in the middle of ethofld_tx(), so the driver grabs a temporary reference on the tag around that loop to defer the free to the end of the function in case it sends the last mbuf to the queue after the inp has dropped its reference on the tag. - mlx5 preallocates send tags and was using the ifp pointer even when the send tag wasn't in use. Explicitly use the ifp from other data structures instead. - Sprinkle some assertions in various places to assert that received packets don't have a send tag, and that other places that overwrite rcvif (e.g. 802.11 transmit) don't clobber a send tag pointer. Reviewed by: gallatin, hselasky, rgrimes, ae Sponsored by: Netflix Differential Revision: https://reviews.freebsd.org/D20117
2019-05-24 22:30:40 +00:00
MPASS((mcopy->m_pkthdr.csum_flags & CSUM_SND_TAG) == 0);
mcopy->m_pkthdr.rcvif = (void *) ni;
err = ieee80211_parent_xmitpkt(ic, mcopy);
IEEE80211_TX_UNLOCK(ic);
if (!err) {
if_inc_counter(ifp, IFCOUNTER_OPACKETS, 1);
if_inc_counter(ifp, IFCOUNTER_OMCASTS, 1);
if_inc_counter(ifp, IFCOUNTER_OBYTES,
m->m_pkthdr.len);
}
}
}
/*
* Handle DWDS discovery on receipt of a 4-address frame in
* ap mode. Queue the frame and post an event for someone
* to plumb the necessary WDS vap for this station. Frames
* received prior to the vap set running will then be reprocessed
* as if they were just received.
*/
void
ieee80211_dwds_discover(struct ieee80211_node *ni, struct mbuf *m)
{
struct ieee80211com *ic = ni->ni_ic;
/*
* Save the frame with an aging interval 4 times
* the listen interval specified by the station.
* Frames that sit around too long are reclaimed
* using this information.
* XXX handle overflow?
* XXX per/vap beacon interval?
*/
Restructure mbuf send tags to provide stronger guarantees. - Perform ifp mismatch checks (to determine if a send tag is allocated for a different ifp than the one the packet is being output on), in ip_output() and ip6_output(). This avoids sending packets with send tags to ifnet drivers that don't support send tags. Since we are now checking for ifp mismatches before invoking if_output, we can now try to allocate a new tag before invoking if_output sending the original packet on the new tag if allocation succeeds. To avoid code duplication for the fragment and unfragmented cases, add ip_output_send() and ip6_output_send() as wrappers around if_output and nd6_output_ifp, respectively. All of the logic for setting send tags and dealing with send tag-related errors is done in these wrapper functions. For pseudo interfaces that wrap other network interfaces (vlan and lagg), wrapper send tags are now allocated so that ip*_output see the wrapper ifp as the ifp in the send tag. The if_transmit routines rewrite the send tags after performing an ifp mismatch check. If an ifp mismatch is detected, the transmit routines fail with EAGAIN. - To provide clearer life cycle management of send tags, especially in the presence of vlan and lagg wrapper tags, add a reference count to send tags managed via m_snd_tag_ref() and m_snd_tag_rele(). Provide a helper function (m_snd_tag_init()) for use by drivers supporting send tags. m_snd_tag_init() takes care of the if_ref on the ifp meaning that code alloating send tags via if_snd_tag_alloc no longer has to manage that manually. Similarly, m_snd_tag_rele drops the refcount on the ifp after invoking if_snd_tag_free when the last reference to a send tag is dropped. This also closes use after free races if there are pending packets in driver tx rings after the socket is closed (e.g. from tcpdrop). In order for m_free to work reliably, add a new CSUM_SND_TAG flag in csum_flags to indicate 'snd_tag' is set (rather than 'rcvif'). Drivers now also check this flag instead of checking snd_tag against NULL. This avoids false positive matches when a forwarded packet has a non-NULL rcvif that was treated as a send tag. - cxgbe was relying on snd_tag_free being called when the inp was detached so that it could kick the firmware to flush any pending work on the flow. This is because the driver doesn't require ACK messages from the firmware for every request, but instead does a kind of manual interrupt coalescing by only setting a flag to request a completion on a subset of requests. If all of the in-flight requests don't have the flag when the tag is detached from the inp, the flow might never return the credits. The current snd_tag_free command issues a flush command to force the credits to return. However, the credit return is what also frees the mbufs, and since those mbufs now hold references on the tag, this meant that snd_tag_free would never be called. To fix, explicitly drop the mbuf's reference on the snd tag when the mbuf is queued in the firmware work queue. This means that once the inp's reference on the tag goes away and all in-flight mbufs have been queued to the firmware, tag's refcount will drop to zero and snd_tag_free will kick in and send the flush request. Note that we need to avoid doing this in the middle of ethofld_tx(), so the driver grabs a temporary reference on the tag around that loop to defer the free to the end of the function in case it sends the last mbuf to the queue after the inp has dropped its reference on the tag. - mlx5 preallocates send tags and was using the ifp pointer even when the send tag wasn't in use. Explicitly use the ifp from other data structures instead. - Sprinkle some assertions in various places to assert that received packets don't have a send tag, and that other places that overwrite rcvif (e.g. 802.11 transmit) don't clobber a send tag pointer. Reviewed by: gallatin, hselasky, rgrimes, ae Sponsored by: Netflix Differential Revision: https://reviews.freebsd.org/D20117
2019-05-24 22:30:40 +00:00
MPASS((m->m_pkthdr.csum_flags & CSUM_SND_TAG) == 0);
m->m_pkthdr.rcvif = (void *)(uintptr_t)
ieee80211_mac_hash(ic, ni->ni_macaddr);
(void) ieee80211_ageq_append(&ic->ic_stageq, m,
((ni->ni_intval * ic->ic_lintval) << 2) / 1024);
ieee80211_notify_wds_discover(ni);
}
/*
* IEEE80211_M_WDS vap state machine handler.
*/
static int
wds_newstate(struct ieee80211vap *vap, enum ieee80211_state nstate, int arg)
{
struct ieee80211com *ic = vap->iv_ic;
enum ieee80211_state ostate;
int error;
IEEE80211_LOCK_ASSERT(ic);
ostate = vap->iv_state;
IEEE80211_DPRINTF(vap, IEEE80211_MSG_STATE, "%s: %s -> %s\n", __func__,
ieee80211_state_name[ostate], ieee80211_state_name[nstate]);
vap->iv_state = nstate; /* state transition */
callout_stop(&vap->iv_mgtsend); /* XXX callout_drain */
if (ostate != IEEE80211_S_SCAN)
ieee80211_cancel_scan(vap); /* background scan */
error = 0;
switch (nstate) {
case IEEE80211_S_INIT:
switch (ostate) {
case IEEE80211_S_SCAN:
ieee80211_cancel_scan(vap);
break;
default:
break;
}
if (ostate != IEEE80211_S_INIT) {
/* NB: optimize INIT -> INIT case */
ieee80211_reset_bss(vap);
}
break;
case IEEE80211_S_SCAN:
switch (ostate) {
case IEEE80211_S_INIT:
ieee80211_check_scan_current(vap);
break;
default:
break;
}
break;
case IEEE80211_S_RUN:
if (ostate == IEEE80211_S_INIT) {
/*
* Already have a channel; bypass the scan
* and startup immediately.
*/
error = ieee80211_create_wds(vap, ic->ic_curchan);
}
break;
default:
break;
}
return error;
}
/*
* Process a received frame. The node associated with the sender
* should be supplied. If nothing was found in the node table then
* the caller is assumed to supply a reference to iv_bss instead.
* The RSSI and a timestamp are also supplied. The RSSI data is used
* during AP scanning to select a AP to associate with; it can have
* any units so long as values have consistent units and higher values
* mean ``better signal''. The receive timestamp is currently not used
* by the 802.11 layer.
*/
static int
Begin plumbing ieee80211_rx_stats through the receive path. Smart NICs with firmware (eg wpi, iwn, the new atheros parts, the intel 7260 series, etc) support doing a lot of things in firmware. This includes but isn't limited to things like scanning, sending probe requests and receiving probe responses. However, net80211 doesn't know about any of this - it still drives the whole scan/probe infrastructure itself. In order to move towards suppoting smart NICs, the receive path needs to know about the channel/details for each received packet. In at least the iwn and 7260 firmware (and I believe wpi, but I haven't tried it yet) it will do the scanning, power-save and off-channel buffering for you - all you need to do is handle receiving beacons and probe responses on channels that aren't what you're currently on. However the whole receive path is peppered with ic->ic_curchan and manual scan/powersave handling. The beacon parsing code also checks ic->ic_curchan to determine if the received beacon is on the correct channel or not.[1] So: * add freq/ieee values to ieee80211_rx_stats; * change ieee80211_parse_beacon() to accept the 'current' channel as an argument; * modify the iv_input() and iv_recv_mgmt() methods to include the rx_stats; * add a new method - ieee80211_lookup_channel_rxstats() - that looks up a channel based on the contents of ieee80211_rx_stats; * if it exists, use it in the mgmt path to switch the current channel (which still defaults to ic->ic_curchan) over to something determined by rx_stats. This is enough to kick-start scan offload support in the Intel 7260 driver that Rui/I are working on. It also is a good start for scan offload support for a handful of existing NICs (wpi, iwn, some USB parts) and it'll very likely dramatically improve stability/performance there. It's not the whole thing - notably, we don't need to do powersave, we should not scan all channels, and we should leave probe request sending to the firmware and not do it ourselves. But, this allows for continued development on the above features whilst actually having a somewhat working NIC. TODO: * Finish tidying up how the net80211 input path works. Right now ieee80211_input / ieee80211_input_all act as the top-level that everything feeds into; it should change so the MIMO input routines are those and the legacy routines are phased out. * The band selection should be done by the driver, not by the net80211 layer. * ieee80211_lookup_channel_rxstats() only determines 11b or 11g channels for now - this is enough for scanning, but not 100% true in all cases. If we ever need to handle off-channel scan support for things like static-40MHz or static-80MHz, or turbo-G, or half/quarter rates, then we should extend this. [1] This is a side effect of frequency-hopping and CCK modes - you can receive beacons when you think you're on a different channel. In particular, CCK (which is used by the low 11b rates, eg beacons!) is decodable from adjacent channels - just at a low SNR. FH is a side effect of having the hardware/firmware do the frequency hopping - it may pick up beacons transmitted from other FH networks that are in a different phase of hopping frequencies.
2015-05-25 16:37:41 +00:00
wds_input(struct ieee80211_node *ni, struct mbuf *m,
const struct ieee80211_rx_stats *rxs, int rssi, int nf)
{
struct ieee80211vap *vap = ni->ni_vap;
struct ieee80211com *ic = ni->ni_ic;
struct ifnet *ifp = vap->iv_ifp;
struct ieee80211_frame *wh;
struct ieee80211_key *key;
struct ether_header *eh;
int hdrspace, need_tap = 1; /* mbuf need to be tapped. */
uint8_t dir, type, subtype, qos;
int is_hw_decrypted = 0;
int has_decrypted = 0;
/*
* Some devices do hardware decryption all the way through
* to pretending the frame wasn't encrypted in the first place.
* So, tag it appropriately so it isn't discarded inappropriately.
*/
if ((rxs != NULL) && (rxs->c_pktflags & IEEE80211_RX_F_DECRYPTED))
is_hw_decrypted = 1;
if (m->m_flags & M_AMPDU_MPDU) {
/*
* Fastpath for A-MPDU reorder q resubmission. Frames
* w/ M_AMPDU_MPDU marked have already passed through
* here but were received out of order and been held on
* the reorder queue. When resubmitted they are marked
* with the M_AMPDU_MPDU flag and we can bypass most of
* the normal processing.
*/
wh = mtod(m, struct ieee80211_frame *);
type = IEEE80211_FC0_TYPE_DATA;
dir = wh->i_fc[1] & IEEE80211_FC1_DIR_MASK;
subtype = IEEE80211_FC0_SUBTYPE_QOS_DATA;
hdrspace = ieee80211_hdrspace(ic, wh); /* XXX optimize? */
goto resubmit_ampdu;
}
KASSERT(ni != NULL, ("null node"));
type = -1; /* undefined */
if (m->m_pkthdr.len < sizeof(struct ieee80211_frame_min)) {
IEEE80211_DISCARD_MAC(vap, IEEE80211_MSG_ANY,
ni->ni_macaddr, NULL,
"too short (1): len %u", m->m_pkthdr.len);
vap->iv_stats.is_rx_tooshort++;
goto out;
}
/*
* Bit of a cheat here, we use a pointer for a 3-address
* frame format but don't reference fields past outside
* ieee80211_frame_min w/o first validating the data is
* present.
*/
wh = mtod(m, struct ieee80211_frame *);
if (!IEEE80211_IS_MULTICAST(wh->i_addr1))
ni->ni_inact = ni->ni_inact_reload;
if ((wh->i_fc[0] & IEEE80211_FC0_VERSION_MASK) !=
IEEE80211_FC0_VERSION_0) {
IEEE80211_DISCARD_MAC(vap, IEEE80211_MSG_ANY,
ni->ni_macaddr, NULL, "wrong version, fc %02x:%02x",
wh->i_fc[0], wh->i_fc[1]);
vap->iv_stats.is_rx_badversion++;
goto err;
}
dir = wh->i_fc[1] & IEEE80211_FC1_DIR_MASK;
type = wh->i_fc[0] & IEEE80211_FC0_TYPE_MASK;
subtype = wh->i_fc[0] & IEEE80211_FC0_SUBTYPE_MASK;
/* NB: WDS vap's do not scan */
if (m->m_pkthdr.len < sizeof(struct ieee80211_frame_addr4)) {
IEEE80211_DISCARD_MAC(vap,
IEEE80211_MSG_ANY, ni->ni_macaddr, NULL,
"too short (3): len %u", m->m_pkthdr.len);
vap->iv_stats.is_rx_tooshort++;
goto out;
}
/* NB: the TA is implicitly verified by finding the wds peer node */
if (!IEEE80211_ADDR_EQ(wh->i_addr1, vap->iv_myaddr) &&
!IEEE80211_ADDR_EQ(wh->i_addr1, ifp->if_broadcastaddr)) {
/* not interested in */
IEEE80211_DISCARD_MAC(vap, IEEE80211_MSG_INPUT,
wh->i_addr1, NULL, "%s", "not to bss");
vap->iv_stats.is_rx_wrongbss++;
goto out;
}
IEEE80211_RSSI_LPF(ni->ni_avgrssi, rssi);
ni->ni_noise = nf;
if (IEEE80211_HAS_SEQ(type, subtype)) {
uint8_t tid = ieee80211_gettid(wh);
if (IEEE80211_QOS_HAS_SEQ(wh) &&
TID_TO_WME_AC(tid) >= WME_AC_VI)
ic->ic_wme.wme_hipri_traffic++;
if (! ieee80211_check_rxseq(ni, wh, wh->i_addr1, rxs))
goto out;
}
switch (type) {
case IEEE80211_FC0_TYPE_DATA:
hdrspace = ieee80211_hdrspace(ic, wh);
if (m->m_len < hdrspace &&
(m = m_pullup(m, hdrspace)) == NULL) {
IEEE80211_DISCARD_MAC(vap, IEEE80211_MSG_ANY,
ni->ni_macaddr, NULL,
"data too short: expecting %u", hdrspace);
vap->iv_stats.is_rx_tooshort++;
goto out; /* XXX */
}
if (dir != IEEE80211_FC1_DIR_DSTODS) {
IEEE80211_DISCARD(vap, IEEE80211_MSG_INPUT,
wh, "data", "incorrect dir 0x%x", dir);
vap->iv_stats.is_rx_wrongdir++;
goto out;
}
/*
* Only legacy WDS traffic should take this path.
*/
if ((vap->iv_flags_ext & IEEE80211_FEXT_WDSLEGACY) == 0) {
IEEE80211_DISCARD(vap, IEEE80211_MSG_INPUT,
wh, "data", "%s", "not legacy wds");
vap->iv_stats.is_rx_wrongdir++;/*XXX*/
goto out;
}
/*
* Handle A-MPDU re-ordering. If the frame is to be
* processed directly then ieee80211_ampdu_reorder
* will return 0; otherwise it has consumed the mbuf
* and we should do nothing more with it.
*/
if ((m->m_flags & M_AMPDU) &&
ieee80211_ampdu_reorder(ni, m, rxs) != 0) {
m = NULL;
goto out;
}
resubmit_ampdu:
/*
* Handle privacy requirements. Note that we
* must not be preempted from here until after
* we (potentially) call ieee80211_crypto_demic;
* otherwise we may violate assumptions in the
* crypto cipher modules used to do delayed update
* of replay sequence numbers.
*/
if (is_hw_decrypted || IEEE80211_IS_PROTECTED(wh)) {
if ((vap->iv_flags & IEEE80211_F_PRIVACY) == 0) {
/*
* Discard encrypted frames when privacy is off.
*/
IEEE80211_DISCARD(vap, IEEE80211_MSG_INPUT,
wh, "WEP", "%s", "PRIVACY off");
vap->iv_stats.is_rx_noprivacy++;
IEEE80211_NODE_STAT(ni, rx_noprivacy);
goto out;
}
if (ieee80211_crypto_decap(ni, m, hdrspace, &key) == 0) {
/* NB: stats+msgs handled in crypto_decap */
IEEE80211_NODE_STAT(ni, rx_wepfail);
goto out;
}
wh = mtod(m, struct ieee80211_frame *);
wh->i_fc[1] &= ~IEEE80211_FC1_PROTECTED;
has_decrypted = 1;
} else {
/* XXX M_WEP and IEEE80211_F_PRIVACY */
key = NULL;
}
/*
* Save QoS bits for use below--before we strip the header.
*/
if (subtype == IEEE80211_FC0_SUBTYPE_QOS_DATA)
qos = ieee80211_getqos(wh)[0];
else
qos = 0;
/*
* Next up, any fragmentation.
*/
if (!IEEE80211_IS_MULTICAST(wh->i_addr1)) {
m = ieee80211_defrag(ni, m, hdrspace, has_decrypted);
if (m == NULL) {
/* Fragment dropped or frame not complete yet */
goto out;
}
}
wh = NULL; /* no longer valid, catch any uses */
/*
* Next strip any MSDU crypto bits.
*/
if (!ieee80211_crypto_demic(vap, key, m, 0)) {
IEEE80211_DISCARD_MAC(vap, IEEE80211_MSG_INPUT,
ni->ni_macaddr, "data", "%s", "demic error");
vap->iv_stats.is_rx_demicfail++;
IEEE80211_NODE_STAT(ni, rx_demicfail);
goto out;
}
/* copy to listener after decrypt */
if (ieee80211_radiotap_active_vap(vap))
ieee80211_radiotap_rx(vap, m);
need_tap = 0;
/*
* Finally, strip the 802.11 header.
*/
m = ieee80211_decap(vap, m, hdrspace, qos);
if (m == NULL) {
/* XXX mask bit to check for both */
/* don't count Null data frames as errors */
if (subtype == IEEE80211_FC0_SUBTYPE_NODATA ||
subtype == IEEE80211_FC0_SUBTYPE_QOS_NULL)
goto out;
IEEE80211_DISCARD_MAC(vap, IEEE80211_MSG_INPUT,
ni->ni_macaddr, "data", "%s", "decap error");
vap->iv_stats.is_rx_decap++;
IEEE80211_NODE_STAT(ni, rx_decap);
goto err;
}
if (!(qos & IEEE80211_QOS_AMSDU))
eh = mtod(m, struct ether_header *);
else
eh = NULL;
if (!ieee80211_node_is_authorized(ni)) {
/*
* Deny any non-PAE frames received prior to
* authorization. For open/shared-key
* authentication the port is mark authorized
* after authentication completes. For 802.1x
* the port is not marked authorized by the
* authenticator until the handshake has completed.
*/
if (eh == NULL ||
eh->ether_type != htons(ETHERTYPE_PAE)) {
IEEE80211_DISCARD_MAC(vap, IEEE80211_MSG_INPUT,
ni->ni_macaddr, "data", "unauthorized or "
"unknown port: ether type 0x%x len %u",
eh == NULL ? -1 : eh->ether_type,
m->m_pkthdr.len);
vap->iv_stats.is_rx_unauth++;
IEEE80211_NODE_STAT(ni, rx_unauth);
goto err;
}
} else {
/*
* When denying unencrypted frames, discard
* any non-PAE frames received without encryption.
*/
if ((vap->iv_flags & IEEE80211_F_DROPUNENC) &&
((has_decrypted == 0) && (m->m_flags & M_WEP) == 0) &&
(is_hw_decrypted == 0) &&
(eh == NULL ||
eh->ether_type != htons(ETHERTYPE_PAE))) {
/*
* Drop unencrypted frames.
*/
vap->iv_stats.is_rx_unencrypted++;
IEEE80211_NODE_STAT(ni, rx_unencrypted);
goto out;
}
}
/* XXX require HT? */
if (qos & IEEE80211_QOS_AMSDU) {
m = ieee80211_decap_amsdu(ni, m);
if (m == NULL)
return IEEE80211_FC0_TYPE_DATA;
} else {
#ifdef IEEE80211_SUPPORT_SUPERG
m = ieee80211_decap_fastframe(vap, ni, m);
if (m == NULL)
return IEEE80211_FC0_TYPE_DATA;
#endif
}
ieee80211_deliver_data(vap, ni, m);
return IEEE80211_FC0_TYPE_DATA;
case IEEE80211_FC0_TYPE_MGT:
vap->iv_stats.is_rx_mgmt++;
IEEE80211_NODE_STAT(ni, rx_mgmt);
if (dir != IEEE80211_FC1_DIR_NODS) {
IEEE80211_DISCARD(vap, IEEE80211_MSG_INPUT,
wh, "data", "incorrect dir 0x%x", dir);
vap->iv_stats.is_rx_wrongdir++;
goto err;
}
if (m->m_pkthdr.len < sizeof(struct ieee80211_frame)) {
IEEE80211_DISCARD_MAC(vap, IEEE80211_MSG_ANY,
ni->ni_macaddr, "mgt", "too short: len %u",
m->m_pkthdr.len);
vap->iv_stats.is_rx_tooshort++;
goto out;
}
#ifdef IEEE80211_DEBUG
if (ieee80211_msg_debug(vap) || ieee80211_msg_dumppkts(vap)) {
if_printf(ifp, "received %s from %s rssi %d\n",
ieee80211_mgt_subtype_name(subtype),
ether_sprintf(wh->i_addr2), rssi);
}
#endif
if (IEEE80211_IS_PROTECTED(wh)) {
IEEE80211_DISCARD(vap, IEEE80211_MSG_INPUT,
wh, NULL, "%s", "WEP set but not permitted");
vap->iv_stats.is_rx_mgtdiscard++; /* XXX */
goto out;
}
Begin plumbing ieee80211_rx_stats through the receive path. Smart NICs with firmware (eg wpi, iwn, the new atheros parts, the intel 7260 series, etc) support doing a lot of things in firmware. This includes but isn't limited to things like scanning, sending probe requests and receiving probe responses. However, net80211 doesn't know about any of this - it still drives the whole scan/probe infrastructure itself. In order to move towards suppoting smart NICs, the receive path needs to know about the channel/details for each received packet. In at least the iwn and 7260 firmware (and I believe wpi, but I haven't tried it yet) it will do the scanning, power-save and off-channel buffering for you - all you need to do is handle receiving beacons and probe responses on channels that aren't what you're currently on. However the whole receive path is peppered with ic->ic_curchan and manual scan/powersave handling. The beacon parsing code also checks ic->ic_curchan to determine if the received beacon is on the correct channel or not.[1] So: * add freq/ieee values to ieee80211_rx_stats; * change ieee80211_parse_beacon() to accept the 'current' channel as an argument; * modify the iv_input() and iv_recv_mgmt() methods to include the rx_stats; * add a new method - ieee80211_lookup_channel_rxstats() - that looks up a channel based on the contents of ieee80211_rx_stats; * if it exists, use it in the mgmt path to switch the current channel (which still defaults to ic->ic_curchan) over to something determined by rx_stats. This is enough to kick-start scan offload support in the Intel 7260 driver that Rui/I are working on. It also is a good start for scan offload support for a handful of existing NICs (wpi, iwn, some USB parts) and it'll very likely dramatically improve stability/performance there. It's not the whole thing - notably, we don't need to do powersave, we should not scan all channels, and we should leave probe request sending to the firmware and not do it ourselves. But, this allows for continued development on the above features whilst actually having a somewhat working NIC. TODO: * Finish tidying up how the net80211 input path works. Right now ieee80211_input / ieee80211_input_all act as the top-level that everything feeds into; it should change so the MIMO input routines are those and the legacy routines are phased out. * The band selection should be done by the driver, not by the net80211 layer. * ieee80211_lookup_channel_rxstats() only determines 11b or 11g channels for now - this is enough for scanning, but not 100% true in all cases. If we ever need to handle off-channel scan support for things like static-40MHz or static-80MHz, or turbo-G, or half/quarter rates, then we should extend this. [1] This is a side effect of frequency-hopping and CCK modes - you can receive beacons when you think you're on a different channel. In particular, CCK (which is used by the low 11b rates, eg beacons!) is decodable from adjacent channels - just at a low SNR. FH is a side effect of having the hardware/firmware do the frequency hopping - it may pick up beacons transmitted from other FH networks that are in a different phase of hopping frequencies.
2015-05-25 16:37:41 +00:00
vap->iv_recv_mgmt(ni, m, subtype, rxs, rssi, nf);
goto out;
case IEEE80211_FC0_TYPE_CTL:
vap->iv_stats.is_rx_ctl++;
IEEE80211_NODE_STAT(ni, rx_ctrl);
goto out;
default:
IEEE80211_DISCARD(vap, IEEE80211_MSG_ANY,
wh, "bad", "frame type 0x%x", type);
/* should not come here */
break;
}
err:
if_inc_counter(ifp, IFCOUNTER_IERRORS, 1);
out:
if (m != NULL) {
if (need_tap && ieee80211_radiotap_active_vap(vap))
ieee80211_radiotap_rx(vap, m);
m_freem(m);
}
return type;
}
static void
Begin plumbing ieee80211_rx_stats through the receive path. Smart NICs with firmware (eg wpi, iwn, the new atheros parts, the intel 7260 series, etc) support doing a lot of things in firmware. This includes but isn't limited to things like scanning, sending probe requests and receiving probe responses. However, net80211 doesn't know about any of this - it still drives the whole scan/probe infrastructure itself. In order to move towards suppoting smart NICs, the receive path needs to know about the channel/details for each received packet. In at least the iwn and 7260 firmware (and I believe wpi, but I haven't tried it yet) it will do the scanning, power-save and off-channel buffering for you - all you need to do is handle receiving beacons and probe responses on channels that aren't what you're currently on. However the whole receive path is peppered with ic->ic_curchan and manual scan/powersave handling. The beacon parsing code also checks ic->ic_curchan to determine if the received beacon is on the correct channel or not.[1] So: * add freq/ieee values to ieee80211_rx_stats; * change ieee80211_parse_beacon() to accept the 'current' channel as an argument; * modify the iv_input() and iv_recv_mgmt() methods to include the rx_stats; * add a new method - ieee80211_lookup_channel_rxstats() - that looks up a channel based on the contents of ieee80211_rx_stats; * if it exists, use it in the mgmt path to switch the current channel (which still defaults to ic->ic_curchan) over to something determined by rx_stats. This is enough to kick-start scan offload support in the Intel 7260 driver that Rui/I are working on. It also is a good start for scan offload support for a handful of existing NICs (wpi, iwn, some USB parts) and it'll very likely dramatically improve stability/performance there. It's not the whole thing - notably, we don't need to do powersave, we should not scan all channels, and we should leave probe request sending to the firmware and not do it ourselves. But, this allows for continued development on the above features whilst actually having a somewhat working NIC. TODO: * Finish tidying up how the net80211 input path works. Right now ieee80211_input / ieee80211_input_all act as the top-level that everything feeds into; it should change so the MIMO input routines are those and the legacy routines are phased out. * The band selection should be done by the driver, not by the net80211 layer. * ieee80211_lookup_channel_rxstats() only determines 11b or 11g channels for now - this is enough for scanning, but not 100% true in all cases. If we ever need to handle off-channel scan support for things like static-40MHz or static-80MHz, or turbo-G, or half/quarter rates, then we should extend this. [1] This is a side effect of frequency-hopping and CCK modes - you can receive beacons when you think you're on a different channel. In particular, CCK (which is used by the low 11b rates, eg beacons!) is decodable from adjacent channels - just at a low SNR. FH is a side effect of having the hardware/firmware do the frequency hopping - it may pick up beacons transmitted from other FH networks that are in a different phase of hopping frequencies.
2015-05-25 16:37:41 +00:00
wds_recv_mgmt(struct ieee80211_node *ni, struct mbuf *m0, int subtype,
const struct ieee80211_rx_stats *rxs, int rssi, int nf)
{
struct ieee80211vap *vap = ni->ni_vap;
struct ieee80211com *ic = ni->ni_ic;
struct ieee80211_frame *wh;
u_int8_t *frm, *efrm;
wh = mtod(m0, struct ieee80211_frame *);
frm = (u_int8_t *)&wh[1];
efrm = mtod(m0, u_int8_t *) + m0->m_len;
switch (subtype) {
case IEEE80211_FC0_SUBTYPE_ACTION:
case IEEE80211_FC0_SUBTYPE_ACTION_NOACK:
if (ni == vap->iv_bss) {
IEEE80211_DISCARD(vap, IEEE80211_MSG_INPUT,
wh, NULL, "%s", "unknown node");
vap->iv_stats.is_rx_mgtdiscard++;
} else if (!IEEE80211_ADDR_EQ(vap->iv_myaddr, wh->i_addr1)) {
/* NB: not interested in multicast frames. */
IEEE80211_DISCARD(vap, IEEE80211_MSG_INPUT,
wh, NULL, "%s", "not for us");
vap->iv_stats.is_rx_mgtdiscard++;
} else if (vap->iv_state != IEEE80211_S_RUN) {
IEEE80211_DISCARD(vap, IEEE80211_MSG_INPUT,
wh, NULL, "wrong state %s",
ieee80211_state_name[vap->iv_state]);
vap->iv_stats.is_rx_mgtdiscard++;
} else {
if (ieee80211_parse_action(ni, m0) == 0)
(void)ic->ic_recv_action(ni, wh, frm, efrm);
}
break;
case IEEE80211_FC0_SUBTYPE_ASSOC_REQ:
case IEEE80211_FC0_SUBTYPE_ASSOC_RESP:
case IEEE80211_FC0_SUBTYPE_REASSOC_REQ:
case IEEE80211_FC0_SUBTYPE_REASSOC_RESP:
case IEEE80211_FC0_SUBTYPE_PROBE_REQ:
case IEEE80211_FC0_SUBTYPE_PROBE_RESP:
case IEEE80211_FC0_SUBTYPE_TIMING_ADV:
case IEEE80211_FC0_SUBTYPE_BEACON:
case IEEE80211_FC0_SUBTYPE_ATIM:
case IEEE80211_FC0_SUBTYPE_DISASSOC:
case IEEE80211_FC0_SUBTYPE_AUTH:
case IEEE80211_FC0_SUBTYPE_DEAUTH:
IEEE80211_DISCARD(vap, IEEE80211_MSG_INPUT,
wh, NULL, "%s", "not handled");
vap->iv_stats.is_rx_mgtdiscard++;
break;
default:
IEEE80211_DISCARD(vap, IEEE80211_MSG_ANY,
wh, "mgt", "subtype 0x%x not handled", subtype);
vap->iv_stats.is_rx_badsubtype++;
break;
}
}