Skip to contentSkip to conversation

cfg80211: introduce critical protocol indication from user-space

RFC Superseded Arend van Spriel 2 participants 2 messages Apr 2, 2013 802.11 (including cfg80211/NL80211)

v3 of 7 Superseded Latest is v7 →

lore
Share & tools
lore
#0 lore raw

RFC v3 1/1 cfg80211: introduce critical protocol indication from user-space

Some protocols need a more reliable connection to complete successful in reasonable time. This patch adds a user-space API to indicate the wireless driver that a critical protocol is about to commence and when it is done, using nl80211 primitives NL80211_CMD_CRIT_PROTOCOL_START and NL80211_CRIT_PROTOCOL_STOP.

The driver can support this by implementing the cfg80211 callbacks .crit_proto_start() and .crit_proto_stop(). Examples of protocols that can benefit from this are DHCP, EAPOL, APIPA. Exactly how the link can/should be made more reliable is up to the driver. Things to consider are avoid scanning, no multi-channel operations, and alter coexistence schemes.

---
Here is version 3. I tried to deal with all the review comments I received, except for the one below:

Ben Greear said: "
Also, you would probably need to enforce in the kernel that only x out of y time in any given period can be locked, otherwise lots of different dhclient processes (perhaps erroneously spawned..or running on lots of different VIFs) could basically disable scanning or channel changes..."

Not sure how or even if we should deal with this. Any thoughts?

Gr. AvS

Changelog:
----------
V3:
- remove protocol identifier.
- remove delayed work from cfg80211.
- guard maximum limit for duration.
- do .crit_proto_stop() upon netlink socket release. V2:
- subject changed. Below previous subject is given for reference:
[RFC] cfg80211: configuration of Bluetooth coexistence mode
- introduced dedicated nl80211 API.
V1:
- initial proposal.

---
 include/net/cfg80211.h       |   10 +++++++
 include/uapi/linux/nl80211.h |   12 ++++++++
 net/wireless/mlme.c          |    3 ++
 net/wireless/nl80211.c       |   63 ++++++++++++++++++++++++++++++++++++++++++
 net/wireless/rdev-ops.h      |   28 ++++++++++++++++++-
 net/wireless/trace.h         |   33 ++++++++++++++++++++++
 6 files changed, 148 insertions(+), 1 deletion(-)
include/net/cfg80211.h +10 -0
index 57870b6..354d49c 100644
@@ -2002,6 +2002,9 @@ struct cfg80211_update_ft_ies_params {
  * @update_ft_ies: Provide updated Fast BSS Transition information to the
  *	driver. If the SME is in the driver/firmware, this information can be
  *	used in building Authentication and Reassociation Request frames.
+ * @crit_proto_start: Indicates a critical protocol needs more link reliability.
+ * @crit_proto_stop: Indicates critical protocol no longer needs increased link
+ *	reliability.
  */
 struct cfg80211_ops {
 	int	(*suspend)(struct wiphy *wiphy, struct cfg80211_wowlan *wow);
@@ -2231,6 +2234,10 @@ struct cfg80211_ops {
 					 struct cfg80211_chan_def *chandef);
 	int	(*update_ft_ies)(struct wiphy *wiphy, struct net_device *dev,
 				 struct cfg80211_update_ft_ies_params *ftie);
+	int	(*crit_proto_start)(struct wiphy *wiphy,
+				    struct wireless_dev *wdev, u16 duration);
+	int	(*crit_proto_stop)(struct wiphy *wiphy,
+				   struct wireless_dev *wdev);
 };
 
 /*
@@ -2830,6 +2837,7 @@ struct cfg80211_cached_keys;
  * @p2p_started: true if this is a P2P Device that has been started
  * @cac_started: true if DFS channel availability check has been started
  * @cac_start_time: timestamp (jiffies) when the dfs state was entered.
+ * @crit_proto_started: true if crit_proto_start() has been done.
  */
 struct wireless_dev {
 	struct wiphy *wiphy;
@@ -2884,6 +2892,8 @@ struct wireless_dev {
 	bool cac_started;
 	unsigned long cac_start_time;
 
+	bool crit_proto_started;
+
 #ifdef CONFIG_CFG80211_WEXT
 	/* wext data */
 	struct {
include/uapi/linux/nl80211.h +12 -0
index 79da871..518fc9a 100644
@@ -639,6 +639,13 @@
  *	with the relevant Information Elements. This event is used to report
  *	received FT IEs (MDIE, FTIE, RSN IE, TIE, RICIE).
  *
+ * @NL80211_CMD_CRIT_PROTOCOL_START: Indicates user-space will start running
+ *	a critical protocol that needs more reliability in the connection to
+ *	complete.
+ *
+ * @NL80211_CMD_CRIT_PROTOCOL_STOP: Indicates the connection reliability can
+ *	return back to normal.
+ *
  * @NL80211_CMD_MAX: highest used command number
  * @__NL80211_CMD_AFTER_LAST: internal use
  */
@@ -798,6 +805,9 @@ enum nl80211_commands {
 	NL80211_CMD_UPDATE_FT_IES,
 	NL80211_CMD_FT_EVENT,
 
+	NL80211_CMD_CRIT_PROTOCOL_START,
+	NL80211_CMD_CRIT_PROTOCOL_STOP,
+
 	/* add new commands above here */
 
 	/* used to define NL80211_CMD_MAX below */
@@ -1709,6 +1719,8 @@ enum nl80211_attrs {
 	NL80211_ATTR_MDID,
 	NL80211_ATTR_IE_RIC,
 
+	NL80211_ATTR_MAX_CRIT_PROT_DURATION,
+
 	/* add attributes here, update the policy in nl80211.c */
 
 	__NL80211_ATTR_AFTER_LAST,
net/wireless/mlme.c +3 -0
index 390198b..7c42b04 100644
@@ -648,6 +648,9 @@ void cfg80211_mlme_unregister_socket(struct wireless_dev *wdev, u32 nlportid)
 
 	spin_unlock_bh(&wdev->mgmt_registrations_lock);
 
+	if (rdev->ops->crit_proto_stop)
+		rdev_crit_proto_stop(rdev, wdev);
+
 	if (nlportid == wdev->ap_unexpected_nlportid)
 		wdev->ap_unexpected_nlportid = 0;
 }
net/wireless/nl80211.c +63 -0

87 lines not on this page · raw message

net/wireless/rdev-ops.h +27 -1
index d77e1c1..c032592 100644
@@ -875,7 +875,7 @@ static inline void rdev_stop_p2p_device(struct cfg80211_registered_device *rdev,
 	trace_rdev_stop_p2p_device(&rdev->wiphy, wdev);
 	rdev->ops->stop_p2p_device(&rdev->wiphy, wdev);
 	trace_rdev_return_void(&rdev->wiphy);
-}					
+}
 
 static inline int rdev_set_mac_acl(struct cfg80211_registered_device *rdev,
 				   struct net_device *dev,
@@ -901,4 +901,30 @@ static inline int rdev_update_ft_ies(struct cfg80211_registered_device *rdev,
 	return ret;
 }
 
+static inline int rdev_crit_proto_start(struct cfg80211_registered_device *rdev,
+					struct wireless_dev *wdev, u16 duration)
+{
+	int ret;
+
+	trace_rdev_crit_proto_start(&rdev->wiphy, wdev, duration);
+	ret = rdev->ops->crit_proto_start(&rdev->wiphy, wdev, duration);
+	wdev->crit_proto_started = !ret;
+	trace_rdev_return_int(&rdev->wiphy, ret);
+	return ret;
+}
+
+static inline int rdev_crit_proto_stop(struct cfg80211_registered_device *rdev,
+				       struct wireless_dev *wdev)
+{
+	int ret = 0;
+
+	trace_rdev_crit_proto_stop(&rdev->wiphy, wdev);
+	if (wdev->crit_proto_started) {
+		ret = rdev->ops->crit_proto_stop(&rdev->wiphy, wdev);
+		wdev->crit_proto_started = ret != 0;
+	}
+	trace_rdev_return_int(&rdev->wiphy, ret);
+	return ret;
+}
+
 #endif /* __CFG80211_RDEV_OPS */
net/wireless/trace.h +33 -0
index ccadef2..06766d0 100644
@@ -1805,6 +1805,39 @@ TRACE_EVENT(rdev_update_ft_ies,
 		  WIPHY_PR_ARG, NETDEV_PR_ARG, __entry->md)
 );
 
+TRACE_EVENT(rdev_crit_proto_start,
+	TP_PROTO(struct wiphy *wiphy, struct wireless_dev *wdev,
+		 u16 duration),
+	TP_ARGS(wiphy, wdev, duration),
+	TP_STRUCT__entry(
+		WIPHY_ENTRY
+		WDEV_ENTRY
+		__field(u16, duration)
+	),
+	TP_fast_assign(
+		WIPHY_ASSIGN;
+		WDEV_ASSIGN;
+		__entry->duration = duration;
+	),
+	TP_printk(WIPHY_PR_FMT ", " WDEV_PR_FMT ", duration=%x",
+		  WIPHY_PR_ARG, WDEV_PR_ARG, __entry->duration)
+);
+
+TRACE_EVENT(rdev_crit_proto_stop,
+	TP_PROTO(struct wiphy *wiphy, struct wireless_dev *wdev),
+	TP_ARGS(wiphy, wdev),
+	TP_STRUCT__entry(
+		WIPHY_ENTRY
+		WDEV_ENTRY
+	),
+	TP_fast_assign(
+		WIPHY_ASSIGN;
+		WDEV_ASSIGN;
+	),
+	TP_printk(WIPHY_PR_FMT ", " WDEV_PR_FMT,
+		  WIPHY_PR_ARG, WDEV_PR_ARG)
+);
+
 /*************************************************************
  *	     cfg80211 exported functions traces		     *
  *************************************************************/
Signature · 1 line
1.7.10.4
#1 lore raw
Quoted 22 lines · Arend van Spriel

Some protocols need a more reliable connection to complete successful in reasonable time. This patch adds a user-space API to indicate the wireless driver that a critical protocol is about to commence and when it is done, using nl80211 primitives NL80211_CMD_CRIT_PROTOCOL_START and NL80211_CRIT_PROTOCOL_STOP.

The driver can support this by implementing the cfg80211 callbacks .crit_proto_start() and .crit_proto_stop(). Examples of protocols that can benefit from this are DHCP, EAPOL, APIPA. Exactly how the link can/should be made more reliable is up to the driver. Things to consider are avoid scanning, no multi-channel operations, and alter coexistence schemes.

---
Here is version 3. I tried to deal with all the review comments I received, except for the one below:

Quoted 8 lines

Ben Greear said: "
Also, you would probably need to enforce in the kernel that only x out of y time in any given period can be locked, otherwise lots of different dhclient processes (perhaps erroneously spawned..or running on lots of different VIFs) could basically disable scanning or channel changes..."

Not sure how or even if we should deal with this. Any thoughts?

I think you could treat the start-critical-section as requests that may be silently (or otherwise) ignored by the driver.

At a minimum, you could keep a timestamp of the last 'stop' time, and ignore any requests that come in too soon after that stop.

Or to be more clever, keep last 5 or so start/stop times and make sure the system is on average available for non-critical stuff at least XX% of the time.

I can imagine wanting to support this in ath9k (or mac80211) with lots of virtual interfaces, and in this case we would definitely need some constraints to keep the system from constantly being in critical section. I guess this could be a driver issue, however...NICs that support only a single interface may not care about this...

Thanks,
Ben

Signature · 2 lines
Ben Greear <[email protected]>
Candela Technologies Inc  http://www.candelatech.com