| Message ID | 1279795286.12439.8.camel@jlt3.sipsolutions.net |
|---|---|
| State | RFC, archived |
| Delegated to: | David Miller |
| Headers | show |
Johannes Berg <johannes@sipsolutions.net> writes: > On Thu, 2010-07-22 at 03:35 -0700, Eric W. Biederman wrote: > >> >> The warning patch just makes things fail faster. Although I get some of the >> >> wireless interfaces for hwsim when I use this one. >> > >> > Hmm, I didn't. >> >> To be clear I just get hwsim0. Not wlan0 or wlan1. > > Ah, yes, but that's just a regular netdev, you can pretty much ignore > it. It just shows all hwsim traffic as it is on the "air" for sniffing. > >> > Right, it actually starts working again with that patch you sent. >> > However, netns support is really broken: >> > >> > <create net namespace, put phy0/wlan0 into it> >> >> Do we have a convenient command line tool to do this? >> I remember there being a different netlink message from >> normal network devices. > > iw phy0 set netns <pid> > > http://git.sipsolutions.net/iw.git > >> > root@kvm:~# ip link >> > 3: wlan0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 1000 >> > link/ether 02:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff >> > 7: lo: <LOOPBACK> mtu 16436 qdisc noop state DOWN >> > link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 >> > root@kvm:~# ls /sys/class/net/ >> > eth0 hwsim0 lo wlan1 wlan2 >> >> I think this is actually the output of something working. >> >> I expect after you created a new netns you didn't mount >> a new instance of /sys. /sys remembers which netns you >> had when you mounted it. So you have to mount /sys again >> so you can see the /sys/class/net for the network namespace >> you are in. > > Ohh, oops! I saw all the "current->" references in the code and somehow > expected the same instance of sysfs to show the right thing. > > Yes, it works now. But the patch below doesn't seem to work, am I > missing something? You are trying to move the phy devices as well? My guess is that at least part of the problem is that you don't have a ieee80211 directory under hwsim. My apologies for not thinking about the peculiarities of the wireless drivers. Eric > --- > include/linux/netdevice.h | 2 ++ > net/core/net-sysfs.c | 3 ++- > net/wireless/sysfs.c | 9 +++++++++ > 3 files changed, 13 insertions(+), 1 deletion(-) > > --- wireless-testing.orig/include/linux/netdevice.h 2010-07-22 10:01:22.000000000 +0200 > +++ wireless-testing/include/linux/netdevice.h 2010-07-22 10:11:00.000000000 +0200 > @@ -2148,6 +2148,8 @@ extern void dev_seq_stop(struct seq_file > extern int netdev_class_create_file(struct class_attribute *class_attr); > extern void netdev_class_remove_file(struct class_attribute *class_attr); > > +extern struct kobj_ns_type_operations net_ns_type_operations; > + > extern char *netdev_drivername(const struct net_device *dev, char *buffer, int len); > > extern void linkwatch_run_queue(void); > --- wireless-testing.orig/net/core/net-sysfs.c 2010-07-22 10:01:22.000000000 +0200 > +++ wireless-testing/net/core/net-sysfs.c 2010-07-22 10:11:51.000000000 +0200 > @@ -785,12 +785,13 @@ static const void *net_netlink_ns(struct > return sock_net(sk); > } > > -static struct kobj_ns_type_operations net_ns_type_operations = { > +struct kobj_ns_type_operations net_ns_type_operations = { > .type = KOBJ_NS_TYPE_NET, > .current_ns = net_current_ns, > .netlink_ns = net_netlink_ns, > .initial_ns = net_initial_ns, > }; > +EXPORT_SYMBOL_GPL(net_ns_type_operations); > > static void net_kobj_ns_exit(struct net *net) > { > --- wireless-testing.orig/net/wireless/sysfs.c 2010-07-22 10:01:22.000000000 +0200 > +++ wireless-testing/net/wireless/sysfs.c 2010-07-22 10:13:08.000000000 +0200 > @@ -110,6 +110,13 @@ static int wiphy_resume(struct device *d > return ret; > } > > +static const void *wiphy_namespace(struct device *d) > +{ > + struct wiphy *wiphy = container_of(d, struct wiphy, dev); > + > + return wiphy_net(wiphy); > +} > + > struct class ieee80211_class = { > .name = "ieee80211", > .owner = THIS_MODULE, > @@ -120,6 +127,8 @@ struct class ieee80211_class = { > #endif > .suspend = wiphy_suspend, > .resume = wiphy_resume, > + .ns_type = &net_ns_type_operations, > + .namespace = wiphy_namespace, > }; > > int wiphy_sysfs_init(void) -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Thu, 2010-07-22 at 04:27 -0700, Eric W. Biederman wrote: > >> Do we have a convenient command line tool to do this? > >> I remember there being a different netlink message from > >> normal network devices. > > > > iw phy0 set netns <pid> > > > > http://git.sipsolutions.net/iw.git > > > >> > root@kvm:~# ip link > >> > 3: wlan0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 1000 > >> > link/ether 02:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff > >> > 7: lo: <LOOPBACK> mtu 16436 qdisc noop state DOWN > >> > link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 > >> > root@kvm:~# ls /sys/class/net/ > >> > eth0 hwsim0 lo wlan1 wlan2 > >> > >> I think this is actually the output of something working. > >> > >> I expect after you created a new netns you didn't mount > >> a new instance of /sys. /sys remembers which netns you > >> had when you mounted it. So you have to mount /sys again > >> so you can see the /sys/class/net for the network namespace > >> you are in. > > > > Ohh, oops! I saw all the "current->" references in the code and somehow > > expected the same instance of sysfs to show the right thing. > > > > Yes, it works now. But the patch below doesn't seem to work, am I > > missing something? > > You are trying to move the phy devices as well? Yes. The intent is that each wireless phy lives in a netns along with all of its child devices. > My guess is that at least part of the problem is that you don't have a > ieee80211 directory under hwsim. But I should have? 'ieee80211' is a class just like 'net', no? > My apologies for not thinking about the peculiarities of the wireless > drivers. No worries. johannes -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
--- wireless-testing.orig/include/linux/netdevice.h 2010-07-22 10:01:22.000000000 +0200 +++ wireless-testing/include/linux/netdevice.h 2010-07-22 10:11:00.000000000 +0200 @@ -2148,6 +2148,8 @@ extern void dev_seq_stop(struct seq_file extern int netdev_class_create_file(struct class_attribute *class_attr); extern void netdev_class_remove_file(struct class_attribute *class_attr); +extern struct kobj_ns_type_operations net_ns_type_operations; + extern char *netdev_drivername(const struct net_device *dev, char *buffer, int len); extern void linkwatch_run_queue(void); --- wireless-testing.orig/net/core/net-sysfs.c 2010-07-22 10:01:22.000000000 +0200 +++ wireless-testing/net/core/net-sysfs.c 2010-07-22 10:11:51.000000000 +0200 @@ -785,12 +785,13 @@ static const void *net_netlink_ns(struct return sock_net(sk); } -static struct kobj_ns_type_operations net_ns_type_operations = { +struct kobj_ns_type_operations net_ns_type_operations = { .type = KOBJ_NS_TYPE_NET, .current_ns = net_current_ns, .netlink_ns = net_netlink_ns, .initial_ns = net_initial_ns, }; +EXPORT_SYMBOL_GPL(net_ns_type_operations); static void net_kobj_ns_exit(struct net *net) { --- wireless-testing.orig/net/wireless/sysfs.c 2010-07-22 10:01:22.000000000 +0200 +++ wireless-testing/net/wireless/sysfs.c 2010-07-22 10:13:08.000000000 +0200 @@ -110,6 +110,13 @@ static int wiphy_resume(struct device *d return ret; } +static const void *wiphy_namespace(struct device *d) +{ + struct wiphy *wiphy = container_of(d, struct wiphy, dev); + + return wiphy_net(wiphy); +} + struct class ieee80211_class = { .name = "ieee80211", .owner = THIS_MODULE, @@ -120,6 +127,8 @@ struct class ieee80211_class = { #endif .suspend = wiphy_suspend, .resume = wiphy_resume, + .ns_type = &net_ns_type_operations, + .namespace = wiphy_namespace, }; int wiphy_sysfs_init(void)