Re: Ralink broken after last update

[Erki Malling]

Sure.

With rssadapt:

# uptime
4:25AM  up 3 days, 16:38, 2 users, load averages: 0.26, 0.17, 0.14
# netstat -i
NameMtu   Network Address  Ipkts IerrsOpkts Oerrs Colls
lo0 33224 Link   26923 026923 0 0
lo0 33224 loopbacklocalhost.my.doma26923 026923 0 0
lo0 33224 localhost.m localhost.my.doma26923 026923 0 0
lo0 33224 fe80::%lo0/ fe80::1%lo0  26923 026923 0 0
rl0 1500  Link  00:50:bf:67:4b:c5  6226370 0  7564081 0 0
rl0 1500  fe80::%rl0/ fe80::250:bfff:fe  6226370 0  7564081 0 0
rl0 1500  ip0.cab60.m ip41.cab60.mus.st  6226370 0  7564081 0 0
ral01500  Link  00:0e:2e:86:87:7621153  836324635  1796 0
ral01500  192.168.2/2 192.168.2.1  21153  836324635  1796 0
ral01500  fe80::%ral0 fe80::20e:2eff:fe21153  836324635  1796 0
fxp01500  Link  00:02:b3:95:b3:a0   990951 0  1620722 0 0
fxp01500  192.168.1/2 192.168.1.1 990951 0  1620722 0 0
fxp01500  fe80::%fxp0 fe80::202:b3ff:fe   990951 0  1620722 0 0
pfsync0 1460  Link   0 00 0 0
enc0*   1536  Link   0 00 0 0
# ifconfig -M ral0
ral0: flags=8943UP,BROADCAST,RUNNING,PROMISC,SIMPLEX,MULTICAST mtu 1500
   lladdr 00:0e:2e:86:87:76
   media: IEEE802.11 autoselect mode 11b hostap
   status: active
   ieee80211: nwid OPENBOX chan 1 bssid 00:0e:2e:86:87:76 100dBm
   none
   inet 192.168.2.1 netmask 0xff00 broadcast 192.168.2.255
   inet6 fe80::20e:2eff:fe86:8776%ral0 prefixlen 64 scopeid 0x2

With AMRR. Rebuilt from CVS couple of hours ago.

# uptime
1:37PM  up  1:14, 2 users, load averages: 0.56, 0.50, 0.54
# netstat -i
NameMtu   Network Address  Ipkts IerrsOpkts Oerrs Colls
lo0 33224 Link 457 0  457 0 0
lo0 33224 loopbacklocalhost.my.doma  457 0  457 0 0
lo0 33224 localhost.m localhost.my.doma  457 0  457 0 0
lo0 33224 fe80::%lo0/ fe80::1%lo0457 0  457 0 0
rl0 1500  Link  00:50:bf:67:4b:c510357 0 9590 0 0
rl0 1500  fe80::%rl0/ fe80::250:bfff:fe10357 0 9590 0 0
rl0 1500  ip0.cab16.m ip220.cab18.mus.s10357 0 9590 0 0
ral01500  Link  00:0e:2e:86:87:76   501773   422  1002191   159 0
ral01500  192.168.2/2 192.168.2.1 501773   422  1002191   159 0
ral01500  fe80::%ral0 fe80::20e:2eff:fe   501773   422  1002191   159 0
fxp01500  Link  00:02:b3:95:b3:a0 2593 0 2883 0 0
fxp01500  192.168.1/2 192.168.1.1   2593 0 2883 0 0
fxp01500  fe80::%fxp0 fe80::202:b3ff:fe 2593 0 2883 0 0
pfsync0 1460  Link   0 00 0 0
enc0*   1536  Link   0 00 0 0
# ifconfig -M ral0
ral0: flags=8843UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST mtu 1500
   lladdr 00:0e:2e:86:87:76
   media: IEEE802.11 autoselect mode 11b hostap
   status: active
   ieee80211: nwid OPENBOX chan 1 bssid 00:0e:2e:86:87:76 100dBm
   lladdr 00:16:ce:21:e9:39 112dB 11M
short_preamble,short_slottime assoc
   inet 192.168.2.1 netmask 0xff00 broadcast 192.168.2.255
   inet6 fe80::20e:2eff:fe86:8776%ral0 prefixlen 64 scopeid 0x2

I left the lappy downloading stuff and it's been going at rock solid 750KB/s
for the last hour or so.

I think it has never worked so well.

Thank you


Could you please run the following commands on your ral-based
access point, preferably  before and after the rssadapt-AMRR
change:

# netstat -i
# ifconfig -M ral0

Ralink broken after last update

[Erki Malling]

I'm running Canyon CN-WF511 wireless PCI card (Ralink RT2561S)
in hostap mode in my 30m2 flat, serving one laptop next room.
The wireless quality has always been less than stellar, even
when few meters from the OpenBSD box antenna. But it works for
reading mail and news and thats all that matters in this case.

Unfortunately, after updating last week (I'm following -current)
things got a bit worse. When in the same room with the AP, it works
as craptastically as before, but further away (say 5 meters and a
concrete wall between) things slowly cripple to the point when laptop
can't even find the AP.

Long story short, looks like this update is to blame here:

replace rssadapt(9) with amrr for automatic rate control.
as a side-effect, this should fix all the bogus xmit rate panics
users have been complaining about for some time when operating in
HostAP mode.

I grabbed previous versions of relevant files from cvsweb and
rebuilt the kernel.

Things work well now.

Btw, I, too, suffered kernel panics before but forcing mode 11b in
hostname.ral0 took care of that.

Here's my dmesg: http://doktor.kompro.ee/dmesg

PS. I stared at the output of tcpdump -i ral0 -y IEEE802_11_RADIO
when it was broken but could'nt make any sense of it.
If someone would tell me what to look for there, I'm willing
to break things again and investigate.