Re: wpa_supplicant causes panic in ieee80211_newstate

> > The good news is that ndisgen and kldload now don't panic the system. The
> > bad news is that it still doesn't work with the WG311 drivers.
> 
> P.S. I also got the latest drivers from the web:
> http://kbserver.netgear.com/products/WG311v2.asp
> ftp://downloads.netgear.com/files/wg311v2_v2_0_0_7.zip   [about 16MB]
> 
> but these turn out to be identical to the ones on the CD (and unpack into a
> subdirectory called 1.2B6, which is rather strange for version 2.0.0.7)
> 
> I tried renaming /compat/ndis to /compat/ndis.hold, and reloading the
> module. No additional error messages are reported - so it looks like it's
> not even getting far enough to need to load the firmware files.
> 
> I also tried using the Windows XP driver, instead of Win2K. I get almost the
> same:
> 
> May  9 17:48:56 foo kernel: ndis0: <NETGEAR WG311v2 802.11g Wireless PCI Adapter> mem 0xfaffc000-0xfaffdfff,0xfafc0000-0xfafdffff irq 11 at device 11.0 on pci2
> May  9 17:48:56 foo kernel: ndis0: [GIANT-LOCKED]
> May  9 17:48:56 foo kernel: ndis0: NDIS API version: 5.1
> May  9 17:48:56 foo kernel: ndis0: init handler failed
> May  9 17:48:56 foo kernel: device_attach: ndis0 attach returned 6
> 
> The only difference is the NDIS API is version 5.1 instead of 5.0
> 
> Regards,
> 
> Brian.

Uhm... hm. I don't know how to tell you this, but I can't reproduce your
problem. I plugged my WG311v2 NIC into my dual Opteron system running
6.0-SNAP002 (current snapshot from march) and the latest code from
/sys/compat/ndis and /sys/dev/if_ndis (as of today), and the NIC is
detected and works fine:

ndis0: <NETGEAR WG311v2 802.11g Wireless PCI Adapter> mem 0xe2020000-0xe203ffff,
0xe2000000-0xe2001fff irq 36 at device 3.0 on pci19
ndis0: Reserved 0x20000 bytes for rid 0x14 type 3 at 0xe2020000
ndis0: Reserved 0x2000 bytes for rid 0x10 type 3 at 0xe2000000
ioapic4: routing intpin 0 (PCI IRQ 36) to cluster 0
ndis0: [MPSAFE]
ndis0: NDIS API version: 5.1
ndis0: bpf attached
ndis0: Ethernet address: 00:09:5b:bb:7b:a0
ndis0: bpf attached

ndis0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
        inet6 fe80::209:5bff:febb:7ba0%ndis0 prefixlen 64 scopeid 0x4 
        inet 10.0.1.2 netmask 0xffffff00 broadcast 10.0.1.255
        ether 00:09:5b:bb:7b:a0
        media: IEEE 802.11 Wireless Ethernet autoselect (DS/11Mbps <adhoc>)
        status: associated
        ssid "SHIPLEY SUCKS" bssid 56:4f:91:fa:a4:33
        authmode OPEN privacy OFF txpowmax 100

opteron# ifconfig ndis0 scan
SSID            BSSID              CHAN RATE  S:N   INT CAPS
SHIPLEY SUCKS   56:4f:91:fa:a4:33    1   54M 190:0   100 I    ??? ??? ??? ???

(Even better, it looks like I fixed the problem with the damn signal
strength values being all messed up.)

I used the Windows XP driver from the wg311v2_v2_0_0_7.zip distribution
and gave ndisgen all three .BIN files. (Note: experimentation has shown
that the driver really only opens one of the files. I think it picks
the right image depending on which Texas Instruments chip it finds.)

opteron# pwd
/u/wpaul/1.2B6/Driver/Windows XP
opteron# ls -l
total 2464
-rw-rw-rw-  1 wpaul  wpaul   76568 Apr  4  2004 FW1130.BIN
-rwxr-xr-x  1 wpaul  wpaul   78112 May  9 12:18 FW1130.BIN.ko
-rw-r--r--  1 wpaul  wpaul   76953 May  9 12:18 FW1130.BIN.o
-rw-rw-rw-  1 wpaul  wpaul   83320 Apr  4  2004 FwRad16.bin
-rwxr-xr-x  1 wpaul  wpaul   84870 May  9 12:18 FwRad16.bin.ko
-rw-r--r--  1 wpaul  wpaul   83707 May  9 12:18 FwRad16.bin.o
-rw-rw-rw-  1 wpaul  wpaul   84912 Apr  4  2004 FwRad17.bin
-rwxr-xr-x  1 wpaul  wpaul   86462 May  9 12:18 FwRad17.bin.ko
-rw-r--r--  1 wpaul  wpaul   85299 May  9 12:18 FwRad17.bin.o
-rw-rw-rw-  1 wpaul  wpaul    9135 Oct  4  2004 netwg311.cat
-rw-rw-rw-  1 wpaul  wpaul  386688 Jun 17  2004 netwg311_XP.sys
-rwxr-xr-x  1 wpaul  wpaul  644894 May  9 12:18 netwg311_XP_sys.ko
-rw-r--r--  1 wpaul  wpaul  642337 May  9 12:18 netwg311_XP_sys.o
-rw-rw-rw-  1 wpaul  wpaul    9157 Sep  1  2004 wg311v2.inf

I'm not sure what to tell you at this point. It's possible there's
something broken in -current that's hosing Project Evil. I won't be
sure until I can get a -current snapshot running at home. You might
want to try some simple diagnostics, like moving the NIC to another
PCI slot.

Speaking of snapshots, does anyone know of a FreeBSD snapshot building
host that actually works? snapshots.jp.freebsd.org hasn't built a 
proper -current snapshot distribution in over 5 months, and
snapshots.se.freebsd.org seems to have vanished.

-Bill

--
=============================================================================
-Bill Paul            (510) 749-2329 | Senior Engineer, Master of Unix-Fu
                 wpaul_at_windriver.com | Wind River Systems
=============================================================================
              <adamw> you're just BEGGING to face the moose
=============================================================================