Interrupt routing and friends ...

[Sepherosa Ziehau]

On Mon, Nov 10, 2008 at 1:09 PM, Bill Hacker <wbh at conducive.org> wrote:
> Sepherosa Ziehau wrote:
>
> *trimmed*
>
>
>>>>>  Lenovo 3000 G400, DFLY 2.0.1_REL not using Broadcom WiFi or 10/100
>>>>> NIC)
>
>> What's the vendor id and product id of this bge?
>>
>
> From such dmesg as I *can* get sight of, I see:
>
> =======
>
> The WiFi:
>
> NetBSD 4.0.1 - reports, doesn't use:
>
> Broadcom BCM4311 (miscellaneous network. revision 0x01) at pci2 dev 0
> function 0 not configured
>
> OpenBSD 4.4 release reports and uses:
>
> bwi0 at pci2 dev 0 function 0 "Broadcom BCM4311" rev 0x01: irq 11, address

We have bwi(4); this driver was actualy written by me, i.e. it is from dfly

> ......
>
> http://www.broadcom.com/products/Wireless-LAN/802.11-Wireless-LAN-Solutions/BCM4311
>
> ======
>
> The 10/100 NIC:
>
> NetBSD 4.0.1 - reports, doesn't use:
>
> Broadcom BCM5906M netlink Fast Ethernet (ethernet network, revision 0x02) at
> pci 3 dev 0 function 0 not configured
>
> OpenBSD 4.4 release reports and uses:
>
> bge0 at pci3 dev 0 function 0 "Broadcom BCM5906M" rev 0x02, BCM A2
> (0x0c002): irq 11, address ......

OK, we have this on HEAD.  Thank you for the information.

>
> ======
>
> ~bridge chipset is reported as Intel 82801GBM - various rev levels for each
> of the functions as they report in.
>
>>
>> If possible, could you post the backtrace?  I think its probably the
>> same as we had seen about aue(4) several months ago.
>>
>> Best Regards,
>> sephe
>>
>
> 'working on it'. I need to get around the lack of working external I/O.
>
> Meanwhile - thoigh I'm sure the issue is on the planar & silicon, not the
> plug-in, here is the ID of the old Eagletec USB NIC that DFLY set up, then
> dropped into debugger when an outbound  'scp ...' was initiated

Due to missing assertion in if_start in various USB NIC drivers, we
didn't catch the possible bug.  I will try to fix it before next
release.

Best Regards,
sephe

-- 
Live Free or Die