Re: 2.2.14pre17 hangup with SMP during ifconfig

[Wakko Warner]

> > > I'm using a 3x905C with a driver module 3c90x supplied
> > > by 3com, so may be there is a bug within this module which is not
> > > part of the official linux kernel. I traced it down to the last
> > > system call of "ifconfig eth0 down":
> > 
> > That's a bug in the 3Com driver: someone else reported that a few days
> > ago, and IIRC the bug is caused by a wrong disable_irq() call. Could you
> > try the default driver from the kernel code?
> 
> The kernel driver does not support the 905C. Currently this card cannot be
> used on an SMP Linux box. Swapping the disable_irq for a disable_irq_nosync
> will probably fix it. I passed the original thing on to 3Com but I've had
> no ack for it yet (possibly due to christmas etc)

Are you sure about this alan?  I've used 2 3c905c cards under linux /w the
kernel drivers.  I'm not sure what 2.2 version started to support it. 
2.3.31 and up also supports it.  Please tell me I'm dreaming:

eth1: 3Com 3c905C Tornado at 0x6800,  00:50:da:84:52:9a, IRQ 12
  8K byte-wide RAM 5:3 Rx:Tx split, autoselect/Autonegotiate interface.
  MII transceiver found at address 24, status 782d.
  Enabling bus-master transmits and whole-frame receives.

# lspci -S 0b -n
00:0b.0 Class 0200: 10b7:9200 (rev 6c)
# lspci -S 0b -v
00:0b.0 Ethernet controller: 3Com Corporation: Unknown device 9200 (rev 6c)
        Flags: bus master, medium devsel, latency 64, IRQ 12
        I/O ports at 6800
        Memory at e0000000 (32-bit, non-prefetchable)

I was using an out of date lspci by the way.

-- 
 Lab tests show that use of micro$oft causes cancer in lab animals