tg3/Broadcom gigabit driver just got worse in 2.4.23-pre3

[walt]

Hi Jeff et all,

I just tried 2.4.23-pre3 with results that are disastrous, for me at least.

As you will remember, I'm the one who has to do an ifconfig down/up cycle
on my asus A7V8X mobo with built-in Broadcom chip.  But after the updates
in -pre3 the chip no longer will work at all.

In fact, if I try 'ifconfig eth0 down' the command hangs forever and chews
up 99.9% of the CPU.  No packets are ever transmitted in spite of a normal
'ifconfig' output after bootup.  The chip is correctly identified in dmesg:

eth0: Tigon3 [partno(BCM95702A20) rev 1002 PHY(5703)] (PCI:33MHz:32-bit)
10/100/1000BaseT Ethernet 00:e0:18:d2:a6:c1

and lspci:
00:09.0 Ethernet controller: Broadcom Corporation NetXtreme BCM5702 Gigabit Ethernet (rev 02)
        Subsystem: Asustek Computer, Inc.: Unknown device 80a9
        Flags: bus master, 66Mhz, medium devsel, latency 64, IRQ 10
        Memory at f1800000 (64-bit, non-prefetchable) [size=64K]
        Expansion ROM at f7ff0000 [disabled] [size=64K]
        Capabilities: [40] PCI-X non-bridge device.
        Capabilities: [48] Power Management version 2
        Capabilities: [50] Vital Product Data
        Capabilities: [58] Message Signalled Interrupts: 64bit+ Queue=0/3 Enable-

The data above were taken from 2.6.0-test4, but are the same in 2.4.23-pre3.

I would ask that you please not apple these recent changes to 2.6.0 until
we figure out what the problem is.

Again, I offer to do whatever testing/patching/whatever that I can do to
get this resolved, but you'll need to give me some specific instructions.

Thanks!

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/