On Sat, 3 Sep 2005, Matthias Scheler wrote:

> In article <B137DA0C-534E-4AB1-948D-B330A5C2D785@dtcc.edu>,
> 	Chris Tribo <ctribo@dtcc.edu> writes:
> > PXE boot rom and Windows dual-boot are working fine.
>
> Does NetBSD works with the card if you enable PXE, let fail and the
> system boots from disk afterwards?

I haven't yet been able to reproduce the failure. I was running NetBSD
without the network cable plugged in, brought it back to my desk and
plugged in the cable and then it wouldn't work until I booted windows.
I've had no previous problems with the card for the better part of 5
months.

> > ...
> > bge0: watchdog timeout -- resetting
> > bge0: block failed to stop: reg 0x1400, bit 0x00000002
> > bge0: watchdog timeout -- resetting
> > ...
>
> Could you please post the message that the kernel prints out when it
> detects the card?

Sure:

ppb1 at pci0 dev 28 function 0: Intel 82801FB/FR PCI Express Port #0 (rev.
0x03)
pci2 at ppb1 bus 2
pci2: i/o space, memory space enabled, rd/line, wr/inv ok
bge0 at pci2 dev 0 function 0: Broadcom BCM5751 Gigabit Ethernet
bge0: interrupting at ioapic0 pin 16 (irq 11)
bge0: ASIC BCM5750 A1 (0x4001), Ethernet address 00:11:43:xx:xx:xx
brgphy0 at bge0 phy 1: BCM5750 1000BASE-T media interface, rev. 0
brgphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseT,
1000baseT-FDX, auto

> > To make matters more strange, booting windows seems to have fixed bge
> > in NetBSD. Now it's working again when it wasn't previously, after
> > about 10 cold boots of trying different kernels.
>
> I don't think that is strange. I suspect that our bge driver doesn't
> initialize the chip properly.
>
> 	Kind regards

bge0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
capabilities=3f00<IP4CSUM_Rx,IP4CSUM_Tx,TCP4CSUM_Rx,TCP4CSUM_Tx,UDP4CSUM_Rx,
UDP4CSUM_Tx>
enabled=3f00<IP4CSUM_Rx,IP4CSUM_Tx,TCP4CSUM_Rx,TCP4CSUM_Tx,UDP4CSUM_Rx,
UDP4CSUM_Tx>