bge0: EEPROM read timed

[pluknet]

2009/3/16 pluknet <pluknet at gmail.com>:
> Hi.
>
> I got this on today's RELENG_6 with Broadcom BCM5722 A0, ASIC rev. 0xa200.
>
> From dmesg (bge related):
>
> bge0: <Broadcom BCM5722 A0, ASIC rev. 0xa200> mem
> 0xe8400000-0xe840ffff irq 16 at device 0.0 on pci2
> bge0: firmware handshake timed out, found 0x4b657654
> bge0: firmware handshake timed out, found 0x4b657654
> bge0: EEPROM read timed out
> bge0: failed to read EEPROM
> device_attach: bge0 attach returned 6
> bge1: <Broadcom BCM5703 B0, ASIC rev. 0x1100> mem
> 0xe8600000-0xe860ffff irq 21 at device 1.0 on pci3
> miibus0: <MII bus> on bge1
> brgphy0: <BCM5703 10/100/1000baseTX PHY> on miibus0
> brgphy0:  10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseT,
> 1000baseT-FDX, auto
> bge1: Ethernet address: 00:21:5e:4d:05:c8
>
>
> any hints?
>
> P.S. I see EEPROM timeout fixes were already merged to RELENG_6 (I
> have post-fix version certainly).
> May that issue be somehow related?
>

I guess it's a regression. Below are my speculations.

I tried to build on 6.2-R the bge(4) sources checked from later RELENG_6
just after BCM5722 support (from if_bgereg.h 1.36.2.11/ if_bge.c1.91.2.26)
in order to backport BCM5722 support into 6.2-R. After some tweaks it
was built, so..

What I got in dmesg (after native statically built bge(4) replacement
in boot loader prompt) is:
FreeBSD 6.2-RELEASE-p8 #13: Thu Feb 19 14:52:30 MSK 2009
[..snip..]
bgex0: <Broadcom BCM5722 A0, ASIC rev. 0xa200> mem
0xe8400000-0xe840ffff irq 16 at device 0.0 on pci2
bgex0: Ethernet address: 00:21:5e:4d:05:c7

Whoohoo..

So, here it even reached macaddr designation. (and going to panic at
nfs boot time (probably
to different locking scheme between 6.2 and 6.4, but it's an another story)).


-- 
wbr,
pluknet