7.1-R to RELENG_7 upgrade breaks re nic

[Pyun YongHyeon]

On Wed, Feb 25, 2009 at 11:06:45PM -0600, Matthew D. Fuller wrote:
> On Wed, Feb 25, 2009 at 10:47:07PM -0500 I heard the voice of
> Steve Wills, and lo! it spake thus:
> > 
> > re0: <RealTek 8168/8168B/8168C/8168CP/8168D/8111B/8111C/8111CP PCIe  
> > Gigabit Ethernet> port 0x7e00-0x7eff mem 0xfd3ff000-0xfd3fffff, 
> > 0xfd3f8000-0xfd3fbfff irq 16 at device 0.0 on pci8
> > re0: Chip rev. 0x28000000
> > re0: MAC rev. 0x00100000
> 
> For a data point, I have
> 
> re0: <RealTek 8168/8168B/8168C/8168CP/8168D/8111B/8111C/8111CP PCIe
>      Gigabit Ethernet> port 0xdc00-0xdcff mem 0xfdfff000-0xfdffffff
>      irq 19 at device 0.0 on pci2
> re0: turning off MSI enable bit.
> re0: Chip rev. 0x38000000
> re0: MAC rev. 0x00000000
> 
> on a recent 7-STABLE (if_re.c,v 1.95.2.41) which is working OK.  I
> have gotten a few Tx interrupt watchdog timeouts, which I don't recall
> seeing in any quantity before, but they don't seem to cause any

If the watchdog timeout message contains "missed Tx interrupts" you
can safely ignore them. It's not real watchdog timeouts.

> problems.  It was previously on a ~Dec 2007 RELENG_7 and worked fine
> there (and had MSI enabled, as well).
> 

Try re(4) in HEAD. You may not see watchdog timeouts anymore.