Re: [E1000-devel] igb driver failing at probe

[Flavio L. <fb...@re...>]

On Thu, Jan 31, 2013 at 07:00:08PM -0200, Flavio Leitner wrote:
> 
> Hi,
> 
> It was told about a system failing to probe igb device:
> ...
> igb 0000:04:00.1: PCI INT B -> GSI 44 (level, low) -> IRQ 44
> mlx4_en: eth0: Link Down
> igb 0000:04:00.1: setting latency timer to 64
> igb 0000:04:00.1: PCI INT B disabled
> igb: probe of 0000:04:00.1 failed with error -2
>   alloc irq_desc for 45 on node -1
>   alloc kstat_irqs on node -1
> igb 0000:04:00.2: PCI INT C -> GSI 45 (level, low) -> IRQ 45
> igb 0000:04:00.2: setting latency timer to 64
> igb 0000:04:00.2: PCI INT C disabled
> igb: probe of 0000:04:00.2 failed with error -2
> ...
> 
> The error -2 seems to not be the standard error, but instead
> I suspect it is E1000_ERR_PHY:
> 
> #define E1000_ERR_PHY      2
> 
> and it is returned if igb_get_invariants() fails to identify
> the phy->id or hw->mac.type.

I found this while searching a bit in the mailing list archives 
http://www.mail-archive.com/e10...@li.../msg04674.html

In my case, the hardware is Radisys XE100 ATCA and Emerson 7370 ATCA blades
so I believe the problem happens due to the unsupported Broadcom PHY.
Is that correct?

If so, what about that statement about having no plans to support it?
Is it still valid?

Thank you!
-- 
fbl