2008-12-05 15:58:18

by Alex Howells

Subject: sata_svw breakage

Afternoon Gents (and Ladies),

I'm setting an interesting problem with the sata_svw driver on a fairly
new (in terms of model) blade server from HP, a BL495c.

[ 3.462135] PCI: setting IRQ 14 as level-triggered
[ 3.462142] sata_svw 0000:01:0e.0: PCI INT A -> Link[ISF0] -> GSI 14
(level,
low) -> IRQ 14
[ 3.465367] sata_svw 0000:01:0e.0: PCI INT A disabled
[ 3.467302] sata_svw: probe of 0000:01:0e.0 failed with error -16

Having trouble tracking down what 'error -16' might imply, any pointers
for what the root cause may be and how to address it?

Regards,
Alex

2008-12-05 16:17:41

by Alex Howells

Subject: Re: sata_svw breakage

Alex Howells wrote:
> Afternoon Gents (and Ladies),
>
> I'm setting an interesting problem with the sata_svw driver on a fairly
> new (in terms of model) blade server from HP, a BL495c.
>
> [ 3.462135] PCI: setting IRQ 14 as level-triggered
> [ 3.462142] sata_svw 0000:01:0e.0: PCI INT A -> Link[ISF0] -> GSI 14
> (level,
> low) -> IRQ 14
> [ 3.465367] sata_svw 0000:01:0e.0: PCI INT A disabled
> [ 3.467302] sata_svw: probe of 0000:01:0e.0 failed with error -16
>
> Having trouble tracking down what 'error -16' might imply, any pointers
> for what the root cause may be and how to address it?

I should have noted this happens in the latest -stable and the problem
appears to persist if I go back a couple of versions to 2.6.25, etc.

Kernel config is over at:

http://havok.aeode.net/~ahowells/2.6.27.7-config

Cheers,
Alex

2008-12-05 17:00:20

by Jeff Garzik

Subject: Re: sata_svw breakage

Alex Howells wrote:
> Afternoon Gents (and Ladies),
>
> I'm setting an interesting problem with the sata_svw driver on a fairly
> new (in terms of model) blade server from HP, a BL495c.
>
> [ 3.462135] PCI: setting IRQ 14 as level-triggered
> [ 3.462142] sata_svw 0000:01:0e.0: PCI INT A -> Link[ISF0] -> GSI 14
> (level,
> low) -> IRQ 14
> [ 3.465367] sata_svw 0000:01:0e.0: PCI INT A disabled
> [ 3.467302] sata_svw: probe of 0000:01:0e.0 failed with error -16
>
> Having trouble tracking down what 'error -16' might imply, any pointers
> for what the root cause may be and how to address it?

That's the EBUSY error (include/asm-generic/errno-base.h).

What kernel? I don't see a 'probe of ... failed' message in drivers/ata
nor drivers/pci.

Jeff

2008-12-05 17:18:40

by Alex Howells

Subject: Re: sata_svw breakage

Jeff Garzik wrote:
> Alex Howells wrote:
>> Afternoon Gents (and Ladies),
>>
>> I'm setting an interesting problem with the sata_svw driver on a fairly
>> new (in terms of model) blade server from HP, a BL495c.
>>
>> [ 3.462135] PCI: setting IRQ 14 as level-triggered
>> [ 3.462142] sata_svw 0000:01:0e.0: PCI INT A -> Link[ISF0] -> GSI 14
>> (level,
>> low) -> IRQ 14
>> [ 3.465367] sata_svw 0000:01:0e.0: PCI INT A disabled
>> [ 3.467302] sata_svw: probe of 0000:01:0e.0 failed with error -16
>>
>> Having trouble tracking down what 'error -16' might imply, any pointers
>> for what the root cause may be and how to address it?
>
> That's the EBUSY error (include/asm-generic/errno-base.h).
>
> What kernel? I don't see a 'probe of ... failed' message in drivers/ata
> nor drivers/pci.

It's running on 2.6.27.7 at the moment...