Re: [2.6.26.*] boot problem (ahci/irq related?)

["Rafael J. Wysocki" <rjw@sisk.pl>]

On Tuesday, 12 of August 2008, Maciej Rutecki wrote:
> 2008/8/12 Alan Cox <alan@lxorguk.ukuu.org.uk>:
> >> > +sr 4:0:0:0: Attached scsi CD-ROM sr0
> >>
> >> looks like you are using ata driver for pata device instead of ide driver.
> >>
> >> so could be caused by bug in ata_piix...
> >
> > See the original bug report - he tried both and got the same kind of
> > behaviour.
> >
> >
> 
> Yes I have similar behaviour, very often kernel stops on this message:
> http://unixy.pl/maciek/download/kernel/dupa/2.6.26.2_config_ide/img_0001.jpg

In both cases one of the last messages is

ACPI: PCI Interrupt 0000:00:1f.2[B] -> GSI 17 (level, low) -> IRQ 17

and that comes from AHCI,  or rather from pci_enable_device() invoked by
AHCI.  Then, there are no more messages from AHCI, although there should
be at least 

ahci 0000:00:1f.2: AHCI 0001.0100 32 slots 4 ports 1.5 Gbps 0x1 impl SATA mode
ahci 0000:00:1f.2: flags: 64bit ncq ilck stag pm led clo pmp pio slum part 
PCI: Setting latency timer of device 0000:00:1f.2 to 64

So, it seems like the enabling of AHCI interrupt causes the failure and since
that's not 100% reproducible, it indicates a timing issue.

Thanks,
Rafael