Re: [PATCH 3/7] PCI PM: Fix saving of device state in pci_legacy_suspend

[Frans Pop <elendil@planet.nl>]

Benjamin Herrenschmidt wrote:
> Minor nit: Should the warning be preceeded by a message ? The reason is,
> right now, all we get is a backtrace, it doesn't actually tell you which
> device or driver caused it which makes it pretty pointless.
> 
> I think you should add a printk(KERN_ERR... just before that which gives
> those informations along with a little blurb along the lines of "driver
> changed device state without saving config space state").

That is actually already included in the patch (through drv->suspend):

> -             if (WARN_ON_ONCE(pci_dev->current_state != PCI_D0))
> +             if (pci_dev->current_state != PCI_D0
> +                 && pci_dev->current_state != PCI_UNKNOWN) {
> +                     WARN_ONCE(pci_dev->current_state != prev,
> +                             "PCI PM: Device state not saved by %pF\n",
> +                             drv->suspend);

Cheers,
FJP