On 18/08/08 22:13, Rafael J. Wysocki wrote:
> On Monday, 18 of August 2008, Andrew Morton wrote:
>> On Tue, 05 Aug 2008 20:29:44 +0100
>> Simon Arlott <simon@fire.lp0.eu> wrote:
>> >
>> 
>> [two weeks pass...]
> 
> [Must have missed this message.]
> 
> I have a box with forcedeth that evidently works after a resume from
> hibernation.  Unfortunately, so far I haven't been able to make the box

With hibernation would it re-initialise the devices differently?

> resume from suspend to RAM.  I'll do my best to try again tomorrow, but there's
> a little hope. :-(

Mine doesn't either, try standby?

>> > My NIC stops working after resuming from standby, it's not receiving any interrupts:
>> > Commit 25d90810ff49d2a63475776f24c74c6bb49b045f ([netdrvr] forcedeth: reorder suspend/resume code) 
>> > introduces pci_disable_device to nv_suspend, but there's no corresponding pci_enable_device in 
>> > nv_resume - so I added one (copied from e1000):
>> > 
>> > diff --git a/drivers/net/forcedeth.c b/drivers/net/forcedeth.c
>> > index 01b38b0..db4f875 100644
>> > --- a/drivers/net/forcedeth.c
>> > +++ b/drivers/net/forcedeth.c
>> > @@ -5922,6 +5922,13 @@ static int nv_resume(struct pci_dev *pdev)
>> >  
>> >  	pci_set_power_state(pdev, PCI_D0);
>> >  	pci_restore_state(pdev);
>> > +	rc = pci_enable_device(pdev);
>> > +	if (rc) {
>> > +		printk(KERN_ERR "forcedeth: Cannot enable PCI device from suspend\n");
>> > +		return rc;
>> > +	}
>> > +	pci_set_master(pdev);
>> > +
>> >  	/* ack any pending wake events, disable PME */
>> >  	pci_enable_wake(pdev, PCI_D0, 0);
>> >  
>> 
>> That seems like a sensible change.
>> 

With this applied, I can resume from standby *without MSI* and the 
NIC still works. I haven't tested it without MSI and without the 
patch... mostly because I got a BUG when I tried to recompile.

>> > This results in interrupts being re-enabled after suspend:
>> > However, the NIC still doesn't work after resume.
> 
> Simon, I'd prefer the full dmesg to the grepped forcedeth messages.
> 

Attached. (The previous email has all the standby/resume log output.)

> I guess this was resume from suspend to RAM?

Standby - resume from RAM is completely broken for my system.

>> Is it still broken in current kernels?
> 
> It probably is.

If I have MSI enabled, it still doesn't work.

-- 
Simon Arlott