Re: [PATCH] 8139too not running s3 suspend/resume pci fix

[Felipe W Damasio <felipewd@terra.com.br>]

	Hi Adrian,

Adrian Yee wrote:
> +	pci_set_power_state (pdev, 3);
> +	pci_save_state (pdev, tp->pci_state);
> +
>  	if (!netif_running (dev))
>  		return 0;
>  
> @@ -2571,9 +2574,6 @@ static int rtl8139_suspend (struct pci_d
>  
>  	spin_unlock_irqrestore (&tp->lock, flags);
>  
> -	pci_set_power_state (pdev, 3);
> -	pci_save_state (pdev, tp->pci_state);
> -
>  	return 0;
>  }

	IMHO, there's no problem in doing "pci_save_state (pdev, 
tp->pci_state)" before the suspend code..but I'm more confortable with 
leaving the set_power_state at the end of that path, since if the 
interface is down, we don't want to leave it in cold state.

> @@ -2583,10 +2583,10 @@ static int rtl8139_resume (struct pci_de
>  	struct net_device *dev = pci_get_drvdata (pdev);
>  	struct rtl8139_private *tp = dev->priv;
>  
> -	if (!netif_running (dev))
> -		return 0;
>  	pci_restore_state (pdev, tp->pci_state);
>  	pci_set_power_state (pdev, 0);
> +	if (!netif_running (dev))
> +		return 0;
>  	rtl8139_init_ring (dev);
>  	rtl8139_hw_start (dev);
>  	netif_device_attach (dev);

	Same thing here.

	Cheers,

Felipe