Re: [PATCH net v3] r8169:fix "rtl_counters_cond == 1 (loop: 1000, delay: 10)" log spam.

[Francois Romieu]

Chunhao Lin <hau@xxxxxxxxxxx> :
[...]
> diff --git a/drivers/net/ethernet/realtek/r8169.c b/drivers/net/ethernet/realtek/r8169.c
> index 537974c..404be51 100644
> --- a/drivers/net/ethernet/realtek/r8169.c
> +++ b/drivers/net/ethernet/realtek/r8169.c
[...]
> @@ -7853,6 +7859,11 @@ static int rtl8169_runtime_suspend(struct device *device)
>  
>  	rtl8169_net_suspend(dev);
>  
> +	/* Update counters before going runtime suspend */
> +	if (netif_running(dev))

This test is useless (always true):

- rtl_open
  [...]
  pm_runtime_get_sync(&pdev->dev);
  [...]
  tp->TxDescArray = blah
  [...]

- rtl8169_close
  [...]
  pm_runtime_get_sync(&pdev->dev);
  [...]
  tp->TxDescArray = NULL;

- rtl8169_runtime_suspend
  [...]
  if (!tp->TxDescArray)
  	  return 0;

(the implicit smp barriers are mildly obvious, ok)

-- 
Ueimor