Re: [PATCH v3 1/4] PCI: Keep AER status in pci_restore_state()

[Sathyanarayanan Kuppuswamy <sathyanarayanan.kuppuswamy@linux.intel.com>]

Hi Kai,

On 4/20/23 5:59 AM, Kai-Heng Feng wrote:
> When AER is using the same IRQ as PME, AER interrupt is treated as a
> wakeup event and it can disrupt system suspend process.
> 
> If that happens, the system will report it's woken up by PME IRQ without
> indicating any AER error since AER status is cleared on resume.
> 
> So keep the AER status so users can know the system is woken up by AER
> instead of PME.
> 
> Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
> Signed-off-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
> ---

Any history on why it is cleared before? Is it done to hide some resume
issues?

> v3:
>  - No change.
> 
> v2:
>  - New patch.
> 
>  drivers/pci/pci.c | 1 -
>  1 file changed, 1 deletion(-)
> 
> diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
> index 7a67611dc5f4..71aead00fc20 100644
> --- a/drivers/pci/pci.c
> +++ b/drivers/pci/pci.c
> @@ -1778,7 +1778,6 @@ void pci_restore_state(struct pci_dev *dev)
>  	pci_restore_dpc_state(dev);
>  	pci_restore_ptm_state(dev);
>  
> -	pci_aer_clear_status(dev);
>  	pci_restore_aer_state(dev);
>  
>  	pci_restore_config_space(dev);

-- 
Sathyanarayanan Kuppuswamy
Linux Kernel Developer