Re: [PATCH 1/1] pci: pciehp: Handle MRL interrupts to enable slot for hotplug.

[Kuppuswamy, Sathyanarayanan]

Hi,

On 11/19/20 2:08 PM, Raj, Ashok wrote:
> > If an Attention Button is present, the current behavior is to bring up
> > the hotplug slot as soon as presence or link is detected.  We don't wait
> > for a button press.  This is intended as a convience feature to bring up
> > slots as quickly as possible, but the behavior can be changed if the
> > button press and 5 second delay is preferred.
> No personal preference, I thought that is how the code in Linux was
> earlier.
>
> Looks like we still don't subscribe to PDC if ATTN is present? So you don't
> get an event until someone presses the button to process hotplug right?
>
> pciehp_hpc.c:pcie_enable_notification()
> {
> ....
>
>           * Always enable link events: thus link-up and link-down shall
>           * always be treated as hotplug and unplug respectively. Enable
>           * presence detect only if Attention Button is not present.
>           */
>          cmd = PCI_EXP_SLTCTL_DLLSCE;
>          if (ATTN_BUTTN(ctrl))
>                  cmd |= PCI_EXP_SLTCTL_ABPE;
>          else
>                  cmd |= PCI_EXP_SLTCTL_PDCE;
> ....
> }
>
>
> Looks like we still wait for button press to process. When you don't have a
> power control yes the DLLSC would kick in and we would process them. but if
> you have power control, you won't turn on until the button press?
Yes, as per current logic, if attention button is present, then we don't
subscribe to PDC event changes. we wait for button press  to turn on/off
the slot.

--
Sathyanarayanan Kuppuswamy
Linux Kernel Developer