Re: Regression (sort of): PCI/portdrv: Turn off PCIe services during shutdown

[Sinan Kaya <okaya@codeaurora.org>]

Hi Lukas,

On 1/12/2018 5:49 AM, Lukas Wunner wrote:
> Hi Sinan,
> 
> I updated to 4.15 this week and noticed a regression (if one may call it
> that, it's more like a very noticeable annoyance) affecting machines
> with Thunderbolt controllers wherein a reboot or shutdown is delayed by
> 8 seconds, caused by

Sorry for that.

I wonder if we can separate remove from shutdown and just disable the IRQs
in shutdown case rather than turning off the slot power etc.

Let me see if I can come up with a quick patch.

> On the one hand, your commit message sounds as though the change merely
> addresses a *potential* issue, so one could argue that if it causes
> *real* issues such as delayed reboots, it's probably not a good idea.
> On the other hand I do see your point since a user might surprise-remove
> a device from a hotplug slot during shutdown, and we clearly wouldn't want
> to act on that.

We are seeing the problem on our platform (QDF2400). It is just sporadic. 
That's why, the commit message was loosely worded.

Sinan

-- 
Sinan Kaya
Qualcomm Datacenter Technologies, Inc. as an affiliate of Qualcomm Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum, a Linux Foundation Collaborative Project.