Re: [BUG] ASUS ProArt PX13 HN7306WU: amd_pmc s2idle S0ix corrupts AMD 1022:150b root port, NVIDIA dGPU returns header type 7f

[Bjorn Helgaas <helgaas@kernel.org>]

On Fri, Apr 03, 2026 at 01:41:04PM -0500, Mario Limonciello wrote:
> On 4/3/26 1:04 PM, Bjorn Helgaas wrote:
> ...

> > But as you point out, it doesn't close the issue.  Somebody else is
> > going to trip over the same issue.  Most likely they will not report
> > it and have no idea how to fix it.  Even if they do report it, we'll
> > have to go through this whole debug process again.
> > 
> > The kernel should work correctly (possibly with increased power
> > consumption or some other non-functional issue) regardless of whether
> > CONFIG_HOTPLUG_PCI_PCIE is enabled.
> 
> I do hope as part of this we can reconsider why CONFIG_HOTPLUG_PCI_PCIE
> isn't part of the defconfig in the first place.
> 
> defconfig doesn't work on any hardware of mine by default and it's too much
> work to figure out what to add to it.  So I always start at distro configs
> and peel back for my own use.
> 
> But, if we could actually make defconfig *usable* for general purpose kernel
> users maybe more people would use it.
> 
> > How can we make Linux smart enough that if we're lacking pciehp or
> > whatever is necessary, we automatically avoid s2idle or S0ix or
> > whatever causes this problem?
> 
> I suppose we /could/ have CONFIG_AMD_PMC depend on CONFIG_HOTPLUG_PCI_PCIE
> but it feels like using super glue on a wound until we know why this
> happens.

IMO making CONFIG_HOTPLUG_PCI_PCIE part of defconfig is definitely not
a solution.  It just makes it slightly less likely that people will
hit this.

Making CONFIG_AMD_PMC depend on CONFIG_HOTPLUG_PCI_PCIE might be a
little better, but I agree that it's pretty janky to do that without
understanding the mechanism.  Seems like we have a good chance to
understand this better, so I'm not ready to paper over it yet.