Re: [PATCH] PCI: Extend Apple Thunderbolt power quirk to Alpine Ridge

[Francisco Beltrán Millalén]

Hi Darrell,

Thank you for testing all three patches on your MacBookPro14,3 and for
such a careful report.  It's very good to see the same results on a
second machine, and the details you sent were useful.

On Thu, Oct 08, 2026 at 11:55:42AM -0700, Darrell Gum wrote:
> - Real S3 resumed on every attempt, about half a dozen cycles
>   including lid closes on battery. noirq resume is about 16 s, with
>   about 11 s of it in each upstream bridge.

That 16 s is the ACPICA problem you mentioned in the other thread, and I
see the same thing here.  One warning before you try that ACPICA change,
though: together with this version of the quirk it makes things worse
when something is plugged in: the controller with the device attached,
and sometimes the other one too, is lost on resume until the next
reboot.  Even with nothing plugged in, the ports came back in my tests
but the thunderbolt driver logged timeouts.  I've explained why in my
reply to Bjorn in this thread; v2 of the quirk will handle it.  Until
then, I'd suggest not combining the ACPICA change with this version of
the quirk.

> - On that cycle the empty side's xHCI logged "xhci_hcd 0000:07:00.0:
>   xHC error in resume, USBSTS 0x401, Reinit" and recovered.

I see the same message here on the side with nothing attached.  With
the quirk, that controller loses power during suspend, and the xHCI
driver notices on resume and reinitialises it, so as far as I can tell
it's expected.

I'll copy you on v2.

Thanks again,
Francisco