Revert rev 1.129:

When detaching devices when we suspend, we need to continue processing
command completion events.  So only return early in xhci_softintr() if
the controller is dead instead of dying.  This fixes USB suspend/resume
in Apple M1/M2.

ok mlarkin@, deraadt@
-----

The change does not only allow completion events to be processed but also
events that get processed by the drivers attached to this usb bus. As a
result I see a uvm panic on suspend in urtwn(4) which is triggered by a
bad rx eof.

OK kettenis@

diff --git a/sys/dev/usb/xhci.c b/sys/dev/usb/xhci.c
index 8b78d5f..324bc01 100644 (file)
--- a/sys/dev/usb/xhci.c
+++ b/sys/dev/usb/xhci.c
@@ -1,4 +1,4 @@
-/* $OpenBSD: xhci.c,v 1.129 2023/07/16 09:33:18 kettenis Exp $ */
+/* $OpenBSD: xhci.c,v 1.130 2023/07/20 09:43:00 claudio Exp $ */
 
 /*
  * Copyright (c) 2014-2015 Martin Pieuchot
@@ -656,7 +656,7 @@ xhci_softintr(void *v)
 {
        struct xhci_softc *sc = v;
 
-       if (sc->sc_dead)
+       if (sc->sc_bus.dying)
                return;
 
        sc->sc_bus.intr_context++;