From: mpi <mpi@openbsd.org>
Date: Wed, 21 Jan 2015 14:02:33 +0000 (+0000)
Subject: Do not try to stop and reset endpoints if USB transfers are aborted
X-Git-Url: http://artulab.com/gitweb/?a=commitdiff_plain;h=7ff307ae954db12cebe37bdf6861a1cc9696014e;p=openbsd

Do not try to stop and reset endpoints if USB transfers are aborted
because the HC has been shut down (during suspend/hibernate) or
removed (PCIe card).

In both cases the hardware wont complete the commands, resulting in
timeouts.  Instead just do the software part of the abort process.

Unbreak suspend/resume with USB a device connected to xhci(4) as
reported by Fabian Raetz on bugs@.
---

diff --git a/sys/dev/usb/xhci.c b/sys/dev/usb/xhci.c
index d0e725097fc..774bd87b759 100644
--- a/sys/dev/usb/xhci.c
+++ b/sys/dev/usb/xhci.c
@@ -1,4 +1,4 @@
-/* $OpenBSD: xhci.c,v 1.57 2015/01/18 20:35:11 mpi Exp $ */
+/* $OpenBSD: xhci.c,v 1.58 2015/01/21 14:02:33 mpi Exp $ */
 
 /*
  * Copyright (c) 2014-2015 Martin Pieuchot
@@ -1912,8 +1912,10 @@ xhci_abort_xfer(struct usbd_xfer *xfer, usbd_status status)
 	    xfer->actlen, xfer->length, ((struct xhci_xfer *)xfer)->index));
 
 	/* XXX The stack should not call abort() in this case. */
-	if (xfer->status == USBD_NOT_STARTED) {
+	if (sc->sc_bus.dying || xfer->status == USBD_NOT_STARTED) {
 		xfer->status = status;
+		timeout_del(&xfer->timeout_handle);
+		usb_rem_task(xfer->device, &xfer->abort_task);
 		usb_transfer_complete(xfer);
 		return;
 	}