svn commit: r360274 - stable/12/sys/dev/pci

[Alexander Motin]

Author: mav
Date: Fri Apr 24 16:39:34 2020
New Revision: 360274
URL: https://svnweb.freebsd.org/changeset/base/360274

Log:
  MFC r351514: Don't consider PCIe hot-plug command timeout fatal.
  
  According to my tests and errata to several generations of Intel CPUs,
  PCIe hot-plug command completion reporting is not very reliable thing.
  At least on my Supermicro X11DPi-NT board I never saw it reported.
  Before this change timeout code detached devices and tried to disable
  the slot, that in my case resulted in hot-plugged device being detached
  just a second after it was successfully detected and attached.  This
  change removes that, so in case of timeout it just prints the error and
  continue operation.  Linux does the same.

Modified:
  stable/12/sys/dev/pci/pci_pci.c
Directory Properties:
  stable/12/   (props changed)

Modified: stable/12/sys/dev/pci/pci_pci.c
==============================================================================
--- stable/12/sys/dev/pci/pci_pci.c	Fri Apr 24 16:33:07 2020	(r360273)
+++ stable/12/sys/dev/pci/pci_pci.c	Fri Apr 24 16:39:34 2020	(r360274)
@@ -1269,11 +1269,8 @@ pcib_pcie_cc_timeout(void *arg)
 	mtx_assert(&Giant, MA_OWNED);
 	sta = pcie_read_config(dev, PCIER_SLOT_STA, 2);
 	if (!(sta & PCIEM_SLOT_STA_CC)) {
-		device_printf(dev,
-		    "HotPlug Command Timed Out - forcing detach\n");
-		sc->flags &= ~(PCIB_HOTPLUG_CMD_PENDING | PCIB_DETACH_PENDING);
-		sc->flags |= PCIB_DETACHING;
-		pcib_pcie_hotplug_update(sc, 0, 0, true);
+		device_printf(dev, "HotPlug Command Timed Out\n");
+		sc->flags &= ~PCIB_HOTPLUG_CMD_PENDING;
 	} else {
 		device_printf(dev,
 	    "Missed HotPlug interrupt waiting for Command Completion\n");