PSARC/2005/469 X86 Energy Star compliance

PSARC/2006/632 PSMI extension for state save and restore
6330209 nge needs to support DDI_SUSPEND/DDI_RESUME
6381827 Suspend to RAM on x86
6393154 audio810 needs to support DDI_SUSPEND/DDI_RESUME
6397047 fd, fdc needs to support Suspend/Resume
6401974 cannot enter S3 with ohci PME enable set on Tyan 2865 with Sun or Tyan 2.01 BIOS
6422613 memscrubber doesn't re-acquire lock before CALLB_CPR_EXIT
6455736 ata/dadk/cmdk should support DDI_SUSPEND/DDI_RESUME
6511370 CPR on SPARC regression
6586018 TODOP Macros in i86pc/sys/machclock.h not in sun4u/sun4v equivilent (Sparc only)
6610124 It takes more than 3 minutes after printing "pci_pre_resume nv_sata:0"
6617143 powerd/pmconfig emits a different default message for an existing on or off action.

--HG--
rename : usr/src/cmd/power/power.conf => usr/src/cmd/power/power.conf.sparc

1 files changed, 19 insertions, 1 deletions

diff --git a/usr/src/uts/i86pc/io/mc/mcamd_drv.c b/usr/src/uts/i86pc/io/mc/mcamd_drv.c
index d7c9371795..284de242a9 100644
--- a/usr/src/uts/i86pc/io/mc/mcamd_drv.c
+++ b/usr/src/uts/i86pc/io/mc/mcamd_drv.c
@@ -1491,6 +1491,17 @@ mc_attach(dev_info_t *dip, ddi_attach_cmd_t cmd)
 	int chipid, rc;
 	mc_t *mc;
 
+	/*
+	 * This driver has no hardware state, but does
+	 * claim to have a reg property, so it will be
+	 * called on suspend.  It is probably better to
+	 * make sure it doesn't get called on suspend,
+	 * but it is just as easy to make sure we just
+	 * return DDI_SUCCESS if called.
+	 */
+	if (cmd == DDI_RESUME)
+		return (DDI_SUCCESS);
+
 	if (cmd != DDI_ATTACH || mc_no_attach != 0)
 		return (DDI_FAILURE);
 
@@ -1672,7 +1683,14 @@ mc_attach(dev_info_t *dip, ddi_attach_cmd_t cmd)
 static int
 mc_detach(dev_info_t *dip, ddi_detach_cmd_t cmd)
 {
-	return (DDI_FAILURE);
+	/*
+	 * See the comment about suspend in
+	 * mc_attach().
+	 */
+	if (cmd == DDI_SUSPEND)
+		return (DDI_SUCCESS);
+	else
+		return (DDI_FAILURE);
 }
 
 static struct dev_ops mc_ops = {