diff mbox series

[1/2] ACPI: processor_idle: Skip dummy wait if kernel is in guest

Message ID 20221004033548.2508712-2-jeffrey.lane@canonical.com
State New
Headers show
Series limit "Dummy wait" to old Intel systems | expand

Commit Message

Jeff Lane Oct. 4, 2022, 3:35 a.m. UTC 
From: Yin Fengwei <fengwei.yin@intel.com>

BugLink: https://bugs.launchpad.net/bugs/1990985

In function acpi_idle_do_entry(), an ioport access is used for
dummy wait to guarantee hardware behavior. But it could trigger
unnecessary VMexit if kernel is running as guest in virtualization
environment.

If it's in virtualization environment, the deeper C state enter
operation (inb()) will trap to hypervisor. It's not needed to do
dummy wait after the inb() call. So we could just remove the
dummy io port access to avoid unnecessary VMexit.

And keep dummy io port access to maintain timing for native
environment.

Signed-off-by: Yin Fengwei <fengwei.yin@intel.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
(cherry picked from commit fa583f71a99c85e52781ed877c82c8757437b680)
Signed-off-by: Jeff Lane <jeffrey.lane@canonical.com>
---
 drivers/acpi/processor_idle.c | 21 +++++++++++++++------
 1 file changed, 15 insertions(+), 6 deletions(-)
diff mbox series

Patch

diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c
index 0ea5e677f00e..a1f982a6f882 100644
--- a/drivers/acpi/processor_idle.c
+++ b/drivers/acpi/processor_idle.c
@@ -682,6 +682,19 @@  static int acpi_idle_bm_check(void)
 	return bm_status;
 }
 
+static void wait_for_freeze(void)
+{
+#ifdef	CONFIG_X86
+	/* No delay is needed if we are in guest */
+	if (boot_cpu_has(X86_FEATURE_HYPERVISOR))
+		return;
+#endif
+	/* Dummy wait op - must do something useless after P_LVL2 read
+	   because chipsets cannot guarantee that STPCLK# signal
+	   gets asserted in time to freeze execution properly. */
+	inl(acpi_gbl_FADT.xpm_timer_block.address);
+}
+
 /**
  * acpi_idle_do_entry - enter idle state using the appropriate method
  * @cx: cstate data
@@ -698,10 +711,7 @@  static void __cpuidle acpi_idle_do_entry(struct acpi_processor_cx *cx)
 	} else {
 		/* IO port based C-state */
 		inb(cx->address);
-		/* Dummy wait op - must do something useless after P_LVL2 read
-		   because chipsets cannot guarantee that STPCLK# signal
-		   gets asserted in time to freeze execution properly. */
-		inl(acpi_gbl_FADT.xpm_timer_block.address);
+		wait_for_freeze();
 	}
 }
 
@@ -722,8 +732,7 @@  static int acpi_idle_play_dead(struct cpuidle_device *dev, int index)
 			safe_halt();
 		else if (cx->entry_method == ACPI_CSTATE_SYSTEMIO) {
 			inb(cx->address);
-			/* See comment in acpi_idle_do_entry() */
-			inl(acpi_gbl_FADT.xpm_timer_block.address);
+			wait_for_freeze();
 		} else
 			return -ENODEV;
 	}