Skip to contentSkip to conversation

Drivers: hv: vmbus: Disallow the freeze PM operation

PATCH In review Dexuan Cui 2 participants 5 messages Apr 7, 2020 – Apr 8, 2020

Subsystems (2)
lore
Share & tools
lore

Dexuan Cui

author maintainer
#0 lore raw

PATCH 1/1 Drivers: hv: vmbus: Disallow the freeze PM operation

Before the hibernation patchset (e.g. f53335e3289f), a Linux VM on Hyper-V can run "echo freeze > /sys/power/state" (or "systemctl suspend") to freeze the system. The user can press the keyboard or move the mouse to wake up the VM. Note: the two aforementioned commands are equivalent here, because Hyper-V doesn't support the guest ACPI S3 state.

With the hibernation patchset, a Linux VM on Hyper-V can hibernate to disk and resume back; however, the 'freeze' operation is broken for Hyper-V Generation-2 VM (which doesn't have a legacy keyboard/mouse): when the vmbus devices are suspended, the VM can not receive any interrupt from the synthetic keyboard/mouse devices, so there is no way to wake up the VM. This is not an issue for Generaton-1 VM, because it looks the legacy keyboard/mouse devices can still be used to wake up the VM in my test.

IMO 'freeze' in a Linux VM on Hyper-V is not really useful in practice, so let's disallow the operation for both Gen-1 and Gen-2 VMs, even if it's not an issue for Gen-1 VMs.

Fixes f53335e3289f ("Drivers: hv: vmbus: Suspend/resume the vmbus itself for hibernation")
---
 drivers/hv/vmbus_drv.c | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)
drivers/hv/vmbus_drv.c +21 -0
index 029378c..82a4327 100644
@@ -28,6 +28,7 @@
 #include <linux/notifier.h>
 #include <linux/ptrace.h>
 #include <linux/screen_info.h>
+#include <linux/suspend.h>
 #include <linux/kdebug.h>
 #include <linux/efi.h>
 #include <linux/random.h>
@@ -2357,6 +2358,23 @@ static void hv_synic_resume(void)
 	.resume = hv_synic_resume,
 };
 
+/*
+ * Note: "freeze/suspend" here means "systemctl suspend".
+ * "systemctl hibernate" is still supported.
+ */
+static int hv_pm_notify(struct notifier_block *nb,
+			unsigned long val, void *ign)
+{
+	if (val == PM_SUSPEND_PREPARE) {
+		pr_info("freeze/suspend is not supported\n");
+		return NOTIFY_BAD;
+	}
+
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block hv_pm_nb;
+
 static int __init hv_acpi_init(void)
 {
 	int ret, t;
@@ -2389,6 +2407,8 @@ static int __init hv_acpi_init(void)
 	hv_setup_crash_handler(hv_crash_handler);
 
 	register_syscore_ops(&hv_synic_syscore_ops);
+	hv_pm_nb.notifier_call = hv_pm_notify;
+	register_pm_notifier(&hv_pm_nb);
 
 	return 0;
 
@@ -2402,6 +2422,7 @@ static void __exit vmbus_exit(void)
 {
 	int cpu;
 
+	unregister_pm_notifier(&hv_pm_nb);
 	unregister_syscore_ops(&hv_synic_syscore_ops);
 
 	hv_remove_kexec_handler();
Signature · 1 line
1.8.3.1
#1 lore raw
Quoted 14 lines · Dexuan Cui

Before the hibernation patchset (e.g. f53335e3289f), a Linux VM on Hyper-V can run "echo freeze > /sys/power/state" (or "systemctl suspend") to freeze the system. The user can press the keyboard or move the mouse to wake up the VM. Note: the two aforementioned commands are equivalent here, because Hyper-V doesn't support the guest ACPI S3 state.

With the hibernation patchset, a Linux VM on Hyper-V can hibernate to disk and resume back; however, the 'freeze' operation is broken for Hyper-V Generation-2 VM (which doesn't have a legacy keyboard/mouse): when the vmbus devices are suspended, the VM can not receive any interrupt from the synthetic keyboard/mouse devices, so there is no way to wake up the VM. This is not an issue for Generaton-1 VM, because it looks the legacy keyboard/mouse devices can still be used to wake up the VM in my test.

Quoted 3 lines

IMO 'freeze' in a Linux VM on Hyper-V is not really useful in practice, so let's disallow the operation for both Gen-1 and Gen-2 VMs, even if it's not an issue for Gen-1 VMs.

Suspend-to-idle may not be very useful indeed, however, it worked before and I think we can just fix it. In particular, why do we need to do anything when we are not hibernating?

Quoted 20 lines
Fixes f53335e3289f ("Drivers: hv: vmbus: Suspend/resume the vmbus itself for hibernation")
---
 drivers/hv/vmbus_drv.c | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)
drivers/hv/vmbus_drv.c +1 -0
index 029378c..82a4327 100644
@@ -28,6 +28,7 @@
 #include <linux/notifier.h>
 #include <linux/ptrace.h>
 #include <linux/screen_info.h>
+#include <linux/suspend.h>
 #include <linux/kdebug.h>
 #include <linux/efi.h>
 #include <linux/random.h>

drivers/hv/vmbus_drv.c static void hv_synic_resume(void) · line 2363

Quoted 6 lines
drivers/hv/vmbus_drv.c +3 -0
@@ -2357,6 +2358,23 @@ static void hv_synic_resume(void)
 	.resume = hv_synic_resume,
 };
 
+/*
+ * Note: "freeze/suspend" here means "systemctl suspend".
+ * "systemctl hibernate" is still supported.

Let's not use systemd terminology in kernel, let's use the ones from admin-guide/pm/sleep-states.rst (Suspend-to-Idle/Standby/Suspend-to-RAM/ Hibernation).

Quoted 34 lines
drivers/hv/vmbus_drv.c +17 -0
+ */
+static int hv_pm_notify(struct notifier_block *nb,
+			unsigned long val, void *ign)
+{
+	if (val == PM_SUSPEND_PREPARE) {
+		pr_info("freeze/suspend is not supported\n");
+		return NOTIFY_BAD;
+	}
+
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block hv_pm_nb;
+
 static int __init hv_acpi_init(void)
 {
 	int ret, t;
@@ -2389,6 +2407,8 @@ static int __init hv_acpi_init(void)
 	hv_setup_crash_handler(hv_crash_handler);
 
 	register_syscore_ops(&hv_synic_syscore_ops);
+	hv_pm_nb.notifier_call = hv_pm_notify;
+	register_pm_notifier(&hv_pm_nb);
 
 	return 0;
 
@@ -2402,6 +2422,7 @@ static void __exit vmbus_exit(void)
 {
 	int cpu;
 
+	unregister_pm_notifier(&hv_pm_nb);
 	unregister_syscore_ops(&hv_synic_syscore_ops);
 
 	hv_remove_kexec_handler();
Signature · 1 line
Vitaly
author maintainer
#2 lore raw
Quoted 8 lines

From: Vitaly Kuznetsov <[email protected]>
Sent: Wednesday, April 8, 2020 8:47 AM

Quoted 3 lines

IMO 'freeze' in a Linux VM on Hyper-V is not really useful in practice, so let's disallow the operation for both Gen-1 and Gen-2 VMs, even if it's not an issue for Gen-1 VMs.

Suspend-to-idle may not be very useful indeed, however, it worked before and I think we can just fix it.

How can we fix Suspend-to-idle for a Gen-2 VM, in which no device can work as wakeup devices? Note: in the case of Suspend-to-idle, now all the vmbus devices including the synthetic keyboard/mouse are suspended completely.

Are you suggesting hv_vmbus should distinguish Suspend-to-idle from hibernation, and for the former hv_vmbus should not suspend the synthetic keyboard/mouse? This should be doable but IMO this is not a very trivial effort, and I'm trying to avoid it since IMO Suspend-to-idle is not really useful in practice for a Linux VM on Hyper-V. :-)

Quoted 2 lines

In particular, why do we need to do
anything when we are not hibernating?

Are you suggesting hv_vmbus should not suspend the vmbus devices at all in the case of Suspend-to-idle?

Quoted 9 lines
Quoted 3 lines

+/*
+ * Note: "freeze/suspend" here means "systemctl suspend".
+ * "systemctl hibernate" is still supported.

Let's not use systemd terminology in kernel, let's use the ones from admin-guide/pm/sleep-states.rst (Suspend-to-Idle/Standby/Suspend-to-RAM/ Hibernation).
--
Vitaly

Thanks! I'll use the accurate terms.

Thanks,
-- Dexuan

#3 lore raw
Quoted 9 lines · Dexuan Cui
Quoted 8 lines

From: Vitaly Kuznetsov <[email protected]>
Sent: Wednesday, April 8, 2020 8:47 AM

Quoted 3 lines

IMO 'freeze' in a Linux VM on Hyper-V is not really useful in practice, so let's disallow the operation for both Gen-1 and Gen-2 VMs, even if it's not an issue for Gen-1 VMs.

Suspend-to-idle may not be very useful indeed, however, it worked before and I think we can just fix it.

Quoted 7 lines

How can we fix Suspend-to-idle for a Gen-2 VM, in which no device can work as wakeup devices? Note: in the case of Suspend-to-idle, now all the vmbus devices including the synthetic keyboard/mouse are suspended completely.

Are you suggesting hv_vmbus should distinguish Suspend-to-idle from hibernation, and for the former hv_vmbus should not suspend the synthetic keyboard/mouse?

Yes, basically.

Quoted 3 lines

This should be doable but IMO this is not a very trivial
effort, and I'm trying to avoid it since IMO Suspend-to-idle is not really useful in practice for a Linux VM on Hyper-V. :-)

Well, to me it's equally (not) useful in all other cases :-) I think we should Cc: [email protected] and someone will describe a real world usecase to educate us, we'll then see if there is any Hyper-V specifics.

Quoted 6 lines
Quoted 3 lines

In particular, why do we need to do
anything when we are not hibernating?

Are you suggesting hv_vmbus should not suspend the vmbus devices at all in the case of Suspend-to-idle?

That what we were doing prior to the hibernation series, right? AFAIU suspend-to-idle is basically 'no processes are scheduled' mode but we don't really need to do anything with devices.

Quoted 16 lines
Quoted 10 lines
Quoted 4 lines

+/*
+ * Note: "freeze/suspend" here means "systemctl suspend".
+ * "systemctl hibernate" is still supported.

Let's not use systemd terminology in kernel, let's use the ones from admin-guide/pm/sleep-states.rst (Suspend-to-Idle/Standby/Suspend-to-RAM/ Hibernation).
--
Vitaly

Thanks! I'll use the accurate terms.

Thanks,
-- Dexuan

Signature · 1 line
Vitaly
author maintainer
#4 lore raw
Quoted 8 lines

From: Vitaly Kuznetsov <[email protected]>
Sent: Wednesday, April 8, 2020 12:24 PM
To: Dexuan Cui <[email protected]>

Quoted 4 lines

...
This should be doable but IMO this is not a very trivial effort, and I'm trying to avoid it since IMO Suspend-to-idle is not really useful in practice for a Linux VM on Hyper-V. :-)

Quoted 4 lines

Well, to me it's equally (not) useful in all other cases :-) I think we should Cc: [email protected] and someone will describe a real world usecase to educate us, we'll then see if there is any Hyper-V specifics.

Maybe I should support Suspend-to-idle, anyway. :-)

Quoted 7 lines
Quoted 5 lines
Quoted 2 lines

In particular, why do we need to do
anything when we are not hibernating?

Are you suggesting hv_vmbus should not suspend the vmbus devices at all in the case of Suspend-to-idle?

That what we were doing prior to the hibernation series, right? AFAIU

Yes.

Quoted 2 lines

suspend-to-idle is basically 'no processes are scheduled' mode but we don't really need to do anything with devices.

Got it. Let me try to make a patch to revert to the old behavior for Suspend-to-idle.

Thanks,
-- Dexuan