wakeup-from-suspend and system_wakeup changes

changes in v8:
- created 'query-current-machine' API to hold the wakeup-suspend-support flag
- wake-up flag now considers the --no-acpi config option for PC archs
- fixes in patch 3 proposed by Markus.
Previous series link:
https://lists.gnu.org/archive/html/qemu-devel/2018-05/msg04234.html

Daniel Henrique Barboza (3):
qmp: query-current-machine with wakeup-suspend-support
qga: update guest-suspend-ram and guest-suspend-hybrid descriptions qmp hmp: Make system_wakeup check wake-up support and run state

  hmp.c                   |  5 ++++-
  include/sysemu/sysemu.h |  1 +
  qapi/misc.json          | 29 ++++++++++++++++++++++++++++-
  qga/qapi-schema.json    | 12 ++++++------
  qmp.c                   | 11 ++++++++++-
  vl.c                    | 30 ++++++++++++++++++++++++++++++
  6 files changed, 79 insertions(+), 9 deletions(-)

changes in v8:
- created 'query-current-machine' API to hold the wakeup-suspend-support flag
- wake-up flag now considers the --no-acpi config option for PC archs
- fixes in patch 3 proposed by Markus.
Previous series link:
https://lists.gnu.org/archive/html/qemu-devel/2018-05/msg04234.html

Daniel Henrique Barboza (3):
qmp: query-current-machine with wakeup-suspend-support
qga: update guest-suspend-ram and guest-suspend-hybrid descriptions qmp hmp: Make system_wakeup check wake-up support and run state

  hmp.c                   |  5 ++++-
  include/sysemu/sysemu.h |  1 +
  qapi/misc.json          | 29 ++++++++++++++++++++++++++++-
  qga/qapi-schema.json    | 12 ++++++------
  qmp.c                   | 11 ++++++++++-
  vl.c                    | 30 ++++++++++++++++++++++++++++++
  6 files changed, 79 insertions(+), 9 deletions(-)

The qmp/hmp command 'system_wakeup' is simply a direct call to 'qemu_system_wakeup_request' from vl.c. This function verifies if runstate is SUSPENDED and if the wake up reason is valid before proceeding. However, no error or warning is thrown if any of those pre-requirements isn't met. There is no way for the caller to differentiate between a successful wakeup or an error state caused when trying to wake up a guest that wasn't suspended.

This means that system_wakeup is silently failing, which can be considered a bug. Adding error handling isn't an API break in this case - applications that didn't check the result will remain broken, the ones that check it will have a chance to deal with it.

Adding to that, the commit before previous created a new QMP API called query-current-machine, with a new flag called wakeup-suspend-support, that indicates if the guest has the capability of waking up from suspended state. Although such guest will never reach SUSPENDED state and erroring it out in this scenario would suffice, it is more informative for the user to differentiate between a failure because the guest isn't suspended versus a failure because the guest does not have support for wake up at all.

All this considered, this patch changes qmp_system_wakeup to:

- check if the guest has the capability to wake-up from suspend. If not, error out informing about the lack of support;

- make the runstate verification before proceeding to call qemu_system_wakeup_request, firing up an error message if the user tries to wake up a machine that isn't in SUSPENDED state.

After this patch, this is the output of system_wakeup in a guest that does not have wake-up from suspend support (ppc64):

(qemu) system_wakeup
wake-up from suspend is not supported by this guest (qemu)

And this is the output of system_wakeup in a x86 guest that has the support but isn't suspended:

(qemu) system_wakeup
Unable to wake up: guest is not in suspended state (qemu)

Reported-by Balamuruhan S <[email protected]>

Signed-off-by Daniel Henrique Barboza <[email protected]>

---
 hmp.c                   |  5 ++++-
 include/sysemu/sysemu.h |  1 +
 qapi/misc.json          |  5 ++++-
 qmp.c                   | 11 ++++++++++-
 vl.c                    |  2 +-
 5 files changed, 20 insertions(+), 4 deletions(-)

hmp.c +4 -1

index 41f5e39b72..83a8041d11 100644
@@ -1169,7 +1169,10 @@ void hmp_cont(Monitor *mon, const QDict *qdict)
 
 void hmp_system_wakeup(Monitor *mon, const QDict *qdict)
 {
-    qmp_system_wakeup(NULL);
+    Error *err = NULL;
+
+    qmp_system_wakeup(&err);
+    hmp_handle_error(mon, &err);
 }
 
 void hmp_nmi(Monitor *mon, const QDict *qdict)

include/sysemu/sysemu.h +1 -0

index b921c6f3b7..483f804b36 100644
@@ -70,6 +70,7 @@ void qemu_exit_preconfig_request(void);
 void qemu_system_reset_request(ShutdownCause reason);
 void qemu_system_suspend_request(void);
 void qemu_register_suspend_notifier(Notifier *notifier);
+bool qemu_wakeup_suspend_support(void);
 void qemu_system_wakeup_request(WakeupReason reason);
 void qemu_system_wakeup_enable(WakeupReason reason, bool enabled);
 void qemu_register_wakeup_notifier(Notifier *notifier);

qapi/misc.json +4 -1

index 266f88cb09..86195d79e9 100644
@@ -1230,7 +1230,10 @@
 ##
 # @system_wakeup:
 #
-# Wakeup guest from suspend.  Does nothing in case the guest isn't suspended.
+# If the guest has wake-up from suspend support enabled
+# (wakeup-suspend-support flag from query-current-machine), wakeup guest
+# from suspend if the guest is in SUSPENDED state. Returns an error
+# otherwise.
 #
 # Since:  1.1
 #

qmp.c +1 -1

index 73e46d795f..0673ff1fa5 100644
@@ -215,7 +215,16 @@ void qmp_cont(Error **errp)
 
 void qmp_system_wakeup(Error **errp)
 {
-    qemu_system_wakeup_request(QEMU_WAKEUP_REASON_OTHER);
+    if (!qemu_wakeup_suspend_support()) {

qmp.c +0 -0

 ObjectPropertyInfoList *qmp_qom_list(const char *path, Error **errp)

vl.c +1 -1

index 96ad448d57..2af10b75cb 100644
@@ -1769,7 +1769,7 @@ void qemu_register_wakeup_notifier(Notifier *notifier)
     notifier_list_add(&wakeup_notifiers, notifier);
 }
 
-static bool qemu_wakeup_suspend_support(void)
+bool qemu_wakeup_suspend_support(void)
 {
     return !QLIST_EMPTY(&wakeup_notifiers.notifiers) && acpi_enabled;
 }

Signature · 1 line

2.17.1

When issuing the qmp/hmp 'system_wakeup' command, what happens in a nutshell is:

- qmp_system_wakeup_request set runstate to RUNNING, sets a wakeup_reason and notify the event
- in the main_loop, all vcpus are paused, a system reset is issued, all subscribers of wakeup_notifiers receives a notification, vcpus are then resumed and the wake up QAPI event is fired

Note that this procedure alone doesn't ensure that the guest will awake from SUSPENDED state - the subscribers of the wake up event must take action to resume the guest, otherwise the guest will simply reboot.

At this moment, there are only two subscribers of the wake up event: one in hw/acpi/core.c and another in hw/i386/xen/xen-hvm.c. This means that system_wakeup does not work as intended with other architectures.

However, only the presence of 'system_wakeup' is required for QGA to support 'guest-suspend-ram' and 'guest-suspend-hybrid' at this moment. This means that the user/management will expect to suspend the guest using one of those suspend commands and then resume execution using system_wakeup, regardless of the support offered in system_wakeup in the first place.

This patch creates a new API called query-current-machine [1], that holds a new flag called 'wakeup-suspend-support', which indicates if the guest supports wake up from suspend via system_wakeup. The guest is considered to implement wake-up support if:

- there is at least one subscriber for the wake up event;

and, for the PC machine type:

- ACPI is enabled.

This is the expected output of query-current-machine when running a x86 guest:

{"execute" : "query-current-machine"}
{"return": {"wakeup-suspend-support": true}}

Running the same x86 guest, but with the --no-acpi option:

{"execute" : "query-current-machine"}
{"return": {"wakeup-suspend-support": false}}

This is the output when running a pseries guest:

{"execute" : "query-current-machine"}
{"return": {"wakeup-suspend-support": false}}

With this extra tool, management can avoid situations where a guest that does not have proper suspend/wake capabilities ends up in inconsistent state (e.g.
https://github.com/open-power-host-os/qemu/issues/31).

[1] the decision of creating the query-current-machine API is based on discussions in the QEMU mailing list where it was decided that query-target wasn't a proper place to store the wake-up flag, neither was query-machines because this isn't a static property of the machine object. This new API can then be used to store other dynamic machine properties that are scattered around the code ATM. More info at:
https://lists.gnu.org/archive/html/qemu-devel/2018-05/msg04235.html

Reported-by Balamuruhan S <[email protected]>

Signed-off-by Daniel Henrique Barboza <[email protected]>

---
 qapi/misc.json | 24 ++++++++++++++++++++++++
 vl.c           | 30 ++++++++++++++++++++++++++++++
 2 files changed, 54 insertions(+)

qapi/misc.json +0 -0

index 29da7856e3..266f88cb09 100644
@@ -2003,6 +2003,30 @@
 ##
 { 'command': 'query-machines', 'returns': ['MachineInfo'] }

qapi/misc.json +15 -0

+##
+{ 'struct': 'CurrentMachineParams',
+  'data': { 'wakeup-suspend-support': 'bool'} }
+
+##
+# @query-current-machine:
+#
+# Return a list of parameters of the running machine.
+#
+# Returns: CurrentMachineParams
+#
+# Since: 3.0.0
+##
+{ 'command': 'query-current-machine', 'returns': 'CurrentMachineParams' }
+
 ##
 # @CpuDefinitionInfo:
 #

vl.c +16 -0

index ef6cfcec40..96ad448d57 100644
@@ -1747,11 +1747,41 @@ void qemu_system_wakeup_enable(WakeupReason reason, bool enabled)
     }
 }
 
+/*
+ * The existence of a wake-up notifier is being checked in the function
+ * qemu_wakeup_suspend_support and it's used in the logic of the
+ * wakeup-suspend-support flag of QMP 'query-current-machine' command.
+ * The idea of this flag is to indicate whether the guest supports wake-up
+ * from suspend (via system_wakeup QMP/HMP call for example), warning the
+ * user that the guest can't handle both wake-up from suspend and the
+ * suspend itself via QGA guest-suspend-ram and guest-suspend-hybrid (if
+ * it can't wake up, it can't be suspended safely).
+ *
+ * An assumption is made by the wakeup-suspend-support flag that only the
+ * guests that can go to RUN_STATE_SUSPENDED and wake up properly would
+ * be interested in this wakeup_notifier. Adding a wakeup_notifier for
+ * any other reason will break the logic of the wakeup-suspend-support
+ * flag and can lead to user/management confusion about the suspend/wake-up
+ * support of the guest.

vl.c +9 -0

+
+CurrentMachineParams *qmp_query_current_machine(Error **errp)
+{
+    CurrentMachineParams *params = g_malloc0(sizeof(*params));
+    params->wakeup_suspend_support = qemu_wakeup_suspend_support();
+
+    return params;
+}
+
 void qemu_system_killed(int signal, pid_t pid)
 {
     shutdown_signal = signal;

Signature · 1 line

2.17.1

Quoting Daniel Henrique Barboza (2018-07-05 15:08:13)

Quoted 104 lines

The qmp/hmp command 'system_wakeup' is simply a direct call to 'qemu_system_wakeup_request' from vl.c. This function verifies if runstate is SUSPENDED and if the wake up reason is valid before proceeding. However, no error or warning is thrown if any of those pre-requirements isn't met. There is no way for the caller to differentiate between a successful wakeup or an error state caused when trying to wake up a guest that wasn't suspended.

This means that system_wakeup is silently failing, which can be considered a bug. Adding error handling isn't an API break in this case - applications that didn't check the result will remain broken, the ones that check it will have a chance to deal with it.

Adding to that, the commit before previous created a new QMP API called query-current-machine, with a new flag called wakeup-suspend-support, that indicates if the guest has the capability of waking up from suspended state. Although such guest will never reach SUSPENDED state and erroring it out in this scenario would suffice, it is more informative for the user to differentiate between a failure because the guest isn't suspended versus a failure because the guest does not have support for wake up at all.

All this considered, this patch changes qmp_system_wakeup to:

- check if the guest has the capability to wake-up from suspend. If not, error out informing about the lack of support;

- make the runstate verification before proceeding to call qemu_system_wakeup_request, firing up an error message if the user tries to wake up a machine that isn't in SUSPENDED state.

After this patch, this is the output of system_wakeup in a guest that does not have wake-up from suspend support (ppc64):

(qemu) system_wakeup
wake-up from suspend is not supported by this guest (qemu)

And this is the output of system_wakeup in a x86 guest that has the support but isn't suspended:

(qemu) system_wakeup
Unable to wake up: guest is not in suspended state (qemu)

Reported-by Balamuruhan S <[email protected]>

Signed-off-by Daniel Henrique Barboza <[email protected]>

---
  hmp.c                   |  5 ++++-
  include/sysemu/sysemu.h |  1 +
  qapi/misc.json          |  5 ++++-
  qmp.c                   | 11 ++++++++++-
  vl.c                    |  2 +-
  5 files changed, 20 insertions(+), 4 deletions(-)

hmp.c +4 -1

index 41f5e39b72..83a8041d11 100644
@@ -1169,7 +1169,10 @@ void hmp_cont(Monitor *mon, const QDict *qdict)
 
  void hmp_system_wakeup(Monitor *mon, const QDict *qdict)
  {
-    qmp_system_wakeup(NULL);
+    Error *err = NULL;
+
+    qmp_system_wakeup(&err);
+    hmp_handle_error(mon, &err);
  }
 
  void hmp_nmi(Monitor *mon, const QDict *qdict)

include/sysemu/sysemu.h +1 -0

index b921c6f3b7..483f804b36 100644
@@ -70,6 +70,7 @@ void qemu_exit_preconfig_request(void);
  void qemu_system_reset_request(ShutdownCause reason);
  void qemu_system_suspend_request(void);
  void qemu_register_suspend_notifier(Notifier *notifier);
+bool qemu_wakeup_suspend_support(void);
  void qemu_system_wakeup_request(WakeupReason reason);
  void qemu_system_wakeup_enable(WakeupReason reason, bool enabled);
  void qemu_register_wakeup_notifier(Notifier *notifier);

qapi/misc.json +4 -1

index 266f88cb09..86195d79e9 100644
@@ -1230,7 +1230,10 @@
  ##
  # @system_wakeup:
  #
-# Wakeup guest from suspend.  Does nothing in case the guest isn't suspended.
+# If the guest has wake-up from suspend support enabled
+# (wakeup-suspend-support flag from query-current-machine), wakeup guest
+# from suspend if the guest is in SUSPENDED state. Returns an error
+# otherwise.
  #
  # Since:  1.1
  #

qmp.c +0 -0

index 73e46d795f..0673ff1fa5 100644
@@ -215,7 +215,16 @@ void qmp_cont(Error **errp)
 

Quoted 4 lines

qmp.c +1 -1

  void qmp_system_wakeup(Error **errp)
  {
-    qemu_system_wakeup_request(QEMU_WAKEUP_REASON_OTHER);
+    if (!qemu_wakeup_suspend_support()) {

qmp.c void qmp_cont(Error **errp) · line 228

Quoted 10 lines

qmp.c +9 -0

+        error_setg(errp,
+                   "wake-up from suspend is not supported by this guest");
+        return;
+    }
+    if (!runstate_check(RUN_STATE_SUSPENDED)) {
+        error_setg(errp,
+                   "Unable to wake up: guest is not in suspended state");
+        return;
+    }
  }

Was removing qemu_system_wakeup_request() intentional or did that just get lost somehow?

Quoted 18 lines

qmp.c +0 -0

  ObjectPropertyInfoList *qmp_qom_list(const char *path, Error **errp)

vl.c +1 -1

index 96ad448d57..2af10b75cb 100644
@@ -1769,7 +1769,7 @@ void qemu_register_wakeup_notifier(Notifier *notifier)
      notifier_list_add(&wakeup_notifiers, notifier);
  }
 
-static bool qemu_wakeup_suspend_support(void)
+bool qemu_wakeup_suspend_support(void)
  {
      return !QLIST_EMPTY(&wakeup_notifiers.notifiers) && acpi_enabled;
  }

Signature · 1 line

2.17.1

Quoting Daniel Henrique Barboza (2018-07-05 15:08:11)

Quoted 78 lines

When issuing the qmp/hmp 'system_wakeup' command, what happens in a nutshell is:

- qmp_system_wakeup_request set runstate to RUNNING, sets a wakeup_reason and notify the event
- in the main_loop, all vcpus are paused, a system reset is issued, all subscribers of wakeup_notifiers receives a notification, vcpus are then resumed and the wake up QAPI event is fired

Note that this procedure alone doesn't ensure that the guest will awake from SUSPENDED state - the subscribers of the wake up event must take action to resume the guest, otherwise the guest will simply reboot.

At this moment, there are only two subscribers of the wake up event: one in hw/acpi/core.c and another in hw/i386/xen/xen-hvm.c. This means that system_wakeup does not work as intended with other architectures.

However, only the presence of 'system_wakeup' is required for QGA to support 'guest-suspend-ram' and 'guest-suspend-hybrid' at this moment. This means that the user/management will expect to suspend the guest using one of those suspend commands and then resume execution using system_wakeup, regardless of the support offered in system_wakeup in the first place.

This patch creates a new API called query-current-machine [1], that holds a new flag called 'wakeup-suspend-support', which indicates if the guest supports wake up from suspend via system_wakeup. The guest is considered to implement wake-up support if:

- there is at least one subscriber for the wake up event;

and, for the PC machine type:

- ACPI is enabled.

This is the expected output of query-current-machine when running a x86 guest:

{"execute" : "query-current-machine"}
{"return": {"wakeup-suspend-support": true}}

Running the same x86 guest, but with the --no-acpi option:

{"execute" : "query-current-machine"}
{"return": {"wakeup-suspend-support": false}}

This is the output when running a pseries guest:

{"execute" : "query-current-machine"}
{"return": {"wakeup-suspend-support": false}}

With this extra tool, management can avoid situations where a guest that does not have proper suspend/wake capabilities ends up in inconsistent state (e.g.
https://github.com/open-power-host-os/qemu/issues/31).

[1] the decision of creating the query-current-machine API is based on discussions in the QEMU mailing list where it was decided that query-target wasn't a proper place to store the wake-up flag, neither was query-machines because this isn't a static property of the machine object. This new API can then be used to store other dynamic machine properties that are scattered around the code ATM. More info at:
https://lists.gnu.org/archive/html/qemu-devel/2018-05/msg04235.html

Reported-by Balamuruhan S <[email protected]>

Signed-off-by Daniel Henrique Barboza <[email protected]>

---
  qapi/misc.json | 24 ++++++++++++++++++++++++
  vl.c           | 30 ++++++++++++++++++++++++++++++
  2 files changed, 54 insertions(+)

qapi/misc.json +0 -0

index 29da7856e3..266f88cb09 100644
@@ -2003,6 +2003,30 @@
  ##
  { 'command': 'query-machines', 'returns': ['MachineInfo'] }

qapi/misc.json line 2014

Quoted 10 lines

qapi/misc.json +9 -0

 
+##
+# @CurrentMachineParams:
+#
+# Information describing the running machine parameters.
+#
+# @wakeup-suspend-support: true if the target supports wake up from
+#                          suspend
+#
+# Since: 3.0.0

3.1.0 now (hopefully :))

Quoted 48 lines

qapi/misc.json +15 -0

+##
+{ 'struct': 'CurrentMachineParams',
+  'data': { 'wakeup-suspend-support': 'bool'} }
+
+##
+# @query-current-machine:
+#
+# Return a list of parameters of the running machine.
+#
+# Returns: CurrentMachineParams
+#
+# Since: 3.0.0
+##
+{ 'command': 'query-current-machine', 'returns': 'CurrentMachineParams' }
+
  ##
  # @CpuDefinitionInfo:
  #

vl.c +17 -0

index ef6cfcec40..96ad448d57 100644
@@ -1747,11 +1747,41 @@ void qemu_system_wakeup_enable(WakeupReason reason, bool enabled)
      }
  }
 
+/*
+ * The existence of a wake-up notifier is being checked in the function
+ * qemu_wakeup_suspend_support and it's used in the logic of the
+ * wakeup-suspend-support flag of QMP 'query-current-machine' command.
+ * The idea of this flag is to indicate whether the guest supports wake-up
+ * from suspend (via system_wakeup QMP/HMP call for example), warning the
+ * user that the guest can't handle both wake-up from suspend and the
+ * suspend itself via QGA guest-suspend-ram and guest-suspend-hybrid (if
+ * it can't wake up, it can't be suspended safely).
+ *
+ * An assumption is made by the wakeup-suspend-support flag that only the
+ * guests that can go to RUN_STATE_SUSPENDED and wake up properly would
+ * be interested in this wakeup_notifier. Adding a wakeup_notifier for
+ * any other reason will break the logic of the wakeup-suspend-support
+ * flag and can lead to user/management confusion about the suspend/wake-up
+ * support of the guest.
+ */
  void qemu_register_wakeup_notifier(Notifier *notifier)
  {
      notifier_list_add(&wakeup_notifiers, notifier);
  }
 

vl.c void qemu_system_wakeup_enable(WakeupReason reason, bool enabled) · line 1775

Quoted 4 lines

vl.c +4 -0

+static bool qemu_wakeup_suspend_support(void)
+{
+    return !QLIST_EMPTY(&wakeup_notifiers.notifiers) && acpi_enabled;
+}

I think this would need to be re-worked for any machine types that eventually implement wakeup even though acpi_enabled == false. It was maybe a bit hacky, but kind of nice that registering a wakeup notifier implicitly registered support for wakeup, but now that we already have these sorts of edge cases to consider maybe it's best to just add an explicit qemu_register_wakeup_support() that just sets a flag in vl.c and add call that at the appropriate locations for known supported configurations. Maybe one call in acpi_pm1_cnt_init() might be enough?

Though I'm noticing mips malta board also calls that, would be curious if that actually supports wake-from-suspend. If not, maybe the callers are where we should declare support (ich9_pm_init and piix4_pm_realize currently)?

Quoted 15 lines

vl.c +9 -0

+
+CurrentMachineParams *qmp_query_current_machine(Error **errp)
+{
+    CurrentMachineParams *params = g_malloc0(sizeof(*params));
+    params->wakeup_suspend_support = qemu_wakeup_suspend_support();
+
+    return params;
+}
+
  void qemu_system_killed(int signal, pid_t pid)
  {
      shutdown_signal = signal;

Signature · 1 line

2.17.1