qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-devel] [PATCH v6 8/8] qmp-event: add event notification for me


From: Zhu Guihua
Subject: Re: [Qemu-devel] [PATCH v6 8/8] qmp-event: add event notification for memory hot unplug error
Date: Mon, 13 Apr 2015 09:56:23 +0800
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.4.0


On 04/10/2015 11:37 PM, Eric Blake wrote:
On 04/02/2015 03:50 AM, Zhu Guihua wrote:
When memory hot unplug fails, this patch adds support to send
QMP event to notify mgmt about this failure.

Signed-off-by: Zhu Guihua <address@hidden>
---
  docs/qmp/qmp-events.txt  | 17 +++++++++++++++++
  hw/acpi/memory_hotplug.c | 10 +++++++++-
  monitor.c                |  1 +
  qapi/event.json          | 14 ++++++++++++++
  trace-events             |  1 +
  5 files changed, 42 insertions(+), 1 deletion(-)

+++ b/monitor.c
@@ -587,6 +587,7 @@ static void monitor_qapi_event_init(void)
      monitor_qapi_event_throttle(QAPI_EVENT_QUORUM_REPORT_BAD, 1000);
      monitor_qapi_event_throttle(QAPI_EVENT_QUORUM_FAILURE, 1000);
      monitor_qapi_event_throttle(QAPI_EVENT_VSERPORT_CHANGE, 1000);
+    monitor_qapi_event_throttle(QAPI_EVENT_MEM_UNPLUG_ERROR, 1000);
Is this event triggerable under guest control, or only under management
control?  We MUST throttle events that the guest can trigger, but it
makes no sense to trigger something that can only happen under host control.

It's under management control, this event should not be throttled, it's my mistake
and I will delete this. Thanks.

+++ b/qapi/event.json
@@ -330,3 +330,17 @@
  ##
  { 'event': 'VSERPORT_CHANGE',
    'data': { 'id': 'str', 'open': 'bool' } }
+
+##
+# @MEM_UNPLUG_ERROR
+#
+# Emitted when memory hot unplug error occurs.
+#
+# @device: device name
+#
+# @msg: Informative message
Any reason you abbreviated instead of spelling it out as 'message'?

This only refer to the spelling of event BLOCK_IMAGE_CORRUPTED
in docs/qmp/qmp-events.txt.

Thanks,
Zhu





reply via email to

[Prev in Thread] Current Thread [Next in Thread]