Hi Igor,

On 09/04/2014 08:15 PM, Igor Mammedov wrote:
On Wed, 27 Aug 2014 16:08:32 +0800
Tang Chen <tangc...@cn.fujitsu.com> wrote:

From: Hu Tao <hu...@cn.fujitsu.com>

Implement acpi_memory_unplug_cb(), sending an sci to guest to trigger
memory hot-remove, and call it in piix4_device_unplug_cb().

Signed-off-by: Hu Tao <hu...@cn.fujitsu.com>
Signed-off-by: Tang Chen <tangc...@cn.fujitsu.com>
---
  hw/acpi/memory_hotplug.c         | 25 +++++++++++++++++++++++++
  hw/acpi/piix4.c                  |  3 +++
  include/hw/acpi/memory_hotplug.h |  2 ++
  3 files changed, 30 insertions(+)

diff --git a/hw/acpi/memory_hotplug.c b/hw/acpi/memory_hotplug.c
index ed39241..c310b44 100644
--- a/hw/acpi/memory_hotplug.c
+++ b/hw/acpi/memory_hotplug.c
@@ -195,6 +195,31 @@ void acpi_memory_plug_cb(ACPIREGS *ar, qemu_irq irq, 
MemHotplugState *mem_st,
      return;
  }
+void acpi_memory_unplug_cb(ACPIREGS *ar, qemu_irq irq, MemHotplugState *mem_st,
+                           DeviceState *dev, Error **errp)
+{
+    Error *local_err = NULL;
+    int slot = object_property_get_int(OBJECT(dev), "slot", &local_err);
might be worth to use PC_DIMM_SLOT_PROP here, and while at it
do the same to acpi_memory_plug_cb() as well in separate patch.


Followed.

+
+    if (local_err) {
+        error_propagate(errp, local_err);
+        return;
+    }
+
+    if (slot >= mem_st->dev_count) {
+        char *dev_path = object_get_canonical_path(OBJECT(dev));
+        error_setg(errp, "acpi_memory_plug_cb: "
+                   "device [%s] returned invalid memory slot[%d]",
+                   dev_path, slot);
+        g_free(dev_path);
+        return;
+    }
move 3 above blocks into separate function to reduce code duplication
with acpi_memory_plug_cb(), perhaps something like following:

MemStatus *
acpi_memory_get_slot_status_descriptor(MemHotplugState *mem_st,
                                        DeviceState *dev, Error **errp);

Followed.

+
place here also missing removal signaling for selected mdev,
perhaps it's due to wrong patch ordering

It is not a wrong ordering of patches. I meant to define mdev->is_removing
in patch 6/8. But it is OK to define it here. So followed.

+    /* do ACPI magic */
+    ar->gpe.sts[0] |= ACPI_MEMORY_HOTPLUG_STATUS;
+    acpi_update_sci(ar, irq);
worth to make this block a separate function and reuse in
acpi_memory_plug_cb()

Followed.

Thanks.

+}
+
  static const VMStateDescription vmstate_memhp_sts = {
      .name = "memory hotplug device state",
      .version_id = 1,
diff --git a/hw/acpi/piix4.c b/hw/acpi/piix4.c
index b72b34e..c1d5187 100644
--- a/hw/acpi/piix4.c
+++ b/hw/acpi/piix4.c
@@ -362,6 +362,9 @@ static void piix4_device_unplug_cb(HotplugHandler 
*hotplug_dev,
      if (object_dynamic_cast(OBJECT(dev), TYPE_PCI_DEVICE)) {
          acpi_pcihp_device_unplug_cb(&s->ar, s->irq, &s->acpi_pci_hotplug, dev,
                                      errp);
+    } else if (object_dynamic_cast(OBJECT(dev), TYPE_PC_DIMM)) {
+        acpi_memory_unplug_cb(&s->ar, s->irq, &s->acpi_memory_hotplug, dev,
+                              errp);
      } else {
          error_setg(errp, "acpi: device unplug request for not supported 
device"
                     " type: %s", object_get_typename(OBJECT(dev)));
diff --git a/include/hw/acpi/memory_hotplug.h b/include/hw/acpi/memory_hotplug.h
index 7bbf8a0..fc6b868 100644
--- a/include/hw/acpi/memory_hotplug.h
+++ b/include/hw/acpi/memory_hotplug.h
@@ -28,6 +28,8 @@ void acpi_memory_hotplug_init(MemoryRegion *as, Object *owner,
void acpi_memory_plug_cb(ACPIREGS *ar, qemu_irq irq, MemHotplugState *mem_st,
                           DeviceState *dev, Error **errp);
+void acpi_memory_unplug_cb(ACPIREGS *ar, qemu_irq irq, MemHotplugState *mem_st,
+                           DeviceState *dev, Error **errp);
extern const VMStateDescription vmstate_memory_hotplug;
  #define VMSTATE_MEMORY_HOTPLUG(memhp, state) \
.



Reply via email to