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

Re: [libvirt] [PATCH] qemu: Restore the original states of PCI device when restarting daemon

On 10/20/2011 04:45 AM, Osier Yang wrote:
This patch is to solve the problem by introducing internal XML
(won't be dumped to user, only dumped to status XML). The XML is:

Which element will this be added under?

+++ b/src/conf/domain_conf.c
@@ -59,8 +59,12 @@ verify(VIR_DOMAIN_VIRT_LAST<= 32);
  /* Private flags used internally by virDomainSaveStatus and
   * virDomainLoadStatus. */
  typedef enum {
-   VIR_DOMAIN_XML_INTERNAL_STATUS = (1<<16), /* dump internal domain status information */
-   VIR_DOMAIN_XML_INTERNAL_ACTUAL_NET = (1<<17), /* dump/parse<actual>  element */
+   /* dump internal domain status information */
+   /* dump/parse<actual>  element */
+   /* dump/parse original states of host PCI device */

Is this worth a new bit, or does it fit under the already-existing umbrella of VIR_DOMAIN_XML_INTERNAL_STATUS, since we are treating these three bits as internal status? But we can change that later, if we need.

+++ b/src/conf/domain_conf.h
@@ -153,6 +153,31 @@ struct _virDomainDeviceInfo {
      } master;

+typedef struct _virDomainHostdevOrigStates virDomainHostdevOrigStates;
+typedef virDomainHostdevOrigStates *virDomainHostdevOrigStatesPtr;
+struct _virDomainHostdevOrigStates {
+    union {

Don't you need some sort of discrimination field outside of the union which tells you which part of the union is valid? Typically we use a discriminator such as 'int type; /* enum vir... */'. But until we have something to distinguish from, I guess this works.

+        struct {
+            /* Does the device need to ubind from stub when


ACK with spelling nit fixed.

Eric Blake   eblake redhat com    +1-801-349-2682
Libvirt virtualization library http://libvirt.org

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