Heidi Eckhart wrote:
The starting point for this patch series was the missing validation of the client's given object path of ResourceAllocationSettingData in the RAFP provider. This is fixed by patch #1. Patch #2 switches the return code for getInstance from FAILED to NOT_FOUND for ResourceAllocationSettingData. Patch #3 registeres the RASD provider for KVM. Patch #4 switches the return code for getInstance from FAILED to NOT_FOUND for DevicePools.
This patch tested fine for me. Patch #4 confused me at first - wasn't sure why the cu_compare_ref() piece was being removed. But since DevicePools only have one key (which we use to grab the instance anyway), there's not need to revalidate the key again. -- Kaitlin Rupert IBM Linux Technology Center karupert@us.ibm.com