DB> Not your fault, but this schedinfo command is a really dumb wrt to
DB> option handling. Requiring that we add new options for every
DB> possible schedular param for every possible hypervisor is just
DB> crazy.
Agreed. I actually just wrote this patch for my own testing, but
decided to send it out as part of the rest anyway.
DB> it should use
DB> schedinfo --set weight=10
Agreed :)
DB> so it automatically has support for every parameter supported by
DB> the hypervisor in question. It can validate that the user suplied
DB> param name is correct by calling 'virDomainGetSchedularParams' to
DB> see what are available.
I can do that, but since the CIM providers don't depend on virsh for
this, can we move forward with the first two patches before we get
this fixed up?
--
Dan Smith
IBM Linux Technology Center
Open Hypervisor Team
email: danms(a)us.ibm.com