Re: [libvirt] [PATCH 1/6] virNodeGetCpuTime: Expose new API

2011/4/1 Eric Blake <eblake(a)redhat.com&gt;: > On 03/31/2011 07:55 PM, Minoru Usui wrote: >> virNodeGetCpuTime: Expose new API >> >>  include/libvirt/libvirt.h.in |   26 ++++++++++++++++++++++++++ >>  src/libvirt_public.syms      |    1 + >>  2 files changed, 27 insertions(+), 0 deletions(-) > >> >> +/** >> + * virNodeCpuTime: >> + * >> + * a virNodeCpuTime is a structure filled by virNodeGetCpuTime() and providing >> + * the information for the cpu time of Node. >> + */ >> + >> +typedef struct _virNodeCpuTime virNodeCpuTime; >> + >> +struct _virNodeCpuTime { >> +    unsigned long long user; >> +    unsigned long long system; >> +    unsigned long long idle; >> +    unsigned long long iowait; >> +}; > > Can we portably get all of this information on Windows?  If not, how do > you express which values we don't know how to obtain? > In the context of ESX I vote against this absolute CPU time values. ESX provides this values relative to a 20 second timeslots with 1 hour of history. This makes it nearly impossible to obtain the absolute CPU time. The same problem already exists for the domain's virtual CPU time. When you look at virt-top's usage of the domain's virtual CPU time, you see that it actually doesn't really care about the absolute value, but deduces the CPU utilization from it. I suggest that we find a different representation for this information that is not by definition impossible to implement for ESX. Matthias