On Nov 9, 2009, at 3:01 PM, Garrett D'Amore wrote: > Michael Corcoran wrote: >> >> On Nov 3, 2009, at 3:52 PM, Jim Walker wrote: >> >>> Garrett D'Amore wrote: >>>> Jim Walker wrote: >>>>> Garrett D'Amore wrote: >>>>>> Can you add *which* kstats are being used? Individual kstats >>>>>> are normally not documented, and some of them are more volatile >>>>>> than others. If this case is going to use kstats, then I think >>>>>> we need to document which ones are used in case someone decides >>>>>> to change them in the future. >>>>> >>>>> Appendix A. in the proposal file has been updated to include the >>>>> individual kstat information. >>>>> >>>>> Cheers, >>>>> Jim >>>> Thanks. The stats you're using look reasonable to me, but I'd >>>> like to hear from someone more familiar with the VM subsystem -- >>>> the anoninfo and cpu_vminfo stats might be unreliable... >>> >>> Here are the anoninfo and cpu_vminfo stats being used. >>> >>> anoninfo.ani_max; >>> anoninfo.ani_resv; >>> cpu_vminfo.pgpgin; >>> cpu_vminfo.pgpgout; >>> >>> Can someone knowledgeable give an idea of their reliability? >> >> I ran these by a few VM people and we all agree that these should >> be relatively stable. > > Thank you, that makes me happier. Are any of them willing to have > their names listed here? First there's my name, then Blake Jones and Stan Studzinski.
--Mike > > -- Garrett >> >> --Mike >> >>> >>> Currently, we are assuming they are as reliable as any >>> other stat. >>> >>> I would like to wrap up this case. >>> >>> Here's the full proposal: >>> http://arc.opensolaris.org/caselog/PSARC/2009/583/proposal.txt >>> >>> Cheers, >>> Jim >>> _______________________________________________ >>> opensolaris-arc mailing list >>> opensolaris-arc at opensolaris.org >> >
