RFR: 8242459: ForceNUMA and only one available NUMA node hits a guarantee

[stefan.johansson at oracle.com]

On 2020-04-27 15:02, Kim Barrett wrote:
>> On Apr 27, 2020, at 6:06 AM, Kim Barrett <kim.barrett at oracle.com> wrote:
>>
>>> On Apr 27, 2020, at 3:42 AM, stefan.johansson at oracle.com wrote:
>>> Looks good, might be interesting to run a few tier with +ForceNUMA, to make sure we don't hit anything else.
>>
>> Thanks.  I’m running tier1 with +ForceNUMA right now.
> 
> It turned out the way I was configuring things I need to run tier2 to actually use it.
> 
> And it seems to not work on Windows.  But that’s not too surprising, given this from os_windows.cpp:
> 4095  if (UseNUMA && !ForceNUMA) {
> 4096    UseNUMA = false; // We don't fully support this yet
> 4097  }
> 
> I haven’t looked further; I think that’s out of scope for JDK-8242459.
> I’ll followup with a separate bug.
> 

Sound good to me.