david.holmes at oracle.com
Mon Dec 17 22:33:12 UTC 2018
On 18/12/2018 6:16 am, David Lloyd wrote:
> On Mon, Dec 17, 2018 at 2:00 PM Florian Weimer <fweimer at redhat.com> wrote:
>> * Doug Lea:
>>> Yes. It will make the getProcessorId() method interesting to specify.
>>> Probably Runtime.getAvailableProcessors() would also need some work.
>> We already fake a processor count of two if we can't determine the real
>> count, to prevent Hotspot from activating uniprocessor
>> optimizations. 8-/
> Maybe there needs to be a separate method on Runtime which yields the
> maximum number of processors possible in the system/environment. If
> the maximum possible is 1, then the JVM wouldn't have to worry about a
> sudden change from 1 to 2 online CPUs.
Hotspot no longer concerns itself with this as it assumes it is always
running on a multi-processor (barring a couple of corner cases on ARM
where different instructions would be needed).
More information about the loom-dev