Request for review (s) - 8024366: UseNUMAInterleaving should be available for other GC's.

[Jon Masamitsu]

On 01/27/2014 12:55 AM, Thomas Schatzl wrote:
> Hi Jon,
>
>    the comment
>
> +     // For those collectors or operating systems that do
>        // not support full UseNUMA, we will map to UseNUMAInterleaving for now
>
> indicates to me that the following mapping is done only for "those
> collectors or operating systems that do not support full UseNUMA" while
> it seems to me that this mapping is now done for all collectors/OSes
> when UseNUMA is enabled. (Regardless of support, could not find any
> logic that disables wrongly set UseNUMA).

Are you asking about OS's that have no support for anything
related to NUMA?

What's an example where UseNUMA needs to be disabled?

Jon

On Fri, 2014-01-24 at 14:54 -0800, Jon Masamitsu wrote:
>> UseNUMAInterleaving was previously implemented for platforms
>> and garbage collectors that did not have full support for UseNUMA.
>> This change turns on UseNUMAInterleaving if UseNUMA is on.
>>
>> http://cr.openjdk.java.net/~jmasa/8024366/webrev.00/
> Maybe the note for the release notes in the bug report could be
> improved: it indicates that "UseNUMA and UseNUMAInterleaving are
> independent flags now, and that UseNUMA would turn on
> UseNUMAInterleaving only for parallel gc."
>
> This does not seem to be the case. To me, this change turns on
> UseNUMAInterleaving if UseNUMA is on on *all* platforms and *all*
> collectors (which also means that there is still an ergonomic decision
> involved).
>
> Thanks,
> Thomas
>