RFR: 8202548: Use reservation Object when creating SpeciesData

[Claes Redestad]

Thanks Peter,

I incorporated your changes and pushed. Sadly missed one of the 
editorial changes somehow, but I'll
make a note to have that fixed next time I touch this.

/Claes


On 2018-05-07 20:46, Peter Levart wrote:
> Hi,
>
> This looks good, except that comments in method are now obsolete. I 
> suggest the following rewording:
>
> http://cr.openjdk.java.net/~plevart/jdk-dev/8202548_ClassSpecializer/webrev.02/
>
> Regards, Peter
>
>
> On 05/07/18 18:57, Paul Sandoz wrote:
>> +1 this is better.
>>
>> Paul.
>>
>>> On May 2, 2018, at 6:54 AM, Claes Redestad<claes.redestad at oracle.com>  wrote:
>>>
>>> Hi,
>>>
>>> a small follow-up to JDK-8202184 in line with Peter Levart's suggestion, where we can avoid
>>> instantiating two different SpeciesData objects, and should avoid creating an instance of
>>> the Function on each lookup:
>>>
>>> Webrev:http://cr.openjdk.java.net/~redestad/8202548/open.00/
>>> Bug:https://bugs.openjdk.java.net/browse/JDK-8202548
>>>
>>> This is a small micro-optimization, as well as granting more freedom to future newSpeciesData
>>> implementations.
>>>
>>> Thanks!
>>>
>>> /Claes
>