RFR 6996807 : FieldReflectorKey hash code computation can be improved

[Roger Riggs]

Hi Ivan,

Happy New Year!

This looks fine.

The indentation of line 2311 makes it harder to see all the conditions 
for the true case.

No new webrev needed.

Thanks, Roger



On 12/31/2018 11:12 PM, Ivan Gerasimov wrote:
> Gentle ping.
>
> I updated the copyright year, as it seems unlikely this fix is going 
> to be reviewed in 2018 :)
>
> Happy New Year to everyone!
>
> Ivan
>
>
> On 12/18/18 2:42 AM, Ivan Gerasimov wrote:
>> Hello!
>>
>> FieldReflectorKey class contains a combined-string representation of 
>> the fields (their names and signatures).
>>
>> This string is used for calculating the hash code and comparing two 
>> FieldReflectorKey objects for equality, which is expensive.
>>
>> It is proposed to store the names and signatures (which are likely to 
>> be interned) in an array an use this array for both tasks utilizing 
>> Arrays.hashCode() and Arrays.equals().
>>
>> BUGURL: https://bugs.openjdk.java.net/browse/JDK-6996807
>> WEBREV: http://cr.openjdk.java.net/~igerasim/6996807/00/webrev/
>>
>> Would you please help review the fix?
>>
>