RFR (S) 8149459: StringConcatFactory should be synced up with LambdaMetafactory

Vladimir Ivanov vladimir.x.ivanov at oracle.com
Wed Feb 10 13:19:08 UTC 2016


Looks good.

Best regards,
Vladimir Ivanov

On 2/10/16 3:51 PM, Aleksey Shipilev wrote:
> See new webrev that adds a few more tests:
>   http://cr.openjdk.java.net/~shade/8149459/webrev.02/
>
> On 10.02.2016 14:25, Remi Forax wrote:
>> Hi Aleksey, if you have a dumper, you should append a value at the
>> end of the generated class otherwise if you have more than one string
>> concatenation, you will see only one class on the disk.
>
> No, we are actually okay, because our class names are unique since
> Unsafe.defineAnonymousClass handles this:
>
> $ ls -lR dump/java.lang.String\$Concat/
> dump/java.lang.String$Concat/2008017533.class
> dump/java.lang.String$Concat/665576141.class
>
> Cheers,
> -Aleksey
>



More information about the core-libs-dev mailing list