RFR: 8254955: x86: MethodHandlesAdapterBlob is too big [v2]

[Vladimir Kozlov]

On Sat, 17 Oct 2020 22:58:21 GMT, Claes Redestad <redestad at openjdk.org> wrote:

>> At some point JSR 292 was reworked to generate all but a small handful of interpreter stubs lazily, leaving the
>> MethodHandlesAdapterBlob with a bit too much room to spare.
>> The remaining stubs use less than 1000 bytes of memory in product builds, and less than 3k in debug builds. This patch
>> adjust the sizes accordingly.
>> Other platforms (except zero) seem like they could use a similar adjustment, but I don't have hardware available to
>> check how big the interpreter stubs get on anything but x86 so I'll leave them untouched unless someone can run the
>> numbers (`java -XX:+UnlockDiagnosticVMOptions -XX:+VerifyMethodHandles -XX:+LogCompilation` - grep the hotspot_log
>> generated for MethodHandlesAdapterBlob or just blob since it's the first one)
>
> Claes Redestad has updated the pull request incrementally with one additional commit since the last revision:
> 
>   ZGC & Shenandoah requires a little more space

Good.

-------------

Marked as reviewed by kvn (Reviewer).

PR: https://git.openjdk.java.net/jdk/pull/717