RFR: 8239347: Refactor Symbol to make _length a standalone field again

Ioi Lam ioi.lam at oracle.com
Tue Feb 18 17:27:55 UTC 2020


Hi Claes,

This looks good. Thanks for finding and fixing this!

- Ioi

On 2/18/20 8:21 AM, Claes Redestad wrote:
> Hi,
>
> JDK-8207359 merged Symbol::_refcount and _length into a single field. We
> could refactor this to instead pack _refcount and _identity_hash,
> leaving _length as a standalone field.
>
> _length is used in a lot more places, so not having to call a method to
> extract the value using shifts results in a code size reduction and
> potential speed-up. Except for a path only used when dumping CDS
> archives, _identity_hash is only accessed in one place, and there in
> conjunction with an access to length.
>
> Bug:    https://bugs.openjdk.java.net/browse/JDK-8239347
> Webrev: http://cr.openjdk.java.net/~redestad/8239347/open.00/
>
> Testing: tier1-4, verified a small static footprint reduction (~4kb),
>          verified accesses to _length reduced from a mov+shr to a
>          movzwl
>
> /Claes



More information about the hotspot-runtime-dev mailing list