RFR: More cleanup patches
Martin Buchholz
martinrb at google.com
Fri Mar 30 16:39:43 UTC 2018
On Fri, Mar 30, 2018 at 4:55 AM, mandy chung <mandy.chung at oracle.com> wrote:
>
>
> On 3/30/18 7:38 PM, Peter Levart wrote:
>
>
>
> On 03/30/2018 01:31 PM, mandy chung wrote:
>
>
>
> On 3/30/18 6:53 PM, Alan Bateman wrote:
>
>
>
> 8200134: Improve ModuleHashesBuilder
> http://cr.openjdk.java.net/~martin/webrevs/jdk/ModuleHashesBuilder/
> https://bugs.openjdk.java.net/browse/JDK-8200134
>
> I suspect Mandy will want to look at this one. We also need to decide if
> there is a better place for it as it is only used by tools (`jar`, `jmod`,
> ...), it is not part of the runtime support.
>
>
> It's in java.base to avoid duplicating this class in both jdk.jar and
> jdk.jlink module. It's not a small method. It's not part of the runtime
> support. I can't think of any better place.
>
>
> New module jdk.tools.internal ?
>
>
> I considered it while I fear that this would become a dumping ground for
> tools utility although we start with one class. The name would be
> jdk.internal.xxx.
>
Right now the jdk has 10^2 modules. Why not 10^4? ModuleHashesBuilder
could be in its own dedicated module.
The good news is that ModuleHashesBuilder improvements in this change will
itself avoid any future performance problems with a large number of modules!
But that could/should be done in a separate change.
More information about the core-libs-dev
mailing list