RFR: Backport of JDK-8209686: cleanup arguments to PhaseIdealLoop() constructor

Langer, Christoph christoph.langer at sap.com
Thu Feb 6 13:07:47 UTC 2020


Hi Roman,

I've approved this backport. Does that mean your Shenandoah patch for 11u needs to be rebased?

Cheers
Christoph

> -----Original Message-----
> From: jdk-updates-dev <jdk-updates-dev-bounces at openjdk.java.net> On
> Behalf Of Roman Kennke
> Sent: Donnerstag, 6. Februar 2020 12:15
> To: Aleksey Shipilev <shade at redhat.com>; jdk-updates-dev <jdk-updates-
> dev at openjdk.java.net>
> Subject: Re: RFR: Backport of JDK-8209686: cleanup arguments to
> PhaseIdealLoop() constructor
> 
> Hi Aleksey,
> 
> >> The original patch applies cleanly on jdk11u-dev, except for some very
> >> minor changes that I needed to make in loopnode.hpp because
> surrounding
> >> comments/code did not match.
> >
> > The changes are not in the code, though, right? I cannot spot any. I
> presume you just needed to
> > reapply some hunks by hand?
> 
> Yes. I did hg transplant to transplant the original change and then
> re-applied the changes. It didn't apply due to surrounding changes like
> here:
> 
>      _dom_lca_tags(arena()), // Thread::resource_area
>      _verify_me(NULL),
>      _verify_only(true) {
> -    build_and_optimize(false, false);
> +    build_and_optimize(LoopOptsVerify);
>    }
> 
> 
>      _verify_me(NULL),
>      _verify_only(true),
>      _dom_lca_tags(arena()) { // Thread::resource_area
> -    build_and_optimize(false, false);
> +    build_and_optimize(LoopOptsVerify);
>    }
> 
> I.e. the changed constructor args order.
> 
> >> Bug:
> >> https://bugs.openjdk.java.net/browse/JDK-8209686
> >> Webrev:
> >> http://cr.openjdk.java.net/~rkennke/JDK-8209686-jdk11u/webrev.00/
> >> Testing: tier1, tier2
> >
> > The backport looks good.
> 
> Thanks for reviewing!
> Roman



More information about the jdk-updates-dev mailing list