[DMARC FAILURE] [11u] RFR 8208601: Introduce native oop barriers in C2 for OopHandle

[Jaroslav Bachorík]

Hi,

On Mon, Sep 9, 2019 at 11:13 AM Langer, Christoph <christoph.langer at sap.com>
wrote:

> Hi Andrew,
>
> > > I'll add the patches for JDK-8208582, JDK-8208601 (with the added
> > > unsafe) and JDK-8210158 to our test queue and let you know the
> > > results. Please don't push until then.
> >
> > I've looked at the Jira but it's not clear why we need this in 11u.
>
> you mean why "JDK-8208601: Introduce native oop barriers in C2 for
> OopHandle" is needed? It is a prerequisite for "JDK-8210158: Accessorize
> JFR getEventWriter() intrinsics" which ought to resolve a potential JFR
> crash. But if only the patch for 8210158 was used, it would not compile. I
> think it would e.g. miss the definition of Node* access_load in
> graphKit.hpp.
>

Indeed. Instead of manually tinkering around to make JDK-8210158 backport
somehow working it is much better to take in JDK-8208601 and be sure that
things are working as intended.

Regards,

-JB-


>
> Best regards
> Christoph
>
>