RFC: Pick up jdk-13+22 to sh/jdk

[Roman Kennke]

Yes, let's do it.

Thanks,
Roman

> jdk/jdk is somewhat stable at the moment, let's pick it up to sh/jdk. This is the penultimate
> stepping stone for upstreaming x86_32 support. The merges were simple, but not trivial, but I
> verified those are correct by diffing against jdk/jdk. We would check webrevs again once this merge
> lands.
> 
> Changesets:
>   http://cr.openjdk.java.net/~shade/shenandoah/merges/jdk-13%2b22/changesets.txt
> 
> Notable ones:
>   8221507: Implement JFR Events for Shenandoah
>   8224495: Shenandoah: Do not rescan code roots in final mark pause if it is not degenerated GC
>   8224579: ResourceMark not declared in shenandoahRootProcessor.inline.hpp with
> --disable-precompiled-headers
>   8224508: Shenandoah: Need to update thread roots in final mark for piggyback ref update cycle
>   8224525: Shenandoah: Eliminate shenandoah verifier's side-effects
>   8224529: [TESTBUG] JFR TestShenandoahHeapRegion* tests fail on build w/o Shenandoah
>   8224522: Shenandoah should apply barriers on deoptimization
>   8224667: Shenandoah: Post-LRB cleanup
>   8224679: Shenandoah: Make ShenandoahParallelCodeCacheIterator noncopyable
>   8224115: Shenandoah: Eliminate RWLock that protects recorded nmethod data array
>   8224751: Shenandoah: Shenandoah Verifier should select proper roots according to current GC cycle
>   8224584: Shenandoah: Eliminate forwarding pointer word
>   8224496: Shenandoah compilation fails with assert(is_CountedLoopEnd()) failed: invalid node class
>   8224970: ShenandoahRootScanner::roots_do assert is too strong
>   8224932: Shenandoah: Rename ShenandoahHeapLock, make it general purpose lock
>   8224875: Shenandoah: ParallelCleaning code unloading should take lock to protect shared code roots
> 
> Testing: hotspot_gc_shenandoah {x86_32, x86_64}, eyeballing the diff against jdk/jdk
>