RFR: 2020-06-04, Bulk integration of Shenandoah 8u to aarch64-port/jdk8u-shenandoah

[Andrew Hughes]

On 04/06/2020 10:23, Aleksey Shipilev wrote:
> https://cr.openjdk.java.net/~shade/shenandoah/merges/aarch64-port-8u-20200604/webrev.01/
> 
> This is another integration of Shenandoah 8u from our shenandoah/jdk8 staging repository. It was
> generated by plain pull from shenandoah/jdk8/hotspot repository and automatic merge.
> 
> The webrev is not very large, and mostly contains file moves and associated build changes. IMO, the
> only thing that deserves explanation is the addition in src/share/vm/runtime/os.hpp:
>   https://mail.openjdk.java.net/pipermail/shenandoah-dev/2020-June/012413.html
> 
> We would negotiate the push order w.r.t. upstream sync with Andrew Hughes.
> 
> Changes:
>   Shenandoah: move parallelCleaning.* to shenandoah/
>   Shenandoah: fix formats in ShenandoahStringSymbolTableUnlinkTask
>   [backport] 8245812: Shenandoah: compute root phase parallelism
>   [backport] 8245814: Shenandoah: reconsider format specifiers for stats
>   [backport] 8245463: Shenandoah: refine ShenandoahPhaseTimings constructor arguments
>   [backport] 8245754: Shenandoah: ditch ShenandoahAlwaysPreTouch
>   [backport] 8245461: Shenandoah: refine mode name()-s
>   [backport] 8245726: Shenandoah: lift/cleanup ShenandoahHeuristics names and properties
>   [backport] 8245825: Shenandoah: Remove diagnostic flag ShenandoahConcurrentScanCodeRoots
>   Shenandoah: move barrier sets to their proper locations
>   Shenandoah: Fix build failure with +JFR -PCH
>   Shenandoah: fix runtime linking failure due to non-compiled shenandoahBarrierSetC1
>   [backport] 8246162: Shenandoah: full GC does not mark code roots when class unloading is off
>   [backport] 8245757: Shenandoah: AlwaysPreTouch should not disable heap resizing or uncommits
>    Merge
> 
> Testing: hotspot_gc_shenandoah {fastdebug,release}; it was also tested in sh/jdk8 with a few nightlies
> 

Is it possible to get the merge changeset itself, exported in git diff
format? What webrev has produced shows moved files as huge diffs of one
file being removed and another being added.

os.hpp change is ugly, but I guess needed.

Thanks,
-- 
Andrew :)

Senior Free Java Software Engineer
Red Hat, Inc. (http://www.redhat.com)

PGP Key: ed25519/0xCFDA0F9B35964222 (hkp://keys.gnupg.net)
Fingerprint = 5132 579D D154 0ED2 3E04  C5A0 CFDA 0F9B 3596 4222