RFR: 8300915: G1: incomplete SATB because nmethod entry barriers don't get armed [v5]

[Erik Österlund]

On Thu, 26 Jan 2023 22:10:10 GMT, Richard Reingruber <rrich at openjdk.org> wrote:

>> src/hotspot/share/gc/g1/g1CollectedHeap.cpp line 3389:
>> 
>>> 3387:     CodeCache::on_gc_marking_cycle_start();
>>> 3388:   }
>>> 3389:   if (!full_gc) {
>> 
>> Should this be else if?
>
> No, `!full_gc` means we reach here because concurrent marking is about to start and we want to arm unconditionally then. Never arm for the start of a full gc on the other hand.
> 
> Would it be clearer if I renamed `full_gc` to `concurrent_marking_start`?

Yes, I think so. Otherwise, looks good. Thanks again!

-------------

PR: https://git.openjdk.org/jdk/pull/12194