RFR: 8210233: Prepare Klass::is_loader_alive() for concurrent class unloading

Kim Barrett kim.barrett at oracle.com
Tue Sep 4 03:51:35 UTC 2018

> On Sep 3, 2018, at 11:25 AM, Erik Österlund <erik.osterlund at oracle.com> wrote:
> Hi Kim,
> You are right. Thank you for catching that.
> Incremental:
> http://cr.openjdk.java.net/~eosterlund/8210233/webrev.00_01/
> Full:
> http://cr.openjdk.java.net/~eosterlund/8210233/webrev.01/
> Thank you for the review.

The comments I was referring to were these:

is_loader_alive in klass.hpp
657   // Iff the class loader (or mirror for unsafe anonymous classes) is alive the
658   // Klass is considered alive.  Has already been marked as unloading.

471     // We do not display unloading loaders, for now.

I'm once again finding myself confused about the relationship between
alive or not and unloading or not. Is the second sentence of the
comment for is_loader_alive even true now?

The new comment for is_alive is fine.

In the new comment for is_unloading, the use of "unloaded" is
confusing me.  Is that a typo for "unloading", or am I confused?

I think I need to retract my review; I think I need to refresh and
de-confuse my understanding of this area.  Don't wait for me if you
get a second okay from someone else.

More information about the hotspot-runtime-dev mailing list