On Tue, 29 Sep 2020 07:28:20 GMT, Per Liden <pli...@openjdk.org> wrote:
>> FTR, this is the RFE to remove the oop verification from the oop_iterate >> framework: >> https://bugs.openjdk.java.net/browse/JDK-8237363 >> >> I really would like to get rid of it, but got push back because it made GCs >> have to duplicate the effort to provide >> verification. > > I've also has problems with this assert in the past, and I think that the > underlying assumption of the assert is wrong. > I would not miss it. IMO it's ok to remove it. However, it can be argued that is_in() should not check for 'in committed memory' but 'in reserved space for heap', IOW 'is this a pointer into our heap memory region?'. Or maybe this would be CH::is_in_reserved() and we should change the assert to that? Also, IMO we should not do anything that may trip barriers on oop-iterators or similar GC-only paths. ------------- PR: https://git.openjdk.java.net/jdk/pull/296
