Re:Java’s new garbage collector promises low pause times on ...
by pitaj on Thursday August 30, 2018 @11:32AM
(#17877533)
> It also might be worth pointing out that you now need 16x the page mappings (Since every mapping needs 15 duplicates for the possible flag states)
Nope. They went into this in the article:
> Since by design only one of remap, mark0 and mark1 can be 1 at any point in time, it’s possible to do this with three mappings. There’s a nice diagram[1] in the ZGC source for this.
That might not be totally current, as it doesn't cover the finalizable flag, but if it works the same, that would only be four mappings. If it works differently, then it would be a maximum of 6 mappings. Not 16.
Re:Java’s new garbage collector promises low pause times on ...
by shaftway on Thursday August 30, 2018 @12:31PM
(#17878048)
And it looks like you only need 3 mappings for the entire space. It's not like you need one per object in memory. Unless I misunderstood the gist of the article.
Nope. They went into this in the article:
> Since by design only one of remap, mark0 and mark1 can be 1 at any point in time, it’s possible to do this with three mappings. There’s a nice diagram[1] in the ZGC source for this.
[1]: http://hg.openjdk.java.net/zgc/zgc/file/59c07aef65ac/src/hot...
That might not be totally current, as it doesn't cover the finalizable flag, but if it works the same, that would only be four mappings. If it works differently, then it would be a maximum of 6 mappings. Not 16.