RFR: 8030815: Code roots are not accounted for in region prediction [v3]

[Thomas Schatzl]

On Mon, 18 Sep 2023 11:35:02 GMT, Albert Mingkun Yang <ayang at openjdk.org> wrote:

> OK, could we reevaluate its effect once the situation is ready and decide whether to keep this heuristic or not?
> 

I will look into this again as soon as the code root rem set changes settled a bit.

> > I needed to start somewhere with changes.
> 
> Sure; however, given that the code-root duration can be quite erratic, wouldn't incorporating it into the heuristic potentially have some negative impact? It seems to me that resolving the surrounding issues should take precedence. Ofc, this is quite subjective.
> 

I did internal performance testing with no impact (the tests typically do not have many code root sets). The balanced code root set change will be provided for review soon.

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

PR Comment: https://git.openjdk.org/jdk/pull/15632#issuecomment-1725045961