On Sun, 11 Jun 2023 21:01:54 GMT, Oliver Kopp <d...@openjdk.org> wrote:
> Fix for [JDK-8240567](https://bugs.openjdk.org/browse/JDK-8240567): > "MethodTooLargeException thrown while creating a jlink image". > > Java still has a 64kb limit: A method may not be longer than 64kb. The idea > of the fix is to split up the generated methods in several smaller methods > > This is a follow-up to https://github.com/openjdk/jdk/pull/10704. GitHub did > not allow me to re-open the PR, because I did a force-push to have one commit. TL;DR: summary of the changes - In the other PR we missed that `dedupSetBuilder` requires local variables in the method for caching - Code comments: `// Restore all (!) sets from parameter to local variables` and `// Store all new sets to List` - In this PR, these variables are transferred from method to method by a `ArrayList` - `0` is `this` when a method is called. `this` is needed to call the next helper method, thus we moved `BUILDER_VAR` to another place. - Code comments: `MAX_LOCAL_VARS + 1; // we need 0 for "this"` and on `MAX_LOCAL_VARS`: `// Constant chosen for this generator, can be higher in practice` > @koppor Is this ready for review? Yes, it is! We especially checked it with our (rather large) desktop application. > The other PR went through a dozens or so iterations before it was returned to > draft. I remember 😅. Therefore, we first had this as draft, waited for the "on-site checks" and then opened it. > It seems like you were still battling with verifier errors. We missed an important point and the other PR. > The comment on this PR says you it was created because a force-push so I > can't tell if you the changes are ready or not. I am sorry for that. I weighted the requirement to have a single commit higher than reviewing the changes. The "excuse" for this decision is that nearly all code submitted back than has changed. ------------- PR Comment: https://git.openjdk.org/jdk/pull/14408#issuecomment-1586615579
