RFR: 8352003: Support --add-opens with -XX:+AOTClassLinking [v3]

Alan Bateman alanb at openjdk.org
Tue Apr 22 06:11:48 UTC 2025


On Mon, 21 Apr 2025 22:54:37 GMT, Calvin Cheung <ccheung at openjdk.org> wrote:

>> src/java.base/share/classes/jdk/internal/module/ModuleBootstrap.java line 165:
>> 
>>> 163:             BootLoader.getUnnamedModule(); // trigger <clinit> of BootLoader.
>>> 164:             CDS.defineArchivedModules(ClassLoaders.platformClassLoader(), ClassLoaders.appClassLoader());
>>> 165:             boolean extraExportsOrOpens = addExtraExportsAndOpens(bootLayer);
>> 
>> (1) The returned value of `addExtraExportsAndOpens()` is not used. So I think this function can be changed to `void`, and all occurrences of the local variable `extraExportsOrOpens` can be removed.
>> 
>> (2) I traced the code paths that depend on the effects of `--add-opens` and `--add-exports`. It Looks like some of the effects are recorded in the `java.lang.Module$ReflectionData::export` table:
>> 
>> https://github.com/openjdk/jdk/blob/684d3b336e9cb31707d35e75f9b785e04e1fdbee/src/java.base/share/classes/java/lang/Module.java#L398C2-L412
>> 
>> 
>>         /**
>>          * A module (1st key) exports or opens a package to another module
>>          * (2nd key). The map value is a map of package name to a boolean
>>          * that indicates if the package is opened.
>>          */
>>         static final WeakPairMap<Module, Module, Map<String, Boolean>> exports =
>>             new WeakPairMap<>();
>> 
>> 
>> This table is *not* stored as part of the `ArchivedBootLayer`, so we must re-initialize this table in the production run. @AlanBateman could you confirm that this is correct.
>> 
>> Eventually, we should enhance the `ArchivedBootLayer` to also include the tables in `Module$ReflectionData`. That will obviate the call to `addExtraExportsAndOpens()` and save a few bytecodes during start-up (but the overall impact would be small, so it's not critical in the current PR). Because these tables use WeakReference, we need to wait for [JDK-8354897](https://bugs.openjdk.org/browse/JDK-8354897).
>
> Fixed (1) above.

> (2) I traced the code paths that depend on the effects of `--add-opens` and `--add-exports`. It Looks like some of the effects are recorded in the `java.lang.Module$ReflectionData::export` table:
> 
> https://github.com/openjdk/jdk/blob/684d3b336e9cb31707d35e75f9b785e04e1fdbee/src/java.base/share/classes/java/lang/Module.java#L398C2-L412
> 
> ```
>         /**
>          * A module (1st key) exports or opens a package to another module
>          * (2nd key). The map value is a map of package name to a boolean
>          * that indicates if the package is opened.
>          */
>         static final WeakPairMap<Module, Module, Map<String, Boolean>> exports =
>             new WeakPairMap<>();
> ```
> 
> This table is _not_ stored as part of the `ArchivedBootLayer`, so we must re-initialize this table in the production run. @AlanBateman could you confirm that this is correct.

In the changes for "JEP draft: Prepare final means final", this is changed significantly so that reflectively exporting or opening a package during startup will add to openPackages/exportedPackages. Once the VM is fully initialized then reflective change via the addExports/addOpens API make use of ReflectionData. So no ReflectionData or weak refs setup for --add-exports/--add-opens. We could potential separate this out in advance so that ReflectionData doesn't need to be re-initialized.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/24695#discussion_r2053392827


More information about the core-libs-dev mailing list