RFR: 8301997: Move method resolution information out of the cpCache

[Matias Saavedra Silva]

On Tue, 17 Oct 2023 16:41:41 GMT, Andrew Dinn <adinn at openjdk.org> wrote:

>> The current structure used to store the resolution information for methods, ConstantPoolCacheEntry, is difficult to interpret due to its ambigious fields f1 and f2. This structure previously held information for fields, methods, and invokedynamic calls which were all encoded into f1 and f2. Currently this structure only handles method entries, but it remains obtuse and inconsistent with recent changes. 
>> 
>> This enhancement introduces a new data structure that stores the necessary resolution data in an intuitive an extensible manner. These resolved entries are stored in an array inside the constant pool cache in a very similar manner to invokedynamic entries in JDK-8301995.
>> 
>> Instances of ConstantPoolCache entry related to field resolution have been replaced with the new ResolvedMethodEntry, and ConstantPoolCacheEntry has been removed entirely. The class ResolvedMethodEntry holds resolution information for all types of invoke calls besides invokedynamic, and thus has fields that may be unused depending on the invoke code. 
>> 
>> To streamline the review, please consider these major areas that have been changed:
>> 1. ResolvedMethodEntry class
>> 2. Rewriter for initialization of the structure
>> 3. cpCache for resolution
>> 4. InterpreterRuntime, linkResolver, and templateTable
>> 5. JVMCI
>> 6. SA
>> 
>> Verified with tier 1-9 tests.
>> 
>> This change supports the following platforms: x86, aarch64
>
> src/hotspot/share/ci/ciReplay.cpp line 436:
> 
>> 434: #endif
>> 435:         ResolvedMethodEntry* method_entry = cp->cache()->resolved_method_entry_at(index);
>> 436:         cp->cache()->set_method_handle(index, callInfo);
> 
> It looks a bit odd that you obtain a pointer to the `ResolvedMethodEntry` at `index` by calling `resolved_method_entry_at` and then pass `index` back into `set_method_handle` in order to update it with the call info. Obviously this is done so that the set routine can handle data races. Would it not be better to modify `set_method_handle` so that it handled the race and also returned the `ResolvedMethodEntry` at `index`.
> 
> Likewise you pass `index` back in the call to `appendix_if_resolved` below. Would it not be better to have this method accept a `ResolvedMethodEntry` pointer?
> Likewise

Right, currently these calls seem redundant since it reads the resolved method entry twice. If I understand correctly, you are suggesting that the methods look something like this?
`ResolvedMethodEntry* set_method_handle(int index, const CallInfo &call_info)`
`oop appendix_if_resolved(ResolvedMethodEntry* method_entry)`

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

PR Review Comment: https://git.openjdk.org/jdk/pull/15455#discussion_r1362501180