RFR: 8302351: "assert(!JavaThread::current()->is_interp_only_mode() || !nm->method()->is_continuation_enter_intrinsic() || ContinuationEntry::is_interpreted_call(return_pc)) failed: interp_only_mode but not in enterSpecial interpreted entry" in fixup_callers_callsite [v2]

[Patricio Chilano Mateo]

On Thu, 29 Jun 2023 07:00:05 GMT, Serguei Spitsyn <sspitsyn at openjdk.org> wrote:

> The fix looks pretty good to me. I have one question on the comment:
> 
> ```
> 1288 // TODO: _linkToNative doesn't have an interpreted version so we always
> 1289 // return the compiled code entry point.
> ```
> 
> I guess there is already a bug filed on this, is not it?
>
Thanks for the review Serguei. There are additional changes that need to be made to handle the case of resolving method handle intrinsics, since we cannot just return the c2i as I mentioned above. There are also issues with enterSpecial and doYield since those do not have interpreted versions. I have been testing a patch (https://github.com/pchilano/jdk/tree/JDK-8302351-alt) but might be too risky to backport it to 21 at this stage, so I just reverted to remove the asserts in here. I filed 8311516 to fix the root issue there.

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

PR Comment: https://git.openjdk.org/jdk/pull/14108#issuecomment-1622002445