[jdk21u-dev] RFR: 8325469: Freeze/Thaw code can crash in the presence of OSR frames [v2]

[Paul Hohensee]

On Tue, 11 Jun 2024 12:14:47 GMT, Aleksey Shipilev <shade at openjdk.org> wrote:

>> Unclean backport to fix Loom bug in presence of OSR frames.
>> 
>> The major uncleanliness point comes from [JDK-8329332](https://bugs.openjdk.org/browse/JDK-8329332) that eliminated `CompiledMethod` in JDK 23. But it is only a minor hassle for this PR, as it moves away from `as_compiled_method()` anyway! So I mostly just accepted the incoming chunks untouched.
>> 
>> Additional testing:
>>  - [x] MacOS AArch64 server fastdebug, new regression test fails without the patch, passes with it
>>  - [x] MacOS AArch64 server fastdebug, `jdk_loom hotspot_loom`
>>  - [x] Linux x86_64 server fastdebug, `all`
>>  - [x] Linux AArch64 server fastdebug, `all`
>
> Aleksey Shipilev has updated the pull request with a new target base due to a merge or a rebase. The incremental webrev excludes the unrelated changes brought in by the merge/rebase. The pull request contains three additional commits since the last revision:
> 
>  - Revert the assert change that is not present in original patch; should be no-op
>  - Merge branch 'master' into JDK-8325469-loom-freeze-thaw-crash
>  - Backport fd331ff17330329a656181cb58714f1bd1623fcb

I'd thought the use of is_nmethod intentional since they are indeed pretty much the same. I'm fine either way.

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

Marked as reviewed by phh (Reviewer).

PR Review: https://git.openjdk.org/jdk21u-dev/pull/675#pullrequestreview-2110293058