RFR: 8306767: Concurrent repacking of extra data in MethodData is potentially unsafe [v17]

[Emanuel Peter]

On Mon, 15 Jan 2024 11:55:05 GMT, Emanuel Peter <epeter at openjdk.org> wrote:

>> As explained in a [comment below](https://github.com/openjdk/jdk/pull/16840#issuecomment-1833529561), we have to ensure that reading/writing/cleaning the extra data all needs to be guarded by the `extra_data_lock`, and that no safepoint should happen while holding that lock, so that the lock is not broken.
>> 
>> I introduced `check_extra_data_locked`, where I check that we hold the lock, and if we are a java thread (only those ever safepoint), that we currently are in a `NoSafepointVerifier` scope, hence we verify that no safepoint will be taken.
>> 
>> I placed `check_extra_data_locked` in all the places where we access the extra data, and then placed locks and no-safepoint-verifiers at the call-site of those places.
>> 
>> I also needed to change the rank of `extra_data_lock` to `nosafepoint` and set the `Mutex::_no_safepoint_check_flag` when taking the lock. Otherwise I could not take the lock from a VM thread.
>> 
>> **Complications with ttyl**
>> There were a few places in printing code, where did `ttyLocker ttyl;`, and then in that scope we would access the extra data. Now that I introduced locking with `extra_data_lock`, this ran into asserts which check the lock ranks: `ttyl` has a very low rank, and `extra_data_lock` a rather high lock. Hence, we cannot lock `extra_data_lock` inside a `ttyl` scope.
>> 
>> If we were to simply remove the `ttyl` locking, then the many print lines inside that scope might be interrupted and another thread can insert other printing in between. To avoid that, I now first buffer all lines in a `stringStream`, and then print that buffered stream to `tty` all at once, which means no other printing can be injected in between.
>> 
>> **Testing**
>> Testing: tier1-3 and stress.
>
> Emanuel Peter has updated the pull request incrementally with two additional commits since the last revision:
> 
>  - Update src/hotspot/share/runtime/deoptimization.cpp
>    
>    rm empty line
>  - change patch to deoptimization.cpp case brought up by Roland

src/hotspot/share/runtime/deoptimization.cpp line 2337:

> 2335:     NoSafepointVerifier no_safepoint;
> 2336:     ProfileData* pdata = nullptr;
> 2337: 

Suggestion:

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

PR Review Comment: https://git.openjdk.org/jdk/pull/16840#discussion_r1452284238