RFR: 8253611: AArch64: Concurrency problems in JavaFrameAnchor
Andrew Dinn
adinn at openjdk.java.net
Fri Sep 25 08:35:17 UTC 2020
On Fri, 25 Sep 2020 07:00:16 GMT, Aleksey Shipilev <shade at openjdk.org> wrote:
> It looks that `_last_Java_sp` is additionally used as acquire/release flag that guards the rest of the state. Yet, the
> coding in `javaFrameAnchor_aarch64.hpp` is incorrect at least in one place:
> _last_Java_fp = src->_last_Java_fp;
> _last_Java_pc = src->_last_Java_pc;
> // Must be last so profiler will always see valid frame if has_last_frame() is true
> _last_Java_sp = src->_last_Java_sp;
>
> There should be `OrderAccess::release()` where the comment suggesting the ordering constraint actually is -- I believe
> that is a copy-paste error from x86. Other platforms have `release` thre. Additionally, it has to be matched with
> `OrderAccess::acquire()` before `_last_Java_sp` reads, which requires the shared code to use the platform-specific
> getter. Testing:
> - [x] Linux x86_64 tier1
> - [x] Linux AArch64 tier1
> - [x] Linux AArch64 tier2
I also remember the discussion about this being inconclusive.
I can certainly see it mattering for profiling threads trying to initiate a stack walk but I suspect they may have
relied on sanity checks to filter out erroneous reads.
-------------
Marked as reviewed by adinn (Reviewer).
PR: https://git.openjdk.java.net/jdk/pull/352
More information about the hotspot-dev
mailing list