RFR  8140687: Move @Contended to the jdk.internal.vm.annotation package
chris.hegarty at oracle.com
Thu Nov 12 17:23:19 UTC 2015
To me, it sounds like @Contended, in its current form, is not quite ready
for prime-time ( inclusion in Java SE 9 ). There is some concern about its
implementation, and I’m not sure how the loader restriction, and the control
through a -XX flag, would translate into SE spec. There is certainly some
additional work, beyond my primary goal ( preparation for JEP 260 ) that
It seems a reasonable compromise ( until someone can spend some
quality time on it ) to support the sun.misc.Contended annotation as
an alias, or similar, along with the internal Contended annotation (
required by java.base ). The changes to support such are relatively
straight forward, and can easily be refactored in the future.
If there is agreement, I will add @Contended to JEP 260.
> On 12 Nov 2015, at 14:15, Vitaly Davidovich <vitalyd at gmail.com> wrote:
> You didn't explicitly say that, but it came across like that to me;
> apologies if I misread.
> So your concern, then, is that the implementation may not be 100% correct,
> safe, and robust? If it were more readily available to non-JDK users,
> you're likely to see more use of it and thus more "coverage" of the
> feature. The disable flag can continue to exist in case something
> catastrophic is discovered, and it can be disabled by default so people
> have to opt-in (for now). I don't quite get how it spending more time in
> internal packaging (and thus seeing limited application) will help in
> shaking out bugs/exploits.
> On Thu, Nov 12, 2015 at 9:00 AM, Paul Sandoz <paul.sandoz at oracle.com> wrote:
>>> On 12 Nov 2015, at 14:48, Vitaly Davidovich <vitalyd at gmail.com> wrote:
>>> Hi Paul,
>>> There is a very valid concern, since @Contended changes object layout
>> and increases object size, liberal use might tickle an overflow in HotSpot
>> code. Hence why it has remained internal so far.
>>> What overflow? OOM of the heap? How is that a "very valid" concern? Why
>> would it be used liberally? Why are you allowing users to allocate any
>> memory whatsoever? Why not put a cap on the size of objects non-JDK code
>> can allocate? :) I mean seriously, with all due respect, the occasional
>> "our users are dumb and misinformed" sentiment is very off-putting.
>> Where did i say that? You are incorrectly reading between the lines. The
>> concern is actually smart and well informed users that may dliberately
>> exploit a bug/weakness in the VM.
>>> If you think there are, currently, *implementation* issues with how
>> @Contended is handled in the VM which cannot be addresses, I'll buy that.
>> But can we please stop with the dumbing down? :)
More information about the core-libs-dev