EA builds with changes to object monitor implementation to avoid pinning with virtual threads

[Yuval Lombard]

Hi Robert,

Thanks for the heads up!
For the two provided examples, I will indeed have a look, but there are
more libraries I use in which pinning occured, and some more business flows
in our application that were not even tested with virtual threads so it
will be a bit challenging to map all of them

On Wed, Apr 24, 2024, 17:15 robert engels <rengels at ix.netcom.com> wrote:

> The source is available for the two stack-traces you cited in the original
> email. You could edit/provide PRs to those code bases.
>
> On Apr 24, 2024, at 9:10 AM, Yuval Lombard <yuval.l at securithings.com>
> wrote:
>
> Thanks Masoud!
>
> Unfortunately I don't have access to the dependencies' source code
>
> On Wed, 24 Apr 2024 at 15:08, masoud parvari <masoud.parvari at gmail.com>
> wrote:
>
>> Just a side note Yuval. If you cannot wait for object.wait() rework and
>> you have access to the dependencies' source code, you can always shadow
>> those classes that use Object.wait() and replace the implementation with
>> CountDownLatch or similar constructs from j.u.c.
>>
>> That's what I did. Luckily the Object.wait() is a lot less prevalent in
>> codebases compared to synchronized and wouldn't be a lot of work.
>>
>> Kind regards,
>> Masoud
>>
>> On Wed, Apr 24, 2024 at 1:03 PM Yuval Lombard <yuval.l at securithings.com>
>> wrote:
>>
>>> Thanks Alan!
>>>
>>> Are you able to estimate the due date?
>>>
>>> On Wed, 24 Apr 2024 at 11:06, Alan Bateman <Alan.Bateman at oracle.com>
>>> wrote:
>>>
>>>> On 24/04/2024 08:51, Yuval Lombard wrote:
>>>> > :
>>>> >
>>>> > I am well aware that java.lang.Object.wait continues to pin, but
>>>> > wanted to point out a few of the 3rd party libs that are used so you
>>>> > get the impression of the necessity and the anticipation for the more
>>>> > robust solution :)
>>>> >
>>>> Thank you for trying out the EA builds. The changes for Object.wait are
>>>> work in progress. So expect a new EA build at some point that will
>>>> address that issue.
>>>>
>>>> -Alan
>>>>
>>>
>>>
>>> --
>>>
>>> Kind regards,
>>>
>>> *Yuval Lombard*
>>>
>>> *Lead Software Engineer*
>>>
>>> +972.50.548.0111
>>>
>>> yuval.l at securithings.com
>>>
>>> <logo_black.png>
>>>
>>
>
> --
>
> Kind regards,
>
> *Yuval Lombard*
>
> *Lead Software Engineer*
>
> +972.50.548.0111
>
> yuval.l at securithings.com
>
> <logo_black.png>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/loom-dev/attachments/20240424/bd8441f1/attachment-0001.htm>