Hi Levente,


thanks a lot for your answer, this was exactly what I was searching for! :-)


Regarding the suspension points, do we have any good documentation on them, preferably on the image side? I am still having problems understanding when the VM would suspend the active process - does this only apply if a process with a higher process gets scheduled by the VM? Wouldn't it be an option to express this intent explicitly, similar to #valueUnpreemptively?


Best,

Christoph

________________________________
Von: Squeak-dev <squeak-dev-bounces at lists.squeakfoundation.org> im Auftrag von Levente Uzonyi <leves at caesar.elte.hu>
Gesendet: Montag, 22. November 2021 12:50:26
An: The general-purpose Squeak developers list
Betreff: Re: [squeak-dev] Disable inlined message sends?

Hi Marcel,

On Mon, 22 Nov 2021, Marcel Taeumel wrote:

> Hi Levente --
> Thank you for the details. Let me emphasize one of the issues you raised with a quote:
>
> > But such change would break code that relies on the atomicity (actually the lack of suspension points) of those selectors.
>
> That's a tricky one. :-)

Indeed. Sometimes the information is expessed in a comment, like how
DelayWaitTimeout >> #wait or MCHttpRepository >> #webClientDo: does.
But there's code that just relies on the behavior without mentioning it.
E.g. Semaphore >> #waitIfLocked: or Semaphore >> #critical:ifLocked:.
Perhaps the best way would be to introduce a pragma to mark such methods.


Levente
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.squeakfoundation.org/pipermail/squeak-dev/attachments/20211127/a79f6e73/attachment.html>