Re: [xmlvm-users] XMLVM changes to synchronization

[Paul P. <bay...@gm...>]

Hi Arno,


I believe I have slightly misled you.  My attached project in my earlier
email is a Java proof of concept written with the intent ready to easily
port to Objective-C.  So it's not a patch just yet.  I wanted to first make
sure I wasn't going against the grain of the XMLVM community.


Since it's all Java & not yet a patch, I have not implemented my own version
of associative references.  I agree that would be a poor approach & we
should absolutely optimize for the device & not the simulator.  My comments
were mostly just pondering as to the best solution to that portion.  If
associative references work in the SDK4, then that's wonderful.  I'm going
to try that out this weekend.


Regarding dex:monitor-enter vs jam:monitorenter, there's a comment saying
"we can't map this to @synchronized {} because DEX may generate multiple
monitor-exit for one monitor-enter".  Other than that, it wouldn't seem very
difficult to use the DEX instructions instead of JVM.


I think we're in agreement.


If everything goes smoothly, I will prepare a patch hopefully within the
week.


Hi Gergely,


I took a look at your implementation.  I believe we have come up with
similar solutions.  I used NSLock for synchronization (recursive conditions
handled so that a single release can be called when needed for wait*) &
NSConditionLock for wait* & notify*.  Again assuming associative references
work in SDK4, I will try to get our solutions together.


Thanks!

Paul Poley

On Sat, Jul 17, 2010 at 9:36 AM, Panayotis Katsaloulis <
pan...@pa...> wrote:

> 17 Ιουλ 2010, 11:49 π.μ., ο/η Arno Puder <ar...@pu...> έγραψε:
>
> >
> > I also wanted to say something about the associations. Panayotis has
> > reported that they don't work in the simulator but do work on the actual
> > device. Apparently you've implemented your own solution as a
> > consequence. I would strongly advise against this. For one, it only
> > matters if it works correctly on the device. I've heard many times that
> > you cannot get around testing on the real device because of many subtle
> > differences between the simulator and the device. So, we should optimize
> > for the device; not the simulator. If the simulator version leaks
> > memory, so be it. I'm sure Apple will fix that in the next version of
> > the SDK.
>
> Just a quick note:
> The problem with associations in the simulator was in the previous version
> of the SDK. It is possible that with the release of 4.0 this is not the case
> any more. Right now I am practically away from my computer but I will have a
> look at it next week.
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Sprint
> What will you do first with EVO, the first 4G phone?
> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
> _______________________________________________
> xmlvm-users mailing list
> xml...@li...
> https://lists.sourceforge.net/lists/listinfo/xmlvm-users
>