you would write at(my_array.dist(3)) my_array(3)

or you could define

static def get[T](a:DistArray[T](1), i:Int) = at (a.dist(i)) a(i);

and then use get(my_array, 3).


Haris H wrote:
> Now when I am thinking, I was actually looking into 2 different arrays
> containing same elements,
> but distributed over different regions, and way to assemble under some
> rules.
> I did found something in report that will take a look.
>
> But going back to the addressing array elements, am I correct if I conclude
> from your responses that
> my_array(3) can not be accessed directly without knowing exact location of
> that element in the distribution and using "at"?
>
> I was trying to jump to that element regardless of its location.
>
>
> Haris
>
>
>
>
> On Tue, Apr 13, 2010 at 9:58 PM, Vijay Saraswat <vi...@saraswat.org> wrote:
>
>   
>> Haris H wrote:
>>     
>>> This was very helpful, thank you.
>>>
>>> Now, is it possible to find out the *place *of distributed element
>>> data_squares(i) ?
>>>
>>>       
>> data_squares.dist(i)
>>     
>>> (I found "apply" abstract method in documentation, but was not sure what
>>>       
>> to
>>     
>>> make out of it).
>>>
>>> How would I go about accessing for example data_squares(3) without
>>>       
>> iterating
>>     
>>> through the entire array.
>>>
>>> Also, don't ask me why, but can I distribute the same array (e.g. 4
>>>       
>> element)
>>     
>>> to two different set of places (0-4 and 4-8)?
>>>
>>>       
>> You have four array elements. These can go into at most four places. You
>> seem to want eight array elements..?
>>
>> (A single array element is a single mutable location. It exists in a
>> single place, cannot exist in two places.)
>>
>>     



------------------------------------------------------------------------------
Download Intel&#174; Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
X10-users mailing list
X10-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/x10-users

Reply via email to