Charm++ parallel programming system

["Van Der Wijngaart, Rob F" <rob.f.van.der.wijngaart AT intel.com>]

Thanks, Phil. This is actually identical  to the workaround proposed by Gengbin. It would definitely work, though I already had to do the pre-work of linearizing the 2D weight array.

 

Rob

 

From: unmobile AT gmail.com [mailto:unmobile AT gmail.com] On Behalf Of Phil Miller
Sent: Monday, September 29, 2014 3:06 PM
To: Kale, Laxmikant V
Cc: Van Der Wijngaart, Rob F; Kunzman, David M; Hammond, Jeff R; Nikhil Jain; Mattson, Timothy G; Charm Mailing List
Subject: Re: [charm] Global variables in Charm++

 

 

 

From: "<Van Der Wijngaart>", Rob F <rob.f.van.der.wijngaart AT intel.com>

 

Actually, as I wrote last week, I wanted to be able to define something a little more complex, which turned out to be totally unpalatable by charmc:

readonly double weight[2*RADIUS+1][2*RADIUS+1]; 

Why would that be a reasonable thing to do for a readonly variable, one may ask? The reason is that this is an array that specifies the weights of a stencil. In my workload I allow different stencil sizes and different weights. But once defined, weight is a global constant. I let the user specify the constant symbol RADIUS, and use that to compute weight’s dimensions. In the .C file I can do that to my heart’s content, but not in the .ci file. To me that is not reasonable.

I realize after some further thought that things like this may have come up before, with a much more 'pleasant' looking workaround, as follows:
======

// In WEIGH_DIM.h:

#define WEIGHT_DIM ((2*RADIUS+1)*(2*RADIUS+1))
======

// In foo.ci:

include "WEIGHT_DIM.h";

readonly double WEIGHT[WEIGHT_DIM];

======

I just tested that this compiles and links, and it should work fine, since Charm++ will see a simple 1D array.