Re: [X10-users] Running C++ generated code . . .

2010-11-03 Thread Vijay Saraswat
Make sure you set X10_NTHREADS. See http://x10.codehaus.org/For+Users, specifically the "X10 Environment Variables" section. See the FAQ -- How do I control the size of the thread pool in a place. http://x10.codehaus.org/FAQ#FAQ-HowdoIcontrolthesizeofthethreadpoolinaplace%3F On 11/3/2010 5:11 P

Re: [X10-users] Running C++ generated code . . .

2010-11-03 Thread Igor Peshansky
Russel Winder wrote on 11/03/2010 05:11:16 PM: > I am fairly confident I am just missing something very simple but . . . > > I have a (trivial) embarrassingly parallel scatter/gather code which > compiles fine in both Java back-end and C++ back-end modes. The Java > backend-version runs nicely

Re: [X10-users] Updating to 2.1.0

2010-11-03 Thread David P Grove
Russel Winder wrote on 11/03/2010 05:05:33 PM: > > I have a program that worked under 2.0.3 -- though it used some > nastiness of the then Array.lift function which has now gone away (which > is probably for the best). However it leaves me with a problem. The > algorithm I have is basically a

[X10-users] Running C++ generated code . . .

2010-11-03 Thread Russel Winder
I am fairly confident I am just missing something very simple but . . . I have a (trivial) embarrassingly parallel scatter/gather code which compiles fine in both Java back-end and C++ back-end modes. The Java backend-version runs nicely using all the cores it can for the number of places specif

[X10-users] Updating to 2.1.0

2010-11-03 Thread Russel Winder
I have a program that worked under 2.0.3 -- though it used some nastiness of the then Array.lift function which has now gone away (which is probably for the best). However it leaves me with a problem. The algorithm I have is basically a data parallel one so I need a parallel map. However I want

Re: [X10-users] Pattern matching on a point with X10 2.1.0

2010-11-03 Thread David P Grove
Dave Hudak wrote on 11/03/2010 10:36:34 AM: > > I am updating some X10 2.0.5 examples to work with 2.1.0. > > This code worked in 2.0.5 (compiled with -STATIC_CHECKS): > > val blockDist = Dist.makeBlock(R); > val p0Dist = blockDist|here; > val p0Region = p0Dist.region(); > for (p i

Re: [X10-users] Pattern matching on a point with X10 2.1.0

2010-11-03 Thread Dave Hudak
On Nov 3, 2010, at 10:46 AM, Yoav Zibin wrote: > In 2.0.5 the syntax for exploding a point was: > val p(i,j) = [3,4]; // p is a Point(2), i is 3, j is 4 > In 2.1 we changed the exploding syntax to: > val p[i,j] = [3,4]; // p is a Point(2), i is 3, j is 4 > Yes, but in the original code I was

Re: [X10-users] Pattern matching on a point with X10 2.1.0

2010-11-03 Thread Yoav Zibin
In 2.0.5 the syntax for exploding a point was: val p(i,j) = [3,4]; // p is a Point(2), i is 3, j is 4 In 2.1 we changed the exploding syntax to: val p[i,j] = [3,4]; // p is a Point(2), i is 3, j is 4 This message: (Warning) Reached threshold when checking constraints. is a mechanism to make sur

[X10-users] Pattern matching on a point with X10 2.1.0

2010-11-03 Thread Dave Hudak
Hi All, I am updating some X10 2.0.5 examples to work with 2.1.0. This code worked in 2.0.5 (compiled with -STATIC_CHECKS): val blockDist = Dist.makeBlock(R); val p0Dist = blockDist|here; val p0Region = p0Dist.region(); for (p in p0Region) { Console.OUT.println("p0 = "+p);