CODEWITHSUNDEEP
parallel-processingtaskjulia
Pedro G.
Pedro G.

Reputation: 405

variable scope in julia tasks

I have adapted the pmap() implementation to my program to do some scheduling and I have a question about the scope of the variables within the tasks. This is the Julia implementation

function pmap(f, lst)
    np = nprocs()  # determine the number of processes available
    n = length(lst)
    results = cell(n)
    i = 1
    # function to produce the next work item from the queue.
    # in this case it's just an index.
    nextidx() = (idx=i; i+=1; idx)
    @sync begin
        for p=1:np
            if p != myid() || np == 1
                @async begin
                    while true
                        idx = nextidx()
                        if idx > n
                            break
                        end
                        results[idx] = remotecall_fetch(p, f, lst[idx])
                    end
                end
            end
        end
    end
    results
end

If i substitute the line idx = nextidx() by idx=x; i=i+1; , each task updates its local copy of the variable i. However the variable i within the function nextidx() is shared by all the tasks. Why is this?

Upvotes: 4

Views: 315

Answers (1)

Reza Afzalan
Reza Afzalan

Reputation: 5746

Let me first simplifies the above code:

function test()
  i=10
  nexti() = (inx=i;i+=1;inx)
  @sync begin
    @async begin
      i=i+10
      nexti()      
      println("value of i in another thread => $i")
    end
  end
  println("value of i in test() => $i")
end

test()

# value of i in another thread => 20
# value of i in test() => 11

We declare nexti() in the same process as i was declared, and i in nexti() refers to the same location, so any changes to i inside nexti() alerts i value in outer scope.

On the other hand @async macro forces the block inside, to run on different process so this block use a copy of i value and any changes inside this block do not alert i value in outer scope.

Upvotes: 1

Related Questions