UnorderedCopy

Usage

use UnorderedCopy;

or

import UnorderedCopy;

Warning

This module represents work in progress. The API is unstable and likely to change over time.

This module provides an unordered version of copy/assign for trivially copyable types. Trivially copyable types require no special behavior to be copied and merely copying their representation is sufficient. They include numeric and bool types as well as tuples or records consisting only of numeric/bool. Records cannot have user-defined copy-initializers, deinitializers, or assignment overloads.

unorderedCopy() can provide a significant speedup for batch assignment operations that do not require ordering of operations. The results from unorderedCopy() are not visible until task or forall termination or an explicit unorderedCopyTaskFence():

use BlockDist, UnorderedCopy;

const size = 10000;
const space = {0..size};
const D = space dmapped Block(space);
var A, reversedA: [D] int = D;

forall i in D do
  unorderedCopy(reversedA[i], A[size-i]);

// no fence required, fenced at task/forall termination

forall (rA, i) in zip(reversedA, D) do
  assert(rA == size-i);

It’s important to be aware that unordered operations are not consistent with regular operations and updates may not be visible until the task or forall that issued them terminates or an explicit unorderedCopyTaskFence().

var a = 0;
on Locales[1] {
  var b = 1;
  unorderedCopy(b, a);
  writeln(b);        // can print 0 or 1
  unorderedCopyTaskFence();
  writeln(b);        // must print 0
}

Generally speaking they are useful for when you have a large batch of remote assignments to perform and the order of those operations doesn’t matter.

Note

Currently, this is only optimized for CHPL_COMM=ugni. Other communication layers fall back to regular operations. Under ugni, GETs are internally buffered. When the buffers are flushed, the operations are performed all at once. Cray Linux Environment (CLE) 5.2.UP04 or newer is required for best performance. In our experience, unordered copies can achieve up to a 5X performance improvement over ordered copies for CLE 5.2UP04 or newer.

proc unorderedCopy(ref dst, src): void

Unordered copy. Only supported for identical trivially copyable types.

proc unorderedCopyTaskFence(): void

Fence any pending unordered copies issued by the current task.