OpenMP-Examples/program_control/reproducible.tex

\pagebreak
\section{Controlling Concurrency and Reproducibility with 
the \code{order} Clause}
\label{sec:reproducible_modifier}

\index{clauses!order(concurrent)@\code{order(concurrent)}}
\index{order(concurrent) clause@\code{order(concurrent)} clause}

The \code{order} clause is used for controlling the parallel execution of 
loop iterations for one or more loops that are associated with a directive. 
It is specified with a clause argument and optional modifier. 
The only supported argument, introduced in OpenMP 5.0, is the keyword 
\code{concurrent} which indicates that the loop iterations may execute 
concurrently, including iterations in the same chunk per the loop schedule. 
Because of the relaxed execution permitted with an \code{order(concurrent)} 
clause, codes must not assume that any cross-iteration data dependences 
would be preserved or that any two iterations may execute on the same thread.

The following example in this section demonstrates the use of 
the \code{order(concurrent)} clause, without any modifiers, for controlling 
the parallel execution of loop iterations.
The \code{order(concurrent)} clause cannot be used for the second and third 
\code{parallel}~\code{for}/\code{do} constructs because of either having 
data dependences or accessing threadprivate variables.

\cexample[5.0]{reproducible}{1}

\ffreeexample[5.0]{reproducible}{1}

\index{order(concurrent) clause@\code{order(concurrent)} clause!reproducible modifier@\code{reproducible} modifier}
\index{order(concurrent) clause@\code{order(concurrent)} clause!unconstrained modifier@\code{unconstrained} modifier}
Modifiers to the \code{order} clause, introduced in OpenMP 5.1, may be 
specified to control the reproducibility of the loop schedule for 
the associated loop(s). A reproducible loop schedule will consistently 
yield the same mapping of iterations to threads (or SIMD lanes) if the 
directive name, loop schedule, iteration space, and binding region remain 
the same. The \code{reproducible} modifier indicates the loop schedule must 
be reproducible, while the \code{unconstrained} modifier indicates that 
the loop schedule is not reproducible.
If a modifier is not specified, then the \code{order} clause does not affect 
the reproducibility of the loop schedule.

The next example demonstrates the use of the \code{order(concurrent)} clause 
with modifiers for additionally controlling the reproducibility of a loop's 
schedule.
The two worksharing-loop constructs in the first \code{parallel} construct
specify that the loops have reproducible schedules, thus memory effects from iteration \plc{i} from the first loop will be observable to iteration \plc{i}
in the second loop. 
In the second \code{parallel} construct, the \code{order} clause does not 
control reproducibility for the loop schedules. However, since both loops 
specify the same static schedules, the schedules are reproducible and the 
data dependences between the loops are preserved by the execution.
In the third \code{parallel} construct, the \code{order} clause indicates 
that the loops are not reproducible, overriding the default reproducibility
prescribed by the specified static schedule. Consequentially, 
the \code{nowait} clause on the first worksharing-loop construct should not 
be used to ensure that the data dependences are preserved by the execution.

\cexample[5.1]{reproducible}{2}

\ffreeexample[5.1]{reproducible}{2}