Bureaucrats, cc_docs_admin, cc_staff
2,879
edits
OpenACC Tutorial - Adding directives: Difference between revisions
Jump to navigation
Jump to search
m
→The parallel loop directive
| Line 299: | Line 299: | ||
== The <tt>parallel loop</tt> directive == | == The <tt>parallel loop</tt> directive == | ||
With the <tt>kernels</tt> directive, we let the compiler do all of the analysis. This is the ''descriptive'' approach to porting a code. OpenACC | With the <tt>kernels</tt> directive, we let the compiler do all of the analysis. This is the ''descriptive'' approach to porting a code. OpenACC supports a ''prescriptive'' approach through a different directive, called the <tt>parallel</tt> directive. This can be combined with the <tt>loop</tt> directive, to form the <tt>parallel loop</tt> directive. An example would be the following code: | ||
</translate> | </translate> | ||
<syntaxhighlight lang="cpp" line> | <syntaxhighlight lang="cpp" line> | ||
| Line 313: | Line 313: | ||
For reasons that we explain below, in order to use this directive in the matrix-vector product example, we need to introduce additional clauses used to manage the scope of data. The <tt>private</tt> and <tt>reduction</tt> clauses control how the data flows through a parallel region. | For reasons that we explain below, in order to use this directive in the matrix-vector product example, we need to introduce additional clauses used to manage the scope of data. The <tt>private</tt> and <tt>reduction</tt> clauses control how the data flows through a parallel region. | ||
* With the <tt>private</tt> clause, a copy of the variable is made for each loop iteration, making the value of the variable independent from other iterations. | * With the <tt>private</tt> clause, a copy of the variable is made for each loop iteration, making the value of the variable independent from other iterations. | ||
* With the <tt>reduction</tt> clause, the values of a variable in each | * With the <tt>reduction</tt> clause, the values of a variable in each iteration will be ''reduced'' to a single value. It supports addition (+), multiplication (*), maximum (max), minimum (min), among other operations. | ||
These clauses were not required with the <tt>kernels</tt> directive, because the <tt>kernels</tt> directive handles this for you. | These clauses were not required with the <tt>kernels</tt> directive, because the <tt>kernels</tt> directive handles this for you. | ||