https://community.intel.com/t5/Intel-Moderncode-for-Parallel/OpenMP-spec-example/m-p/907467#M4490 <P>My comments on the referenced secton is the author made a typographical error.</P> <P>IMHO lines 34:38 should be placed before line 28 (just inside the opening brace)</P> <P>Anyone else wish to comment?</P> <P>Good catch.</P> <P>Jim Dempsey</P> Sun, 10 Aug 2008 15:01:22 GMT jimdempseyatthecove 2008-08-10T15:01:22Z https://community.intel.com/t5/Intel-Moderncode-for-Parallel/OpenMP-spec-example/m-p/907466#M4489 Consider the example A.2.2c on page 157 of the OpenMP 3.0 spec [not a new feature in openmp3 as far as I know].<BR />(Available from here: <A href="http://www.openmp.org/mp-documents/spec30.pdf)" target="_blank">http://www.openmp.org/mp-documents/spec30.pdf)</A><BR /><BR />Question: Is this example and explanitive text correct about only data being defined at the end? If so, can you please explain to me why this is so?<BR /><BR />Thread 0 has flushed the data and then the flag [ordering imposed by flush directives] in that order.<BR /><BR />Thread 1 cannot cross the line 36 flush pragma until flag is 1, and hence data is 42. how can only data be the only one defined at the final print statement?<BR /><BR />Thanks,<BR />Jonathan.<BR /> Wed, 06 Aug 2008 20:43:00 GMT https://community.intel.com/t5/Intel-Moderncode-for-Parallel/OpenMP-spec-example/m-p/907466#M4489 jhogg41 2008-08-06T20:43:00Z https://community.intel.com/t5/Intel-Moderncode-for-Parallel/OpenMP-spec-example/m-p/907467#M4490 <P>My comments on the referenced secton is the author made a typographical error.</P> <P>IMHO lines 34:38 should be placed before line 28 (just inside the opening brace)</P> <P>Anyone else wish to comment?</P> <P>Good catch.</P> <P>Jim Dempsey</P> Sun, 10 Aug 2008 15:01:22 GMT https://community.intel.com/t5/Intel-Moderncode-for-Parallel/OpenMP-spec-example/m-p/907467#M4490 jimdempseyatthecove 2008-08-10T15:01:22Z