Geek asks and answers

Creating an FFTW Plan Using OpenMP

I am trying to run several FFTs in parallel. I use FFTW and OpenMP. Each FFT is different, so I don't rely on FFTW multithreading, which, as I know, uses OpenMP.

int m;

// assume:
// int numberOfColumns = 100;
// int numberOfRows = 100;

#pragma omp parallel for default(none) private(m) shared(numberOfColumns, numberOfRows)//  num_threads(4)
    for(m = 0; m < 36; m++){

        // create pointers
        double          *inputTest;
        fftw_complex    *outputTest;
        fftw_plan       testPlan;

        // preallocate vectors for FFTW
         outputTest = (fftw_complex*)fftw_malloc(sizeof(fftw_complex)*numberOfRows*numberOfColumns);
         inputTest  = (double *)fftw_malloc(sizeof(double)*numberOfRows*numberOfColumns);

         // confirm that preallocation worked
         if (inputTest == NULL || outputTest == NULL){
             logger_.log_error("\t\t FFTW memory not allocated on m = %i", m);
         }

         // EDIT: insert data into inputTest
         inputTest = someDataSpecificToThisIteration(m); // same size for all m

        // create FFTW plan
        #pragma omp critical (make_plan)
        {
            testPlan = fftw_plan_dft_r2c_2d(numberOfRows, numberOfColumns, inputTest, outputTest, FFTW_ESTIMATE);
        }

         // confirm that plan was created correctly
         if (testPlan == NULL){
             logger_.log_error("\t\t failed to create plan on m = %i", m);
         }

        // execute plan
         fftw_execute(testPlan);

        // clean up
         fftw_free(inputTest);
         fftw_free(outputTest);
         fftw_destroy_plan(testPlan);

    }// end parallelized for loop

It all works great. However, if I remove the critical construct around creating the plan (fftw_plan_dft_r2c_2d), my code will fail. Can someone explain why? fftw_plan_dft_r2c_2d is not really an orphan, is it? Is this due to the fact that two threads can simultaneously try to delete memory cells numberOfRows or numberOfColumns at the same time?

+5
source share
1 answer

Pretty much everything is written in the FFTW documentation on thread safety :

... , (, ) .

, - ( ) FFTW fftw_execute ( ). (, ) . , , ; , . , (FFTW , ), .

FFT , , . , . :

#pragma omp parallel default(none) private(m) shared(numberOfColumns, numberOfRows)
{
   // create pointers
   double          *inputTest;
   fftw_complex    *outputTest;
   fftw_plan       testPlan;

   // preallocate vectors for FFTW
   outputTest = (fftw_complex*)fftw_malloc(sizeof(fftw_complex)*numberOfRows*numberOfColumns);
   inputTest  = (double *)fftw_malloc(sizeof(double)*numberOfRows*numberOfColumns);

   // confirm that preallocation worked
   if (inputTest == NULL || outputTest == NULL){
      logger_.log_error("\t\t FFTW memory not allocated on m = %i", m);
   }

   // create FFTW plan
   #pragma omp critical (make_plan)
   testPlan = fftw_plan_dft_r2c_2d(numberOfRows, numberOfColumns, inputTest, outputTest, FFTW_ESTIMATE);

   #pragma omp for
   for (m = 0; m < 36; m++) {
      // execute plan
      fftw_execute(testPlan);
   }

   // clean up
   fftw_free(inputTest);
   fftw_free(outputTest);
   fftw_destroy_plan(testPlan);
}

, fftw_execute(). NUMA (, AMD64 Intel (post-) Nehalem), .

+7

More articles:


All Articles