I have issues with mcmapply that only occures if an Rcpp-function, testfun(), is first called outside mcmapply and then called through mcmapply. The issue only exists if:

• testfun() uses arma::exp() (or arma::sqrt(), etc) on a matrix with at least 18×18 elements and stores the result as arma::mat
• mcmapply uses parallelization (mc.cores>1)

If I call mcmapply with testfun(), it runs and finishes without errors. If I first call testfun() separately and afterwards mcmapply, then mcmapply stalls.

MWE:

Rcpp::sourceCpp(code=
"  #include <RcppArmadillo.h>;
  // [[Rcpp::depends(RcppArmadillo)]]
  // [[Rcpp::export]]
  void testfun(int n) {
    arma::mat foo = arma::sqrt(arma::zeros(n,n));
  }")
library(parallel)

print("Run mcmapply+testfun: works")
mcmapply(mc.cores=2, function(x) testfun(100), 1:2)

print("Calling testfun separately: works")
testfun(100)

print("Run mcmapply+testfun after having run testfun separately: stalls")
mcmapply(mc.cores=2, function(x) testfun(100), 1:2)

I suspect the issue has to do with storing the matrices, as it only occurs if testfun() assigns the result of arma::sqrt() to an arma::mat.

Setup: linux with R 4.4.0, gcc 14.2.0, and gsl 2.7.1, but issue also replicated on older R versions.