Debugging runtime error in Rcpp

Question

I have converted my sampling algorithm in R to a Rcpp. The Rcpp code is compiling, but giving an error when calling the function from R Error in eval(expr, envir, enclos) : index out of bounds". traceback() is also not informative.

7: stop(list(message = "index out of bounds", call = eval(expr, 
   envir, enclos), cppstack = NULL))  
6: .Call("temp1_eeSNP_regress", PACKAGE = "temp1", x, y, feature, 
   alpHa, gamMa_prob, beTa, estimate_alpha, estimate_beta, B_inv_alpHa, 
   itermax, thin, burnIn)  
5: eeSNP_regress(x = e.obj$x, y = e.obj$y, feature = e.obj$feature, 
   alpHa = e.obj$alpHa, beTa = e.obj$beTa, gamMa_prob = e.obj$gamMa.prob, 
   estimate_alpha = F, estimate_beta = T, B_inv_alpHa = rep(1/100, 
       dim(e.obj$feature)[2] + 1), itermax = 5000, thin = 40, 
   burnIn = 1000) at cmd.R#1  
4: eval(expr, envir, enclos)  
3: eval(ei, envir)  
2: withVisible(eval(ei, envir))

Is there a way to debug such runtime error in Rcpp rather than throwing bunch of cout? Is it possible to call gdb inside R?

Answer 1

Yes, you can use gdb with R.

See for example Section 4.4 "Debugging Compiled Code" in the Writing R Extentions manual.