[Rcpp-devel] Segfault error during simulation in Rcpp

[Matteo Fasiolo]

Thanks for your reply Dirk.

Maybe I have found something.
Hopefully this is reproducible and simple enough:

/*
* C++ file "b.cpp"
* Just copying the input matrix into A and returning A.
*/

#include <Rcpp.h>

using namespace Rcpp;

// [[Rcpp::export]]
NumericMatrix myFun(NumericMatrix input, int n){

NumericMatrix A(n, n);

for(int Row = 0; Row < n; Row++)
  for(int Col = 0; Col < n; Col++)
  {
    A(Row, Col) = input(Row, Col);
  }

return A;
}


///////////////////////////////////////////

Then I open a terminal:

teo at oracolo:~$ R --vanilla

R version 3.0.0 (2013-04-03) -- "Masked Marvel"
Copyright (C) 2013 The R Foundation for Statistical Computing
Platform: x86_64-pc-linux-gnu (64-bit)
[........]
Type 'q()' to quit R.

> library(Rcpp)
> sourceCpp("~/Desktop/b.cpp")
>
> #I run it 10 times and everything is fine.
> n = 10
> x <- 1:n^2
>
> for(ii in 1:10)
+ {
+   means <- matrix(x, n, n)
+   res <- myFun(means, n)
+   a <- res[1, 1]
+ }
> res
      [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10]
 [1,]    1   11   21   31   41   51   61   71   81    91
 [2,]    2   12   22   32   42   52   62   72   82    92
 [3,]    3   13   23   33   43   53   63   73   83    93
 [4,]    4   14   24   34   44   54   64   74   84    94
 [5,]    5   15   25   35   45   55   65   75   85    95
 [6,]    6   16   26   36   46   56   66   76   86    96
 [7,]    7   17   27   37   47   57   67   77   87    97
 [8,]    8   18   28   38   48   58   68   78   88    98
 [9,]    9   19   29   39   49   59   69   79   89    99
[10,]   10   20   30   40   50   60   70   80   90   100
>
>
> #I run it 10^6 times and everything and I get a segfault.
> n = 10
> x <- 1:n^2
>
> for(ii in 1:10^6)
+ {
+   means <- matrix(x, n, n)
+   res <- myFun(means, n)
+   a <- res[1, 1]
+ }

 *** caught segfault ***
address (nil), cause 'unknown'

Traceback:
 1: res[1, 1]


If I run the same code without the --vanilla option it works fine!
Certainly you know why using --vanilla is a problem here, honestly I've
always used that option
because I don't want R to ask me if I want to save the working environment
when I quit.
As you said the problem was coming from R (actually my improper use of R)
and hopefully this is it!





On Thu, May 16, 2013 at 3:42 PM, Dirk Eddelbuettel <edd at debian.org> wrote:

>
> Matteo,
>
> Can you provide a single, self-contained example and calling sequence that
> leads to reproducible crashes?
>
> That would be a bug. And we try to address it in Rcpp.
>
> As for your "issues" with RNGScope, I'd recommend that you write a C(++)
> function called from R __without using Rcpp__ and I very confident that you
> would the exact same issue.  Meaning that that all comes form R, which in
> itself is a pretty big system with numerous temp. allocations. But
> generally
> no known bug.  So please learn more about R and valgrind -- I suspect that
> you are simply getting confused by the copious and somewhat technical
> output
> produced by valgrind when running R.
>
> Dirk
>
> --
> Dirk Eddelbuettel | edd at debian.org | http://dirk.eddelbuettel.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/rcpp-devel/attachments/20130516/574f810f/attachment-0001.html>