This is just a heads-up that the ARRAYLIB library supplied with all versions of BB4W and BBCSDL contains procedures to transpose a matrix (PROC_transpose etc.) which call an assembly language subroutine for speed. This has never been adapted to run on the ARM editions (Raspberry Pi, Android, iOS) so of course any attempt to call the procedure(s) on those platforms will fail. Indeed since iOS does not allow arbitrary code execution an assembly-language implementation isn't even possible.
It is therefore my intention to substitute straightforward nested FOR..NEXT loops to do the transposition in future releases of BBCSDL. This won't be as fast, but at least will be CPU-agnostic and run on iOS as well. Clearly these are not commonly-used functions, otherwise somebody would have noticed and reported this issue long ago, so it's probably not a big deal.
Richard.