On 11-Aug-2007, Fredrik Lingvall wrote:
| From gcc's man page:
|
| `-m32'
| `-m64'
| Generate code for a 32-bit or 64-bit environment. The 32-bit
| environment sets int, long and pointer to 32 bits and generates
| code that runs on any i386 system. The 64-bit environment sets
| int to 32 bits and long and pointer to 64 bits and generates code
| for AMD's x86-64 architecture.
That doesn't make ints 8 bytes wide (nor is that what you probably
want for the C++ parts of Octave).
| From what I can see (in the liboctave/dbleCHOL.cc), octave's chol and
| cholinv also uses dpotrf_(....) and
| dpotri_(...). I have assumed,
|
| extern "C" void dpotrf_(const char *UPLO, const int *N,
| const double *A, const int *lda,
| const int *info);
|
| extern "C" void dpotri_(const char *UPLO, const int *N,
| const double *A, const int *lda,
| const int *info);
|
| that is, N and lda are int's. Perhaps, they should be long's instead?
Yes, probably. Octave uses the typedef octave_idx_type to handle this
so that configure can choose an appropriate type.
| Or, maybe the gfortan
| option -fdefault-integer-8 should be used when compiling LAPACK?
Yes.
jwe