Porting from 32-bit to 64-bit mode
IBM® XL C/C++ for Linux, V16.1 supports only 64-bit compilation mode, which means you can use the XL C/C++ compiler to develop only 64-bit applications.
You might want to port existing 32-bit applications
to the 64-bit IBM XL
C/C++ for Linux, V16.1. However, this can
lead to a number of problems, mostly related to the differences in
C/C++ long and pointer data type sizes and
alignment between the two modes. The following table summarizes these
differences.
| Data type | 32-bit mode | 64-bit mode | ||
|---|---|---|---|---|
| Size | Alignment | Size | Alignment | |
| long, signed long, unsigned long | 4 bytes | 4-byte boundaries | 8 bytes | 8-byte boundaries |
| pointer | 4 bytes | 4-byte boundaries | 8 bytes | 8-byte boundaries |
| size_t (defined in the header file <cstddef>) | 4 bytes | 4-byte boundaries | 8 bytes | 8-byte boundaries |
| ptrdiff_t (defined in the header file <cstddef>) | 4 bytes | 4-byte boundaries | 8 bytes | 8-byte boundaries |
The following sections discuss some of the common pitfalls implied
by these differences, as well as recommended programming practices
to help you avoid most of these issues:
For suggestions on improving performance in 64-bit mode, see Optimize operations in 64-bit mode.
