On 08/17/2010 05:37 PM, Eric Blake wrote: > On 08/17/2010 05:35 PM, Bruno Haible wrote: >>>> What a shame that POSIX omitted an <inttypes.h> PRIu* for size_t. >>> >>> You can define it by yourself ... >> >> Or use uintptr_t instead of size_t. By the definition of these types, >> you can be sure that uintptr_t is at least as wide as size_t.
Actually, you aren't even guaranteed that; uintptr_t is optional on non-XSI systems. But you are correct that uintptr_t is the size of void*, size_t is the maximum size of an object, and an object must necessarily fit within a pointer; therefore sizeof(size_t) <= sizeof(uintptr_t). The <= is important - it is possible to have a system with 32-bit size_t and 64-bit void* (or, in older days, a 16-bit size_t and 32-bit void *). In fact, people are already experimenting with that alternate ABI on x86_64 architectures, with demonstrations of the speedups possible when object sizes are constrained to 4G while still allowing more than 4G of memory accessible to a single process. -- Eric Blake ebl...@redhat.com +1-801-349-2682 Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
