This affects intptr_t, uintptr_t, and size_t, which are variable-sized. The previous mappings were mapped directly to 64-bit values.
@@ -105,9 +105,9 @@ prim_types = {
'uint64_t': 'u64',
'float': 'f32',
'double': 'f64',
- 'uintptr_t': 'u64',
- 'intptr_t': 'i64',
- 'size_t': 'u64'
+ 'uintptr_t': 'c.uintptr_t',
+ 'intptr_t': 'c.intptr_t',
+ 'size_t': 'c.size_t'
}
prim_defaults = {