py/obj: Add "full" and "empty" non-variable-length mp_obj_type_t.
This will always have the maximum/minimum size of a mp_obj_type_t representation and can be used as a member in other structs. Signed-off-by: Jim Mussared <jim.mussared@gmail.com>
This commit is contained in:
committed by
Damien George
parent
5ddf671944
commit
e8355eb163
15
py/obj.h
15
py/obj.h
@@ -562,6 +562,9 @@ typedef mp_int_t (*mp_buffer_fun_t)(mp_obj_t obj, mp_buffer_info_t *bufinfo, mp_
|
||||
bool mp_get_buffer(mp_obj_t obj, mp_buffer_info_t *bufinfo, mp_uint_t flags);
|
||||
void mp_get_buffer_raise(mp_obj_t obj, mp_buffer_info_t *bufinfo, mp_uint_t flags);
|
||||
|
||||
// This struct will be updated to become a variable sized struct. In order to
|
||||
// use this as a member, or allocate dynamically, use the mp_obj_empty_type_t
|
||||
// or mp_obj_full_type_t structs below (which must be kept in sync).
|
||||
struct _mp_obj_type_t {
|
||||
// A type is an object so must start with this entry, which points to mp_type_type.
|
||||
mp_obj_base_t base;
|
||||
@@ -632,6 +635,13 @@ struct _mp_obj_type_t {
|
||||
struct _mp_obj_dict_t *locals_dict;
|
||||
};
|
||||
|
||||
// Non-variable sized versions of mp_obj_type_t to be used as a member
|
||||
// in other structs or for dynamic allocation. The fields are exactly
|
||||
// as in mp_obj_type_t, but with a fixed size for the flexible array
|
||||
// members.
|
||||
typedef mp_obj_type_t mp_obj_empty_type_t;
|
||||
typedef mp_obj_type_t mp_obj_full_type_t;
|
||||
|
||||
#define MP_TYPE_NULL_MAKE_NEW (NULL)
|
||||
|
||||
// Implementation of MP_DEFINE_CONST_OBJ_TYPE for each number of arguments.
|
||||
@@ -657,14 +667,15 @@ struct _mp_obj_type_t {
|
||||
// of the 30th argument (30 is 13*2 + 4).
|
||||
#define MP_DEFINE_CONST_OBJ_TYPE_NARGS(_1, _2, _3, _4, _5, _6, _7, _8, _9, _10, _11, _12, _13, _14, _15, _16, _17, _18, _19, _20, _21, _22, _23, _24, _25, _26, _27, _28, _29, N, ...) MP_DEFINE_CONST_OBJ_TYPE_NARGS_##N
|
||||
|
||||
// This macro is used to define a object type in ROM.
|
||||
// These macros are used to define a object type in ROM.
|
||||
// Invoke as MP_DEFINE_CONST_OBJ_TYPE(_typename, _name, _flags, _make_new [, slot, func]*)
|
||||
// It uses the number of arguments to select which MP_DEFINE_CONST_OBJ_TYPE_*
|
||||
// They use the number of arguments to select which MP_DEFINE_CONST_OBJ_TYPE_*
|
||||
// macro to use based on the number of arguments. It works by shifting the
|
||||
// numeric values 12, 11, ... 0 by the number of arguments, such that the
|
||||
// 30th argument ends up being the number to use. The _INV values are
|
||||
// placeholders because the slot arguments come in pairs.
|
||||
#define MP_DEFINE_CONST_OBJ_TYPE(...) MP_DEFINE_CONST_OBJ_TYPE_EXPAND(MP_DEFINE_CONST_OBJ_TYPE_NARGS(__VA_ARGS__, _INV, 12, _INV, 11, _INV, 10, _INV, 9, _INV, 8, _INV, 7, _INV, 6, _INV, 5, _INV, 4, _INV, 3, _INV, 2, _INV, 1, _INV, 0)(mp_obj_type_t, __VA_ARGS__))
|
||||
#define MP_DEFINE_CONST_OBJ_FULL_TYPE(...) MP_DEFINE_CONST_OBJ_TYPE_EXPAND(MP_DEFINE_CONST_OBJ_TYPE_NARGS(__VA_ARGS__, _INV, 12, _INV, 11, _INV, 10, _INV, 9, _INV, 8, _INV, 7, _INV, 6, _INV, 5, _INV, 4, _INV, 3, _INV, 2, _INV, 1, _INV, 0)(mp_obj_full_type_t, __VA_ARGS__))
|
||||
|
||||
// Constant types, globally accessible
|
||||
extern const mp_obj_type_t mp_type_type;
|
||||
|
||||
@@ -29,7 +29,10 @@
|
||||
#include "py/objtuple.h"
|
||||
|
||||
typedef struct _mp_obj_namedtuple_type_t {
|
||||
mp_obj_type_t base;
|
||||
// Must use the full-size version to avoid this being a variable sized member.
|
||||
// This means that named tuples use slightly more RAM than necessary, but
|
||||
// no worse than if we didn't have slots/split representation.
|
||||
mp_obj_full_type_t base;
|
||||
size_t n_fields;
|
||||
qstr fields[];
|
||||
} mp_obj_namedtuple_type_t;
|
||||
|
||||
@@ -1148,7 +1148,10 @@ mp_obj_t mp_obj_new_type(qstr name, mp_obj_t bases_tuple, mp_obj_t locals_dict)
|
||||
#endif
|
||||
}
|
||||
|
||||
mp_obj_type_t *o = m_new0(mp_obj_type_t, 1);
|
||||
// Allocate a full-sized mp_obj_full_type_t instance (i.e. all slots / extended fields).
|
||||
// Given that Python types use almost all the slots anyway, this doesn't cost anything
|
||||
// extra.
|
||||
mp_obj_type_t *o = (mp_obj_type_t *)m_new0(mp_obj_full_type_t, 1);
|
||||
o->base.type = &mp_type_type;
|
||||
o->flags = base_flags;
|
||||
o->name = name;
|
||||
|
||||
Reference in New Issue
Block a user