| 1 |
.. Copyright (C) 2014-2025 Free Software Foundation, Inc. |
| 2 |
Originally contributed by David Malcolm <[email protected]> |
| 3 |
|
| 4 |
This is free software: you can redistribute it and/or modify it |
| 5 |
under the terms of the GNU General Public License as published by |
| 6 |
the Free Software Foundation, either version 3 of the License, or |
| 7 |
(at your option) any later version. |
| 8 |
|
| 9 |
This program is distributed in the hope that it will be useful, but |
| 10 |
WITHOUT ANY WARRANTY; without even the implied warranty of |
| 11 |
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
| 12 |
General Public License for more details. |
| 13 |
|
| 14 |
You should have received a copy of the GNU General Public License |
| 15 |
along with this program. If not, see |
| 16 |
<https://www.gnu.org/licenses/>. |
| 17 |
|
| 18 |
.. default-domain:: cpp |
| 19 |
|
| 20 |
Types |
| 21 |
===== |
| 22 |
|
| 23 |
.. class:: gccjit::type |
| 24 |
|
| 25 |
gccjit::type represents a type within the library. It is a subclass |
| 26 |
of :class:`gccjit::object`. |
| 27 |
|
| 28 |
Types can be created in several ways: |
| 29 |
|
| 30 |
* fundamental types can be accessed using |
| 31 |
:func:`gccjit::context::get_type`: |
| 32 |
|
| 33 |
.. code-block:: c++ |
| 34 |
|
| 35 |
gccjit::type int_type = ctxt.get_type (GCC_JIT_TYPE_INT); |
| 36 |
|
| 37 |
or using the :func:`gccjit::context::get_int_type<T>` template: |
| 38 |
|
| 39 |
.. code-block:: c++ |
| 40 |
|
| 41 |
gccjit::type t = ctxt.get_int_type <unsigned short> (); |
| 42 |
|
| 43 |
See :c:func:`gcc_jit_context_get_type` for the available types. |
| 44 |
|
| 45 |
* derived types can be accessed by using functions such as |
| 46 |
:func:`gccjit::type::get_pointer` and :func:`gccjit::type::get_const`: |
| 47 |
|
| 48 |
.. code-block:: c++ |
| 49 |
|
| 50 |
gccjit::type const_int_star = int_type.get_const ().get_pointer (); |
| 51 |
gccjit::type int_const_star = int_type.get_pointer ().get_const (); |
| 52 |
|
| 53 |
* by creating structures (see below). |
| 54 |
|
| 55 |
Standard types |
| 56 |
-------------- |
| 57 |
|
| 58 |
.. function:: gccjit::type gccjit::context::get_type (enum gcc_jit_types) |
| 59 |
|
| 60 |
Access a specific type. This is a thin wrapper around |
| 61 |
:c:func:`gcc_jit_context_get_type`; the parameter has the same meaning. |
| 62 |
|
| 63 |
.. function:: gccjit::type \ |
| 64 |
gccjit::context::get_int_type (size_t num_bytes, int is_signed) |
| 65 |
|
| 66 |
Access the integer type of the given size. |
| 67 |
|
| 68 |
.. function:: template<typename T> gccjit::type \ |
| 69 |
gccjit::context::get_int_type <T> () |
| 70 |
|
| 71 |
Access the given integer type. For example, you could map the |
| 72 |
``unsigned short`` type into a gccjit::type via: |
| 73 |
|
| 74 |
.. code-block:: c++ |
| 75 |
|
| 76 |
gccjit::type t = ctxt.get_int_type <unsigned short> (); |
| 77 |
|
| 78 |
Pointers, `const`, and `volatile` |
| 79 |
--------------------------------- |
| 80 |
|
| 81 |
.. function:: gccjit::type gccjit::type::get_pointer () |
| 82 |
|
| 83 |
Given type "T", get type "T*". |
| 84 |
|
| 85 |
.. function:: gccjit::type gccjit::type::get_const () |
| 86 |
|
| 87 |
Given type "T", get type "const T". |
| 88 |
|
| 89 |
.. function:: gccjit::type gccjit::type::get_volatile () |
| 90 |
|
| 91 |
Given type "T", get type "volatile T". |
| 92 |
|
| 93 |
.. function:: gccjit::type gccjit::type::get_aligned (size_t alignment_in_bytes) |
| 94 |
|
| 95 |
Given type "T", get type: |
| 96 |
|
| 97 |
.. code-block:: c |
| 98 |
|
| 99 |
T __attribute__ ((aligned (ALIGNMENT_IN_BYTES))) |
| 100 |
|
| 101 |
The alignment must be a power of two. |
| 102 |
|
| 103 |
.. function:: gccjit::type \ |
| 104 |
gccjit::context::new_array_type (gccjit::type element_type, \ |
| 105 |
int num_elements, \ |
| 106 |
gccjit::location loc) |
| 107 |
|
| 108 |
Given type "T", get type "T[N]" (for a constant N). |
| 109 |
Param "loc" is optional. |
| 110 |
|
| 111 |
|
| 112 |
Vector types |
| 113 |
------------ |
| 114 |
|
| 115 |
.. function:: gccjit::type gccjit::type::get_vector (size_t num_units) |
| 116 |
|
| 117 |
Given type "T", get type: |
| 118 |
|
| 119 |
.. code-block:: c |
| 120 |
|
| 121 |
T __attribute__ ((vector_size (sizeof(T) * num_units)) |
| 122 |
|
| 123 |
T must be integral or floating point; num_units must be a power of two. |
| 124 |
|
| 125 |
|
| 126 |
Structures and unions |
| 127 |
--------------------- |
| 128 |
|
| 129 |
.. class:: gccjit::struct_ |
| 130 |
|
| 131 |
A compound type analagous to a C `struct`. |
| 132 |
|
| 133 |
:class:`gccjit::struct_` is a subclass of :class:`gccjit::type` (and thus |
| 134 |
of :class:`gccjit::object` in turn). |
| 135 |
|
| 136 |
.. class:: gccjit::field |
| 137 |
|
| 138 |
A field within a :class:`gccjit::struct_`. |
| 139 |
|
| 140 |
:class:`gccjit::field` is a subclass of :class:`gccjit::object`. |
| 141 |
|
| 142 |
You can model C `struct` types by creating :class:`gccjit::struct_` and |
| 143 |
:class:`gccjit::field` instances, in either order: |
| 144 |
|
| 145 |
* by creating the fields, then the structure. For example, to model: |
| 146 |
|
| 147 |
.. code-block:: c |
| 148 |
|
| 149 |
struct coord {double x; double y; }; |
| 150 |
|
| 151 |
you could call: |
| 152 |
|
| 153 |
.. code-block:: c++ |
| 154 |
|
| 155 |
gccjit::field field_x = ctxt.new_field (double_type, "x"); |
| 156 |
gccjit::field field_y = ctxt.new_field (double_type, "y"); |
| 157 |
std::vector fields; |
| 158 |
fields.push_back (field_x); |
| 159 |
fields.push_back (field_y); |
| 160 |
gccjit::struct_ coord = ctxt.new_struct_type ("coord", fields); |
| 161 |
|
| 162 |
* by creating the structure, then populating it with fields, typically |
| 163 |
to allow modelling self-referential structs such as: |
| 164 |
|
| 165 |
.. code-block:: c |
| 166 |
|
| 167 |
struct node { int m_hash; struct node *m_next; }; |
| 168 |
|
| 169 |
like this: |
| 170 |
|
| 171 |
.. code-block:: c++ |
| 172 |
|
| 173 |
gccjit::struct_ node = ctxt.new_opaque_struct_type ("node"); |
| 174 |
gccjit::type node_ptr = node.get_pointer (); |
| 175 |
gccjit::field field_hash = ctxt.new_field (int_type, "m_hash"); |
| 176 |
gccjit::field field_next = ctxt.new_field (node_ptr, "m_next"); |
| 177 |
std::vector fields; |
| 178 |
fields.push_back (field_hash); |
| 179 |
fields.push_back (field_next); |
| 180 |
node.set_fields (fields); |
| 181 |
|
| 182 |
.. FIXME: the above API doesn't seem to exist yet |
| 183 |
|
| 184 |
.. function:: gccjit::field \ |
| 185 |
gccjit::context::new_field (gccjit::type type,\ |
| 186 |
const char *name, \ |
| 187 |
gccjit::location loc) |
| 188 |
|
| 189 |
Construct a new field, with the given type and name. |
| 190 |
|
| 191 |
.. function:: gccjit::struct_ \ |
| 192 |
gccjit::context::new_struct_type (const std::string &name,\ |
| 193 |
std::vector<field> &fields,\ |
| 194 |
gccjit::location loc) |
| 195 |
|
| 196 |
Construct a new struct type, with the given name and fields. |
| 197 |
|
| 198 |
.. function:: gccjit::struct_ \ |
| 199 |
gccjit::context::new_opaque_struct (const std::string &name, \ |
| 200 |
gccjit::location loc) |
| 201 |
|
| 202 |
Construct a new struct type, with the given name, but without |
| 203 |
specifying the fields. The fields can be omitted (in which case the |
| 204 |
size of the struct is not known), or later specified using |
| 205 |
:c:func:`gcc_jit_struct_set_fields`. |