1 Concepts
2 ========
3
4 This document describes the underlying concepts employed in micropython.
5
6 * Namespaces and attribute definition
7 * Contexts and values
8 * Tables, attributes and lookups
9 * Objects and structures
10 * Parameters and lookups
11 * Instantiation
12 * Register usage
13 * List and tuple representations
14
15 Namespaces and Attribute Definition
16 ===================================
17
18 Namespaces are any objects which can retain attributes.
19
20 * Module attributes are defined either at the module level or by global
21 statements.
22 * Class attributes are defined only within class statements.
23 * Instance attributes are defined only by assignments to attributes of self
24 within __init__ methods.
25
26 These restrictions apply because such attributes are thus explicitly declared,
27 permitting the use of tables (described below). Module and class attributes
28 can also be finalised in this way in order to permit certain optimisations.
29
30 An additional restriction required for the current implementation of tables
31 (as described below) applies to class definitions: each class must be defined
32 using a unique name; repeated definition of classes having the same name is
33 thus not permitted. This restriction arises from the use of the "full name" of
34 a class as a key to the object table, where the full name is a qualified path
35 via the module hierarchy ending with the name of the class.
36
37 See rejected.txt for complicating mechanisms which could be applied to
38 mitigate the effects of these restrictions on optimisations.
39
40 Contexts and Values
41 ===================
42
43 Values are used as the common reference representation in micropython: as
44 stored representations of attributes (of classes, instances, modules, and
45 other objects supporting attribute-like entities) as well as the stored values
46 associated with names in functions and methods.
47
48 Unlike other implementations, micropython does not create things like bound
49 method objects for individual instances. Instead, all objects are referenced
50 using a context, reference pair:
51
52 Value Layout
53 ------------
54
55 0 1
56 context object
57 reference reference
58
59 Specific implementations might reverse this ordering for optimisation
60 purposes.
61
62 Rationale
63 ---------
64
65 To reduce the number of created objects whilst retaining the ability to
66 support bound method invocations. The context indicates the context in which
67 an invocation is performed, typically the owner of the method.
68
69 Usage
70 -----
71
72 The context may be inserted as the first argument when a value is involved in
73 an invocation. This argument may then be omitted from the invocation if its
74 usage is not appropriate.
75
76 See invocation.txt for details.
77
78 Context Value Types
79 -------------------
80
81 The following types of context value exist:
82
83 Type Usage Transformations
84 ---- ----- ---------------
85
86 Replaceable With functions (not methods) May be replaced with an
87 instance or a class when a
88 value is stored on an
89 instance or class
90
91 Placeholder With classes May not be replaced
92
93 Instance With instances (and constants) May not be replaced
94 or functions as methods
95
96 Class With functions as methods May be replaced when a
97 value is loaded from a
98 class attribute via an
99 instance
100
101 Contexts in Acquired Values
102 ---------------------------
103
104 There are four classes of instructions which provide values:
105
106 Instruction Purpose Context Operations
107 ----------- ------- ------------------
108
109 1) LoadConst Load module, constant Use loaded object with itself
110 as context
111
112 2) LoadFunction Load function Combine replaceable context
113 with loaded object
114
115 3) LoadClass Load class Combine placeholder context
116 with loaded object
117
118 4) LoadAddress* Load attribute from Preserve or override stored
119 LoadAttr* class, module, context (as described in
120 instance assignment.txt)
121
122 In order to comply with traditional Python behaviour, contexts may or may not
123 represent the object from which an attribute has been acquired.
124
125 See assignment.txt for details.
126
127 Contexts in Stored Values
128 -------------------------
129
130 There are two classes of instruction for storing values:
131
132 Instruction Purpose Context Operations
133 ----------- ------- ------------------
134
135 1) StoreAddress Store attribute in a Preserve context; note that no
136 known object test for class attribute
137 assignment should be necessary
138 since this instruction should only
139 be generated for module globals
140
141 StoreAttr Store attribute in an Preserve context; note that no
142 instance test for class attribute
143 assignment should be necessary
144 since this instruction should only
145 be generated for self accesses
146
147 StoreAttrIndex Store attribute in an Preserve context; since the index
148 unknown object lookup could yield a class
149 attribute, a test of the nature of
150 the nature of the structure is
151 necessary in order to prevent
152 assignments to classes
153
154 2) StoreAddressContext Store attribute in a Override context if appropriate;
155 known object if the value has a replaceable
156 context, permit the target to
157 take ownership of the value
158
159 See assignment.txt for details.
160
161 Tables, Attributes and Lookups
162 ==============================
163
164 Attribute lookups, where the exact location of an object attribute is deduced,
165 are performed differently in micropython than in other implementations.
166 Instead of providing attribute dictionaries, in which attributes are found,
167 attributes are located at fixed places in object structures (described below)
168 and their locations are stored using a special representation known as a
169 table.
170
171 For a given program, a table can be considered as being like a matrix mapping
172 classes to attribute names. For example:
173
174 class A:
175 # instances have attributes x, y
176
177 class B(A):
178 # introduces attribute z for instances
179
180 class C:
181 # instances have attributes a, b, z
182
183 This would provide the following table, referred to as an object table in the
184 context of classes and instances:
185
186 Class/attr a b x y z
187
188 A 1 2
189 B 1 2 3
190 C 1 2 3
191
192 A limitation of this representation is that instance attributes may not shadow
193 class attributes: if an attribute with a given name is not defined on an
194 instance, an attribute with the same name cannot be provided by the class of
195 the instance or any superclass of the instance's class.
196
197 The table can be compacted using a representation known as a displacement
198 list (referred to as an object list in this context):
199
200 Classes with attribute offsets
201
202 classcode A
203 attrcode a b x y z
204
205 B
206 a b x y z
207
208 C
209 a b x y z
210
211 List . . 1 2 1 2 3 1 2 . . 3
212
213 Here, the classcode refers to the offset in the list at which a class's
214 attributes are defined, whereas the attrcode defines the offset within a
215 region of attributes corresponding to a single attribute of a given name.
216
217 Attribute Locations
218 -------------------
219
220 The locations stored in table/list elements are generally for instance
221 attributes relative to the location of the instance, whereas those for class
222 attributes and module attributes are generally absolute addresses. Thus, each
223 occupied table cell has the following structure:
224
225 attrcode, uses-absolute-address, address (or location)
226
227 This could be given instead as follows:
228
229 attrcode, is-class-or-module, location
230
231 Since uses-absolute-address corresponds to is-class-or-module, and since there
232 is a need to test for classes and modules to prevent assignment to attributes
233 of such objects, this particular information is always required.
234
235 The __class__ Attribute
236 -----------------------
237
238 The exception to the above general rules about relative locations and absolute
239 addresses involves the __class__ attribute which is defined differently for
240 each class and its instances. Since the table elements can only refer to a
241 single absolute address, thus providing only a single value, such absolute
242 references which are sufficient for most class attributes would not be
243 appropriate for the __class__ attribute. However, using an object-relative
244 location would require both classes and instances to retain an attribute
245 location specifically to hold the value appropriate for each object type.
246
247 Comparing Tables as Matrices with Displacement Lists
248 ----------------------------------------------------
249
250 Although displacement lists can provide reasonable levels of compaction for
251 attribute data, the element size is larger than that required for a simple
252 matrix: the attribute code (attrcode) need not be stored since each element
253 unambiguously refers to the availability of an attribute for a particular
254 class or instance of that class, and so the data at a given element need not
255 be tested for relevance to a given attribute access operation.
256
257 Given a program with 20 object types and 100 attribute types, a matrix would
258 occupy the following amount of space:
259
260 number of object types * number of attribute types * element size
261 = 20 * 100 * 1 (assuming that a single location is sufficient for an element)
262 = 2000
263
264 In contrast, given a compaction to 40% of the matrix size (without considering
265 element size) in a displacement list, the amount of space would be as follows:
266
267 number of elements * element size
268 = 40% * (20 * 100) * 2 (assuming that one additional location is required)
269 = 1600
270
271 Consequently, the principal overhead of using a displacement list is likely to
272 be in the need to check element relevance when retrieving values from such a
273 list.
274
275 Objects and Structures
276 ======================
277
278 As well as references, micropython needs to have actual objects to refer to.
279 Since classes, functions and instances are all objects, it is desirable that
280 certain common features and operations are supported in the same way for all
281 of these things. To permit this, a common data structure format is used.
282
283 Header.................................................... Attributes.................
284
285 Identifier Identifier Address Identifier Size Object Object ...
286
287 0 1 2 3 4 5 6 7
288 classcode attrcode/ invocation funccode size __class__ attribute ...
289 instance reference reference reference
290 status
291
292 Classcode
293 ---------
294
295 Used in attribute lookup.
296
297 Here, the classcode refers to the attribute lookup table for the object (as
298 described above). Classes and instances share the same classcode, and their
299 structures reflect this. Functions all belong to the same type and thus employ
300 the classcode for the function built-in type, whereas modules have distinct
301 types since they must support different sets of attributes.
302
303 Attrcode
304 --------
305
306 Used to test instances for membership of classes (or descendants of classes).
307
308 Since, in traditional Python, classes are only ever instances of some generic
309 built-in type, support for testing such a relationship directly has been
310 removed and the attrcode is not specified for classes: the presence of an
311 attrcode indicates that a given object is an instance. In addition, support
312 has also been removed for testing modules in the same way, meaning that the
313 attrcode is also not specified for modules.
314
315 See the "Testing Instance Compatibility with Classes (Attrcode)" section below
316 for details of attrcodes.
317
318 Invocation Reference
319 --------------------
320
321 Used when an object is called.
322
323 This is the address of the code to be executed when an invocation is performed
324 on the object.
325
326 Funccode
327 --------
328
329 Used to look up argument positions by name.
330
331 The strategy with keyword arguments in micropython is to attempt to position
332 such arguments in the invocation frame as it is being constructed.
333
334 See the "Parameters and Lookups" section for more information.
335
336 Size
337 ----
338
339 Used to indicate the size of an object including attributes.
340
341 Attributes
342 ----------
343
344 For classes, modules and instances, the attributes in the structure correspond
345 to the attributes of each kind of object. For functions, however, the
346 attributes in the structure correspond to the default arguments for each
347 function, if any.
348
349 Structure Types
350 ---------------
351
352 Class C:
353
354 0 1 2 3 4 5 6 7
355 classcode (unused) __new__ funccode size class type attribute ...
356 for C reference for reference reference
357 instantiator
358
359 Instance of C:
360
361 0 1 2 3 4 5 6 7
362 classcode attrcode C.__call__ funccode size class C attribute ...
363 for C for C reference for reference reference
364 (if exists) C.__call__
365
366 Function f:
367
368 0 1 2 3 4 5 6 7
369 classcode attrcode code funccode size class attribute ...
370 for for reference function (default)
371 function function reference reference
372
373 Module m:
374
375 0 1 2 3 4 5 6 7
376 classcode attrcode (unused) (unused) (unused) module type attribute ...
377 for m for m reference (global)
378 reference
379
380 The __class__ Attribute
381 -----------------------
382
383 All objects support the __class__ attribute and this is illustrated above with
384 the first attribute.
385
386 Class: refers to the type class (type.__class__ also refers to the type class)
387 Function: refers to the function class
388 Instance: refers to the class instantiated to make the object
389
390 Lists and Tuples
391 ----------------
392
393 The built-in list and tuple sequences employ variable length structures using
394 the attribute locations to store their elements, where each element is a
395 reference to a separately stored object.
396
397 Testing Instance Compatibility with Classes (Attrcode)
398 ------------------------------------------------------
399
400 Although it would be possible to have a data structure mapping classes to
401 compatible classes, such as a matrix indicating the subclasses (or
402 superclasses) of each class, the need to retain the key to such a data
403 structure for each class might introduce a noticeable overhead.
404
405 Instead of having a separate structure, descendant classes of each class are
406 inserted as special attributes into the object table. This requires an extra
407 key to be retained, since each class must provide its own attribute code such
408 that upon an instance/class compatibility test, the code may be obtained and
409 used in the object table.
410
411 Invocation and Code References
412 ------------------------------
413
414 Modules: there is no meaningful invocation reference since modules cannot be
415 explicitly called.
416
417 Functions: a simple code reference is employed pointing to code implementing
418 the function. Note that the function locals are completely distinct from this
419 structure and are not comparable to attributes. Instead, attributes are
420 reserved for default parameter values, although they do not appear in the
421 object table described above, appearing instead in a separate parameter table
422 described below.
423
424 Classes: given that classes must be invoked in order to create instances, a
425 reference must be provided in class structures. However, this reference does
426 not point directly at the __init__ method of the class. Instead, the
427 referenced code belongs to a special initialiser function, __new__, consisting
428 of the following instructions:
429
430 create instance for C
431 call C.__init__(instance, ...)
432 return instance
433
434 Instances: each instance employs a reference to any __call__ method defined in
435 the class hierarchy for the instance, thus maintaining its callable nature.
436
437 Both classes and modules may contain code in their definitions - the former in
438 the "body" of the class, potentially defining attributes, and the latter as
439 the "top-level" code in the module, potentially defining attributes/globals -
440 but this code is not associated with any invocation target. It is thus
441 generated in order of appearance and is not referenced externally.
442
443 Invocation Operation
444 --------------------
445
446 Consequently, regardless of the object an invocation is always done as
447 follows:
448
449 get invocation reference from the header
450 jump to reference
451
452 Additional preparation is necessary before the above code: positional
453 arguments must be saved in the invocation frame, and keyword arguments must be
454 resolved and saved to the appropriate position in the invocation frame.
455
456 See invocation.txt for details.
457
458 Parameters and Lookups
459 ======================
460
461 Since Python supports keyword arguments when making invocations, it becomes
462 necessary to record the parameter names associated with each function or
463 method. Just as object tables record attributes positions on classes and
464 instances, parameter tables record parameter positions in function or method
465 parameter lists.
466
467 For a given program, a parameter table can be considered as being like a
468 matrix mapping functions/methods to parameter names. For example:
469
470 def f(x, y, z):
471 pass
472
473 def g(a, b, c):
474 pass
475
476 def h(a, x):
477 pass
478
479 This would provide the following table, referred to as a parameter table in
480 the context of functions and methods:
481
482 Function/param a b c x y z
483
484 f 1 2 3
485 g 1 2 3
486 h 1 2
487
488 Confusion can occur when functions are adopted as methods, since the context
489 then occupies the first slot in the invocation frame:
490
491 def f(x, y, z):
492 pass
493
494 f(x=1, y=2, z=3) -> f(<context>, 1, 2, 3)
495 -> f(1, 2, 3)
496
497 class C:
498 f = f
499
500 def g(x, y, z):
501 pass
502
503 c = C()
504
505 c.f(y=2, z=3) -> f(<context>, 2, 3)
506 c.g(y=2, z=3) -> C.g(<context>, 2, 3)
507
508 Just as with parameter tables, a displacement list can be prepared from a
509 parameter table:
510
511 Functions with parameter (attribute) offsets
512
513 funccode f
514 attrcode a b c x y z
515
516 g
517 a b c x y z
518
519 h
520 a b c x y z
521
522 List . . . 1 2 3 1 2 3 1 . . 2 . .
523
524 Here, the funccode refers to the offset in the list at which a function's
525 parameters are defined, whereas the attrcode defines the offset within a
526 region of attributes corresponding to a single parameter of a given name.
527
528 Instantiation
529 =============
530
531 When instantiating classes, memory must be reserved for the header of the
532 resulting instance, along with locations for the attributes of the instance.
533 Since the instance header contains data common to all instances of a class, a
534 template header is copied to the start of the newly reserved memory region.
535
536 Register Usage
537 ==============
538
539 During code generation, much of the evaluation produces results which are
540 implicitly recorded in the "active value" register, and various instructions
541 will consume the active value. In addition, some instructions will consume a
542 separate "active source value" from a register, typically those which are
543 assigning the result of an expression to an assignment target.
544
545 Since values often need to be retained for later use, a set of temporary
546 storage locations are typically employed. However, optimisations may reduce
547 the need to use such temporary storage where instructions which provide the
548 "active value" can be re-executed and will produce the same result.
549
550 List and Tuple Representations
551 ==============================
552
553 Since tuples have a fixed size, the representation of a tuple instance is
554 merely a header describing the size of the entire object, together with a
555 sequence of references to the object "stored" at each position in the
556 structure. Such references consist of the usual context and reference pair.
557
558 Lists, however, have a variable size and must be accessible via an unchanging
559 location even as more memory is allocated elsewhere to accommodate the
560 contents of the list. Consequently, the representation must resemble the
561 following:
562
563 Structure header for list (size == header plus special attribute)
564 Special attribute referencing the underlying sequence
565
566 The underlying sequence has a fixed size, like a tuple, but may contain fewer
567 elements than the size of the sequence permits:
568
569 Special header indicating the current size and allocated size
570 Element
571 ... <-- current size
572 (Unused space)
573 ... <-- allocated size
574
575 This representation permits the allocation of a new sequence when space is
576 exhausted in an existing sequence, with the new sequence address stored in the
577 main list structure. Since access to the contents of the list must go through
578 the main list structure, underlying allocation activities may take place
579 without the users of a list having to be aware of such activities.
