1 ===========================
2 TableGen Language Reference
3 ===========================
5 .. sectionauthor:: Sean Silva <silvas@purdue.edu>
11 This document is extremely rough. If you find something lacking, please
12 fix it, file a documentation bug, or ask about it on llvmdev.
17 This document is meant to be a normative spec about the TableGen language
18 in and of itself (i.e. how to understand a given construct in terms of how
19 it affects the final set of records represented by the TableGen file). If
20 you are unsure if this document is really what you are looking for, please
21 read :doc:`/TableGenFundamentals` first.
26 The lexical and syntax notation used here is intended to imitate
27 `Python's`_. In particular, for lexical definitions, the productions
28 operate at the character level and there is no implied whitespace between
29 elements. The syntax definitions operate at the token level, so there is
30 implied whitespace between tokens.
32 .. _`Python's`: http://docs.python.org/py3k/reference/introduction.html#notation
37 TableGen supports BCPL (``// ...``) and nestable C-style (``/* ... */``)
40 The following is a listing of the basic punctuation tokens::
42 - + [ ] { } ( ) < > : ; . = ? #
44 Numeric literals take one of the following forms:
46 .. TableGen actually will lex some pretty strange sequences an interpret
47 them as numbers. What is shown here is an attempt to approximate what it
51 TokInteger: `DecimalInteger` | `HexInteger` | `BinInteger`
52 DecimalInteger: ["+" | "-"] ("0"..."9")+
53 HexInteger: "0x" ("0"..."9" | "a"..."f" | "A"..."F")+
54 BinInteger: "0b" ("0" | "1")+
56 One aspect to note is that the :token:`DecimalInteger` token *includes* the
57 ``+`` or ``-``, as opposed to having ``+`` and ``-`` be unary operators as
60 TableGen has identifier-like tokens:
63 ualpha: "a"..."z" | "A"..."Z" | "_"
64 TokIdentifier: ("0"..."9")* `ualpha` (`ualpha` | "0"..."9")*
65 TokVarName: "$" `ualpha` (`ualpha` | "0"..."9")*
67 Note that unlike most languages, TableGen allows :token:`TokIdentifier` to
68 begin with a number. In case of ambiguity, a token will be interpreted as a
69 numeric literal rather than an identifier.
71 TableGen also has two string-like literals:
74 TokString: '"' <non-'"' characters and C-like escapes> '"'
75 TokCodeFragment: "[{" <shortest text not containing "}]"> "}]"
77 TableGen also has the following keywords::
79 bit bits class code dag
80 def foreach defm field in
81 int let list multiclass string
83 TableGen also has "bang operators" which have a
84 wide variety of meanings::
86 !eq !if !head !tail !con
88 !cast !empty !subst !foreach !strconcat
93 TableGen has an ``include`` mechanism. It does not play a role in the
94 syntax per se, since it is lexically replaced with the contents of the
98 IncludeDirective: "include" `TokString`
100 TableGen's top-level production consists of "objects".
103 TableGenFile: `Object`*
104 Object: `Class` | `Def` | `Defm` | `Let` | `MultiClass` | `Foreach`
110 Class: "class" `TokIdentifier` [`TemplateArgList`] `ObjectBody`
112 A ``class`` declaration creates a record which other records can inherit
113 from. A class can be parametrized by a list of "template arguments", whose
114 values can be used in the class body.
116 A given class can only be defined once. A ``class`` declaration is
117 considered to define the class if any of the following is true:
119 .. break ObjectBody into its consituents so that they are present here?
121 #. The :token:`TemplateArgList` is present.
122 #. The :token:`Body` in the :token:`ObjectBody` is present and is not empty.
123 #. The :token:`BaseClassList` in the :token:`ObjectBody` is present.
125 You can declare an empty class by giving and empty :token:`TemplateArgList`
126 and an empty :token:`ObjectBody`. This can serve as a restricted form of
127 forward declaration: note that records deriving from the forward-declared
128 class will inherit no fields from it since the record expansion is done
129 when the record is parsed.
132 TemplateArgList: "<" `Declaration` ("," `Declaration`)* ">"
137 .. Omitting mention of arcane "field" prefix to discourage its use.
139 The declaration syntax is pretty much what you would expect as a C++
143 Declaration: `Type` `TokIdentifier` ["=" `Value`]
145 It assigns the value to the identifer.
151 Type: "string" | "code" | "bit" | "int" | "dag"
152 :| "bits" "<" `TokInteger` ">"
153 :| "list" "<" `Type` ">"
155 ClassID: `TokIdentifier`
157 Both ``string`` and ``code`` correspond to the string type; the difference
158 is purely to indicate programmer intention.
160 The :token:`ClassID` must identify a class that has been previously
167 Value: `SimpleValue` `ValueSuffix`*
168 ValueSuffix: "{" `RangeList` "}"
169 :| "[" `RangeList` "]"
170 :| "." `TokIdentifier`
171 RangeList: `RangePiece` ("," `RangePiece`)*
172 RangePiece: `TokInteger`
173 :| `TokInteger` "-" `TokInteger`
174 :| `TokInteger` `TokInteger`
176 The peculiar last form of :token:`RangePiece` is due to the fact that the
177 "``-``" is included in the :token:`TokInteger`, hence ``1-5`` gets lexed as
178 two consecutive :token:`TokInteger`'s, with values ``1`` and ``-5``,
179 instead of "1", "-", and "5".
180 The :token:`RangeList` can be thought of as specifying "list slice" in some
184 :token:`SimpleValue` has a number of forms:
188 SimpleValue: `TokIdentifier`
190 The value will be the variable referenced by the identifier. It can be one
193 .. The code for this is exceptionally abstruse. These examples are a
196 * name of a ``def``, such as the use of ``Bar`` in::
198 def Bar : SomeClass {
206 * value local to a ``def``, such as the use of ``Bar`` in::
213 * a template arg of a ``class``, such as the use of ``Bar`` in::
219 * value local to a ``multiclass``, such as the use of ``Bar`` in::
226 * a template arg to a ``multiclass``, such as the use of ``Bar`` in::
228 multiclass Foo<int Bar> {
233 SimpleValue: `TokInteger`
235 This represents the numeric value of the integer.
238 SimpleValue: `TokString`+
240 Multiple adjacent string literals are concatenated like in C/C++. The value
241 is the concatenation of the strings.
244 SimpleValue: `TokCodeFragment`
246 The value is the string value of the code fragment.
251 ``?`` represents an "unset" initializer.
254 SimpleValue: "{" `ValueList` "}"
255 ValueList: [`ValueListNE`]
256 ValueListNE: `Value` ("," `Value`)*
258 This represents a sequence of bits, as would be used to initialize a
259 ``bits<n>`` field (where ``n`` is the number of bits).
262 SimpleValue: `ClassID` "<" `ValueListNE` ">"
264 This generates a new anonymous record definition (as would be created by an
265 unnamed ``def`` inheriting from the given class with the given template
266 arguments) and the value is the value of that record definition.
269 SimpleValue: "[" `ValueList` "]" ["<" `Type` ">"]
271 A list initializer. The optional :token:`Type` can be used to indicate a
272 specific element type, otherwise the element type will be deduced from the
275 .. The initial `DagArg` of the dag must start with an identifier or
276 !cast, but this is more of an implementation detail and so for now just
280 SimpleValue: "(" `DagArg` `DagArgList` ")"
281 DagArgList: `DagArg` ("," `DagArg`)*
282 DagArg: `Value` [":" `TokVarName`]
284 The initial :token:`DagArg` is called the "operator" of the dag.
287 SimpleValue: `BangOperator` ["<" `Type` ">"] "(" `ValueListNE` ")"
293 ObjectBody: `BaseClassList` `Body`
294 BaseClassList: [`BaseClassListNE`]
295 BaseClassListNE: `SubClassRef` ("," `SubClassRef`)*
296 SubClassRef: (`ClassID` | `DefmID`) ["<" `ValueList` ">"]
297 DefmID: `TokIdentifier`
299 The version with the :token:`DefmID` is only valid in the
300 :token:`BaseClassList` of a ``defm``.
301 The :token:`DefmID` should be the name of a ``multiclass``.
303 .. put this somewhere else
305 It is after parsing the base class list that the "let stack" is applied.
308 Body: ";" | "{" BodyList "}"
310 BodyItem: `Declaration` ";"
311 :| "let" `TokIdentifier` [`RangeList`] "=" `Value` ";"
313 The ``let`` form allows overriding the value of an inherited field.
319 There can be pastes in the names here, like ``#NAME#``. Look into that
320 and document it (it boils down to ParseIDValue with IDParseMode ==
321 ParseNameMode). ParseObjectName calls into the general ParseValue, with
322 the only different from "arbitrary expression parsing" being IDParseMode
326 Def: "def" `TokIdentifier` `ObjectBody`
328 Defines a record whose name is given by the :token:`TokIdentifier`. The
329 fields of the record are inherited from the base classes and defined in the
332 Special handling occurs if this ``def`` appears inside a ``multiclass`` or
339 Defm: "defm" `TokIdentifier` ":" `BaseClassList` ";"
341 Note that in the :token:`BaseClassList`, all of the ``multiclass``'s must
342 precede any ``class``'s that appear.
348 Foreach: "foreach" `Declaration` "in" "{" `Object`* "}"
349 :| "foreach" `Declaration` "in" `Object`
351 The value assigned to the variable in the declaration is iterated over and
352 the object or object list is reevaluated with the variable set at each
359 Let: "let" `LetList` "in" "{" `Object`* "}"
360 :| "let" `LetList` "in" `Object`
361 LetList: `LetItem` ("," `LetItem`)*
362 LetItem: `TokIdentifier` [`RangeList`] "=" `Value`
364 This is effectively equivalent to ``let`` inside the body of a record
365 except that it applies to multiple records at a time. The bindings are
366 applied at the end of parsing the base classes of a record.
372 MultiClass: "multiclass" `TokIdentifier` [`TemplateArgList`]
373 : [":" `BaseMultiClassList`] "{" `MultiClassObject`+ "}"
374 BaseMultiClassList: `MultiClassID` ("," `MultiClassID`)*
375 MultiClassID: `TokIdentifier`
376 MultiClassObject: `Def` | `Defm` | `Let` | `Foreach`