Commit | Line | Data |
---|---|---|
f5cb0e57 | 1 | Common Trace Format (CTF) Specification (v1.8.2) |
5ba9f198 MD |
2 | |
3 | Mathieu Desnoyers, EfficiOS Inc. | |
4 | ||
339a7dde | 5 | The goal of the present document is to specify a trace format that suits the |
cc089c3a | 6 | needs of the embedded, telecom, high-performance and kernel communities. It is |
5ba9f198 | 7 | based on the Common Trace Format Requirements (v1.4) document. It is designed to |
cc089c3a | 8 | allow traces to be natively generated by the Linux kernel, Linux user-space |
6c7226e9 MD |
9 | applications written in C/C++, and hardware components. One major element of |
10 | CTF is the Trace Stream Description Language (TSDL) which flexibility | |
11 | enables description of various binary trace stream layouts. | |
cc089c3a MD |
12 | |
13 | The latest version of this document can be found at: | |
14 | ||
15 | git tree: git://git.efficios.com/ctf.git | |
16 | gitweb: http://git.efficios.com/?p=ctf.git | |
5ba9f198 MD |
17 | |
18 | A reference implementation of a library to read and write this trace format is | |
19 | being implemented within the BabelTrace project, a converter between trace | |
20 | formats. The development tree is available at: | |
21 | ||
22 | git tree: git://git.efficios.com/babeltrace.git | |
23 | gitweb: http://git.efficios.com/?p=babeltrace.git | |
24 | ||
339a7dde MD |
25 | The CE Workgroup of the Linux Foundation, Ericsson, and EfficiOS have |
26 | sponsored this work. | |
27 | ||
5ba9f198 | 28 | |
beabf088 MD |
29 | Table of Contents |
30 | ||
31 | 1. Preliminary definitions | |
32 | 2. High-level representation of a trace | |
33 | 3. Event stream | |
34 | 4. Types | |
35 | 4.1 Basic types | |
36 | 4.1.1 Type inheritance | |
37 | 4.1.2 Alignment | |
38 | 4.1.3 Byte order | |
39 | 4.1.4 Size | |
40 | 4.1.5 Integers | |
41 | 4.1.6 GNU/C bitfields | |
42 | 4.1.7 Floating point | |
43 | 4.1.8 Enumerations | |
ddd045b0 MD |
44 | 4.2 Compound types |
45 | 4.2.1 Structures | |
46 | 4.2.2 Variants (Discriminated/Tagged Unions) | |
47 | 4.2.3 Arrays | |
48 | 4.2.4 Sequences | |
49 | 4.2.5 Strings | |
beabf088 MD |
50 | 5. Event Packet Header |
51 | 5.1 Event Packet Header Description | |
52 | 5.2 Event Packet Context Description | |
53 | 6. Event Structure | |
54 | 6.1 Event Header | |
55 | 6.1.1 Type 1 - Few event IDs | |
56 | 6.1.2 Type 2 - Many event IDs | |
0b174d87 | 57 | 6.2 Stream Event Context and Event Context |
beabf088 MD |
58 | 6.3 Event Payload |
59 | 6.3.1 Padding | |
60 | 6.3.2 Alignment | |
61 | 7. Trace Stream Description Language (TSDL) | |
62 | 7.1 Meta-data | |
63 | 7.2 Declaration vs Definition | |
64 | 7.3 TSDL Scopes | |
65 | 7.3.1 Lexical Scope | |
37ab95c3 | 66 | 7.3.2 Static and Dynamic Scopes |
beabf088 | 67 | 7.4 TSDL Examples |
2fa70eba | 68 | 8. Clocks |
beabf088 MD |
69 | |
70 | ||
5ba9f198 MD |
71 | 1. Preliminary definitions |
72 | ||
3bf79539 MD |
73 | - Event Trace: An ordered sequence of events. |
74 | - Event Stream: An ordered sequence of events, containing a subset of the | |
75 | trace event types. | |
76 | - Event Packet: A sequence of physically contiguous events within an event | |
77 | stream. | |
5ba9f198 MD |
78 | - Event: This is the basic entry in a trace. (aka: a trace record). |
79 | - An event identifier (ID) relates to the class (a type) of event within | |
3bf79539 MD |
80 | an event stream. |
81 | e.g. event: irq_entry. | |
5ba9f198 MD |
82 | - An event (or event record) relates to a specific instance of an event |
83 | class. | |
3bf79539 MD |
84 | e.g. event: irq_entry, at time X, on CPU Y |
85 | - Source Architecture: Architecture writing the trace. | |
86 | - Reader Architecture: Architecture reading the trace. | |
5ba9f198 MD |
87 | |
88 | ||
89 | 2. High-level representation of a trace | |
90 | ||
3bf79539 MD |
91 | A trace is divided into multiple event streams. Each event stream contains a |
92 | subset of the trace event types. | |
5ba9f198 | 93 | |
3bf79539 MD |
94 | The final output of the trace, after its generation and optional transport over |
95 | the network, is expected to be either on permanent or temporary storage in a | |
96 | virtual file system. Because each event stream is appended to while a trace is | |
32254806 MD |
97 | being recorded, each is associated with a distinct set of files for |
98 | output. Therefore, a stored trace can be represented as a directory | |
99 | containing zero, one or more files per stream. | |
5ba9f198 | 100 | |
6672e9e1 MD |
101 | Meta-data description associated with the trace contains information on |
102 | trace event types expressed in the Trace Stream Description Language | |
103 | (TSDL). This language describes: | |
5ba9f198 MD |
104 | |
105 | - Trace version. | |
106 | - Types available. | |
6672e9e1 | 107 | - Per-trace event header description. |
3bf79539 | 108 | - Per-stream event header description. |
6672e9e1 | 109 | - Per-stream event context description. |
5ba9f198 | 110 | - Per-event |
3bf79539 | 111 | - Event type to stream mapping. |
5ba9f198 MD |
112 | - Event type to name mapping. |
113 | - Event type to ID mapping. | |
6672e9e1 | 114 | - Event context description. |
5ba9f198 MD |
115 | - Event fields description. |
116 | ||
117 | ||
3bf79539 | 118 | 3. Event stream |
5ba9f198 | 119 | |
6672e9e1 | 120 | An event stream can be divided into contiguous event packets of variable |
12c9069f MD |
121 | size. An event packet can contain a certain amount of padding at the |
122 | end. The stream header is repeated at the beginning of each event | |
123 | packet. The rationale for the event stream design choices is explained | |
124 | in Appendix B. Stream Header Rationale. | |
5ba9f198 | 125 | |
3bf79539 MD |
126 | The event stream header will therefore be referred to as the "event packet |
127 | header" throughout the rest of this document. | |
5ba9f198 MD |
128 | |
129 | ||
130 | 4. Types | |
131 | ||
1fad7a85 MD |
132 | Types are organized as type classes. Each type class belong to either of two |
133 | kind of types: basic types or compound types. | |
134 | ||
5ba9f198 MD |
135 | 4.1 Basic types |
136 | ||
1fad7a85 MD |
137 | A basic type is a scalar type, as described in this section. It includes |
138 | integers, GNU/C bitfields, enumerations, and floating point values. | |
5ba9f198 MD |
139 | |
140 | 4.1.1 Type inheritance | |
141 | ||
80fd2569 MD |
142 | Type specifications can be inherited to allow deriving types from a |
143 | type class. For example, see the uint32_t named type derived from the "integer" | |
144 | type class below ("Integers" section). Types have a precise binary | |
145 | representation in the trace. A type class has methods to read and write these | |
146 | types, but must be derived into a type to be usable in an event field. | |
5ba9f198 MD |
147 | |
148 | 4.1.2 Alignment | |
149 | ||
150 | We define "byte-packed" types as aligned on the byte size, namely 8-bit. | |
151 | We define "bit-packed" types as following on the next bit, as defined by the | |
370eae99 | 152 | "Integers" section. |
5ba9f198 | 153 | |
6672e9e1 | 154 | Each basic type must specify its alignment, in bits. Examples of |
ec4404a7 MD |
155 | possible alignments are: bit-packed (align = 1), byte-packed (align = |
156 | 8), or word-aligned (e.g. align = 32 or align = 64). The choice depends | |
157 | on the architecture preference and compactness vs performance trade-offs | |
158 | of the implementation. Architectures providing fast unaligned write | |
159 | byte-packed basic types to save space, aligning each type on byte | |
160 | boundaries (8-bit). Architectures with slow unaligned writes align types | |
161 | on specific alignment values. If no specific alignment is declared for a | |
162 | type, it is assumed to be bit-packed for integers with size not multiple | |
163 | of 8 bits and for gcc bitfields. All other basic types are byte-packed | |
164 | by default. It is however recommended to always specify the alignment | |
165 | explicitly. Alignment values must be power of two. Compound types are | |
166 | aligned as specified in their individual specification. | |
5ba9f198 | 167 | |
96bf5a3b MD |
168 | The base offset used for field alignment is the start of the packet |
169 | containing the field. For instance, a field aligned on 32-bit needs to | |
170 | be at an offset multiple of 32-bit from the start of the packet that | |
171 | contains it. | |
172 | ||
6672e9e1 | 173 | TSDL meta-data attribute representation of a specific alignment: |
5ba9f198 MD |
174 | |
175 | align = value; /* value in bits */ | |
176 | ||
177 | 4.1.3 Byte order | |
178 | ||
237324a2 MD |
179 | By default, byte order of a basic type is the byte order described in |
180 | the trace description. It can be overridden by specifying a | |
181 | "byte_order" attribute for a basic type. Typical use-case is to specify | |
182 | the network byte order (big endian: "be") to save data captured from the | |
183 | network into the trace without conversion. | |
5ba9f198 | 184 | |
6672e9e1 | 185 | TSDL meta-data representation: |
5ba9f198 MD |
186 | |
187 | byte_order = native OR network OR be OR le; /* network and be are aliases */ | |
188 | ||
237324a2 MD |
189 | The "native" keyword selects the byte order described in the trace |
190 | description. The "network" byte order is an alias for big endian. | |
191 | ||
7e58b8bf | 192 | Even though the trace description section is not per se a type, for sake |
237324a2 MD |
193 | of clarity, it should be noted that "native" and "network" byte orders |
194 | are only allowed within type declaration. The byte_order specified in | |
195 | the trace description section only accepts "be" or "le" values. | |
7e58b8bf | 196 | |
5ba9f198 MD |
197 | 4.1.4 Size |
198 | ||
199 | Type size, in bits, for integers and floats is that returned by "sizeof()" in C | |
200 | multiplied by CHAR_BIT. | |
201 | We require the size of "char" and "unsigned char" types (CHAR_BIT) to be fixed | |
202 | to 8 bits for cross-endianness compatibility. | |
203 | ||
6672e9e1 | 204 | TSDL meta-data representation: |
5ba9f198 MD |
205 | |
206 | size = value; (value is in bits) | |
207 | ||
208 | 4.1.5 Integers | |
209 | ||
6672e9e1 MD |
210 | Signed integers are represented in two-complement. Integer alignment, |
211 | size, signedness and byte ordering are defined in the TSDL meta-data. | |
212 | Integers aligned on byte size (8-bit) and with length multiple of byte | |
213 | size (8-bit) correspond to the C99 standard integers. In addition, | |
214 | integers with alignment and/or size that are _not_ a multiple of the | |
215 | byte size are permitted; these correspond to the C99 standard bitfields, | |
216 | with the added specification that the CTF integer bitfields have a fixed | |
7e5cdbfc MD |
217 | binary representation. Integer size needs to be a positive integer. |
218 | Integers of size 0 are forbidden. A MIT-licensed reference | |
219 | implementation of the CTF portable bitfields is available at: | |
5ba9f198 MD |
220 | |
221 | http://git.efficios.com/?p=babeltrace.git;a=blob;f=include/babeltrace/bitfield.h | |
222 | ||
223 | Binary representation of integers: | |
224 | ||
225 | - On little and big endian: | |
226 | - Within a byte, high bits correspond to an integer high bits, and low bits | |
227 | correspond to low bits. | |
228 | - On little endian: | |
229 | - Integer across multiple bytes are placed from the less significant to the | |
230 | most significant. | |
231 | - Consecutive integers are placed from lower bits to higher bits (even within | |
232 | a byte). | |
233 | - On big endian: | |
234 | - Integer across multiple bytes are placed from the most significant to the | |
235 | less significant. | |
236 | - Consecutive integers are placed from higher bits to lower bits (even within | |
237 | a byte). | |
238 | ||
239 | This binary representation is derived from the bitfield implementation in GCC | |
240 | for little and big endian. However, contrary to what GCC does, integers can | |
6672e9e1 | 241 | cross units boundaries (no padding is required). Padding can be explicitly |
5ba9f198 MD |
242 | added (see 4.1.6 GNU/C bitfields) to follow the GCC layout if needed. |
243 | ||
6672e9e1 | 244 | TSDL meta-data representation: |
5ba9f198 | 245 | |
80fd2569 | 246 | integer { |
5ba9f198 MD |
247 | signed = true OR false; /* default false */ |
248 | byte_order = native OR network OR be OR le; /* default native */ | |
249 | size = value; /* value in bits, no default */ | |
250 | align = value; /* value in bits */ | |
56b77c06 | 251 | /* based used for pretty-printing output, default: decimal. */ |
b46cfbf6 | 252 | base = decimal OR dec OR d OR i OR u OR 10 OR hexadecimal OR hex OR x OR X OR p OR 16 |
56b77c06 | 253 | OR octal OR oct OR o OR 8 OR binary OR b OR 2; |
13441bf5 MD |
254 | /* character encoding, default: none */ |
255 | encoding = none or UTF8 or ASCII; | |
2152348f | 256 | } |
5ba9f198 | 257 | |
80fd2569 | 258 | Example of type inheritance (creation of a uint32_t named type): |
5ba9f198 | 259 | |
359894ac | 260 | typealias integer { |
9e4e34e9 | 261 | size = 32; |
5ba9f198 MD |
262 | signed = false; |
263 | align = 32; | |
38b8da21 | 264 | } := uint32_t; |
5ba9f198 | 265 | |
80fd2569 | 266 | Definition of a named 5-bit signed bitfield: |
5ba9f198 | 267 | |
359894ac | 268 | typealias integer { |
5ba9f198 MD |
269 | size = 5; |
270 | signed = true; | |
271 | align = 1; | |
38b8da21 | 272 | } := int5_t; |
5ba9f198 | 273 | |
13441bf5 MD |
274 | The character encoding field can be used to specify that the integer |
275 | must be printed as a text character when read. e.g.: | |
276 | ||
277 | typealias integer { | |
278 | size = 8; | |
279 | align = 8; | |
280 | signed = false; | |
281 | encoding = UTF8; | |
282 | } := utf_char; | |
283 | ||
284 | ||
5ba9f198 MD |
285 | 4.1.6 GNU/C bitfields |
286 | ||
287 | The GNU/C bitfields follow closely the integer representation, with a | |
288 | particularity on alignment: if a bitfield cannot fit in the current unit, the | |
80fd2569 MD |
289 | unit is padded and the bitfield starts at the following unit. The unit size is |
290 | defined by the size of the type "unit_type". | |
5ba9f198 | 291 | |
6672e9e1 | 292 | TSDL meta-data representation: |
80fd2569 | 293 | |
d674f4b8 | 294 | unit_type name:size; |
80fd2569 | 295 | |
5ba9f198 MD |
296 | As an example, the following structure declared in C compiled by GCC: |
297 | ||
298 | struct example { | |
299 | short a:12; | |
300 | short b:5; | |
301 | }; | |
302 | ||
2152348f MD |
303 | The example structure is aligned on the largest element (short). The second |
304 | bitfield would be aligned on the next unit boundary, because it would not fit in | |
305 | the current unit. | |
5ba9f198 MD |
306 | |
307 | 4.1.7 Floating point | |
308 | ||
6672e9e1 | 309 | The floating point values byte ordering is defined in the TSDL meta-data. |
5ba9f198 MD |
310 | |
311 | Floating point values follow the IEEE 754-2008 standard interchange formats. | |
312 | Description of the floating point values include the exponent and mantissa size | |
313 | in bits. Some requirements are imposed on the floating point values: | |
314 | ||
315 | - FLT_RADIX must be 2. | |
316 | - mant_dig is the number of digits represented in the mantissa. It is specified | |
317 | by the ISO C99 standard, section 5.2.4, as FLT_MANT_DIG, DBL_MANT_DIG and | |
318 | LDBL_MANT_DIG as defined by <float.h>. | |
319 | - exp_dig is the number of digits represented in the exponent. Given that | |
320 | mant_dig is one bit more than its actual size in bits (leading 1 is not | |
321 | needed) and also given that the sign bit always takes one bit, exp_dig can be | |
322 | specified as: | |
323 | ||
324 | - sizeof(float) * CHAR_BIT - FLT_MANT_DIG | |
325 | - sizeof(double) * CHAR_BIT - DBL_MANT_DIG | |
326 | - sizeof(long double) * CHAR_BIT - LDBL_MANT_DIG | |
327 | ||
6672e9e1 | 328 | TSDL meta-data representation: |
5ba9f198 | 329 | |
80fd2569 | 330 | floating_point { |
ec4404a7 MD |
331 | exp_dig = value; |
332 | mant_dig = value; | |
333 | byte_order = native OR network OR be OR le; | |
334 | align = value; | |
2152348f | 335 | } |
5ba9f198 MD |
336 | |
337 | Example of type inheritance: | |
338 | ||
359894ac | 339 | typealias floating_point { |
5ba9f198 MD |
340 | exp_dig = 8; /* sizeof(float) * CHAR_BIT - FLT_MANT_DIG */ |
341 | mant_dig = 24; /* FLT_MANT_DIG */ | |
342 | byte_order = native; | |
ec4404a7 | 343 | align = 32; |
38b8da21 | 344 | } := float; |
5ba9f198 MD |
345 | |
346 | TODO: define NaN, +inf, -inf behavior. | |
347 | ||
ec4404a7 MD |
348 | Bit-packed, byte-packed or larger alignments can be used for floating |
349 | point values, similarly to integers. | |
350 | ||
5ba9f198 MD |
351 | 4.1.8 Enumerations |
352 | ||
353 | Enumerations are a mapping between an integer type and a table of strings. The | |
354 | numerical representation of the enumeration follows the integer type specified | |
6672e9e1 MD |
355 | by the meta-data. The enumeration mapping table is detailed in the enumeration |
356 | description within the meta-data. The mapping table maps inclusive value | |
357 | ranges (or single values) to strings. Instead of being limited to simple | |
3bf79539 | 358 | "value -> string" mappings, these enumerations map |
80fd2569 | 359 | "[ start_value ... end_value ] -> string", which map inclusive ranges of |
3bf79539 | 360 | values to strings. An enumeration from the C language can be represented in |
c23c2d0d MD |
361 | this format by having the same start_value and end_value for each |
362 | mapping, which is in fact a range of size 1. This single-value range is | |
363 | supported without repeating the start and end values with the value = | |
364 | string declaration. Enumerations need to contain at least one entry. | |
80fd2569 | 365 | |
a9b83695 | 366 | enum name : integer_type { |
359894ac | 367 | somestring = start_value1 ... end_value1, |
80fd2569 MD |
368 | "other string" = start_value2 ... end_value2, |
369 | yet_another_string, /* will be assigned to end_value2 + 1 */ | |
370 | "some other string" = value, | |
371 | ... | |
372 | }; | |
373 | ||
374 | If the values are omitted, the enumeration starts at 0 and increment of 1 for | |
0cedf6e5 MD |
375 | each entry. An entry with omitted value that follows a range entry takes |
376 | as value the end_value of the previous range + 1: | |
80fd2569 | 377 | |
a9b83695 | 378 | enum name : unsigned int { |
80fd2569 MD |
379 | ZERO, |
380 | ONE, | |
381 | TWO, | |
382 | TEN = 10, | |
383 | ELEVEN, | |
3bf79539 | 384 | }; |
5ba9f198 | 385 | |
80fd2569 | 386 | Overlapping ranges within a single enumeration are implementation defined. |
5ba9f198 | 387 | |
2152348f MD |
388 | A nameless enumeration can be declared as a field type or as part of a typedef: |
389 | ||
a9b83695 | 390 | enum : integer_type { |
2152348f MD |
391 | ... |
392 | } | |
393 | ||
c2742c56 MD |
394 | Enumerations omitting the container type ": integer_type" use the "int" |
395 | type (for compatibility with C99). The "int" type must be previously | |
396 | declared. E.g.: | |
397 | ||
82cbdb16 | 398 | typealias integer { size = 32; align = 32; signed = true; } := int; |
c2742c56 MD |
399 | |
400 | enum { | |
401 | ... | |
402 | } | |
403 | ||
1fad7a85 | 404 | |
5ba9f198 MD |
405 | 4.2 Compound types |
406 | ||
1fad7a85 MD |
407 | Compound are aggregation of type declarations. Compound types include |
408 | structures, variant, arrays, sequences, and strings. | |
409 | ||
5ba9f198 MD |
410 | 4.2.1 Structures |
411 | ||
412 | Structures are aligned on the largest alignment required by basic types | |
413 | contained within the structure. (This follows the ISO/C standard for structures) | |
414 | ||
6672e9e1 | 415 | TSDL meta-data representation of a named structure: |
5ba9f198 | 416 | |
80fd2569 MD |
417 | struct name { |
418 | field_type field_name; | |
419 | field_type field_name; | |
420 | ... | |
421 | }; | |
5ba9f198 MD |
422 | |
423 | Example: | |
424 | ||
80fd2569 MD |
425 | struct example { |
426 | integer { /* Nameless type */ | |
427 | size = 16; | |
428 | signed = true; | |
429 | align = 16; | |
430 | } first_field_name; | |
6672e9e1 | 431 | uint64_t second_field_name; /* Named type declared in the meta-data */ |
3bf79539 | 432 | }; |
5ba9f198 | 433 | |
4cac83ee MD |
434 | The fields are placed in a sequence next to each other. They each |
435 | possess a field name, which is a unique identifier within the structure. | |
436 | The identifier is not allowed to use any reserved keyword | |
437 | (see Section C.1.2). Replacing reserved keywords with | |
70375f92 | 438 | underscore-prefixed field names is recommended. Fields starting with an |
92250c71 MD |
439 | underscore should have their leading underscore removed by the CTF trace |
440 | readers. | |
5ba9f198 | 441 | |
2152348f | 442 | A nameless structure can be declared as a field type or as part of a typedef: |
80fd2569 MD |
443 | |
444 | struct { | |
445 | ... | |
2152348f | 446 | } |
80fd2569 | 447 | |
ec4404a7 MD |
448 | Alignment for a structure compound type can be forced to a minimum value |
449 | by adding an "align" specifier after the declaration of a structure | |
450 | body. This attribute is read as: align(value). The value is specified in | |
451 | bits. The structure will be aligned on the maximum value between this | |
452 | attribute and the alignment required by the basic types contained within | |
453 | the structure. e.g. | |
454 | ||
455 | struct { | |
456 | ... | |
457 | } align(32) | |
458 | ||
77a98c82 | 459 | 4.2.2 Variants (Discriminated/Tagged Unions) |
fcba70d4 | 460 | |
fdf2bb05 MD |
461 | A CTF variant is a selection between different types. A CTF variant must |
462 | always be defined within the scope of a structure or within fields | |
463 | contained within a structure (defined recursively). A "tag" enumeration | |
37ab95c3 | 464 | field must appear in either the same static scope, prior to the variant |
c23c2d0d | 465 | field (in field declaration order), in an upper static scope, or in an |
37ab95c3 MD |
466 | upper dynamic scope (see Section 7.3.2). The type selection is indicated |
467 | by the mapping from the enumeration value to the string used as variant | |
468 | type selector. The field to use as tag is specified by the "tag_field", | |
469 | specified between "< >" after the "variant" keyword for unnamed | |
c23c2d0d | 470 | variants, and after "variant name" for named variants. It is not |
23bb140f MD |
471 | required that each enumeration mapping appears as variant type tag |
472 | field. It is also not required that each variant type tag appears as | |
473 | enumeration mapping. However, it is required that any enumeration | |
474 | mapping encountered within a stream has a matching variant type tag | |
475 | field. | |
fcba70d4 | 476 | |
0bed08fc MD |
477 | The alignment of the variant is the alignment of the type as selected by |
478 | the tag value for the specific instance of the variant. The size of the | |
479 | variant is the size as selected by the tag value for the specific | |
480 | instance of the variant. | |
481 | ||
482 | The alignment of the type containing the variant is independent of the | |
483 | variant alignment. For instance, if a structure contains two fields, a | |
484 | 32-bit integer, aligned on 32 bits, and a variant, which contains two | |
485 | choices: either a 32-bit field, aligned on 32 bits, or a 64-bit field, | |
486 | aligned on 64 bits, the alignment of the outmost structure will be | |
487 | 32-bit (the alignment of its largest field, disregarding the alignment | |
488 | of the variant). The alignment of the variant will depend on the | |
489 | selector: if the variant's 32-bit field is selected, its alignment will | |
490 | be 32-bit, or 64-bit otherwise. It is important to note that variants | |
491 | are specifically tailored for compactness in a stream. Therefore, the | |
492 | relative offsets of compound type fields can vary depending on | |
493 | the offset at which the compound type starts if it contains a variant | |
494 | that itself contains a type with alignment larger than the largest field | |
495 | contained within the compound type. This is caused by the fact that the | |
496 | compound type may contain the enumeration that select the variant's | |
497 | choice, and therefore the alignment to be applied to the compound type | |
498 | cannot be determined before encountering the enumeration. | |
fcba70d4 | 499 | |
4cac83ee MD |
500 | Each variant type selector possess a field name, which is a unique |
501 | identifier within the variant. The identifier is not allowed to use any | |
502 | reserved keyword (see Section C.1.2). Replacing reserved keywords with | |
70375f92 | 503 | underscore-prefixed field names is recommended. Fields starting with an |
92250c71 MD |
504 | underscore should have their leading underscore removed by the CTF trace |
505 | readers. | |
70375f92 | 506 | |
4cac83ee | 507 | |
fcba70d4 MD |
508 | A named variant declaration followed by its definition within a structure |
509 | declaration: | |
510 | ||
511 | variant name { | |
512 | field_type sel1; | |
513 | field_type sel2; | |
514 | field_type sel3; | |
515 | ... | |
516 | }; | |
517 | ||
518 | struct { | |
a9b83695 | 519 | enum : integer_type { sel1, sel2, sel3, ... } tag_field; |
fcba70d4 MD |
520 | ... |
521 | variant name <tag_field> v; | |
522 | } | |
523 | ||
524 | An unnamed variant definition within a structure is expressed by the following | |
6672e9e1 | 525 | TSDL meta-data: |
fcba70d4 MD |
526 | |
527 | struct { | |
a9b83695 | 528 | enum : integer_type { sel1, sel2, sel3, ... } tag_field; |
fcba70d4 MD |
529 | ... |
530 | variant <tag_field> { | |
531 | field_type sel1; | |
532 | field_type sel2; | |
533 | field_type sel3; | |
534 | ... | |
535 | } v; | |
536 | } | |
537 | ||
538 | Example of a named variant within a sequence that refers to a single tag field: | |
539 | ||
540 | variant example { | |
541 | uint32_t a; | |
542 | uint64_t b; | |
543 | short c; | |
544 | }; | |
545 | ||
546 | struct { | |
a9b83695 | 547 | enum : uint2_t { a, b, c } choice; |
1ab22b2a MD |
548 | unsigned int seqlen; |
549 | variant example <choice> v[seqlen]; | |
fcba70d4 MD |
550 | } |
551 | ||
552 | Example of an unnamed variant: | |
553 | ||
554 | struct { | |
a9b83695 | 555 | enum : uint2_t { a, b, c, d } choice; |
fcba70d4 MD |
556 | /* Unrelated fields can be added between the variant and its tag */ |
557 | int32_t somevalue; | |
558 | variant <choice> { | |
559 | uint32_t a; | |
560 | uint64_t b; | |
561 | short c; | |
562 | struct { | |
563 | unsigned int field1; | |
564 | uint64_t field2; | |
565 | } d; | |
566 | } s; | |
567 | } | |
568 | ||
569 | Example of an unnamed variant within an array: | |
570 | ||
571 | struct { | |
a9b83695 | 572 | enum : uint2_t { a, b, c } choice; |
fcba70d4 MD |
573 | variant <choice> { |
574 | uint32_t a; | |
575 | uint64_t b; | |
576 | short c; | |
15850440 | 577 | } v[10]; |
fcba70d4 MD |
578 | } |
579 | ||
580 | Example of a variant type definition within a structure, where the defined type | |
581 | is then declared within an array of structures. This variant refers to a tag | |
37ab95c3 | 582 | located in an upper static scope. This example clearly shows that a variant |
fcba70d4 | 583 | type definition referring to the tag "x" uses the closest preceding field from |
37ab95c3 | 584 | the static scope of the type definition. |
fcba70d4 MD |
585 | |
586 | struct { | |
a9b83695 | 587 | enum : uint2_t { a, b, c, d } x; |
fcba70d4 MD |
588 | |
589 | typedef variant <x> { /* | |
590 | * "x" refers to the preceding "x" enumeration in the | |
37ab95c3 | 591 | * static scope of the type definition. |
fcba70d4 MD |
592 | */ |
593 | uint32_t a; | |
594 | uint64_t b; | |
595 | short c; | |
596 | } example_variant; | |
597 | ||
598 | struct { | |
a9b83695 | 599 | enum : int { x, y, z } x; /* This enumeration is not used by "v". */ |
fcba70d4 | 600 | example_variant v; /* |
a9b83695 | 601 | * "v" uses the "enum : uint2_t { a, b, c, d }" |
fcba70d4 MD |
602 | * tag. |
603 | */ | |
604 | } a[10]; | |
605 | } | |
606 | ||
607 | 4.2.3 Arrays | |
5ba9f198 | 608 | |
6672e9e1 MD |
609 | Arrays are fixed-length. Their length is declared in the type |
610 | declaration within the meta-data. They contain an array of "inner type" | |
611 | elements, which can refer to any type not containing the type of the | |
612 | array being declared (no circular dependency). The length is the number | |
613 | of elements in an array. | |
5ba9f198 | 614 | |
6672e9e1 | 615 | TSDL meta-data representation of a named array: |
80fd2569 MD |
616 | |
617 | typedef elem_type name[length]; | |
5ba9f198 | 618 | |
2152348f | 619 | A nameless array can be declared as a field type within a structure, e.g.: |
5ba9f198 | 620 | |
2152348f | 621 | uint8_t field_name[10]; |
80fd2569 | 622 | |
ec4404a7 | 623 | Arrays are always aligned on their element alignment requirement. |
5ba9f198 | 624 | |
fcba70d4 | 625 | 4.2.4 Sequences |
5ba9f198 | 626 | |
f3afd1c7 | 627 | Sequences are dynamically-sized arrays. They refer to a "length" |
37ab95c3 | 628 | unsigned integer field, which must appear in either the same static scope, |
1ab22b2a | 629 | prior to the sequence field (in field declaration order), in an upper |
37ab95c3 MD |
630 | static scope, or in an upper dynamic scope (see Section 7.3.2). This |
631 | length field represents the number of elements in the sequence. The | |
632 | sequence per se is an array of "inner type" elements. | |
5ba9f198 | 633 | |
1ab22b2a | 634 | TSDL meta-data representation for a sequence type definition: |
80fd2569 | 635 | |
1ab22b2a MD |
636 | struct { |
637 | unsigned int length_field; | |
638 | typedef elem_type typename[length_field]; | |
639 | typename seq_field_name; | |
640 | } | |
641 | ||
642 | A sequence can also be declared as a field type, e.g.: | |
80fd2569 | 643 | |
1ab22b2a MD |
644 | struct { |
645 | unsigned int length_field; | |
646 | long seq_field_name[length_field]; | |
647 | } | |
80fd2569 | 648 | |
1ab22b2a MD |
649 | Multiple sequences can refer to the same length field, and these length |
650 | fields can be in a different upper dynamic scope: | |
651 | ||
652 | e.g., assuming the stream.event.header defines: | |
653 | ||
654 | stream { | |
655 | ... | |
656 | id = 1; | |
657 | event.header := struct { | |
658 | uint16_t seq_len; | |
659 | }; | |
660 | }; | |
661 | ||
662 | event { | |
663 | ... | |
664 | stream_id = 1; | |
665 | fields := struct { | |
666 | long seq_a[stream.event.header.seq_len]; | |
667 | char seq_b[stream.event.header.seq_len]; | |
668 | }; | |
669 | }; | |
80fd2569 | 670 | |
1ab22b2a | 671 | The sequence elements follow the "array" specifications. |
5ba9f198 | 672 | |
fcba70d4 | 673 | 4.2.5 Strings |
5ba9f198 MD |
674 | |
675 | Strings are an array of bytes of variable size and are terminated by a '\0' | |
6672e9e1 MD |
676 | "NULL" character. Their encoding is described in the TSDL meta-data. In |
677 | absence of encoding attribute information, the default encoding is | |
678 | UTF-8. | |
5ba9f198 | 679 | |
6672e9e1 | 680 | TSDL meta-data representation of a named string type: |
80fd2569 | 681 | |
359894ac | 682 | typealias string { |
5ba9f198 | 683 | encoding = UTF8 OR ASCII; |
38b8da21 | 684 | } := name; |
5ba9f198 | 685 | |
80fd2569 MD |
686 | A nameless string type can be declared as a field type: |
687 | ||
688 | string field_name; /* Use default UTF8 encoding */ | |
5ba9f198 | 689 | |
ec4404a7 MD |
690 | Strings are always aligned on byte size. |
691 | ||
3bf79539 MD |
692 | 5. Event Packet Header |
693 | ||
6672e9e1 MD |
694 | The event packet header consists of two parts: the "event packet header" |
695 | is the same for all streams of a trace. The second part, the "event | |
696 | packet context", is described on a per-stream basis. Both are described | |
96bf5a3b | 697 | in the TSDL meta-data. |
3bf79539 | 698 | |
6672e9e1 | 699 | Event packet header (all fields are optional, specified by TSDL meta-data): |
3bf79539 | 700 | |
6672e9e1 MD |
701 | - Magic number (CTF magic number: 0xC1FC1FC1) specifies that this is a |
702 | CTF packet. This magic number is optional, but when present, it should | |
703 | come at the very beginning of the packet. | |
704 | - Trace UUID, used to ensure the event packet match the meta-data used. | |
705 | (note: we cannot use a meta-data checksum in every cases instead of a | |
706 | UUID because meta-data can be appended to while tracing is active) | |
707 | This field is optional. | |
708 | - Stream ID, used as reference to stream description in meta-data. | |
709 | This field is optional if there is only one stream description in the | |
710 | meta-data, but becomes required if there are more than one stream in | |
711 | the TSDL meta-data description. | |
3bf79539 | 712 | |
6672e9e1 | 713 | Event packet context (all fields are optional, specified by TSDL meta-data): |
3bf79539 | 714 | |
4dfca05b MD |
715 | - Event packet content size (in bits). |
716 | - Event packet size (in bits, includes padding). | |
cda89682 | 717 | - Event packet content checksum. Checksum excludes the event packet |
3bf79539 MD |
718 | header. |
719 | - Per-stream event packet sequence count (to deal with UDP packet loss). The | |
720 | number of significant sequence counter bits should also be present, so | |
b11853af | 721 | wrap-arounds are dealt with correctly. |
6672e9e1 | 722 | - Time-stamp at the beginning and time-stamp at the end of the event packet. |
3bf79539 MD |
723 | Both timestamps are written in the packet header, but sampled respectively |
724 | while (or before) writing the first event and while (or after) writing the | |
725 | last event in the packet. The inclusive range between these timestamps should | |
726 | include all event timestamps assigned to events contained within the packet. | |
3f584fe3 MD |
727 | The timestamp at the beginning of an event packet is guaranteed to be |
728 | below or equal the timestamp at the end of that event packet. | |
729 | The timestamp at the end of an event packet is guaranteed to be below | |
730 | or equal the timestamps at the end of any following packet within the | |
731 | same stream. See Section 8. Clocks for more detail. | |
5ba9f198 | 732 | - Events discarded count |
3bf79539 | 733 | - Snapshot of a per-stream free-running counter, counting the number of |
8f34dd4f MD |
734 | events discarded that were supposed to be written in the stream after |
735 | the last event in the event packet. | |
736 | * Note: producer-consumer buffer full condition can fill the current | |
3bf79539 | 737 | event packet with padding so we know exactly where events have been |
8f34dd4f MD |
738 | discarded. However, if the buffer full condition chooses not |
739 | to fill the current event packet with padding, all we know | |
740 | about the timestamp range in which the events have been | |
741 | discarded is that it is somewhere between the beginning and | |
742 | the end of the packet. | |
3bf79539 MD |
743 | - Lossless compression scheme used for the event packet content. Applied |
744 | directly to raw data. New types of compression can be added in following | |
745 | versions of the format. | |
5ba9f198 MD |
746 | 0: no compression scheme |
747 | 1: bzip2 | |
748 | 2: gzip | |
3bf79539 MD |
749 | 3: xz |
750 | - Cypher used for the event packet content. Applied after compression. | |
5ba9f198 MD |
751 | 0: no encryption |
752 | 1: AES | |
3bf79539 | 753 | - Checksum scheme used for the event packet content. Applied after encryption. |
5ba9f198 MD |
754 | 0: no checksum |
755 | 1: md5 | |
756 | 2: sha1 | |
757 | 3: crc32 | |
758 | ||
6672e9e1 | 759 | 5.1 Event Packet Header Description |
3bf79539 | 760 | |
fc5425db | 761 | The event packet header layout is indicated by the trace packet.header |
6672e9e1 MD |
762 | field. Here is a recommended structure type for the packet header with |
763 | the fields typically expected (although these fields are each optional): | |
fc5425db | 764 | |
80fd2569 MD |
765 | struct event_packet_header { |
766 | uint32_t magic; | |
3fde5da1 | 767 | uint8_t uuid[16]; |
3bf79539 | 768 | uint32_t stream_id; |
80fd2569 | 769 | }; |
5ba9f198 | 770 | |
fc5425db MD |
771 | trace { |
772 | ... | |
773 | packet.header := struct event_packet_header; | |
774 | }; | |
775 | ||
6672e9e1 MD |
776 | If the magic number is not present, tools such as "file" will have no |
777 | mean to discover the file type. | |
778 | ||
3fde5da1 MD |
779 | If the uuid is not present, no validation that the meta-data actually |
780 | corresponds to the stream is performed. | |
37b39e47 MD |
781 | |
782 | If the stream_id packet header field is missing, the trace can only | |
783 | contain a single stream. Its "id" field can be left out, and its events | |
784 | don't need to declare a "stream_id" field. | |
785 | ||
786 | ||
3bf79539 MD |
787 | 5.2 Event Packet Context Description |
788 | ||
789 | Event packet context example. These are declared within the stream declaration | |
6672e9e1 | 790 | in the meta-data. All these fields are optional. If the packet size field is |
6a7c61df | 791 | missing, the whole stream only contains a single packet. If the content |
f6b81726 MD |
792 | size field is missing, the packet is filled (no padding). The content |
793 | and packet sizes include all headers. | |
3bf79539 MD |
794 | |
795 | An example event packet context type: | |
796 | ||
80fd2569 | 797 | struct event_packet_context { |
3bf79539 MD |
798 | uint64_t timestamp_begin; |
799 | uint64_t timestamp_end; | |
800 | uint32_t checksum; | |
801 | uint32_t stream_packet_count; | |
802 | uint32_t events_discarded; | |
803 | uint32_t cpu_id; | |
47911ef2 MD |
804 | uint64_t/uint32_t/uint16_t content_size; |
805 | uint64_t/uint32_t/uint16_t packet_size; | |
3bf79539 MD |
806 | uint8_t compression_scheme; |
807 | uint8_t encryption_scheme; | |
3b0f8e4d | 808 | uint8_t checksum_scheme; |
3bf79539 | 809 | }; |
5ba9f198 | 810 | |
fcba70d4 | 811 | |
5ba9f198 MD |
812 | 6. Event Structure |
813 | ||
814 | The overall structure of an event is: | |
815 | ||
0b174d87 MD |
816 | 1 - Event Header (as specified by the stream meta-data) |
817 | 2 - Stream Event Context (as specified by the stream meta-data) | |
818 | 3 - Event Context (as specified by the event meta-data) | |
819 | 4 - Event Payload (as specified by the event meta-data) | |
5ba9f198 | 820 | |
fdf2bb05 | 821 | This structure defines an implicit dynamic scoping, where variants |
7d9d7e92 MD |
822 | located in inner structures (those with a higher number in the listing |
823 | above) can refer to the fields of outer structures (with lower number in | |
6c7226e9 | 824 | the listing above). See Section 7.3 TSDL Scopes for more detail. |
5ba9f198 | 825 | |
7409082d MD |
826 | The total length of an event is defined as the difference between the |
827 | end of its Event Payload and the end of the previous event's Event | |
828 | Payload. Therefore, it includes the event header alignment padding, and | |
829 | all its fields and their respective alignment padding. Events of length | |
830 | 0 are forbidden. | |
831 | ||
fdf2bb05 | 832 | 6.1 Event Header |
fcba70d4 | 833 | |
6672e9e1 | 834 | Event headers can be described within the meta-data. We hereby propose, as an |
fcba70d4 MD |
835 | example, two types of events headers. Type 1 accommodates streams with less than |
836 | 31 event IDs. Type 2 accommodates streams with 31 or more event IDs. | |
5ba9f198 | 837 | |
3bf79539 MD |
838 | One major factor can vary between streams: the number of event IDs assigned to |
839 | a stream. Luckily, this information tends to stay relatively constant (modulo | |
5ba9f198 | 840 | event registration while trace is being recorded), so we can specify different |
3bf79539 | 841 | representations for streams containing few event IDs and streams containing |
6672e9e1 MD |
842 | many event IDs, so we end up representing the event ID and time-stamp as |
843 | densely as possible in each case. | |
5ba9f198 | 844 | |
fcba70d4 MD |
845 | The header is extended in the rare occasions where the information cannot be |
846 | represented in the ranges available in the standard event header. They are also | |
3bf79539 MD |
847 | used in the rare occasions where the data required for a field could not be |
848 | collected: the flag corresponding to the missing field within the missing_fields | |
849 | array is then set to 1. | |
5ba9f198 | 850 | |
6672e9e1 MD |
851 | Types uintX_t represent an X-bit unsigned integer, as declared with |
852 | either: | |
5ba9f198 | 853 | |
82cbdb16 | 854 | typealias integer { size = X; align = X; signed = false; } := uintX_t; |
6672e9e1 MD |
855 | |
856 | or | |
857 | ||
82cbdb16 | 858 | typealias integer { size = X; align = 1; signed = false; } := uintX_t; |
5ba9f198 | 859 | |
2a0bae9e MD |
860 | For more information about timestamp fields, see Section 8. Clocks. |
861 | ||
fdf2bb05 | 862 | 6.1.1 Type 1 - Few event IDs |
5ba9f198 MD |
863 | |
864 | - Aligned on 32-bit (or 8-bit if byte-packed, depending on the architecture | |
865 | preference). | |
5ba9f198 | 866 | - Native architecture byte ordering. |
fcba70d4 MD |
867 | - For "compact" selection |
868 | - Fixed size: 32 bits. | |
869 | - For "extended" selection | |
870 | - Size depends on the architecture and variant alignment. | |
5ba9f198 | 871 | |
80fd2569 | 872 | struct event_header_1 { |
fcba70d4 MD |
873 | /* |
874 | * id: range: 0 - 30. | |
875 | * id 31 is reserved to indicate an extended header. | |
876 | */ | |
a9b83695 | 877 | enum : uint5_t { compact = 0 ... 30, extended = 31 } id; |
fcba70d4 MD |
878 | variant <id> { |
879 | struct { | |
880 | uint27_t timestamp; | |
881 | } compact; | |
882 | struct { | |
883 | uint32_t id; /* 32-bit event IDs */ | |
884 | uint64_t timestamp; /* 64-bit timestamps */ | |
885 | } extended; | |
886 | } v; | |
cb108fea | 887 | } align(32); /* or align(8) */ |
5ba9f198 | 888 | |
5ba9f198 | 889 | |
fdf2bb05 | 890 | 6.1.2 Type 2 - Many event IDs |
5ba9f198 | 891 | |
fcba70d4 | 892 | - Aligned on 16-bit (or 8-bit if byte-packed, depending on the architecture |
5ba9f198 | 893 | preference). |
5ba9f198 | 894 | - Native architecture byte ordering. |
fcba70d4 MD |
895 | - For "compact" selection |
896 | - Size depends on the architecture and variant alignment. | |
897 | - For "extended" selection | |
898 | - Size depends on the architecture and variant alignment. | |
5ba9f198 | 899 | |
80fd2569 | 900 | struct event_header_2 { |
fcba70d4 MD |
901 | /* |
902 | * id: range: 0 - 65534. | |
903 | * id 65535 is reserved to indicate an extended header. | |
904 | */ | |
a9b83695 | 905 | enum : uint16_t { compact = 0 ... 65534, extended = 65535 } id; |
fcba70d4 MD |
906 | variant <id> { |
907 | struct { | |
908 | uint32_t timestamp; | |
909 | } compact; | |
910 | struct { | |
911 | uint32_t id; /* 32-bit event IDs */ | |
912 | uint64_t timestamp; /* 64-bit timestamps */ | |
913 | } extended; | |
914 | } v; | |
cb108fea | 915 | } align(16); /* or align(8) */ |
5ba9f198 | 916 | |
5ba9f198 | 917 | |
0b174d87 | 918 | 6.2 Stream Event Context and Event Context |
5ba9f198 | 919 | |
6672e9e1 MD |
920 | The event context contains information relative to the current event. |
921 | The choice and meaning of this information is specified by the TSDL | |
922 | stream and event meta-data descriptions. The stream context is applied | |
923 | to all events within the stream. The stream context structure follows | |
924 | the event header. The event context is applied to specific events. Its | |
925 | structure follows the stream context structure. | |
5ba9f198 | 926 | |
fcba70d4 MD |
927 | An example of stream-level event context is to save the event payload size with |
928 | each event, or to save the current PID with each event. These are declared | |
6672e9e1 | 929 | within the stream declaration within the meta-data: |
5ba9f198 | 930 | |
fcba70d4 MD |
931 | stream { |
932 | ... | |
6672e9e1 | 933 | event.context := struct { |
80fd2569 MD |
934 | uint pid; |
935 | uint16_t payload_size; | |
6672e9e1 | 936 | }; |
fcba70d4 MD |
937 | }; |
938 | ||
939 | An example of event-specific event context is to declare a bitmap of missing | |
940 | fields, only appended after the stream event context if the extended event | |
941 | header is selected. NR_FIELDS is the number of fields within the event (a | |
942 | numeric value). | |
5ba9f198 | 943 | |
fcba70d4 MD |
944 | event { |
945 | context = struct { | |
946 | variant <id> { | |
947 | struct { } compact; | |
948 | struct { | |
949 | uint1_t missing_fields[NR_FIELDS]; /* missing event fields bitmap */ | |
950 | } extended; | |
951 | } v; | |
952 | }; | |
953 | ... | |
954 | } | |
5ba9f198 MD |
955 | |
956 | 6.3 Event Payload | |
957 | ||
958 | An event payload contains fields specific to a given event type. The fields | |
6672e9e1 | 959 | belonging to an event type are described in the event-specific meta-data |
5ba9f198 MD |
960 | within a structure type. |
961 | ||
962 | 6.3.1 Padding | |
963 | ||
964 | No padding at the end of the event payload. This differs from the ISO/C standard | |
965 | for structures, but follows the CTF standard for structures. In a trace, even | |
966 | though it makes sense to align the beginning of a structure, it really makes no | |
967 | sense to add padding at the end of the structure, because structures are usually | |
968 | not followed by a structure of the same type. | |
969 | ||
970 | This trick can be done by adding a zero-length "end" field at the end of the C | |
971 | structures, and by using the offset of this field rather than using sizeof() | |
3bf79539 | 972 | when calculating the size of a structure (see Appendix "A. Helper macros"). |
5ba9f198 MD |
973 | |
974 | 6.3.2 Alignment | |
975 | ||
976 | The event payload is aligned on the largest alignment required by types | |
977 | contained within the payload. (This follows the ISO/C standard for structures) | |
978 | ||
979 | ||
6c7226e9 MD |
980 | 7. Trace Stream Description Language (TSDL) |
981 | ||
982 | The Trace Stream Description Language (TSDL) allows expression of the | |
983 | binary trace streams layout in a C99-like Domain Specific Language | |
984 | (DSL). | |
985 | ||
986 | ||
6672e9e1 | 987 | 7.1 Meta-data |
6c7226e9 MD |
988 | |
989 | The trace stream layout description is located in the trace meta-data. | |
990 | The meta-data is itself located in a stream identified by its name: | |
991 | "metadata". | |
5ba9f198 | 992 | |
6672e9e1 MD |
993 | The meta-data description can be expressed in two different formats: |
994 | text-only and packet-based. The text-only description facilitates | |
995 | generation of meta-data and provides a convenient way to enter the | |
996 | meta-data information by hand. The packet-based meta-data provides the | |
997 | CTF stream packet facilities (checksumming, compression, encryption, | |
998 | network-readiness) for meta-data stream generated and transported by a | |
999 | tracer. | |
1000 | ||
1b4d35eb MD |
1001 | The text-only meta-data file is a plain-text TSDL description. This file |
1002 | must begin with the following characters to identify the file as a CTF | |
9486a18c | 1003 | TSDL text-based metadata file (without the double-quotes) : |
1b4d35eb | 1004 | |
ec2b4db8 | 1005 | "/* CTF" |
1b4d35eb | 1006 | |
ec2b4db8 MD |
1007 | It must be followed by a space, and the version of the specification |
1008 | followed by the CTF trace, e.g.: | |
1009 | ||
1010 | " 1.8" | |
1011 | ||
1012 | These characters allow automated discovery of file type and CTF | |
1013 | specification version. They are interpreted as a the beginning of a | |
1014 | comment by the TSDL metadata parser. The comment can be continued to | |
1015 | contain extra commented characters before it is closed. | |
6672e9e1 MD |
1016 | |
1017 | The packet-based meta-data is made of "meta-data packets", which each | |
1018 | start with a meta-data packet header. The packet-based meta-data | |
1019 | description is detected by reading the magic number "0x75D11D57" at the | |
1020 | beginning of the file. This magic number is also used to detect the | |
1021 | endianness of the architecture by trying to read the CTF magic number | |
1022 | and its counterpart in reversed endianness. The events within the | |
1023 | meta-data stream have no event header nor event context. Each event only | |
46400bde MD |
1024 | contains a special "sequence" payload, which is a sequence of bits which |
1025 | length is implicitly calculated by using the | |
1026 | "trace.packet.header.content_size" field, minus the packet header size. | |
1027 | The formatting of this sequence of bits is a plain-text representation | |
1028 | of the TSDL description. Each meta-data packet start with a special | |
1029 | packet header, specific to the meta-data stream, which contains, | |
1030 | exactly: | |
6672e9e1 MD |
1031 | |
1032 | struct metadata_packet_header { | |
2daeaa3a | 1033 | uint32_t magic; /* 0x75D11D57 */ |
3fde5da1 | 1034 | uint8_t uuid[16]; /* Unique Universal Identifier */ |
6672e9e1 MD |
1035 | uint32_t checksum; /* 0 if unused */ |
1036 | uint32_t content_size; /* in bits */ | |
1037 | uint32_t packet_size; /* in bits */ | |
1038 | uint8_t compression_scheme; /* 0 if unused */ | |
1039 | uint8_t encryption_scheme; /* 0 if unused */ | |
1040 | uint8_t checksum_scheme; /* 0 if unused */ | |
ec2b4db8 MD |
1041 | uint8_t major; /* CTF spec version major number */ |
1042 | uint8_t minor; /* CTF spec version minor number */ | |
6672e9e1 MD |
1043 | }; |
1044 | ||
1045 | The packet-based meta-data can be converted to a text-only meta-data by | |
3568031f | 1046 | concatenating all the strings it contains. |
4fafe1ad | 1047 | |
6672e9e1 MD |
1048 | In the textual representation of the meta-data, the text contained |
1049 | within "/*" and "*/", as well as within "//" and end of line, are | |
1050 | treated as comments. Boolean values can be represented as true, TRUE, | |
1051 | or 1 for true, and false, FALSE, or 0 for false. Within the string-based | |
1052 | meta-data description, the trace UUID is represented as a string of | |
1053 | hexadecimal digits and dashes "-". In the event packet header, the trace | |
1054 | UUID is represented as an array of bytes. | |
fcba70d4 | 1055 | |
fdf2bb05 | 1056 | |
6c7226e9 | 1057 | 7.2 Declaration vs Definition |
fdf2bb05 MD |
1058 | |
1059 | A declaration associates a layout to a type, without specifying where | |
1060 | this type is located in the event structure hierarchy (see Section 6). | |
1061 | This therefore includes typedef, typealias, as well as all type | |
1062 | specifiers. In certain circumstances (typedef, structure field and | |
1063 | variant field), a declaration is followed by a declarator, which specify | |
1064 | the newly defined type name (for typedef), or the field name (for | |
1065 | declarations located within structure and variants). Array and sequence, | |
1066 | declared with square brackets ("[" "]"), are part of the declarator, | |
a9b83695 | 1067 | similarly to C99. The enumeration base type is specified by |
6c7226e9 | 1068 | ": enum_base", which is part of the type specifier. The variant tag |
a9b83695 | 1069 | name, specified between "<" ">", is also part of the type specifier. |
fdf2bb05 MD |
1070 | |
1071 | A definition associates a type to a location in the event structure | |
b9606a77 MD |
1072 | hierarchy (see Section 6). This association is denoted by ":=", as shown |
1073 | in Section 7.3. | |
fdf2bb05 MD |
1074 | |
1075 | ||
6c7226e9 | 1076 | 7.3 TSDL Scopes |
fdf2bb05 | 1077 | |
37ab95c3 MD |
1078 | TSDL uses three different types of scoping: a lexical scope is used for |
1079 | declarations and type definitions, and static and dynamic scopes are | |
1080 | used for variants references to tag fields (with relative and absolute | |
1081 | path lookups) and for sequence references to length fields. | |
fdf2bb05 | 1082 | |
6c7226e9 | 1083 | 7.3.1 Lexical Scope |
fdf2bb05 | 1084 | |
570ecabe MD |
1085 | Each of "trace", "env", "stream", "event", "struct" and "variant" have |
1086 | their own nestable declaration scope, within which types can be declared | |
1087 | using "typedef" and "typealias". A root declaration scope also contains | |
1088 | all declarations located outside of any of the aforementioned | |
1089 | declarations. An inner declaration scope can refer to type declared | |
1090 | within its container lexical scope prior to the inner declaration scope. | |
1091 | Redefinition of a typedef or typealias is not valid, although hiding an | |
1092 | upper scope typedef or typealias is allowed within a sub-scope. | |
fdf2bb05 | 1093 | |
37ab95c3 | 1094 | 7.3.2 Static and Dynamic Scopes |
fdf2bb05 | 1095 | |
37ab95c3 MD |
1096 | A local static scope consists in the scope generated by the declaration |
1097 | of fields within a compound type. A static scope is a local static scope | |
1098 | augmented with the nested sub-static-scopes it contains. | |
1099 | ||
1100 | A dynamic scope consists in the static scope augmented with the | |
7d9d7e92 | 1101 | implicit event structure definition hierarchy presented at Section 6. |
fdf2bb05 | 1102 | |
37ab95c3 MD |
1103 | Multiple declarations of the same field name within a local static scope |
1104 | is not valid. It is however valid to re-use the same field name in | |
1105 | different local scopes. | |
1106 | ||
1107 | Nested static and dynamic scopes form lookup paths. These are used for | |
1108 | variant tag and sequence length references. They are used at the variant | |
1109 | and sequence definition site to look up the location of the tag field | |
1110 | associated with a variant, and to lookup up the location of the length | |
1111 | field associated with a sequence. | |
1112 | ||
1113 | Variants and sequences can refer to a tag field either using a relative | |
4cac83ee MD |
1114 | path or an absolute path. The relative path is relative to the scope in |
1115 | which the variant or sequence performing the lookup is located. | |
37ab95c3 MD |
1116 | Relative paths are only allowed to lookup within the same static scope, |
1117 | which includes its nested static scopes. Lookups targeting parent static | |
1118 | scopes need to be performed with an absolute path. | |
1119 | ||
1120 | Absolute path lookups use the full path including the dynamic scope | |
1121 | followed by a "." and then the static scope. Therefore, variants (or | |
1122 | sequences) in lower levels in the dynamic scope (e.g. event context) can | |
1123 | refer to a tag (or length) field located in upper levels (e.g. in the | |
1124 | event header) by specifying, in this case, the associated tag with | |
1125 | <stream.event.header.field_name>. This allows, for instance, the event | |
1126 | context to define a variant referring to the "id" field of the event | |
1127 | header as selector. | |
1128 | ||
284724ae | 1129 | The dynamic scope prefixes are thus: |
fdf2bb05 | 1130 | |
570ecabe | 1131 | - Trace Environment: <env. >, |
e0d9e2c7 | 1132 | - Trace Packet Header: <trace.packet.header. >, |
7d9d7e92 MD |
1133 | - Stream Packet Context: <stream.packet.context. >, |
1134 | - Event Header: <stream.event.header. >, | |
1135 | - Stream Event Context: <stream.event.context. >, | |
1136 | - Event Context: <event.context. >, | |
1137 | - Event Payload: <event.fields. >. | |
fdf2bb05 | 1138 | |
37ab95c3 MD |
1139 | |
1140 | The target dynamic scope must be specified explicitly when referring to | |
4cac83ee MD |
1141 | a field outside of the static scope (absolute scope reference). No |
1142 | conflict can occur between relative and dynamic paths, because the | |
1143 | keywords "trace", "stream", and "event" are reserved, and thus | |
1144 | not permitted as field names. It is recommended that field names | |
1145 | clashing with CTF and C99 reserved keywords use an underscore prefix to | |
1146 | eliminate the risk of generating a description containing an invalid | |
70375f92 | 1147 | field name. Consequently, fields starting with an underscore should have |
92250c71 | 1148 | their leading underscore removed by the CTF trace readers. |
70375f92 | 1149 | |
fdf2bb05 | 1150 | |
457d8b0a MD |
1151 | The information available in the dynamic scopes can be thought of as the |
1152 | current tracing context. At trace production, information about the | |
1153 | current context is saved into the specified scope field levels. At trace | |
1154 | consumption, for each event, the current trace context is therefore | |
1155 | readable by accessing the upper dynamic scopes. | |
1156 | ||
fdf2bb05 | 1157 | |
6c7226e9 | 1158 | 7.4 TSDL Examples |
d285084f | 1159 | |
6672e9e1 | 1160 | The grammar representing the TSDL meta-data is presented in Appendix C. |
7df6b93a | 1161 | TSDL Grammar. This section presents a rather lighter reading that |
6672e9e1 | 1162 | consists in examples of TSDL meta-data, with template values. |
969f30c0 | 1163 | |
350bb633 MD |
1164 | The stream "id" can be left out if there is only one stream in the |
1165 | trace. The event "id" field can be left out if there is only one event | |
1166 | in a stream. | |
1167 | ||
5ba9f198 | 1168 | trace { |
1e9de2d1 MD |
1169 | major = value; /* CTF spec version major number */ |
1170 | minor = value; /* CTF spec version minor number */ | |
fdf2bb05 | 1171 | uuid = "aaaaaaaa-aaaa-aaaa-aaaa-aaaaaaaaaaaa"; /* Trace UUID */ |
58997e9e | 1172 | byte_order = be OR le; /* Endianness (required) */ |
fc5425db MD |
1173 | packet.header := struct { |
1174 | uint32_t magic; | |
3fde5da1 | 1175 | uint8_t uuid[16]; |
fc5425db MD |
1176 | uint32_t stream_id; |
1177 | }; | |
3bf79539 | 1178 | }; |
5ba9f198 | 1179 | |
570ecabe MD |
1180 | /* |
1181 | * The "env" (environment) scope contains assignment expressions. The | |
1182 | * field names and content are implementation-defined. | |
1183 | */ | |
1184 | env { | |
1185 | pid = value; /* example */ | |
1186 | proc_name = "name"; /* example */ | |
1187 | ... | |
1188 | }; | |
1189 | ||
3bf79539 MD |
1190 | stream { |
1191 | id = stream_id; | |
fdf2bb05 | 1192 | /* Type 1 - Few event IDs; Type 2 - Many event IDs. See section 6.1. */ |
4fa992a5 MD |
1193 | event.header := event_header_1 OR event_header_2; |
1194 | event.context := struct { | |
77a98c82 | 1195 | ... |
3bf79539 | 1196 | }; |
4fa992a5 | 1197 | packet.context := struct { |
77a98c82 | 1198 | ... |
3bf79539 MD |
1199 | }; |
1200 | }; | |
5ba9f198 MD |
1201 | |
1202 | event { | |
980015f9 | 1203 | name = "event_name"; |
3bf79539 | 1204 | id = value; /* Numeric identifier within the stream */ |
67f02e24 | 1205 | stream_id = stream_id; |
dc56f167 | 1206 | loglevel = value; |
8e9060f2 | 1207 | model.emf.uri = "string"; |
4fa992a5 | 1208 | context := struct { |
fcba70d4 MD |
1209 | ... |
1210 | }; | |
4fa992a5 | 1211 | fields := struct { |
80fd2569 MD |
1212 | ... |
1213 | }; | |
3bf79539 | 1214 | }; |
5ba9f198 | 1215 | |
555f54e6 MD |
1216 | callsite { |
1217 | name = "event_name"; | |
1218 | func = "func_name"; | |
1219 | file = "myfile.c"; | |
1220 | line = 39; | |
5a6b4ee1 | 1221 | ip = 0x40096c; |
555f54e6 MD |
1222 | }; |
1223 | ||
5ba9f198 MD |
1224 | /* More detail on types in section 4. Types */ |
1225 | ||
3d13ef1a MD |
1226 | /* |
1227 | * Named types: | |
1228 | * | |
4fa992a5 | 1229 | * Type declarations behave similarly to the C standard. |
3d13ef1a MD |
1230 | */ |
1231 | ||
80af8ac6 | 1232 | typedef aliased_type_specifiers new_type_declarators; |
2152348f | 1233 | |
3d13ef1a | 1234 | /* e.g.: typedef struct example new_type_name[10]; */ |
80fd2569 | 1235 | |
4fa992a5 MD |
1236 | /* |
1237 | * typealias | |
1238 | * | |
1239 | * The "typealias" declaration can be used to give a name (including | |
80af8ac6 MD |
1240 | * pointer declarator specifier) to a type. It should also be used to |
1241 | * map basic C types (float, int, unsigned long, ...) to a CTF type. | |
1242 | * Typealias is a superset of "typedef": it also allows assignment of a | |
38b8da21 | 1243 | * simple variable identifier to a type. |
4fa992a5 MD |
1244 | */ |
1245 | ||
1246 | typealias type_class { | |
80fd2569 | 1247 | ... |
38b8da21 | 1248 | } := type_specifiers type_declarator; |
2152348f | 1249 | |
3d13ef1a MD |
1250 | /* |
1251 | * e.g.: | |
4fa992a5 | 1252 | * typealias integer { |
3d13ef1a MD |
1253 | * size = 32; |
1254 | * align = 32; | |
1255 | * signed = false; | |
38b8da21 | 1256 | * } := struct page *; |
359894ac MD |
1257 | * |
1258 | * typealias integer { | |
1259 | * size = 32; | |
1260 | * align = 32; | |
1261 | * signed = true; | |
38b8da21 | 1262 | * } := int; |
3d13ef1a | 1263 | */ |
80fd2569 MD |
1264 | |
1265 | struct name { | |
3bf79539 MD |
1266 | ... |
1267 | }; | |
5ba9f198 | 1268 | |
fcba70d4 MD |
1269 | variant name { |
1270 | ... | |
1271 | }; | |
1272 | ||
a9b83695 | 1273 | enum name : integer_type { |
3bf79539 MD |
1274 | ... |
1275 | }; | |
1276 | ||
2152348f | 1277 | |
4fa992a5 MD |
1278 | /* |
1279 | * Unnamed types, contained within compound type fields, typedef or typealias. | |
1280 | */ | |
2152348f | 1281 | |
80fd2569 MD |
1282 | struct { |
1283 | ... | |
2152348f | 1284 | } |
5ba9f198 | 1285 | |
ec4404a7 MD |
1286 | struct { |
1287 | ... | |
1288 | } align(value) | |
1289 | ||
fcba70d4 MD |
1290 | variant { |
1291 | ... | |
1292 | } | |
1293 | ||
a9b83695 | 1294 | enum : integer_type { |
80fd2569 | 1295 | ... |
2152348f MD |
1296 | } |
1297 | ||
1298 | typedef type new_type[length]; | |
3bf79539 | 1299 | |
2152348f MD |
1300 | struct { |
1301 | type field_name[length]; | |
1302 | } | |
1303 | ||
1304 | typedef type new_type[length_type]; | |
1305 | ||
1306 | struct { | |
1307 | type field_name[length_type]; | |
1308 | } | |
1309 | ||
1310 | integer { | |
80fd2569 | 1311 | ... |
2152348f | 1312 | } |
3bf79539 | 1313 | |
2152348f | 1314 | floating_point { |
80fd2569 | 1315 | ... |
2152348f MD |
1316 | } |
1317 | ||
1318 | struct { | |
1319 | integer_type field_name:size; /* GNU/C bitfield */ | |
1320 | } | |
1321 | ||
1322 | struct { | |
1323 | string field_name; | |
1324 | } | |
3bf79539 | 1325 | |
fcba70d4 | 1326 | |
2fa70eba MD |
1327 | 8. Clocks |
1328 | ||
1329 | Clock metadata allows to describe the clock topology of the system, as | |
1330 | well as to detail each clock parameter. In absence of clock description, | |
1331 | it is assumed that all fields named "timestamp" use the same clock | |
aed18b5e | 1332 | source, which increments once per nanosecond. |
2fa70eba MD |
1333 | |
1334 | Describing a clock and how it is used by streams is threefold: first, | |
1335 | the clock and clock topology should be described in a "clock" | |
1336 | description block, e.g.: | |
1337 | ||
d803bfcb | 1338 | clock { |
58262d97 | 1339 | name = cycle_counter_sync; |
2fa70eba | 1340 | uuid = "62189bee-96dc-11e0-91a8-cfa3d89f3923"; |
58262d97 | 1341 | description = "Cycle counter synchronized across CPUs"; |
2fa70eba MD |
1342 | freq = 1000000000; /* frequency, in Hz */ |
1343 | /* precision in seconds is: 1000 * (1/freq) */ | |
1344 | precision = 1000; | |
a40cccda MD |
1345 | /* |
1346 | * clock value offset from Epoch is: | |
1347 | * offset_s + (offset * (1/freq)) | |
1348 | */ | |
1349 | offset_s = 1326476837; | |
1350 | offset = 897235420; | |
ce0fadbd | 1351 | absolute = FALSE; |
2fa70eba MD |
1352 | }; |
1353 | ||
d803bfcb MD |
1354 | The mandatory "name" field specifies the name of the clock identifier, |
1355 | which can later be used as a reference. The optional field "uuid" is the | |
1356 | unique identifier of the clock. It can be used to correlate different | |
1357 | traces that use the same clock. An optional textual description string | |
1358 | can be added with the "description" field. The "freq" field is the | |
1359 | initial frequency of the clock, in Hz. If the "freq" field is not | |
1360 | present, the frequency is assumed to be 1000000000 (providing clock | |
1361 | increment of 1 ns). The optional "precision" field details the | |
a40cccda MD |
1362 | uncertainty on the clock measurements, in (1/freq) units. The "offset_s" |
1363 | and "offset" fields indicate the offset from POSIX.1 Epoch, 1970-01-01 | |
1364 | 00:00:00 +0000 (UTC), to the zero of value of the clock. The "offset_s" | |
1365 | field is in seconds. The "offset" field is in (1/freq) units. If any of | |
1366 | the "offset_s" or "offset" field is not present, it is assigned the 0 | |
ce0fadbd MD |
1367 | value. The field "absolute" is TRUE if the clock is a global reference |
1368 | across different clock uuid (e.g. NTP time). Otherwise, "absolute" is | |
1369 | FALSE, and the clock can be considered as synchronized only with other | |
1370 | clocks that have the same uuid. | |
1371 | ||
2fa70eba MD |
1372 | |
1373 | Secondly, a reference to this clock should be added within an integer | |
1374 | type: | |
1375 | ||
1376 | typealias integer { | |
1377 | size = 64; align = 1; signed = false; | |
58262d97 | 1378 | map = clock.cycle_counter_sync.value; |
2fa70eba MD |
1379 | } := uint64_ccnt_t; |
1380 | ||
1381 | Thirdly, stream declarations can reference the clock they use as a | |
1382 | time-stamp source: | |
1383 | ||
1384 | struct packet_context { | |
1385 | uint64_ccnt_t ccnt_begin; | |
1386 | uint64_ccnt_t ccnt_end; | |
1387 | /* ... */ | |
1388 | }; | |
1389 | ||
1390 | stream { | |
1391 | /* ... */ | |
1392 | event.header := struct { | |
1393 | uint64_ccnt_t timestamp; | |
1394 | /* ... */ | |
1395 | } | |
1396 | packet.context := struct packet_context; | |
1397 | }; | |
1398 | ||
1399 | For a N-bit integer type referring to a clock, if the integer overflows | |
2a0bae9e MD |
1400 | compared to the N low order bits of the clock prior value found in the |
1401 | same stream, then it is assumed that one, and only one, overflow | |
1402 | occurred. It is therefore important that events encoding time on a small | |
1403 | number of bits happen frequently enough to detect when more than one | |
1404 | N-bit overflow occurs. | |
2fa70eba MD |
1405 | |
1406 | In a packet context, clock field names ending with "_begin" and "_end" | |
1407 | have a special meaning: this refers to the time-stamps at, respectively, | |
1408 | the beginning and the end of each packet. | |
1409 | ||
1410 | ||
3bf79539 | 1411 | A. Helper macros |
5ba9f198 MD |
1412 | |
1413 | The two following macros keep track of the size of a GNU/C structure without | |
1414 | padding at the end by placing HEADER_END as the last field. A one byte end field | |
1415 | is used for C90 compatibility (C99 flexible arrays could be used here). Note | |
1416 | that this does not affect the effective structure size, which should always be | |
1417 | calculated with the header_sizeof() helper. | |
1418 | ||
1419 | #define HEADER_END char end_field | |
1420 | #define header_sizeof(type) offsetof(typeof(type), end_field) | |
3bf79539 MD |
1421 | |
1422 | ||
1423 | B. Stream Header Rationale | |
1424 | ||
1425 | An event stream is divided in contiguous event packets of variable size. These | |
1426 | subdivisions allow the trace analyzer to perform a fast binary search by time | |
1427 | within the stream (typically requiring to index only the event packet headers) | |
1428 | without reading the whole stream. These subdivisions have a variable size to | |
1429 | eliminate the need to transfer the event packet padding when partially filled | |
1430 | event packets must be sent when streaming a trace for live viewing/analysis. | |
1431 | An event packet can contain a certain amount of padding at the end. Dividing | |
1432 | streams into event packets is also useful for network streaming over UDP and | |
1433 | flight recorder mode tracing (a whole event packet can be swapped out of the | |
1434 | buffer atomically for reading). | |
1435 | ||
1436 | The stream header is repeated at the beginning of each event packet to allow | |
1437 | flexibility in terms of: | |
1438 | ||
1439 | - streaming support, | |
1440 | - allowing arbitrary buffers to be discarded without making the trace | |
1441 | unreadable, | |
1442 | - allow UDP packet loss handling by either dealing with missing event packet | |
1443 | or asking for re-transmission. | |
1444 | - transparently support flight recorder mode, | |
1445 | - transparently support crash dump. | |
1446 | ||
6c7226e9 MD |
1447 | |
1448 | C. TSDL Grammar | |
fcba70d4 | 1449 | |
4fa992a5 | 1450 | /* |
6c7226e9 | 1451 | * Common Trace Format (CTF) Trace Stream Description Language (TSDL) Grammar. |
4fa992a5 MD |
1452 | * |
1453 | * Inspired from the C99 grammar: | |
1454 | * http://www.open-std.org/jtc1/sc22/wg14/www/docs/n1124.pdf (Annex A) | |
6c7226e9 MD |
1455 | * and c++1x grammar (draft) |
1456 | * http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2011/n3291.pdf (Annex A) | |
4fa992a5 MD |
1457 | * |
1458 | * Specialized for CTF needs by including only constant and declarations from | |
1459 | * C99 (excluding function declarations), and by adding support for variants, | |
6c7226e9 MD |
1460 | * sequences and CTF-specific specifiers. Enumeration container types |
1461 | * semantic is inspired from c++1x enum-base. | |
4fa992a5 MD |
1462 | */ |
1463 | ||
1464 | 1) Lexical grammar | |
1465 | ||
1466 | 1.1) Lexical elements | |
1467 | ||
1468 | token: | |
1469 | keyword | |
1470 | identifier | |
1471 | constant | |
1472 | string-literal | |
1473 | punctuator | |
1474 | ||
1475 | 1.2) Keywords | |
1476 | ||
1477 | keyword: is one of | |
1478 | ||
ec4404a7 | 1479 | align |
dbb8b280 | 1480 | callsite |
4fa992a5 MD |
1481 | const |
1482 | char | |
2fa70eba | 1483 | clock |
4fa992a5 MD |
1484 | double |
1485 | enum | |
570ecabe | 1486 | env |
4fa992a5 MD |
1487 | event |
1488 | floating_point | |
1489 | float | |
1490 | integer | |
1491 | int | |
1492 | long | |
1493 | short | |
1494 | signed | |
1495 | stream | |
1496 | string | |
1497 | struct | |
1498 | trace | |
3e1e1a78 | 1499 | typealias |
4fa992a5 MD |
1500 | typedef |
1501 | unsigned | |
1502 | variant | |
1503 | void | |
1504 | _Bool | |
1505 | _Complex | |
1506 | _Imaginary | |
1507 | ||
1508 | ||
1509 | 1.3) Identifiers | |
1510 | ||
1511 | identifier: | |
1512 | identifier-nondigit | |
1513 | identifier identifier-nondigit | |
1514 | identifier digit | |
1515 | ||
1516 | identifier-nondigit: | |
1517 | nondigit | |
1518 | universal-character-name | |
1519 | any other implementation-defined characters | |
1520 | ||
1521 | nondigit: | |
1522 | _ | |
1523 | [a-zA-Z] /* regular expression */ | |
1524 | ||
1525 | digit: | |
1526 | [0-9] /* regular expression */ | |
1527 | ||
1528 | 1.4) Universal character names | |
1529 | ||
1530 | universal-character-name: | |
1531 | \u hex-quad | |
1532 | \U hex-quad hex-quad | |
1533 | ||
1534 | hex-quad: | |
1535 | hexadecimal-digit hexadecimal-digit hexadecimal-digit hexadecimal-digit | |
1536 | ||
1537 | 1.5) Constants | |
1538 | ||
1539 | constant: | |
1540 | integer-constant | |
1541 | enumeration-constant | |
1542 | character-constant | |
1543 | ||
1544 | integer-constant: | |
1545 | decimal-constant integer-suffix-opt | |
1546 | octal-constant integer-suffix-opt | |
1547 | hexadecimal-constant integer-suffix-opt | |
1548 | ||
1549 | decimal-constant: | |
1550 | nonzero-digit | |
1551 | decimal-constant digit | |
1552 | ||
1553 | octal-constant: | |
1554 | 0 | |
1555 | octal-constant octal-digit | |
1556 | ||
1557 | hexadecimal-constant: | |
1558 | hexadecimal-prefix hexadecimal-digit | |
1559 | hexadecimal-constant hexadecimal-digit | |
1560 | ||
1561 | hexadecimal-prefix: | |
1562 | 0x | |
1563 | 0X | |
1564 | ||
1565 | nonzero-digit: | |
1566 | [1-9] | |
1567 | ||
1568 | integer-suffix: | |
1569 | unsigned-suffix long-suffix-opt | |
1570 | unsigned-suffix long-long-suffix | |
1571 | long-suffix unsigned-suffix-opt | |
1572 | long-long-suffix unsigned-suffix-opt | |
1573 | ||
1574 | unsigned-suffix: | |
1575 | u | |
1576 | U | |
1577 | ||
1578 | long-suffix: | |
1579 | l | |
1580 | L | |
1581 | ||
1582 | long-long-suffix: | |
1583 | ll | |
1584 | LL | |
1585 | ||
4fa992a5 MD |
1586 | enumeration-constant: |
1587 | identifier | |
1588 | string-literal | |
1589 | ||
1590 | character-constant: | |
1591 | ' c-char-sequence ' | |
1592 | L' c-char-sequence ' | |
1593 | ||
1594 | c-char-sequence: | |
1595 | c-char | |
1596 | c-char-sequence c-char | |
1597 | ||
1598 | c-char: | |
1599 | any member of source charset except single-quote ('), backslash | |
1600 | (\), or new-line character. | |
1601 | escape-sequence | |
1602 | ||
1603 | escape-sequence: | |
1604 | simple-escape-sequence | |
1605 | octal-escape-sequence | |
1606 | hexadecimal-escape-sequence | |
1607 | universal-character-name | |
1608 | ||
1609 | simple-escape-sequence: one of | |
1610 | \' \" \? \\ \a \b \f \n \r \t \v | |
1611 | ||
1612 | octal-escape-sequence: | |
1613 | \ octal-digit | |
1614 | \ octal-digit octal-digit | |
1615 | \ octal-digit octal-digit octal-digit | |
1616 | ||
1617 | hexadecimal-escape-sequence: | |
1618 | \x hexadecimal-digit | |
1619 | hexadecimal-escape-sequence hexadecimal-digit | |
1620 | ||
1621 | 1.6) String literals | |
1622 | ||
1623 | string-literal: | |
1624 | " s-char-sequence-opt " | |
1625 | L" s-char-sequence-opt " | |
1626 | ||
1627 | s-char-sequence: | |
1628 | s-char | |
1629 | s-char-sequence s-char | |
1630 | ||
1631 | s-char: | |
1632 | any member of source charset except double-quote ("), backslash | |
1633 | (\), or new-line character. | |
1634 | escape-sequence | |
1635 | ||
1636 | 1.7) Punctuators | |
1637 | ||
1638 | punctuator: one of | |
1639 | [ ] ( ) { } . -> * + - < > : ; ... = , | |
1640 | ||
1641 | ||
1642 | 2) Phrase structure grammar | |
1643 | ||
1644 | primary-expression: | |
1645 | identifier | |
1646 | constant | |
1647 | string-literal | |
1648 | ( unary-expression ) | |
1649 | ||
1650 | postfix-expression: | |
1651 | primary-expression | |
1652 | postfix-expression [ unary-expression ] | |
1653 | postfix-expression . identifier | |
1654 | postfix-expressoin -> identifier | |
1655 | ||
1656 | unary-expression: | |
1657 | postfix-expression | |
1658 | unary-operator postfix-expression | |
1659 | ||
1660 | unary-operator: one of | |
1661 | + - | |
1662 | ||
4fa992a5 MD |
1663 | assignment-operator: |
1664 | = | |
1665 | ||
b9606a77 MD |
1666 | type-assignment-operator: |
1667 | := | |
1668 | ||
4fa992a5 | 1669 | constant-expression-range: |
73d61ac3 | 1670 | unary-expression ... unary-expression |
4fa992a5 MD |
1671 | |
1672 | 2.2) Declarations: | |
1673 | ||
1674 | declaration: | |
689e04b4 | 1675 | declaration-specifiers declarator-list-opt ; |
4fa992a5 MD |
1676 | ctf-specifier ; |
1677 | ||
1678 | declaration-specifiers: | |
689e04b4 | 1679 | storage-class-specifier declaration-specifiers-opt |
4fa992a5 MD |
1680 | type-specifier declaration-specifiers-opt |
1681 | type-qualifier declaration-specifiers-opt | |
1682 | ||
1683 | declarator-list: | |
1684 | declarator | |
1685 | declarator-list , declarator | |
1686 | ||
d285084f MD |
1687 | abstract-declarator-list: |
1688 | abstract-declarator | |
1689 | abstract-declarator-list , abstract-declarator | |
1690 | ||
4fa992a5 MD |
1691 | storage-class-specifier: |
1692 | typedef | |
1693 | ||
1694 | type-specifier: | |
1695 | void | |
1696 | char | |
1697 | short | |
1698 | int | |
1699 | long | |
1700 | float | |
1701 | double | |
1702 | signed | |
1703 | unsigned | |
1704 | _Bool | |
1705 | _Complex | |
cfdd51ec | 1706 | _Imaginary |
9dfcfc0f MD |
1707 | struct-specifier |
1708 | variant-specifier | |
4fa992a5 MD |
1709 | enum-specifier |
1710 | typedef-name | |
1711 | ctf-type-specifier | |
1712 | ||
ec4404a7 | 1713 | align-attribute: |
73d61ac3 | 1714 | align ( unary-expression ) |
ec4404a7 | 1715 | |
4fa992a5 | 1716 | struct-specifier: |
ec4404a7 MD |
1717 | struct identifier-opt { struct-or-variant-declaration-list-opt } align-attribute-opt |
1718 | struct identifier align-attribute-opt | |
4fa992a5 MD |
1719 | |
1720 | struct-or-variant-declaration-list: | |
1721 | struct-or-variant-declaration | |
1722 | struct-or-variant-declaration-list struct-or-variant-declaration | |
1723 | ||
1724 | struct-or-variant-declaration: | |
1725 | specifier-qualifier-list struct-or-variant-declarator-list ; | |
eacb16d1 | 1726 | declaration-specifiers-opt storage-class-specifier declaration-specifiers-opt declarator-list ; |
6143bab7 MD |
1727 | typealias declaration-specifiers abstract-declarator-list type-assignment-operator declaration-specifiers abstract-declarator-list ; |
1728 | typealias declaration-specifiers abstract-declarator-list type-assignment-operator declarator-list ; | |
4fa992a5 MD |
1729 | |
1730 | specifier-qualifier-list: | |
1731 | type-specifier specifier-qualifier-list-opt | |
1732 | type-qualifier specifier-qualifier-list-opt | |
1733 | ||
1734 | struct-or-variant-declarator-list: | |
1735 | struct-or-variant-declarator | |
1736 | struct-or-variant-declarator-list , struct-or-variant-declarator | |
1737 | ||
1738 | struct-or-variant-declarator: | |
1739 | declarator | |
73d61ac3 | 1740 | declarator-opt : unary-expression |
4fa992a5 MD |
1741 | |
1742 | variant-specifier: | |
1743 | variant identifier-opt variant-tag-opt { struct-or-variant-declaration-list } | |
1744 | variant identifier variant-tag | |
1745 | ||
1746 | variant-tag: | |
37ab95c3 | 1747 | < unary-expression > |
4fa992a5 MD |
1748 | |
1749 | enum-specifier: | |
1750 | enum identifier-opt { enumerator-list } | |
1751 | enum identifier-opt { enumerator-list , } | |
1752 | enum identifier | |
a9b83695 MD |
1753 | enum identifier-opt : declaration-specifiers { enumerator-list } |
1754 | enum identifier-opt : declaration-specifiers { enumerator-list , } | |
4fa992a5 MD |
1755 | |
1756 | enumerator-list: | |
1757 | enumerator | |
1758 | enumerator-list , enumerator | |
1759 | ||
1760 | enumerator: | |
1761 | enumeration-constant | |
8d2d41f7 MD |
1762 | enumeration-constant assignment-operator unary-expression |
1763 | enumeration-constant assignment-operator constant-expression-range | |
4fa992a5 MD |
1764 | |
1765 | type-qualifier: | |
1766 | const | |
1767 | ||
1768 | declarator: | |
1769 | pointer-opt direct-declarator | |
1770 | ||
1771 | direct-declarator: | |
1772 | identifier | |
1773 | ( declarator ) | |
1ab22b2a | 1774 | direct-declarator [ unary-expression ] |
4fa992a5 | 1775 | |
d285084f MD |
1776 | abstract-declarator: |
1777 | pointer-opt direct-abstract-declarator | |
1778 | ||
1779 | direct-abstract-declarator: | |
1780 | identifier-opt | |
1781 | ( abstract-declarator ) | |
1ab22b2a | 1782 | direct-abstract-declarator [ unary-expression ] |
d285084f MD |
1783 | direct-abstract-declarator [ ] |
1784 | ||
4fa992a5 | 1785 | pointer: |
3b0f8e4d MD |
1786 | * type-qualifier-list-opt |
1787 | * type-qualifier-list-opt pointer | |
4fa992a5 MD |
1788 | |
1789 | type-qualifier-list: | |
1790 | type-qualifier | |
1791 | type-qualifier-list type-qualifier | |
1792 | ||
4fa992a5 MD |
1793 | typedef-name: |
1794 | identifier | |
1795 | ||
1796 | 2.3) CTF-specific declarations | |
1797 | ||
1798 | ctf-specifier: | |
662baf84 | 1799 | clock { ctf-assignment-expression-list-opt } |
4fa992a5 MD |
1800 | event { ctf-assignment-expression-list-opt } |
1801 | stream { ctf-assignment-expression-list-opt } | |
570ecabe | 1802 | env { ctf-assignment-expression-list-opt } |
4fa992a5 | 1803 | trace { ctf-assignment-expression-list-opt } |
555f54e6 | 1804 | callsite { ctf-assignment-expression-list-opt } |
b12919a5 MD |
1805 | typealias declaration-specifiers abstract-declarator-list type-assignment-operator declaration-specifiers abstract-declarator-list |
1806 | typealias declaration-specifiers abstract-declarator-list type-assignment-operator declarator-list | |
4fa992a5 MD |
1807 | |
1808 | ctf-type-specifier: | |
1809 | floating_point { ctf-assignment-expression-list-opt } | |
1810 | integer { ctf-assignment-expression-list-opt } | |
1811 | string { ctf-assignment-expression-list-opt } | |
7609d3c7 | 1812 | string |
4fa992a5 MD |
1813 | |
1814 | ctf-assignment-expression-list: | |
13441bf5 MD |
1815 | ctf-assignment-expression ; |
1816 | ctf-assignment-expression-list ctf-assignment-expression ; | |
4fa992a5 MD |
1817 | |
1818 | ctf-assignment-expression: | |
1819 | unary-expression assignment-operator unary-expression | |
1820 | unary-expression type-assignment-operator type-specifier | |
eacb16d1 | 1821 | declaration-specifiers-opt storage-class-specifier declaration-specifiers-opt declarator-list |
6143bab7 MD |
1822 | typealias declaration-specifiers abstract-declarator-list type-assignment-operator declaration-specifiers abstract-declarator-list |
1823 | typealias declaration-specifiers abstract-declarator-list type-assignment-operator declarator-list |