Update: packet context sizes expressed in bits
[ctf.git] / common-trace-format-specification.txt
CommitLineData
4dfca05b 1Common Trace Format (CTF) Specification (pre-v1.8)
5ba9f198
MD
2
3Mathieu Desnoyers, EfficiOS Inc.
4
339a7dde 5The goal of the present document is to specify a trace format that suits the
cc089c3a 6needs of the embedded, telecom, high-performance and kernel communities. It is
5ba9f198 7based on the Common Trace Format Requirements (v1.4) document. It is designed to
cc089c3a 8allow traces to be natively generated by the Linux kernel, Linux user-space
6c7226e9
MD
9applications written in C/C++, and hardware components. One major element of
10CTF is the Trace Stream Description Language (TSDL) which flexibility
11enables description of various binary trace stream layouts.
cc089c3a
MD
12
13The 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
18A reference implementation of a library to read and write this trace format is
19being implemented within the BabelTrace project, a converter between trace
20formats. 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
25The CE Workgroup of the Linux Foundation, Ericsson, and EfficiOS have
26sponsored this work.
27
5ba9f198 28
beabf088
MD
29Table of Contents
30
311. Preliminary definitions
322. High-level representation of a trace
333. Event stream
344. 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
444.2 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
505. Event Packet Header
51 5.1 Event Packet Header Description
52 5.2 Event Packet Context Description
536. 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
57 6.2 Event Context
58 6.3 Event Payload
59 6.3.1 Padding
60 6.3.2 Alignment
617. 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
66 7.3.2 Dynamic Scope
67 7.4 TSDL Examples
68
69
5ba9f198
MD
701. Preliminary definitions
71
3bf79539
MD
72 - Event Trace: An ordered sequence of events.
73 - Event Stream: An ordered sequence of events, containing a subset of the
74 trace event types.
75 - Event Packet: A sequence of physically contiguous events within an event
76 stream.
5ba9f198
MD
77 - Event: This is the basic entry in a trace. (aka: a trace record).
78 - An event identifier (ID) relates to the class (a type) of event within
3bf79539
MD
79 an event stream.
80 e.g. event: irq_entry.
5ba9f198
MD
81 - An event (or event record) relates to a specific instance of an event
82 class.
3bf79539
MD
83 e.g. event: irq_entry, at time X, on CPU Y
84 - Source Architecture: Architecture writing the trace.
85 - Reader Architecture: Architecture reading the trace.
5ba9f198
MD
86
87
882. High-level representation of a trace
89
3bf79539
MD
90A trace is divided into multiple event streams. Each event stream contains a
91subset of the trace event types.
5ba9f198 92
3bf79539
MD
93The final output of the trace, after its generation and optional transport over
94the network, is expected to be either on permanent or temporary storage in a
95virtual file system. Because each event stream is appended to while a trace is
96being recorded, each is associated with a separate file for output. Therefore,
97a stored trace can be represented as a directory containing one file per stream.
5ba9f198 98
6672e9e1
MD
99Meta-data description associated with the trace contains information on
100trace event types expressed in the Trace Stream Description Language
101(TSDL). This language describes:
5ba9f198
MD
102
103- Trace version.
104- Types available.
6672e9e1 105- Per-trace event header description.
3bf79539 106- Per-stream event header description.
6672e9e1 107- Per-stream event context description.
5ba9f198 108- Per-event
3bf79539 109 - Event type to stream mapping.
5ba9f198
MD
110 - Event type to name mapping.
111 - Event type to ID mapping.
6672e9e1 112 - Event context description.
5ba9f198
MD
113 - Event fields description.
114
115
3bf79539 1163. Event stream
5ba9f198 117
6672e9e1
MD
118An event stream can be divided into contiguous event packets of variable
119size. These subdivisions have a variable size. An event packet can
120contain a certain amount of padding at the end. The stream header is
121repeated at the beginning of each event packet. The rationale for the
122event stream design choices is explained in Appendix B. Stream Header
123Rationale.
5ba9f198 124
3bf79539
MD
125The event stream header will therefore be referred to as the "event packet
126header" throughout the rest of this document.
5ba9f198
MD
127
128
1294. Types
130
1fad7a85
MD
131Types are organized as type classes. Each type class belong to either of two
132kind of types: basic types or compound types.
133
5ba9f198
MD
1344.1 Basic types
135
1fad7a85
MD
136A basic type is a scalar type, as described in this section. It includes
137integers, GNU/C bitfields, enumerations, and floating point values.
5ba9f198
MD
138
1394.1.1 Type inheritance
140
80fd2569
MD
141Type specifications can be inherited to allow deriving types from a
142type class. For example, see the uint32_t named type derived from the "integer"
143type class below ("Integers" section). Types have a precise binary
144representation in the trace. A type class has methods to read and write these
145types, but must be derived into a type to be usable in an event field.
5ba9f198
MD
146
1474.1.2 Alignment
148
149We define "byte-packed" types as aligned on the byte size, namely 8-bit.
150We define "bit-packed" types as following on the next bit, as defined by the
370eae99 151"Integers" section.
5ba9f198 152
6672e9e1 153Each basic type must specify its alignment, in bits. Examples of
ec4404a7
MD
154possible alignments are: bit-packed (align = 1), byte-packed (align =
1558), or word-aligned (e.g. align = 32 or align = 64). The choice depends
156on the architecture preference and compactness vs performance trade-offs
157of the implementation. Architectures providing fast unaligned write
158byte-packed basic types to save space, aligning each type on byte
159boundaries (8-bit). Architectures with slow unaligned writes align types
160on specific alignment values. If no specific alignment is declared for a
161type, it is assumed to be bit-packed for integers with size not multiple
162of 8 bits and for gcc bitfields. All other basic types are byte-packed
163by default. It is however recommended to always specify the alignment
164explicitly. Alignment values must be power of two. Compound types are
165aligned as specified in their individual specification.
5ba9f198 166
6672e9e1 167TSDL meta-data attribute representation of a specific alignment:
5ba9f198
MD
168
169 align = value; /* value in bits */
170
1714.1.3 Byte order
172
3bf79539
MD
173By default, the native endianness of the source architecture the trace is used.
174Byte order can be overridden for a basic type by specifying a "byte_order"
175attribute. Typical use-case is to specify the network byte order (big endian:
176"be") to save data captured from the network into the trace without conversion.
177If not specified, the byte order is native.
5ba9f198 178
6672e9e1 179TSDL meta-data representation:
5ba9f198
MD
180
181 byte_order = native OR network OR be OR le; /* network and be are aliases */
182
1834.1.4 Size
184
185Type size, in bits, for integers and floats is that returned by "sizeof()" in C
186multiplied by CHAR_BIT.
187We require the size of "char" and "unsigned char" types (CHAR_BIT) to be fixed
188to 8 bits for cross-endianness compatibility.
189
6672e9e1 190TSDL meta-data representation:
5ba9f198
MD
191
192 size = value; (value is in bits)
193
1944.1.5 Integers
195
6672e9e1
MD
196Signed integers are represented in two-complement. Integer alignment,
197size, signedness and byte ordering are defined in the TSDL meta-data.
198Integers aligned on byte size (8-bit) and with length multiple of byte
199size (8-bit) correspond to the C99 standard integers. In addition,
200integers with alignment and/or size that are _not_ a multiple of the
201byte size are permitted; these correspond to the C99 standard bitfields,
202with the added specification that the CTF integer bitfields have a fixed
203binary representation. A MIT-licensed reference implementation of the
204CTF portable bitfields is available at:
5ba9f198
MD
205
206 http://git.efficios.com/?p=babeltrace.git;a=blob;f=include/babeltrace/bitfield.h
207
208Binary representation of integers:
209
210- On little and big endian:
211 - Within a byte, high bits correspond to an integer high bits, and low bits
212 correspond to low bits.
213- On little endian:
214 - Integer across multiple bytes are placed from the less significant to the
215 most significant.
216 - Consecutive integers are placed from lower bits to higher bits (even within
217 a byte).
218- On big endian:
219 - Integer across multiple bytes are placed from the most significant to the
220 less significant.
221 - Consecutive integers are placed from higher bits to lower bits (even within
222 a byte).
223
224This binary representation is derived from the bitfield implementation in GCC
225for little and big endian. However, contrary to what GCC does, integers can
6672e9e1 226cross units boundaries (no padding is required). Padding can be explicitly
5ba9f198
MD
227added (see 4.1.6 GNU/C bitfields) to follow the GCC layout if needed.
228
6672e9e1 229TSDL meta-data representation:
5ba9f198 230
80fd2569 231 integer {
5ba9f198
MD
232 signed = true OR false; /* default false */
233 byte_order = native OR network OR be OR le; /* default native */
234 size = value; /* value in bits, no default */
235 align = value; /* value in bits */
56b77c06
MD
236 /* based used for pretty-printing output, default: decimal. */
237 base = decimal OR dec OR OR d OR i OR u OR 10 OR hexadecimal OR hex OR x OR X OR p OR 16
238 OR octal OR oct OR o OR 8 OR binary OR b OR 2;
13441bf5
MD
239 /* character encoding, default: none */
240 encoding = none or UTF8 or ASCII;
2152348f 241 }
5ba9f198 242
80fd2569 243Example of type inheritance (creation of a uint32_t named type):
5ba9f198 244
359894ac 245typealias integer {
9e4e34e9 246 size = 32;
5ba9f198
MD
247 signed = false;
248 align = 32;
38b8da21 249} := uint32_t;
5ba9f198 250
80fd2569 251Definition of a named 5-bit signed bitfield:
5ba9f198 252
359894ac 253typealias integer {
5ba9f198
MD
254 size = 5;
255 signed = true;
256 align = 1;
38b8da21 257} := int5_t;
5ba9f198 258
13441bf5
MD
259The character encoding field can be used to specify that the integer
260must be printed as a text character when read. e.g.:
261
262typealias integer {
263 size = 8;
264 align = 8;
265 signed = false;
266 encoding = UTF8;
267} := utf_char;
268
269
5ba9f198
MD
2704.1.6 GNU/C bitfields
271
272The GNU/C bitfields follow closely the integer representation, with a
273particularity on alignment: if a bitfield cannot fit in the current unit, the
80fd2569
MD
274unit is padded and the bitfield starts at the following unit. The unit size is
275defined by the size of the type "unit_type".
5ba9f198 276
6672e9e1 277TSDL meta-data representation:
80fd2569 278
d674f4b8 279 unit_type name:size;
80fd2569 280
5ba9f198
MD
281As an example, the following structure declared in C compiled by GCC:
282
283struct example {
284 short a:12;
285 short b:5;
286};
287
2152348f
MD
288The example structure is aligned on the largest element (short). The second
289bitfield would be aligned on the next unit boundary, because it would not fit in
290the current unit.
5ba9f198
MD
291
2924.1.7 Floating point
293
6672e9e1 294The floating point values byte ordering is defined in the TSDL meta-data.
5ba9f198
MD
295
296Floating point values follow the IEEE 754-2008 standard interchange formats.
297Description of the floating point values include the exponent and mantissa size
298in bits. Some requirements are imposed on the floating point values:
299
300- FLT_RADIX must be 2.
301- mant_dig is the number of digits represented in the mantissa. It is specified
302 by the ISO C99 standard, section 5.2.4, as FLT_MANT_DIG, DBL_MANT_DIG and
303 LDBL_MANT_DIG as defined by <float.h>.
304- exp_dig is the number of digits represented in the exponent. Given that
305 mant_dig is one bit more than its actual size in bits (leading 1 is not
306 needed) and also given that the sign bit always takes one bit, exp_dig can be
307 specified as:
308
309 - sizeof(float) * CHAR_BIT - FLT_MANT_DIG
310 - sizeof(double) * CHAR_BIT - DBL_MANT_DIG
311 - sizeof(long double) * CHAR_BIT - LDBL_MANT_DIG
312
6672e9e1 313TSDL meta-data representation:
5ba9f198 314
80fd2569 315floating_point {
ec4404a7
MD
316 exp_dig = value;
317 mant_dig = value;
318 byte_order = native OR network OR be OR le;
319 align = value;
2152348f 320}
5ba9f198
MD
321
322Example of type inheritance:
323
359894ac 324typealias floating_point {
5ba9f198
MD
325 exp_dig = 8; /* sizeof(float) * CHAR_BIT - FLT_MANT_DIG */
326 mant_dig = 24; /* FLT_MANT_DIG */
327 byte_order = native;
ec4404a7 328 align = 32;
38b8da21 329} := float;
5ba9f198
MD
330
331TODO: define NaN, +inf, -inf behavior.
332
ec4404a7
MD
333Bit-packed, byte-packed or larger alignments can be used for floating
334point values, similarly to integers.
335
5ba9f198
MD
3364.1.8 Enumerations
337
338Enumerations are a mapping between an integer type and a table of strings. The
339numerical representation of the enumeration follows the integer type specified
6672e9e1
MD
340by the meta-data. The enumeration mapping table is detailed in the enumeration
341description within the meta-data. The mapping table maps inclusive value
342ranges (or single values) to strings. Instead of being limited to simple
3bf79539 343"value -> string" mappings, these enumerations map
80fd2569 344"[ start_value ... end_value ] -> string", which map inclusive ranges of
3bf79539
MD
345values to strings. An enumeration from the C language can be represented in
346this format by having the same start_value and end_value for each element, which
347is in fact a range of size 1. This single-value range is supported without
4767a9e7 348repeating the start and end values with the value = string declaration.
80fd2569 349
a9b83695 350enum name : integer_type {
359894ac 351 somestring = start_value1 ... end_value1,
80fd2569
MD
352 "other string" = start_value2 ... end_value2,
353 yet_another_string, /* will be assigned to end_value2 + 1 */
354 "some other string" = value,
355 ...
356};
357
358If the values are omitted, the enumeration starts at 0 and increment of 1 for
359each entry:
360
a9b83695 361enum name : unsigned int {
80fd2569
MD
362 ZERO,
363 ONE,
364 TWO,
365 TEN = 10,
366 ELEVEN,
3bf79539 367};
5ba9f198 368
80fd2569 369Overlapping ranges within a single enumeration are implementation defined.
5ba9f198 370
2152348f
MD
371A nameless enumeration can be declared as a field type or as part of a typedef:
372
a9b83695 373enum : integer_type {
2152348f
MD
374 ...
375}
376
c2742c56
MD
377Enumerations omitting the container type ": integer_type" use the "int"
378type (for compatibility with C99). The "int" type must be previously
379declared. E.g.:
380
381typealias integer { size = 32; align = 32; signed = true } := int;
382
383enum {
384 ...
385}
386
1fad7a85 387
5ba9f198
MD
3884.2 Compound types
389
1fad7a85
MD
390Compound are aggregation of type declarations. Compound types include
391structures, variant, arrays, sequences, and strings.
392
5ba9f198
MD
3934.2.1 Structures
394
395Structures are aligned on the largest alignment required by basic types
396contained within the structure. (This follows the ISO/C standard for structures)
397
6672e9e1 398TSDL meta-data representation of a named structure:
5ba9f198 399
80fd2569
MD
400struct name {
401 field_type field_name;
402 field_type field_name;
403 ...
404};
5ba9f198
MD
405
406Example:
407
80fd2569
MD
408struct example {
409 integer { /* Nameless type */
410 size = 16;
411 signed = true;
412 align = 16;
413 } first_field_name;
6672e9e1 414 uint64_t second_field_name; /* Named type declared in the meta-data */
3bf79539 415};
5ba9f198
MD
416
417The fields are placed in a sequence next to each other. They each possess a
418field name, which is a unique identifier within the structure.
419
2152348f 420A nameless structure can be declared as a field type or as part of a typedef:
80fd2569
MD
421
422struct {
423 ...
2152348f 424}
80fd2569 425
ec4404a7
MD
426Alignment for a structure compound type can be forced to a minimum value
427by adding an "align" specifier after the declaration of a structure
428body. This attribute is read as: align(value). The value is specified in
429bits. The structure will be aligned on the maximum value between this
430attribute and the alignment required by the basic types contained within
431the structure. e.g.
432
433struct {
434 ...
435} align(32)
436
77a98c82 4374.2.2 Variants (Discriminated/Tagged Unions)
fcba70d4 438
fdf2bb05
MD
439A CTF variant is a selection between different types. A CTF variant must
440always be defined within the scope of a structure or within fields
441contained within a structure (defined recursively). A "tag" enumeration
442field must appear in either the same lexical scope, prior to the variant
1ab22b2a
MD
443field (in field declaration order), in an upper lexical scope (see
444Section 7.3.1), or in an upper dynamic scope (see Section 7.3.2). The
445type selection is indicated by the mapping from the enumeration value to
446the string used as variant type selector. The field to use as tag is
447specified by the "tag_field", specified between "< >" after the
fdf2bb05
MD
448"variant" keyword for unnamed variants, and after "variant name" for
449named variants.
fcba70d4
MD
450
451The alignment of the variant is the alignment of the type as selected by the tag
452value for the specific instance of the variant. The alignment of the type
453containing the variant is independent of the variant alignment. The size of the
454variant is the size as selected by the tag value for the specific instance of
455the variant.
456
457A named variant declaration followed by its definition within a structure
458declaration:
459
460variant name {
461 field_type sel1;
462 field_type sel2;
463 field_type sel3;
464 ...
465};
466
467struct {
a9b83695 468 enum : integer_type { sel1, sel2, sel3, ... } tag_field;
fcba70d4
MD
469 ...
470 variant name <tag_field> v;
471}
472
473An unnamed variant definition within a structure is expressed by the following
6672e9e1 474TSDL meta-data:
fcba70d4
MD
475
476struct {
a9b83695 477 enum : integer_type { sel1, sel2, sel3, ... } tag_field;
fcba70d4
MD
478 ...
479 variant <tag_field> {
480 field_type sel1;
481 field_type sel2;
482 field_type sel3;
483 ...
484 } v;
485}
486
487Example of a named variant within a sequence that refers to a single tag field:
488
489variant example {
490 uint32_t a;
491 uint64_t b;
492 short c;
493};
494
495struct {
a9b83695 496 enum : uint2_t { a, b, c } choice;
1ab22b2a
MD
497 unsigned int seqlen;
498 variant example <choice> v[seqlen];
fcba70d4
MD
499}
500
501Example of an unnamed variant:
502
503struct {
a9b83695 504 enum : uint2_t { a, b, c, d } choice;
fcba70d4
MD
505 /* Unrelated fields can be added between the variant and its tag */
506 int32_t somevalue;
507 variant <choice> {
508 uint32_t a;
509 uint64_t b;
510 short c;
511 struct {
512 unsigned int field1;
513 uint64_t field2;
514 } d;
515 } s;
516}
517
518Example of an unnamed variant within an array:
519
520struct {
a9b83695 521 enum : uint2_t { a, b, c } choice;
fcba70d4
MD
522 variant <choice> {
523 uint32_t a;
524 uint64_t b;
525 short c;
15850440 526 } v[10];
fcba70d4
MD
527}
528
529Example of a variant type definition within a structure, where the defined type
530is then declared within an array of structures. This variant refers to a tag
531located in an upper lexical scope. This example clearly shows that a variant
532type definition referring to the tag "x" uses the closest preceding field from
533the lexical scope of the type definition.
534
535struct {
a9b83695 536 enum : uint2_t { a, b, c, d } x;
fcba70d4
MD
537
538 typedef variant <x> { /*
539 * "x" refers to the preceding "x" enumeration in the
540 * lexical scope of the type definition.
541 */
542 uint32_t a;
543 uint64_t b;
544 short c;
545 } example_variant;
546
547 struct {
a9b83695 548 enum : int { x, y, z } x; /* This enumeration is not used by "v". */
fcba70d4 549 example_variant v; /*
a9b83695 550 * "v" uses the "enum : uint2_t { a, b, c, d }"
fcba70d4
MD
551 * tag.
552 */
553 } a[10];
554}
555
5564.2.3 Arrays
5ba9f198 557
6672e9e1
MD
558Arrays are fixed-length. Their length is declared in the type
559declaration within the meta-data. They contain an array of "inner type"
560elements, which can refer to any type not containing the type of the
561array being declared (no circular dependency). The length is the number
562of elements in an array.
5ba9f198 563
6672e9e1 564TSDL meta-data representation of a named array:
80fd2569
MD
565
566typedef elem_type name[length];
5ba9f198 567
2152348f 568A nameless array can be declared as a field type within a structure, e.g.:
5ba9f198 569
2152348f 570 uint8_t field_name[10];
80fd2569 571
ec4404a7 572Arrays are always aligned on their element alignment requirement.
5ba9f198 573
fcba70d4 5744.2.4 Sequences
5ba9f198 575
1ab22b2a
MD
576Sequences are dynamically-sized arrays. They refer to a a "length"
577unsigned integer field, which must appear in either the same lexical scope,
578prior to the sequence field (in field declaration order), in an upper
579lexical scope (see Section 7.3.1), or in an upper dynamic scope (see
580Section 7.3.2). This length field represents the number of elements in
581the sequence. The sequence per se is an array of "inner type" elements.
5ba9f198 582
1ab22b2a 583TSDL meta-data representation for a sequence type definition:
80fd2569 584
1ab22b2a
MD
585struct {
586 unsigned int length_field;
587 typedef elem_type typename[length_field];
588 typename seq_field_name;
589}
590
591A sequence can also be declared as a field type, e.g.:
80fd2569 592
1ab22b2a
MD
593struct {
594 unsigned int length_field;
595 long seq_field_name[length_field];
596}
80fd2569 597
1ab22b2a
MD
598Multiple sequences can refer to the same length field, and these length
599fields can be in a different upper dynamic scope:
600
601e.g., assuming the stream.event.header defines:
602
603stream {
604 ...
605 id = 1;
606 event.header := struct {
607 uint16_t seq_len;
608 };
609};
610
611event {
612 ...
613 stream_id = 1;
614 fields := struct {
615 long seq_a[stream.event.header.seq_len];
616 char seq_b[stream.event.header.seq_len];
617 };
618};
80fd2569 619
1ab22b2a 620The sequence elements follow the "array" specifications.
5ba9f198 621
fcba70d4 6224.2.5 Strings
5ba9f198
MD
623
624Strings are an array of bytes of variable size and are terminated by a '\0'
6672e9e1
MD
625"NULL" character. Their encoding is described in the TSDL meta-data. In
626absence of encoding attribute information, the default encoding is
627UTF-8.
5ba9f198 628
6672e9e1 629TSDL meta-data representation of a named string type:
80fd2569 630
359894ac 631typealias string {
5ba9f198 632 encoding = UTF8 OR ASCII;
38b8da21 633} := name;
5ba9f198 634
80fd2569
MD
635A nameless string type can be declared as a field type:
636
637string field_name; /* Use default UTF8 encoding */
5ba9f198 638
ec4404a7
MD
639Strings are always aligned on byte size.
640
3bf79539
MD
6415. Event Packet Header
642
6672e9e1
MD
643The event packet header consists of two parts: the "event packet header"
644is the same for all streams of a trace. The second part, the "event
645packet context", is described on a per-stream basis. Both are described
646in the TSDL meta-data. The packets are aligned on architecture-page-sized
647addresses.
3bf79539 648
6672e9e1 649Event packet header (all fields are optional, specified by TSDL meta-data):
3bf79539 650
6672e9e1
MD
651- Magic number (CTF magic number: 0xC1FC1FC1) specifies that this is a
652 CTF packet. This magic number is optional, but when present, it should
653 come at the very beginning of the packet.
654- Trace UUID, used to ensure the event packet match the meta-data used.
655 (note: we cannot use a meta-data checksum in every cases instead of a
656 UUID because meta-data can be appended to while tracing is active)
657 This field is optional.
658- Stream ID, used as reference to stream description in meta-data.
659 This field is optional if there is only one stream description in the
660 meta-data, but becomes required if there are more than one stream in
661 the TSDL meta-data description.
3bf79539 662
6672e9e1 663Event packet context (all fields are optional, specified by TSDL meta-data):
3bf79539 664
4dfca05b
MD
665- Event packet content size (in bits).
666- Event packet size (in bits, includes padding).
cda89682 667- Event packet content checksum. Checksum excludes the event packet
3bf79539
MD
668 header.
669- Per-stream event packet sequence count (to deal with UDP packet loss). The
670 number of significant sequence counter bits should also be present, so
b11853af 671 wrap-arounds are dealt with correctly.
6672e9e1 672- Time-stamp at the beginning and time-stamp at the end of the event packet.
3bf79539
MD
673 Both timestamps are written in the packet header, but sampled respectively
674 while (or before) writing the first event and while (or after) writing the
675 last event in the packet. The inclusive range between these timestamps should
676 include all event timestamps assigned to events contained within the packet.
5ba9f198 677- Events discarded count
3bf79539
MD
678 - Snapshot of a per-stream free-running counter, counting the number of
679 events discarded that were supposed to be written in the stream prior to
680 the first event in the event packet.
5ba9f198 681 * Note: producer-consumer buffer full condition should fill the current
3bf79539 682 event packet with padding so we know exactly where events have been
5ba9f198 683 discarded.
3bf79539
MD
684- Lossless compression scheme used for the event packet content. Applied
685 directly to raw data. New types of compression can be added in following
686 versions of the format.
5ba9f198
MD
687 0: no compression scheme
688 1: bzip2
689 2: gzip
3bf79539
MD
690 3: xz
691- Cypher used for the event packet content. Applied after compression.
5ba9f198
MD
692 0: no encryption
693 1: AES
3bf79539 694- Checksum scheme used for the event packet content. Applied after encryption.
5ba9f198
MD
695 0: no checksum
696 1: md5
697 2: sha1
698 3: crc32
699
6672e9e1 7005.1 Event Packet Header Description
3bf79539 701
fc5425db 702The event packet header layout is indicated by the trace packet.header
6672e9e1
MD
703field. Here is a recommended structure type for the packet header with
704the fields typically expected (although these fields are each optional):
fc5425db 705
80fd2569
MD
706struct event_packet_header {
707 uint32_t magic;
3fde5da1 708 uint8_t uuid[16];
3bf79539 709 uint32_t stream_id;
80fd2569 710};
5ba9f198 711
fc5425db
MD
712trace {
713 ...
714 packet.header := struct event_packet_header;
715};
716
6672e9e1
MD
717If the magic number is not present, tools such as "file" will have no
718mean to discover the file type.
719
3fde5da1
MD
720If the uuid is not present, no validation that the meta-data actually
721corresponds to the stream is performed.
37b39e47
MD
722
723If the stream_id packet header field is missing, the trace can only
724contain a single stream. Its "id" field can be left out, and its events
725don't need to declare a "stream_id" field.
726
727
3bf79539
MD
7285.2 Event Packet Context Description
729
730Event packet context example. These are declared within the stream declaration
6672e9e1 731in the meta-data. All these fields are optional. If the packet size field is
6a7c61df 732missing, the whole stream only contains a single packet. If the content
f6b81726
MD
733size field is missing, the packet is filled (no padding). The content
734and packet sizes include all headers.
3bf79539
MD
735
736An example event packet context type:
737
80fd2569 738struct event_packet_context {
3bf79539
MD
739 uint64_t timestamp_begin;
740 uint64_t timestamp_end;
741 uint32_t checksum;
742 uint32_t stream_packet_count;
743 uint32_t events_discarded;
744 uint32_t cpu_id;
745 uint32_t/uint16_t content_size;
746 uint32_t/uint16_t packet_size;
747 uint8_t stream_packet_count_bits; /* Significant counter bits */
748 uint8_t compression_scheme;
749 uint8_t encryption_scheme;
3b0f8e4d 750 uint8_t checksum_scheme;
3bf79539 751};
5ba9f198 752
fcba70d4 753
5ba9f198
MD
7546. Event Structure
755
756The overall structure of an event is:
757
6672e9e1
MD
7581 - Stream Packet Context (as specified by the stream meta-data)
759 2 - Event Header (as specified by the stream meta-data)
760 3 - Stream Event Context (as specified by the stream meta-data)
761 4 - Event Context (as specified by the event meta-data)
762 5 - Event Payload (as specified by the event meta-data)
5ba9f198 763
fdf2bb05 764This structure defines an implicit dynamic scoping, where variants
7d9d7e92
MD
765located in inner structures (those with a higher number in the listing
766above) can refer to the fields of outer structures (with lower number in
6c7226e9 767the listing above). See Section 7.3 TSDL Scopes for more detail.
5ba9f198 768
fdf2bb05 7696.1 Event Header
fcba70d4 770
6672e9e1 771Event headers can be described within the meta-data. We hereby propose, as an
fcba70d4
MD
772example, two types of events headers. Type 1 accommodates streams with less than
77331 event IDs. Type 2 accommodates streams with 31 or more event IDs.
5ba9f198 774
3bf79539
MD
775One major factor can vary between streams: the number of event IDs assigned to
776a stream. Luckily, this information tends to stay relatively constant (modulo
5ba9f198 777event registration while trace is being recorded), so we can specify different
3bf79539 778representations for streams containing few event IDs and streams containing
6672e9e1
MD
779many event IDs, so we end up representing the event ID and time-stamp as
780densely as possible in each case.
5ba9f198 781
fcba70d4
MD
782The header is extended in the rare occasions where the information cannot be
783represented in the ranges available in the standard event header. They are also
3bf79539
MD
784used in the rare occasions where the data required for a field could not be
785collected: the flag corresponding to the missing field within the missing_fields
786array is then set to 1.
5ba9f198 787
6672e9e1
MD
788Types uintX_t represent an X-bit unsigned integer, as declared with
789either:
5ba9f198 790
6672e9e1
MD
791 typealias integer { size = X; align = X; signed = false } := uintX_t;
792
793 or
794
795 typealias integer { size = X; align = 1; signed = false } := uintX_t;
5ba9f198 796
fdf2bb05 7976.1.1 Type 1 - Few event IDs
5ba9f198
MD
798
799 - Aligned on 32-bit (or 8-bit if byte-packed, depending on the architecture
800 preference).
5ba9f198 801 - Native architecture byte ordering.
fcba70d4
MD
802 - For "compact" selection
803 - Fixed size: 32 bits.
804 - For "extended" selection
805 - Size depends on the architecture and variant alignment.
5ba9f198 806
80fd2569 807struct event_header_1 {
fcba70d4
MD
808 /*
809 * id: range: 0 - 30.
810 * id 31 is reserved to indicate an extended header.
811 */
a9b83695 812 enum : uint5_t { compact = 0 ... 30, extended = 31 } id;
fcba70d4
MD
813 variant <id> {
814 struct {
815 uint27_t timestamp;
816 } compact;
817 struct {
818 uint32_t id; /* 32-bit event IDs */
819 uint64_t timestamp; /* 64-bit timestamps */
820 } extended;
821 } v;
cb108fea 822} align(32); /* or align(8) */
5ba9f198 823
5ba9f198 824
fdf2bb05 8256.1.2 Type 2 - Many event IDs
5ba9f198 826
fcba70d4 827 - Aligned on 16-bit (or 8-bit if byte-packed, depending on the architecture
5ba9f198 828 preference).
5ba9f198 829 - Native architecture byte ordering.
fcba70d4
MD
830 - For "compact" selection
831 - Size depends on the architecture and variant alignment.
832 - For "extended" selection
833 - Size depends on the architecture and variant alignment.
5ba9f198 834
80fd2569 835struct event_header_2 {
fcba70d4
MD
836 /*
837 * id: range: 0 - 65534.
838 * id 65535 is reserved to indicate an extended header.
839 */
a9b83695 840 enum : uint16_t { compact = 0 ... 65534, extended = 65535 } id;
fcba70d4
MD
841 variant <id> {
842 struct {
843 uint32_t timestamp;
844 } compact;
845 struct {
846 uint32_t id; /* 32-bit event IDs */
847 uint64_t timestamp; /* 64-bit timestamps */
848 } extended;
849 } v;
cb108fea 850} align(16); /* or align(8) */
5ba9f198 851
5ba9f198
MD
852
8536.2 Event Context
854
6672e9e1
MD
855The event context contains information relative to the current event.
856The choice and meaning of this information is specified by the TSDL
857stream and event meta-data descriptions. The stream context is applied
858to all events within the stream. The stream context structure follows
859the event header. The event context is applied to specific events. Its
860structure follows the stream context structure.
5ba9f198 861
fcba70d4
MD
862An example of stream-level event context is to save the event payload size with
863each event, or to save the current PID with each event. These are declared
6672e9e1 864within the stream declaration within the meta-data:
5ba9f198 865
fcba70d4
MD
866 stream {
867 ...
6672e9e1 868 event.context := struct {
80fd2569
MD
869 uint pid;
870 uint16_t payload_size;
6672e9e1 871 };
fcba70d4
MD
872 };
873
874An example of event-specific event context is to declare a bitmap of missing
875fields, only appended after the stream event context if the extended event
876header is selected. NR_FIELDS is the number of fields within the event (a
877numeric value).
5ba9f198 878
fcba70d4
MD
879 event {
880 context = struct {
881 variant <id> {
882 struct { } compact;
883 struct {
884 uint1_t missing_fields[NR_FIELDS]; /* missing event fields bitmap */
885 } extended;
886 } v;
887 };
888 ...
889 }
5ba9f198
MD
890
8916.3 Event Payload
892
893An event payload contains fields specific to a given event type. The fields
6672e9e1 894belonging to an event type are described in the event-specific meta-data
5ba9f198
MD
895within a structure type.
896
8976.3.1 Padding
898
899No padding at the end of the event payload. This differs from the ISO/C standard
900for structures, but follows the CTF standard for structures. In a trace, even
901though it makes sense to align the beginning of a structure, it really makes no
902sense to add padding at the end of the structure, because structures are usually
903not followed by a structure of the same type.
904
905This trick can be done by adding a zero-length "end" field at the end of the C
906structures, and by using the offset of this field rather than using sizeof()
3bf79539 907when calculating the size of a structure (see Appendix "A. Helper macros").
5ba9f198
MD
908
9096.3.2 Alignment
910
911The event payload is aligned on the largest alignment required by types
912contained within the payload. (This follows the ISO/C standard for structures)
913
914
6c7226e9
MD
9157. Trace Stream Description Language (TSDL)
916
917The Trace Stream Description Language (TSDL) allows expression of the
918binary trace streams layout in a C99-like Domain Specific Language
919(DSL).
920
921
6672e9e1 9227.1 Meta-data
6c7226e9
MD
923
924The trace stream layout description is located in the trace meta-data.
925The meta-data is itself located in a stream identified by its name:
926"metadata".
5ba9f198 927
6672e9e1
MD
928The meta-data description can be expressed in two different formats:
929text-only and packet-based. The text-only description facilitates
930generation of meta-data and provides a convenient way to enter the
931meta-data information by hand. The packet-based meta-data provides the
932CTF stream packet facilities (checksumming, compression, encryption,
933network-readiness) for meta-data stream generated and transported by a
934tracer.
935
936The text-only meta-data file is a plain text TSDL description.
937
938The packet-based meta-data is made of "meta-data packets", which each
939start with a meta-data packet header. The packet-based meta-data
940description is detected by reading the magic number "0x75D11D57" at the
941beginning of the file. This magic number is also used to detect the
942endianness of the architecture by trying to read the CTF magic number
943and its counterpart in reversed endianness. The events within the
944meta-data stream have no event header nor event context. Each event only
5ae2352d 945contains a "sequence" payload, which is a sequence of bits using the
ce2228da
MD
946"trace.packet.header.content_size" field as a placeholder for its length
947(the packet header size should be substracted). The formatting of this
948sequence of bits is a plain-text representation of the TSDL description.
949Each meta-data packet start with a special packet header, specific to
950the meta-data stream, which contains, exactly:
6672e9e1
MD
951
952struct metadata_packet_header {
2daeaa3a 953 uint32_t magic; /* 0x75D11D57 */
3fde5da1 954 uint8_t uuid[16]; /* Unique Universal Identifier */
6672e9e1
MD
955 uint32_t checksum; /* 0 if unused */
956 uint32_t content_size; /* in bits */
957 uint32_t packet_size; /* in bits */
958 uint8_t compression_scheme; /* 0 if unused */
959 uint8_t encryption_scheme; /* 0 if unused */
960 uint8_t checksum_scheme; /* 0 if unused */
961};
962
963The packet-based meta-data can be converted to a text-only meta-data by
964concatenating all the strings in contains.
4fafe1ad 965
6672e9e1
MD
966In the textual representation of the meta-data, the text contained
967within "/*" and "*/", as well as within "//" and end of line, are
968treated as comments. Boolean values can be represented as true, TRUE,
969or 1 for true, and false, FALSE, or 0 for false. Within the string-based
970meta-data description, the trace UUID is represented as a string of
971hexadecimal digits and dashes "-". In the event packet header, the trace
972UUID is represented as an array of bytes.
fcba70d4 973
fdf2bb05 974
6c7226e9 9757.2 Declaration vs Definition
fdf2bb05
MD
976
977A declaration associates a layout to a type, without specifying where
978this type is located in the event structure hierarchy (see Section 6).
979This therefore includes typedef, typealias, as well as all type
980specifiers. In certain circumstances (typedef, structure field and
981variant field), a declaration is followed by a declarator, which specify
982the newly defined type name (for typedef), or the field name (for
983declarations located within structure and variants). Array and sequence,
984declared with square brackets ("[" "]"), are part of the declarator,
a9b83695 985similarly to C99. The enumeration base type is specified by
6c7226e9 986": enum_base", which is part of the type specifier. The variant tag
a9b83695 987name, specified between "<" ">", is also part of the type specifier.
fdf2bb05
MD
988
989A definition associates a type to a location in the event structure
b9606a77
MD
990hierarchy (see Section 6). This association is denoted by ":=", as shown
991in Section 7.3.
fdf2bb05
MD
992
993
6c7226e9 9947.3 TSDL Scopes
fdf2bb05 995
6c7226e9
MD
996TSDL uses two different types of scoping: a lexical scope is used for
997declarations and type definitions, and a dynamic scope is used for
2e7d7fcb
MD
998variants references to tag fields and for sequence references to length
999fields.
fdf2bb05 1000
6c7226e9 10017.3.1 Lexical Scope
fdf2bb05 1002
d285084f
MD
1003Each of "trace", "stream", "event", "struct" and "variant" have their own
1004nestable declaration scope, within which types can be declared using "typedef"
fdf2bb05 1005and "typealias". A root declaration scope also contains all declarations
7d9d7e92 1006located outside of any of the aforementioned declarations. An inner
fdf2bb05 1007declaration scope can refer to type declared within its container
7d9d7e92
MD
1008lexical scope prior to the inner declaration scope. Redefinition of a
1009typedef or typealias is not valid, although hiding an upper scope
fdf2bb05
MD
1010typedef or typealias is allowed within a sub-scope.
1011
6c7226e9 10127.3.2 Dynamic Scope
fdf2bb05 1013
7d9d7e92
MD
1014A dynamic scope consists in the lexical scope augmented with the
1015implicit event structure definition hierarchy presented at Section 6.
2e7d7fcb
MD
1016The dynamic scope is used for variant tag and sequence length
1017definitions. It is used at definition time to look up the location of
1018the tag field associated with a variant, and to lookup up the location
1019of the length field associated with a sequence.
1020
1021Therefore, variants (or sequences) in lower levels in the dynamic scope
1022(e.g. event context) can refer to a tag (or length) field located in
1023upper levels (e.g. in the event header) by specifying, in this case, the
1024associated tag with <header.field_name>. This allows, for instance, the
1025event context to define a variant referring to the "id" field of the
1026event header as selector.
fdf2bb05
MD
1027
1028The target dynamic scope must be specified explicitly when referring to
1029a field outside of the local static scope. The dynamic scope prefixes
1030are thus:
1031
e0d9e2c7 1032 - Trace Packet Header: <trace.packet.header. >,
7d9d7e92
MD
1033 - Stream Packet Context: <stream.packet.context. >,
1034 - Event Header: <stream.event.header. >,
1035 - Stream Event Context: <stream.event.context. >,
1036 - Event Context: <event.context. >,
1037 - Event Payload: <event.fields. >.
fdf2bb05
MD
1038
1039Multiple declarations of the same field name within a single scope is
1040not valid. It is however valid to re-use the same field name in
1041different scopes. There is no possible conflict, because the dynamic
1042scope must be specified when a variant refers to a tag field located in
1043a different dynamic scope.
1044
457d8b0a
MD
1045The information available in the dynamic scopes can be thought of as the
1046current tracing context. At trace production, information about the
1047current context is saved into the specified scope field levels. At trace
1048consumption, for each event, the current trace context is therefore
1049readable by accessing the upper dynamic scopes.
1050
fdf2bb05 1051
6c7226e9 10527.4 TSDL Examples
d285084f 1053
6672e9e1 1054The grammar representing the TSDL meta-data is presented in Appendix C.
7df6b93a 1055TSDL Grammar. This section presents a rather lighter reading that
6672e9e1 1056consists in examples of TSDL meta-data, with template values.
969f30c0 1057
350bb633
MD
1058The stream "id" can be left out if there is only one stream in the
1059trace. The event "id" field can be left out if there is only one event
1060in a stream.
1061
5ba9f198 1062trace {
fdf2bb05 1063 major = value; /* Trace format version */
5ba9f198 1064 minor = value;
fdf2bb05 1065 uuid = "aaaaaaaa-aaaa-aaaa-aaaa-aaaaaaaaaaaa"; /* Trace UUID */
58997e9e 1066 byte_order = be OR le; /* Endianness (required) */
fc5425db
MD
1067 packet.header := struct {
1068 uint32_t magic;
3fde5da1 1069 uint8_t uuid[16];
fc5425db
MD
1070 uint32_t stream_id;
1071 };
3bf79539 1072};
5ba9f198 1073
3bf79539
MD
1074stream {
1075 id = stream_id;
fdf2bb05 1076 /* Type 1 - Few event IDs; Type 2 - Many event IDs. See section 6.1. */
4fa992a5
MD
1077 event.header := event_header_1 OR event_header_2;
1078 event.context := struct {
77a98c82 1079 ...
3bf79539 1080 };
4fa992a5 1081 packet.context := struct {
77a98c82 1082 ...
3bf79539
MD
1083 };
1084};
5ba9f198
MD
1085
1086event {
3d13ef1a 1087 name = event_name;
3bf79539 1088 id = value; /* Numeric identifier within the stream */
67f02e24 1089 stream_id = stream_id;
4fa992a5 1090 context := struct {
fcba70d4
MD
1091 ...
1092 };
4fa992a5 1093 fields := struct {
80fd2569
MD
1094 ...
1095 };
3bf79539 1096};
5ba9f198
MD
1097
1098/* More detail on types in section 4. Types */
1099
3d13ef1a
MD
1100/*
1101 * Named types:
1102 *
4fa992a5 1103 * Type declarations behave similarly to the C standard.
3d13ef1a
MD
1104 */
1105
80af8ac6 1106typedef aliased_type_specifiers new_type_declarators;
2152348f 1107
3d13ef1a 1108/* e.g.: typedef struct example new_type_name[10]; */
80fd2569 1109
4fa992a5
MD
1110/*
1111 * typealias
1112 *
1113 * The "typealias" declaration can be used to give a name (including
80af8ac6
MD
1114 * pointer declarator specifier) to a type. It should also be used to
1115 * map basic C types (float, int, unsigned long, ...) to a CTF type.
1116 * Typealias is a superset of "typedef": it also allows assignment of a
38b8da21 1117 * simple variable identifier to a type.
4fa992a5
MD
1118 */
1119
1120typealias type_class {
80fd2569 1121 ...
38b8da21 1122} := type_specifiers type_declarator;
2152348f 1123
3d13ef1a
MD
1124/*
1125 * e.g.:
4fa992a5 1126 * typealias integer {
3d13ef1a
MD
1127 * size = 32;
1128 * align = 32;
1129 * signed = false;
38b8da21 1130 * } := struct page *;
359894ac
MD
1131 *
1132 * typealias integer {
1133 * size = 32;
1134 * align = 32;
1135 * signed = true;
38b8da21 1136 * } := int;
3d13ef1a 1137 */
80fd2569
MD
1138
1139struct name {
3bf79539
MD
1140 ...
1141};
5ba9f198 1142
fcba70d4
MD
1143variant name {
1144 ...
1145};
1146
a9b83695 1147enum name : integer_type {
3bf79539
MD
1148 ...
1149};
1150
2152348f 1151
4fa992a5
MD
1152/*
1153 * Unnamed types, contained within compound type fields, typedef or typealias.
1154 */
2152348f 1155
80fd2569
MD
1156struct {
1157 ...
2152348f 1158}
5ba9f198 1159
ec4404a7
MD
1160struct {
1161 ...
1162} align(value)
1163
fcba70d4
MD
1164variant {
1165 ...
1166}
1167
a9b83695 1168enum : integer_type {
80fd2569 1169 ...
2152348f
MD
1170}
1171
1172typedef type new_type[length];
3bf79539 1173
2152348f
MD
1174struct {
1175 type field_name[length];
1176}
1177
1178typedef type new_type[length_type];
1179
1180struct {
1181 type field_name[length_type];
1182}
1183
1184integer {
80fd2569 1185 ...
2152348f 1186}
3bf79539 1187
2152348f 1188floating_point {
80fd2569 1189 ...
2152348f
MD
1190}
1191
1192struct {
1193 integer_type field_name:size; /* GNU/C bitfield */
1194}
1195
1196struct {
1197 string field_name;
1198}
3bf79539 1199
fcba70d4 1200
3bf79539 1201A. Helper macros
5ba9f198
MD
1202
1203The two following macros keep track of the size of a GNU/C structure without
1204padding at the end by placing HEADER_END as the last field. A one byte end field
1205is used for C90 compatibility (C99 flexible arrays could be used here). Note
1206that this does not affect the effective structure size, which should always be
1207calculated with the header_sizeof() helper.
1208
1209#define HEADER_END char end_field
1210#define header_sizeof(type) offsetof(typeof(type), end_field)
3bf79539
MD
1211
1212
1213B. Stream Header Rationale
1214
1215An event stream is divided in contiguous event packets of variable size. These
1216subdivisions allow the trace analyzer to perform a fast binary search by time
1217within the stream (typically requiring to index only the event packet headers)
1218without reading the whole stream. These subdivisions have a variable size to
1219eliminate the need to transfer the event packet padding when partially filled
1220event packets must be sent when streaming a trace for live viewing/analysis.
1221An event packet can contain a certain amount of padding at the end. Dividing
1222streams into event packets is also useful for network streaming over UDP and
1223flight recorder mode tracing (a whole event packet can be swapped out of the
1224buffer atomically for reading).
1225
1226The stream header is repeated at the beginning of each event packet to allow
1227flexibility in terms of:
1228
1229 - streaming support,
1230 - allowing arbitrary buffers to be discarded without making the trace
1231 unreadable,
1232 - allow UDP packet loss handling by either dealing with missing event packet
1233 or asking for re-transmission.
1234 - transparently support flight recorder mode,
1235 - transparently support crash dump.
1236
6c7226e9
MD
1237
1238C. TSDL Grammar
fcba70d4 1239
4fa992a5 1240/*
6c7226e9 1241 * Common Trace Format (CTF) Trace Stream Description Language (TSDL) Grammar.
4fa992a5
MD
1242 *
1243 * Inspired from the C99 grammar:
1244 * http://www.open-std.org/jtc1/sc22/wg14/www/docs/n1124.pdf (Annex A)
6c7226e9
MD
1245 * and c++1x grammar (draft)
1246 * http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2011/n3291.pdf (Annex A)
4fa992a5
MD
1247 *
1248 * Specialized for CTF needs by including only constant and declarations from
1249 * C99 (excluding function declarations), and by adding support for variants,
6c7226e9
MD
1250 * sequences and CTF-specific specifiers. Enumeration container types
1251 * semantic is inspired from c++1x enum-base.
4fa992a5
MD
1252 */
1253
12541) Lexical grammar
1255
12561.1) Lexical elements
1257
1258token:
1259 keyword
1260 identifier
1261 constant
1262 string-literal
1263 punctuator
1264
12651.2) Keywords
1266
1267keyword: is one of
1268
ec4404a7 1269align
4fa992a5
MD
1270const
1271char
1272double
1273enum
1274event
1275floating_point
1276float
1277integer
1278int
1279long
1280short
1281signed
1282stream
1283string
1284struct
1285trace
3e1e1a78 1286typealias
4fa992a5
MD
1287typedef
1288unsigned
1289variant
1290void
1291_Bool
1292_Complex
1293_Imaginary
1294
1295
12961.3) Identifiers
1297
1298identifier:
1299 identifier-nondigit
1300 identifier identifier-nondigit
1301 identifier digit
1302
1303identifier-nondigit:
1304 nondigit
1305 universal-character-name
1306 any other implementation-defined characters
1307
1308nondigit:
1309 _
1310 [a-zA-Z] /* regular expression */
1311
1312digit:
1313 [0-9] /* regular expression */
1314
13151.4) Universal character names
1316
1317universal-character-name:
1318 \u hex-quad
1319 \U hex-quad hex-quad
1320
1321hex-quad:
1322 hexadecimal-digit hexadecimal-digit hexadecimal-digit hexadecimal-digit
1323
13241.5) Constants
1325
1326constant:
1327 integer-constant
1328 enumeration-constant
1329 character-constant
1330
1331integer-constant:
1332 decimal-constant integer-suffix-opt
1333 octal-constant integer-suffix-opt
1334 hexadecimal-constant integer-suffix-opt
1335
1336decimal-constant:
1337 nonzero-digit
1338 decimal-constant digit
1339
1340octal-constant:
1341 0
1342 octal-constant octal-digit
1343
1344hexadecimal-constant:
1345 hexadecimal-prefix hexadecimal-digit
1346 hexadecimal-constant hexadecimal-digit
1347
1348hexadecimal-prefix:
1349 0x
1350 0X
1351
1352nonzero-digit:
1353 [1-9]
1354
1355integer-suffix:
1356 unsigned-suffix long-suffix-opt
1357 unsigned-suffix long-long-suffix
1358 long-suffix unsigned-suffix-opt
1359 long-long-suffix unsigned-suffix-opt
1360
1361unsigned-suffix:
1362 u
1363 U
1364
1365long-suffix:
1366 l
1367 L
1368
1369long-long-suffix:
1370 ll
1371 LL
1372
4fa992a5
MD
1373enumeration-constant:
1374 identifier
1375 string-literal
1376
1377character-constant:
1378 ' c-char-sequence '
1379 L' c-char-sequence '
1380
1381c-char-sequence:
1382 c-char
1383 c-char-sequence c-char
1384
1385c-char:
1386 any member of source charset except single-quote ('), backslash
1387 (\), or new-line character.
1388 escape-sequence
1389
1390escape-sequence:
1391 simple-escape-sequence
1392 octal-escape-sequence
1393 hexadecimal-escape-sequence
1394 universal-character-name
1395
1396simple-escape-sequence: one of
1397 \' \" \? \\ \a \b \f \n \r \t \v
1398
1399octal-escape-sequence:
1400 \ octal-digit
1401 \ octal-digit octal-digit
1402 \ octal-digit octal-digit octal-digit
1403
1404hexadecimal-escape-sequence:
1405 \x hexadecimal-digit
1406 hexadecimal-escape-sequence hexadecimal-digit
1407
14081.6) String literals
1409
1410string-literal:
1411 " s-char-sequence-opt "
1412 L" s-char-sequence-opt "
1413
1414s-char-sequence:
1415 s-char
1416 s-char-sequence s-char
1417
1418s-char:
1419 any member of source charset except double-quote ("), backslash
1420 (\), or new-line character.
1421 escape-sequence
1422
14231.7) Punctuators
1424
1425punctuator: one of
1426 [ ] ( ) { } . -> * + - < > : ; ... = ,
1427
1428
14292) Phrase structure grammar
1430
1431primary-expression:
1432 identifier
1433 constant
1434 string-literal
1435 ( unary-expression )
1436
1437postfix-expression:
1438 primary-expression
1439 postfix-expression [ unary-expression ]
1440 postfix-expression . identifier
1441 postfix-expressoin -> identifier
1442
1443unary-expression:
1444 postfix-expression
1445 unary-operator postfix-expression
1446
1447unary-operator: one of
1448 + -
1449
4fa992a5
MD
1450assignment-operator:
1451 =
1452
b9606a77
MD
1453type-assignment-operator:
1454 :=
1455
4fa992a5 1456constant-expression-range:
73d61ac3 1457 unary-expression ... unary-expression
4fa992a5
MD
1458
14592.2) Declarations:
1460
1461declaration:
689e04b4 1462 declaration-specifiers declarator-list-opt ;
4fa992a5
MD
1463 ctf-specifier ;
1464
1465declaration-specifiers:
689e04b4 1466 storage-class-specifier declaration-specifiers-opt
4fa992a5
MD
1467 type-specifier declaration-specifiers-opt
1468 type-qualifier declaration-specifiers-opt
1469
1470declarator-list:
1471 declarator
1472 declarator-list , declarator
1473
d285084f
MD
1474abstract-declarator-list:
1475 abstract-declarator
1476 abstract-declarator-list , abstract-declarator
1477
4fa992a5
MD
1478storage-class-specifier:
1479 typedef
1480
1481type-specifier:
1482 void
1483 char
1484 short
1485 int
1486 long
1487 float
1488 double
1489 signed
1490 unsigned
1491 _Bool
1492 _Complex
cfdd51ec 1493 _Imaginary
9dfcfc0f
MD
1494 struct-specifier
1495 variant-specifier
4fa992a5
MD
1496 enum-specifier
1497 typedef-name
1498 ctf-type-specifier
1499
ec4404a7 1500align-attribute:
73d61ac3 1501 align ( unary-expression )
ec4404a7 1502
4fa992a5 1503struct-specifier:
ec4404a7
MD
1504 struct identifier-opt { struct-or-variant-declaration-list-opt } align-attribute-opt
1505 struct identifier align-attribute-opt
4fa992a5
MD
1506
1507struct-or-variant-declaration-list:
1508 struct-or-variant-declaration
1509 struct-or-variant-declaration-list struct-or-variant-declaration
1510
1511struct-or-variant-declaration:
1512 specifier-qualifier-list struct-or-variant-declarator-list ;
eacb16d1 1513 declaration-specifiers-opt storage-class-specifier declaration-specifiers-opt declarator-list ;
6143bab7
MD
1514 typealias declaration-specifiers abstract-declarator-list type-assignment-operator declaration-specifiers abstract-declarator-list ;
1515 typealias declaration-specifiers abstract-declarator-list type-assignment-operator declarator-list ;
4fa992a5
MD
1516
1517specifier-qualifier-list:
1518 type-specifier specifier-qualifier-list-opt
1519 type-qualifier specifier-qualifier-list-opt
1520
1521struct-or-variant-declarator-list:
1522 struct-or-variant-declarator
1523 struct-or-variant-declarator-list , struct-or-variant-declarator
1524
1525struct-or-variant-declarator:
1526 declarator
73d61ac3 1527 declarator-opt : unary-expression
4fa992a5
MD
1528
1529variant-specifier:
1530 variant identifier-opt variant-tag-opt { struct-or-variant-declaration-list }
1531 variant identifier variant-tag
1532
1533variant-tag:
1534 < identifier >
1535
1536enum-specifier:
1537 enum identifier-opt { enumerator-list }
1538 enum identifier-opt { enumerator-list , }
1539 enum identifier
a9b83695
MD
1540 enum identifier-opt : declaration-specifiers { enumerator-list }
1541 enum identifier-opt : declaration-specifiers { enumerator-list , }
4fa992a5
MD
1542
1543enumerator-list:
1544 enumerator
1545 enumerator-list , enumerator
1546
1547enumerator:
1548 enumeration-constant
8d2d41f7
MD
1549 enumeration-constant assignment-operator unary-expression
1550 enumeration-constant assignment-operator constant-expression-range
4fa992a5
MD
1551
1552type-qualifier:
1553 const
1554
1555declarator:
1556 pointer-opt direct-declarator
1557
1558direct-declarator:
1559 identifier
1560 ( declarator )
1ab22b2a 1561 direct-declarator [ unary-expression ]
4fa992a5 1562
d285084f
MD
1563abstract-declarator:
1564 pointer-opt direct-abstract-declarator
1565
1566direct-abstract-declarator:
1567 identifier-opt
1568 ( abstract-declarator )
1ab22b2a 1569 direct-abstract-declarator [ unary-expression ]
d285084f
MD
1570 direct-abstract-declarator [ ]
1571
4fa992a5 1572pointer:
3b0f8e4d
MD
1573 * type-qualifier-list-opt
1574 * type-qualifier-list-opt pointer
4fa992a5
MD
1575
1576type-qualifier-list:
1577 type-qualifier
1578 type-qualifier-list type-qualifier
1579
4fa992a5
MD
1580typedef-name:
1581 identifier
1582
15832.3) CTF-specific declarations
1584
1585ctf-specifier:
1586 event { ctf-assignment-expression-list-opt }
1587 stream { ctf-assignment-expression-list-opt }
1588 trace { ctf-assignment-expression-list-opt }
b12919a5
MD
1589 typealias declaration-specifiers abstract-declarator-list type-assignment-operator declaration-specifiers abstract-declarator-list
1590 typealias declaration-specifiers abstract-declarator-list type-assignment-operator declarator-list
4fa992a5
MD
1591
1592ctf-type-specifier:
1593 floating_point { ctf-assignment-expression-list-opt }
1594 integer { ctf-assignment-expression-list-opt }
1595 string { ctf-assignment-expression-list-opt }
7609d3c7 1596 string
4fa992a5
MD
1597
1598ctf-assignment-expression-list:
13441bf5
MD
1599 ctf-assignment-expression ;
1600 ctf-assignment-expression-list ctf-assignment-expression ;
4fa992a5
MD
1601
1602ctf-assignment-expression:
1603 unary-expression assignment-operator unary-expression
1604 unary-expression type-assignment-operator type-specifier
eacb16d1 1605 declaration-specifiers-opt storage-class-specifier declaration-specifiers-opt declarator-list
6143bab7
MD
1606 typealias declaration-specifiers abstract-declarator-list type-assignment-operator declaration-specifiers abstract-declarator-list
1607 typealias declaration-specifiers abstract-declarator-list type-assignment-operator declarator-list
This page took 0.186159 seconds and 4 git commands to generate.