Update struct event_packet_context
[ctf.git] / common-trace-format-specification.txt
1 Common Trace Format (CTF) Specification (pre-v1.8)
2
3 Mathieu Desnoyers, EfficiOS Inc.
4
5 The goal of the present document is to specify a trace format that suits the
6 needs of the embedded, telecom, high-performance and kernel communities. It is
7 based on the Common Trace Format Requirements (v1.4) document. It is designed to
8 allow traces to be natively generated by the Linux kernel, Linux user-space
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.
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
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
25 The CE Workgroup of the Linux Foundation, Ericsson, and EfficiOS have
26 sponsored this work.
27
28
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
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
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
57 6.2 Event Context
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
66 7.3.2 Static and Dynamic Scopes
67 7.4 TSDL Examples
68
69
70 1. Preliminary definitions
71
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.
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
79 an event stream.
80 e.g. event: irq_entry.
81 - An event (or event record) relates to a specific instance of an event
82 class.
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.
86
87
88 2. High-level representation of a trace
89
90 A trace is divided into multiple event streams. Each event stream contains a
91 subset of the trace event types.
92
93 The final output of the trace, after its generation and optional transport over
94 the network, is expected to be either on permanent or temporary storage in a
95 virtual file system. Because each event stream is appended to while a trace is
96 being recorded, each is associated with a distinct set of files for
97 output. Therefore, a stored trace can be represented as a directory
98 containing zero, one or more files per stream.
99
100 Meta-data description associated with the trace contains information on
101 trace event types expressed in the Trace Stream Description Language
102 (TSDL). This language describes:
103
104 - Trace version.
105 - Types available.
106 - Per-trace event header description.
107 - Per-stream event header description.
108 - Per-stream event context description.
109 - Per-event
110 - Event type to stream mapping.
111 - Event type to name mapping.
112 - Event type to ID mapping.
113 - Event context description.
114 - Event fields description.
115
116
117 3. Event stream
118
119 An event stream can be divided into contiguous event packets of variable
120 size. These subdivisions have a variable size. An event packet can
121 contain a certain amount of padding at the end. The stream header is
122 repeated at the beginning of each event packet. The rationale for the
123 event stream design choices is explained in Appendix B. Stream Header
124 Rationale.
125
126 The event stream header will therefore be referred to as the "event packet
127 header" throughout the rest of this document.
128
129
130 4. Types
131
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
135 4.1 Basic types
136
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.
139
140 4.1.1 Type inheritance
141
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.
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
152 "Integers" section.
153
154 Each basic type must specify its alignment, in bits. Examples of
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.
167
168 TSDL meta-data attribute representation of a specific alignment:
169
170 align = value; /* value in bits */
171
172 4.1.3 Byte order
173
174 By default, the native endianness of the source architecture the trace is used.
175 Byte order can be overridden for a basic type by specifying a "byte_order"
176 attribute. Typical use-case is to specify the network byte order (big endian:
177 "be") to save data captured from the network into the trace without conversion.
178 If not specified, the byte order is native.
179
180 TSDL meta-data representation:
181
182 byte_order = native OR network OR be OR le; /* network and be are aliases */
183
184 4.1.4 Size
185
186 Type size, in bits, for integers and floats is that returned by "sizeof()" in C
187 multiplied by CHAR_BIT.
188 We require the size of "char" and "unsigned char" types (CHAR_BIT) to be fixed
189 to 8 bits for cross-endianness compatibility.
190
191 TSDL meta-data representation:
192
193 size = value; (value is in bits)
194
195 4.1.5 Integers
196
197 Signed integers are represented in two-complement. Integer alignment,
198 size, signedness and byte ordering are defined in the TSDL meta-data.
199 Integers aligned on byte size (8-bit) and with length multiple of byte
200 size (8-bit) correspond to the C99 standard integers. In addition,
201 integers with alignment and/or size that are _not_ a multiple of the
202 byte size are permitted; these correspond to the C99 standard bitfields,
203 with the added specification that the CTF integer bitfields have a fixed
204 binary representation. A MIT-licensed reference implementation of the
205 CTF portable bitfields is available at:
206
207 http://git.efficios.com/?p=babeltrace.git;a=blob;f=include/babeltrace/bitfield.h
208
209 Binary representation of integers:
210
211 - On little and big endian:
212 - Within a byte, high bits correspond to an integer high bits, and low bits
213 correspond to low bits.
214 - On little endian:
215 - Integer across multiple bytes are placed from the less significant to the
216 most significant.
217 - Consecutive integers are placed from lower bits to higher bits (even within
218 a byte).
219 - On big endian:
220 - Integer across multiple bytes are placed from the most significant to the
221 less significant.
222 - Consecutive integers are placed from higher bits to lower bits (even within
223 a byte).
224
225 This binary representation is derived from the bitfield implementation in GCC
226 for little and big endian. However, contrary to what GCC does, integers can
227 cross units boundaries (no padding is required). Padding can be explicitly
228 added (see 4.1.6 GNU/C bitfields) to follow the GCC layout if needed.
229
230 TSDL meta-data representation:
231
232 integer {
233 signed = true OR false; /* default false */
234 byte_order = native OR network OR be OR le; /* default native */
235 size = value; /* value in bits, no default */
236 align = value; /* value in bits */
237 /* based used for pretty-printing output, default: decimal. */
238 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
239 OR octal OR oct OR o OR 8 OR binary OR b OR 2;
240 /* character encoding, default: none */
241 encoding = none or UTF8 or ASCII;
242 }
243
244 Example of type inheritance (creation of a uint32_t named type):
245
246 typealias integer {
247 size = 32;
248 signed = false;
249 align = 32;
250 } := uint32_t;
251
252 Definition of a named 5-bit signed bitfield:
253
254 typealias integer {
255 size = 5;
256 signed = true;
257 align = 1;
258 } := int5_t;
259
260 The character encoding field can be used to specify that the integer
261 must be printed as a text character when read. e.g.:
262
263 typealias integer {
264 size = 8;
265 align = 8;
266 signed = false;
267 encoding = UTF8;
268 } := utf_char;
269
270
271 4.1.6 GNU/C bitfields
272
273 The GNU/C bitfields follow closely the integer representation, with a
274 particularity on alignment: if a bitfield cannot fit in the current unit, the
275 unit is padded and the bitfield starts at the following unit. The unit size is
276 defined by the size of the type "unit_type".
277
278 TSDL meta-data representation:
279
280 unit_type name:size;
281
282 As an example, the following structure declared in C compiled by GCC:
283
284 struct example {
285 short a:12;
286 short b:5;
287 };
288
289 The example structure is aligned on the largest element (short). The second
290 bitfield would be aligned on the next unit boundary, because it would not fit in
291 the current unit.
292
293 4.1.7 Floating point
294
295 The floating point values byte ordering is defined in the TSDL meta-data.
296
297 Floating point values follow the IEEE 754-2008 standard interchange formats.
298 Description of the floating point values include the exponent and mantissa size
299 in bits. Some requirements are imposed on the floating point values:
300
301 - FLT_RADIX must be 2.
302 - mant_dig is the number of digits represented in the mantissa. It is specified
303 by the ISO C99 standard, section 5.2.4, as FLT_MANT_DIG, DBL_MANT_DIG and
304 LDBL_MANT_DIG as defined by <float.h>.
305 - exp_dig is the number of digits represented in the exponent. Given that
306 mant_dig is one bit more than its actual size in bits (leading 1 is not
307 needed) and also given that the sign bit always takes one bit, exp_dig can be
308 specified as:
309
310 - sizeof(float) * CHAR_BIT - FLT_MANT_DIG
311 - sizeof(double) * CHAR_BIT - DBL_MANT_DIG
312 - sizeof(long double) * CHAR_BIT - LDBL_MANT_DIG
313
314 TSDL meta-data representation:
315
316 floating_point {
317 exp_dig = value;
318 mant_dig = value;
319 byte_order = native OR network OR be OR le;
320 align = value;
321 }
322
323 Example of type inheritance:
324
325 typealias floating_point {
326 exp_dig = 8; /* sizeof(float) * CHAR_BIT - FLT_MANT_DIG */
327 mant_dig = 24; /* FLT_MANT_DIG */
328 byte_order = native;
329 align = 32;
330 } := float;
331
332 TODO: define NaN, +inf, -inf behavior.
333
334 Bit-packed, byte-packed or larger alignments can be used for floating
335 point values, similarly to integers.
336
337 4.1.8 Enumerations
338
339 Enumerations are a mapping between an integer type and a table of strings. The
340 numerical representation of the enumeration follows the integer type specified
341 by the meta-data. The enumeration mapping table is detailed in the enumeration
342 description within the meta-data. The mapping table maps inclusive value
343 ranges (or single values) to strings. Instead of being limited to simple
344 "value -> string" mappings, these enumerations map
345 "[ start_value ... end_value ] -> string", which map inclusive ranges of
346 values to strings. An enumeration from the C language can be represented in
347 this format by having the same start_value and end_value for each element, which
348 is in fact a range of size 1. This single-value range is supported without
349 repeating the start and end values with the value = string declaration.
350
351 enum name : integer_type {
352 somestring = start_value1 ... end_value1,
353 "other string" = start_value2 ... end_value2,
354 yet_another_string, /* will be assigned to end_value2 + 1 */
355 "some other string" = value,
356 ...
357 };
358
359 If the values are omitted, the enumeration starts at 0 and increment of 1 for
360 each entry:
361
362 enum name : unsigned int {
363 ZERO,
364 ONE,
365 TWO,
366 TEN = 10,
367 ELEVEN,
368 };
369
370 Overlapping ranges within a single enumeration are implementation defined.
371
372 A nameless enumeration can be declared as a field type or as part of a typedef:
373
374 enum : integer_type {
375 ...
376 }
377
378 Enumerations omitting the container type ": integer_type" use the "int"
379 type (for compatibility with C99). The "int" type must be previously
380 declared. E.g.:
381
382 typealias integer { size = 32; align = 32; signed = true } := int;
383
384 enum {
385 ...
386 }
387
388
389 4.2 Compound types
390
391 Compound are aggregation of type declarations. Compound types include
392 structures, variant, arrays, sequences, and strings.
393
394 4.2.1 Structures
395
396 Structures are aligned on the largest alignment required by basic types
397 contained within the structure. (This follows the ISO/C standard for structures)
398
399 TSDL meta-data representation of a named structure:
400
401 struct name {
402 field_type field_name;
403 field_type field_name;
404 ...
405 };
406
407 Example:
408
409 struct example {
410 integer { /* Nameless type */
411 size = 16;
412 signed = true;
413 align = 16;
414 } first_field_name;
415 uint64_t second_field_name; /* Named type declared in the meta-data */
416 };
417
418 The fields are placed in a sequence next to each other. They each
419 possess a field name, which is a unique identifier within the structure.
420 The identifier is not allowed to use any reserved keyword
421 (see Section C.1.2). Replacing reserved keywords with
422 underscore-prefixed field names is recommended.
423
424 A nameless structure can be declared as a field type or as part of a typedef:
425
426 struct {
427 ...
428 }
429
430 Alignment for a structure compound type can be forced to a minimum value
431 by adding an "align" specifier after the declaration of a structure
432 body. This attribute is read as: align(value). The value is specified in
433 bits. The structure will be aligned on the maximum value between this
434 attribute and the alignment required by the basic types contained within
435 the structure. e.g.
436
437 struct {
438 ...
439 } align(32)
440
441 4.2.2 Variants (Discriminated/Tagged Unions)
442
443 A CTF variant is a selection between different types. A CTF variant must
444 always be defined within the scope of a structure or within fields
445 contained within a structure (defined recursively). A "tag" enumeration
446 field must appear in either the same static scope, prior to the variant
447 field (in field declaration order), in an upper static scope , or in an
448 upper dynamic scope (see Section 7.3.2). The type selection is indicated
449 by the mapping from the enumeration value to the string used as variant
450 type selector. The field to use as tag is specified by the "tag_field",
451 specified between "< >" after the "variant" keyword for unnamed
452 variants, and after "variant name" for named variants.
453
454 The alignment of the variant is the alignment of the type as selected by the tag
455 value for the specific instance of the variant. The alignment of the type
456 containing the variant is independent of the variant alignment. The size of the
457 variant is the size as selected by the tag value for the specific instance of
458 the variant.
459
460 Each variant type selector possess a field name, which is a unique
461 identifier within the variant. The identifier is not allowed to use any
462 reserved keyword (see Section C.1.2). Replacing reserved keywords with
463 underscore-prefixed field names is recommended.
464
465 A named variant declaration followed by its definition within a structure
466 declaration:
467
468 variant name {
469 field_type sel1;
470 field_type sel2;
471 field_type sel3;
472 ...
473 };
474
475 struct {
476 enum : integer_type { sel1, sel2, sel3, ... } tag_field;
477 ...
478 variant name <tag_field> v;
479 }
480
481 An unnamed variant definition within a structure is expressed by the following
482 TSDL meta-data:
483
484 struct {
485 enum : integer_type { sel1, sel2, sel3, ... } tag_field;
486 ...
487 variant <tag_field> {
488 field_type sel1;
489 field_type sel2;
490 field_type sel3;
491 ...
492 } v;
493 }
494
495 Example of a named variant within a sequence that refers to a single tag field:
496
497 variant example {
498 uint32_t a;
499 uint64_t b;
500 short c;
501 };
502
503 struct {
504 enum : uint2_t { a, b, c } choice;
505 unsigned int seqlen;
506 variant example <choice> v[seqlen];
507 }
508
509 Example of an unnamed variant:
510
511 struct {
512 enum : uint2_t { a, b, c, d } choice;
513 /* Unrelated fields can be added between the variant and its tag */
514 int32_t somevalue;
515 variant <choice> {
516 uint32_t a;
517 uint64_t b;
518 short c;
519 struct {
520 unsigned int field1;
521 uint64_t field2;
522 } d;
523 } s;
524 }
525
526 Example of an unnamed variant within an array:
527
528 struct {
529 enum : uint2_t { a, b, c } choice;
530 variant <choice> {
531 uint32_t a;
532 uint64_t b;
533 short c;
534 } v[10];
535 }
536
537 Example of a variant type definition within a structure, where the defined type
538 is then declared within an array of structures. This variant refers to a tag
539 located in an upper static scope. This example clearly shows that a variant
540 type definition referring to the tag "x" uses the closest preceding field from
541 the static scope of the type definition.
542
543 struct {
544 enum : uint2_t { a, b, c, d } x;
545
546 typedef variant <x> { /*
547 * "x" refers to the preceding "x" enumeration in the
548 * static scope of the type definition.
549 */
550 uint32_t a;
551 uint64_t b;
552 short c;
553 } example_variant;
554
555 struct {
556 enum : int { x, y, z } x; /* This enumeration is not used by "v". */
557 example_variant v; /*
558 * "v" uses the "enum : uint2_t { a, b, c, d }"
559 * tag.
560 */
561 } a[10];
562 }
563
564 4.2.3 Arrays
565
566 Arrays are fixed-length. Their length is declared in the type
567 declaration within the meta-data. They contain an array of "inner type"
568 elements, which can refer to any type not containing the type of the
569 array being declared (no circular dependency). The length is the number
570 of elements in an array.
571
572 TSDL meta-data representation of a named array:
573
574 typedef elem_type name[length];
575
576 A nameless array can be declared as a field type within a structure, e.g.:
577
578 uint8_t field_name[10];
579
580 Arrays are always aligned on their element alignment requirement.
581
582 4.2.4 Sequences
583
584 Sequences are dynamically-sized arrays. They refer to a a "length"
585 unsigned integer field, which must appear in either the same static scope,
586 prior to the sequence field (in field declaration order), in an upper
587 static scope, or in an upper dynamic scope (see Section 7.3.2). This
588 length field represents the number of elements in the sequence. The
589 sequence per se is an array of "inner type" elements.
590
591 TSDL meta-data representation for a sequence type definition:
592
593 struct {
594 unsigned int length_field;
595 typedef elem_type typename[length_field];
596 typename seq_field_name;
597 }
598
599 A sequence can also be declared as a field type, e.g.:
600
601 struct {
602 unsigned int length_field;
603 long seq_field_name[length_field];
604 }
605
606 Multiple sequences can refer to the same length field, and these length
607 fields can be in a different upper dynamic scope:
608
609 e.g., assuming the stream.event.header defines:
610
611 stream {
612 ...
613 id = 1;
614 event.header := struct {
615 uint16_t seq_len;
616 };
617 };
618
619 event {
620 ...
621 stream_id = 1;
622 fields := struct {
623 long seq_a[stream.event.header.seq_len];
624 char seq_b[stream.event.header.seq_len];
625 };
626 };
627
628 The sequence elements follow the "array" specifications.
629
630 4.2.5 Strings
631
632 Strings are an array of bytes of variable size and are terminated by a '\0'
633 "NULL" character. Their encoding is described in the TSDL meta-data. In
634 absence of encoding attribute information, the default encoding is
635 UTF-8.
636
637 TSDL meta-data representation of a named string type:
638
639 typealias string {
640 encoding = UTF8 OR ASCII;
641 } := name;
642
643 A nameless string type can be declared as a field type:
644
645 string field_name; /* Use default UTF8 encoding */
646
647 Strings are always aligned on byte size.
648
649 5. Event Packet Header
650
651 The event packet header consists of two parts: the "event packet header"
652 is the same for all streams of a trace. The second part, the "event
653 packet context", is described on a per-stream basis. Both are described
654 in the TSDL meta-data. The packets are aligned on architecture-page-sized
655 addresses.
656
657 Event packet header (all fields are optional, specified by TSDL meta-data):
658
659 - Magic number (CTF magic number: 0xC1FC1FC1) specifies that this is a
660 CTF packet. This magic number is optional, but when present, it should
661 come at the very beginning of the packet.
662 - Trace UUID, used to ensure the event packet match the meta-data used.
663 (note: we cannot use a meta-data checksum in every cases instead of a
664 UUID because meta-data can be appended to while tracing is active)
665 This field is optional.
666 - Stream ID, used as reference to stream description in meta-data.
667 This field is optional if there is only one stream description in the
668 meta-data, but becomes required if there are more than one stream in
669 the TSDL meta-data description.
670
671 Event packet context (all fields are optional, specified by TSDL meta-data):
672
673 - Event packet content size (in bits).
674 - Event packet size (in bits, includes padding).
675 - Event packet content checksum. Checksum excludes the event packet
676 header.
677 - Per-stream event packet sequence count (to deal with UDP packet loss). The
678 number of significant sequence counter bits should also be present, so
679 wrap-arounds are dealt with correctly.
680 - Time-stamp at the beginning and time-stamp at the end of the event packet.
681 Both timestamps are written in the packet header, but sampled respectively
682 while (or before) writing the first event and while (or after) writing the
683 last event in the packet. The inclusive range between these timestamps should
684 include all event timestamps assigned to events contained within the packet.
685 - Events discarded count
686 - Snapshot of a per-stream free-running counter, counting the number of
687 events discarded that were supposed to be written in the stream prior to
688 the first event in the event packet.
689 * Note: producer-consumer buffer full condition should fill the current
690 event packet with padding so we know exactly where events have been
691 discarded.
692 - Lossless compression scheme used for the event packet content. Applied
693 directly to raw data. New types of compression can be added in following
694 versions of the format.
695 0: no compression scheme
696 1: bzip2
697 2: gzip
698 3: xz
699 - Cypher used for the event packet content. Applied after compression.
700 0: no encryption
701 1: AES
702 - Checksum scheme used for the event packet content. Applied after encryption.
703 0: no checksum
704 1: md5
705 2: sha1
706 3: crc32
707
708 5.1 Event Packet Header Description
709
710 The event packet header layout is indicated by the trace packet.header
711 field. Here is a recommended structure type for the packet header with
712 the fields typically expected (although these fields are each optional):
713
714 struct event_packet_header {
715 uint32_t magic;
716 uint8_t uuid[16];
717 uint32_t stream_id;
718 };
719
720 trace {
721 ...
722 packet.header := struct event_packet_header;
723 };
724
725 If the magic number is not present, tools such as "file" will have no
726 mean to discover the file type.
727
728 If the uuid is not present, no validation that the meta-data actually
729 corresponds to the stream is performed.
730
731 If the stream_id packet header field is missing, the trace can only
732 contain a single stream. Its "id" field can be left out, and its events
733 don't need to declare a "stream_id" field.
734
735
736 5.2 Event Packet Context Description
737
738 Event packet context example. These are declared within the stream declaration
739 in the meta-data. All these fields are optional. If the packet size field is
740 missing, the whole stream only contains a single packet. If the content
741 size field is missing, the packet is filled (no padding). The content
742 and packet sizes include all headers.
743
744 An example event packet context type:
745
746 struct event_packet_context {
747 uint64_t timestamp_begin;
748 uint64_t timestamp_end;
749 uint32_t checksum;
750 uint32_t stream_packet_count;
751 uint32_t events_discarded;
752 uint32_t cpu_id;
753 uint32_t/uint16_t content_size;
754 uint32_t/uint16_t packet_size;
755 uint8_t compression_scheme;
756 uint8_t encryption_scheme;
757 uint8_t checksum_scheme;
758 };
759
760
761 6. Event Structure
762
763 The overall structure of an event is:
764
765 1 - Stream Packet Context (as specified by the stream meta-data)
766 2 - Event Header (as specified by the stream meta-data)
767 3 - Stream Event Context (as specified by the stream meta-data)
768 4 - Event Context (as specified by the event meta-data)
769 5 - Event Payload (as specified by the event meta-data)
770
771 This structure defines an implicit dynamic scoping, where variants
772 located in inner structures (those with a higher number in the listing
773 above) can refer to the fields of outer structures (with lower number in
774 the listing above). See Section 7.3 TSDL Scopes for more detail.
775
776 6.1 Event Header
777
778 Event headers can be described within the meta-data. We hereby propose, as an
779 example, two types of events headers. Type 1 accommodates streams with less than
780 31 event IDs. Type 2 accommodates streams with 31 or more event IDs.
781
782 One major factor can vary between streams: the number of event IDs assigned to
783 a stream. Luckily, this information tends to stay relatively constant (modulo
784 event registration while trace is being recorded), so we can specify different
785 representations for streams containing few event IDs and streams containing
786 many event IDs, so we end up representing the event ID and time-stamp as
787 densely as possible in each case.
788
789 The header is extended in the rare occasions where the information cannot be
790 represented in the ranges available in the standard event header. They are also
791 used in the rare occasions where the data required for a field could not be
792 collected: the flag corresponding to the missing field within the missing_fields
793 array is then set to 1.
794
795 Types uintX_t represent an X-bit unsigned integer, as declared with
796 either:
797
798 typealias integer { size = X; align = X; signed = false } := uintX_t;
799
800 or
801
802 typealias integer { size = X; align = 1; signed = false } := uintX_t;
803
804 6.1.1 Type 1 - Few event IDs
805
806 - Aligned on 32-bit (or 8-bit if byte-packed, depending on the architecture
807 preference).
808 - Native architecture byte ordering.
809 - For "compact" selection
810 - Fixed size: 32 bits.
811 - For "extended" selection
812 - Size depends on the architecture and variant alignment.
813
814 struct event_header_1 {
815 /*
816 * id: range: 0 - 30.
817 * id 31 is reserved to indicate an extended header.
818 */
819 enum : uint5_t { compact = 0 ... 30, extended = 31 } id;
820 variant <id> {
821 struct {
822 uint27_t timestamp;
823 } compact;
824 struct {
825 uint32_t id; /* 32-bit event IDs */
826 uint64_t timestamp; /* 64-bit timestamps */
827 } extended;
828 } v;
829 } align(32); /* or align(8) */
830
831
832 6.1.2 Type 2 - Many event IDs
833
834 - Aligned on 16-bit (or 8-bit if byte-packed, depending on the architecture
835 preference).
836 - Native architecture byte ordering.
837 - For "compact" selection
838 - Size depends on the architecture and variant alignment.
839 - For "extended" selection
840 - Size depends on the architecture and variant alignment.
841
842 struct event_header_2 {
843 /*
844 * id: range: 0 - 65534.
845 * id 65535 is reserved to indicate an extended header.
846 */
847 enum : uint16_t { compact = 0 ... 65534, extended = 65535 } id;
848 variant <id> {
849 struct {
850 uint32_t timestamp;
851 } compact;
852 struct {
853 uint32_t id; /* 32-bit event IDs */
854 uint64_t timestamp; /* 64-bit timestamps */
855 } extended;
856 } v;
857 } align(16); /* or align(8) */
858
859
860 6.2 Event Context
861
862 The event context contains information relative to the current event.
863 The choice and meaning of this information is specified by the TSDL
864 stream and event meta-data descriptions. The stream context is applied
865 to all events within the stream. The stream context structure follows
866 the event header. The event context is applied to specific events. Its
867 structure follows the stream context structure.
868
869 An example of stream-level event context is to save the event payload size with
870 each event, or to save the current PID with each event. These are declared
871 within the stream declaration within the meta-data:
872
873 stream {
874 ...
875 event.context := struct {
876 uint pid;
877 uint16_t payload_size;
878 };
879 };
880
881 An example of event-specific event context is to declare a bitmap of missing
882 fields, only appended after the stream event context if the extended event
883 header is selected. NR_FIELDS is the number of fields within the event (a
884 numeric value).
885
886 event {
887 context = struct {
888 variant <id> {
889 struct { } compact;
890 struct {
891 uint1_t missing_fields[NR_FIELDS]; /* missing event fields bitmap */
892 } extended;
893 } v;
894 };
895 ...
896 }
897
898 6.3 Event Payload
899
900 An event payload contains fields specific to a given event type. The fields
901 belonging to an event type are described in the event-specific meta-data
902 within a structure type.
903
904 6.3.1 Padding
905
906 No padding at the end of the event payload. This differs from the ISO/C standard
907 for structures, but follows the CTF standard for structures. In a trace, even
908 though it makes sense to align the beginning of a structure, it really makes no
909 sense to add padding at the end of the structure, because structures are usually
910 not followed by a structure of the same type.
911
912 This trick can be done by adding a zero-length "end" field at the end of the C
913 structures, and by using the offset of this field rather than using sizeof()
914 when calculating the size of a structure (see Appendix "A. Helper macros").
915
916 6.3.2 Alignment
917
918 The event payload is aligned on the largest alignment required by types
919 contained within the payload. (This follows the ISO/C standard for structures)
920
921
922 7. Trace Stream Description Language (TSDL)
923
924 The Trace Stream Description Language (TSDL) allows expression of the
925 binary trace streams layout in a C99-like Domain Specific Language
926 (DSL).
927
928
929 7.1 Meta-data
930
931 The trace stream layout description is located in the trace meta-data.
932 The meta-data is itself located in a stream identified by its name:
933 "metadata".
934
935 The meta-data description can be expressed in two different formats:
936 text-only and packet-based. The text-only description facilitates
937 generation of meta-data and provides a convenient way to enter the
938 meta-data information by hand. The packet-based meta-data provides the
939 CTF stream packet facilities (checksumming, compression, encryption,
940 network-readiness) for meta-data stream generated and transported by a
941 tracer.
942
943 The text-only meta-data file is a plain-text TSDL description. This file
944 must begin with the following characters to identify the file as a CTF
945 TSDL text-based metadata file (without the double-quotes) :
946
947 "/* CTF"
948
949 It must be followed by a space, and the version of the specification
950 followed by the CTF trace, e.g.:
951
952 " 1.8"
953
954 These characters allow automated discovery of file type and CTF
955 specification version. They are interpreted as a the beginning of a
956 comment by the TSDL metadata parser. The comment can be continued to
957 contain extra commented characters before it is closed.
958
959 The packet-based meta-data is made of "meta-data packets", which each
960 start with a meta-data packet header. The packet-based meta-data
961 description is detected by reading the magic number "0x75D11D57" at the
962 beginning of the file. This magic number is also used to detect the
963 endianness of the architecture by trying to read the CTF magic number
964 and its counterpart in reversed endianness. The events within the
965 meta-data stream have no event header nor event context. Each event only
966 contains a "sequence" payload, which is a sequence of bits using the
967 "trace.packet.header.content_size" field as a placeholder for its length
968 (the packet header size should be substracted). The formatting of this
969 sequence of bits is a plain-text representation of the TSDL description.
970 Each meta-data packet start with a special packet header, specific to
971 the meta-data stream, which contains, exactly:
972
973 struct metadata_packet_header {
974 uint32_t magic; /* 0x75D11D57 */
975 uint8_t uuid[16]; /* Unique Universal Identifier */
976 uint32_t checksum; /* 0 if unused */
977 uint32_t content_size; /* in bits */
978 uint32_t packet_size; /* in bits */
979 uint8_t compression_scheme; /* 0 if unused */
980 uint8_t encryption_scheme; /* 0 if unused */
981 uint8_t checksum_scheme; /* 0 if unused */
982 uint8_t major; /* CTF spec version major number */
983 uint8_t minor; /* CTF spec version minor number */
984 };
985
986 The packet-based meta-data can be converted to a text-only meta-data by
987 concatenating all the strings in contains.
988
989 In the textual representation of the meta-data, the text contained
990 within "/*" and "*/", as well as within "//" and end of line, are
991 treated as comments. Boolean values can be represented as true, TRUE,
992 or 1 for true, and false, FALSE, or 0 for false. Within the string-based
993 meta-data description, the trace UUID is represented as a string of
994 hexadecimal digits and dashes "-". In the event packet header, the trace
995 UUID is represented as an array of bytes.
996
997
998 7.2 Declaration vs Definition
999
1000 A declaration associates a layout to a type, without specifying where
1001 this type is located in the event structure hierarchy (see Section 6).
1002 This therefore includes typedef, typealias, as well as all type
1003 specifiers. In certain circumstances (typedef, structure field and
1004 variant field), a declaration is followed by a declarator, which specify
1005 the newly defined type name (for typedef), or the field name (for
1006 declarations located within structure and variants). Array and sequence,
1007 declared with square brackets ("[" "]"), are part of the declarator,
1008 similarly to C99. The enumeration base type is specified by
1009 ": enum_base", which is part of the type specifier. The variant tag
1010 name, specified between "<" ">", is also part of the type specifier.
1011
1012 A definition associates a type to a location in the event structure
1013 hierarchy (see Section 6). This association is denoted by ":=", as shown
1014 in Section 7.3.
1015
1016
1017 7.3 TSDL Scopes
1018
1019 TSDL uses three different types of scoping: a lexical scope is used for
1020 declarations and type definitions, and static and dynamic scopes are
1021 used for variants references to tag fields (with relative and absolute
1022 path lookups) and for sequence references to length fields.
1023
1024 7.3.1 Lexical Scope
1025
1026 Each of "trace", "stream", "event", "struct" and "variant" have their own
1027 nestable declaration scope, within which types can be declared using "typedef"
1028 and "typealias". A root declaration scope also contains all declarations
1029 located outside of any of the aforementioned declarations. An inner
1030 declaration scope can refer to type declared within its container
1031 lexical scope prior to the inner declaration scope. Redefinition of a
1032 typedef or typealias is not valid, although hiding an upper scope
1033 typedef or typealias is allowed within a sub-scope.
1034
1035 7.3.2 Static and Dynamic Scopes
1036
1037 A local static scope consists in the scope generated by the declaration
1038 of fields within a compound type. A static scope is a local static scope
1039 augmented with the nested sub-static-scopes it contains.
1040
1041 A dynamic scope consists in the static scope augmented with the
1042 implicit event structure definition hierarchy presented at Section 6.
1043
1044 Multiple declarations of the same field name within a local static scope
1045 is not valid. It is however valid to re-use the same field name in
1046 different local scopes.
1047
1048 Nested static and dynamic scopes form lookup paths. These are used for
1049 variant tag and sequence length references. They are used at the variant
1050 and sequence definition site to look up the location of the tag field
1051 associated with a variant, and to lookup up the location of the length
1052 field associated with a sequence.
1053
1054 Variants and sequences can refer to a tag field either using a relative
1055 path or an absolute path. The relative path is relative to the scope in
1056 which the variant or sequence performing the lookup is located.
1057 Relative paths are only allowed to lookup within the same static scope,
1058 which includes its nested static scopes. Lookups targeting parent static
1059 scopes need to be performed with an absolute path.
1060
1061 Absolute path lookups use the full path including the dynamic scope
1062 followed by a "." and then the static scope. Therefore, variants (or
1063 sequences) in lower levels in the dynamic scope (e.g. event context) can
1064 refer to a tag (or length) field located in upper levels (e.g. in the
1065 event header) by specifying, in this case, the associated tag with
1066 <stream.event.header.field_name>. This allows, for instance, the event
1067 context to define a variant referring to the "id" field of the event
1068 header as selector.
1069
1070 The dynamic scope prefixes are thus:
1071
1072 - Trace Packet Header: <trace.packet.header. >,
1073 - Stream Packet Context: <stream.packet.context. >,
1074 - Event Header: <stream.event.header. >,
1075 - Stream Event Context: <stream.event.context. >,
1076 - Event Context: <event.context. >,
1077 - Event Payload: <event.fields. >.
1078
1079
1080 The target dynamic scope must be specified explicitly when referring to
1081 a field outside of the static scope (absolute scope reference). No
1082 conflict can occur between relative and dynamic paths, because the
1083 keywords "trace", "stream", and "event" are reserved, and thus
1084 not permitted as field names. It is recommended that field names
1085 clashing with CTF and C99 reserved keywords use an underscore prefix to
1086 eliminate the risk of generating a description containing an invalid
1087 field name.
1088
1089 The information available in the dynamic scopes can be thought of as the
1090 current tracing context. At trace production, information about the
1091 current context is saved into the specified scope field levels. At trace
1092 consumption, for each event, the current trace context is therefore
1093 readable by accessing the upper dynamic scopes.
1094
1095
1096 7.4 TSDL Examples
1097
1098 The grammar representing the TSDL meta-data is presented in Appendix C.
1099 TSDL Grammar. This section presents a rather lighter reading that
1100 consists in examples of TSDL meta-data, with template values.
1101
1102 The stream "id" can be left out if there is only one stream in the
1103 trace. The event "id" field can be left out if there is only one event
1104 in a stream.
1105
1106 trace {
1107 major = value; /* Trace format version */
1108 minor = value;
1109 uuid = "aaaaaaaa-aaaa-aaaa-aaaa-aaaaaaaaaaaa"; /* Trace UUID */
1110 byte_order = be OR le; /* Endianness (required) */
1111 packet.header := struct {
1112 uint32_t magic;
1113 uint8_t uuid[16];
1114 uint32_t stream_id;
1115 };
1116 };
1117
1118 stream {
1119 id = stream_id;
1120 /* Type 1 - Few event IDs; Type 2 - Many event IDs. See section 6.1. */
1121 event.header := event_header_1 OR event_header_2;
1122 event.context := struct {
1123 ...
1124 };
1125 packet.context := struct {
1126 ...
1127 };
1128 };
1129
1130 event {
1131 name = "event_name";
1132 id = value; /* Numeric identifier within the stream */
1133 stream_id = stream_id;
1134 loglevel.identifier = "loglevel_identifier";
1135 loglevel.value = value;
1136 context := struct {
1137 ...
1138 };
1139 fields := struct {
1140 ...
1141 };
1142 };
1143
1144 /* More detail on types in section 4. Types */
1145
1146 /*
1147 * Named types:
1148 *
1149 * Type declarations behave similarly to the C standard.
1150 */
1151
1152 typedef aliased_type_specifiers new_type_declarators;
1153
1154 /* e.g.: typedef struct example new_type_name[10]; */
1155
1156 /*
1157 * typealias
1158 *
1159 * The "typealias" declaration can be used to give a name (including
1160 * pointer declarator specifier) to a type. It should also be used to
1161 * map basic C types (float, int, unsigned long, ...) to a CTF type.
1162 * Typealias is a superset of "typedef": it also allows assignment of a
1163 * simple variable identifier to a type.
1164 */
1165
1166 typealias type_class {
1167 ...
1168 } := type_specifiers type_declarator;
1169
1170 /*
1171 * e.g.:
1172 * typealias integer {
1173 * size = 32;
1174 * align = 32;
1175 * signed = false;
1176 * } := struct page *;
1177 *
1178 * typealias integer {
1179 * size = 32;
1180 * align = 32;
1181 * signed = true;
1182 * } := int;
1183 */
1184
1185 struct name {
1186 ...
1187 };
1188
1189 variant name {
1190 ...
1191 };
1192
1193 enum name : integer_type {
1194 ...
1195 };
1196
1197
1198 /*
1199 * Unnamed types, contained within compound type fields, typedef or typealias.
1200 */
1201
1202 struct {
1203 ...
1204 }
1205
1206 struct {
1207 ...
1208 } align(value)
1209
1210 variant {
1211 ...
1212 }
1213
1214 enum : integer_type {
1215 ...
1216 }
1217
1218 typedef type new_type[length];
1219
1220 struct {
1221 type field_name[length];
1222 }
1223
1224 typedef type new_type[length_type];
1225
1226 struct {
1227 type field_name[length_type];
1228 }
1229
1230 integer {
1231 ...
1232 }
1233
1234 floating_point {
1235 ...
1236 }
1237
1238 struct {
1239 integer_type field_name:size; /* GNU/C bitfield */
1240 }
1241
1242 struct {
1243 string field_name;
1244 }
1245
1246
1247 A. Helper macros
1248
1249 The two following macros keep track of the size of a GNU/C structure without
1250 padding at the end by placing HEADER_END as the last field. A one byte end field
1251 is used for C90 compatibility (C99 flexible arrays could be used here). Note
1252 that this does not affect the effective structure size, which should always be
1253 calculated with the header_sizeof() helper.
1254
1255 #define HEADER_END char end_field
1256 #define header_sizeof(type) offsetof(typeof(type), end_field)
1257
1258
1259 B. Stream Header Rationale
1260
1261 An event stream is divided in contiguous event packets of variable size. These
1262 subdivisions allow the trace analyzer to perform a fast binary search by time
1263 within the stream (typically requiring to index only the event packet headers)
1264 without reading the whole stream. These subdivisions have a variable size to
1265 eliminate the need to transfer the event packet padding when partially filled
1266 event packets must be sent when streaming a trace for live viewing/analysis.
1267 An event packet can contain a certain amount of padding at the end. Dividing
1268 streams into event packets is also useful for network streaming over UDP and
1269 flight recorder mode tracing (a whole event packet can be swapped out of the
1270 buffer atomically for reading).
1271
1272 The stream header is repeated at the beginning of each event packet to allow
1273 flexibility in terms of:
1274
1275 - streaming support,
1276 - allowing arbitrary buffers to be discarded without making the trace
1277 unreadable,
1278 - allow UDP packet loss handling by either dealing with missing event packet
1279 or asking for re-transmission.
1280 - transparently support flight recorder mode,
1281 - transparently support crash dump.
1282
1283
1284 C. TSDL Grammar
1285
1286 /*
1287 * Common Trace Format (CTF) Trace Stream Description Language (TSDL) Grammar.
1288 *
1289 * Inspired from the C99 grammar:
1290 * http://www.open-std.org/jtc1/sc22/wg14/www/docs/n1124.pdf (Annex A)
1291 * and c++1x grammar (draft)
1292 * http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2011/n3291.pdf (Annex A)
1293 *
1294 * Specialized for CTF needs by including only constant and declarations from
1295 * C99 (excluding function declarations), and by adding support for variants,
1296 * sequences and CTF-specific specifiers. Enumeration container types
1297 * semantic is inspired from c++1x enum-base.
1298 */
1299
1300 1) Lexical grammar
1301
1302 1.1) Lexical elements
1303
1304 token:
1305 keyword
1306 identifier
1307 constant
1308 string-literal
1309 punctuator
1310
1311 1.2) Keywords
1312
1313 keyword: is one of
1314
1315 align
1316 const
1317 char
1318 double
1319 enum
1320 event
1321 floating_point
1322 float
1323 integer
1324 int
1325 long
1326 short
1327 signed
1328 stream
1329 string
1330 struct
1331 trace
1332 typealias
1333 typedef
1334 unsigned
1335 variant
1336 void
1337 _Bool
1338 _Complex
1339 _Imaginary
1340
1341
1342 1.3) Identifiers
1343
1344 identifier:
1345 identifier-nondigit
1346 identifier identifier-nondigit
1347 identifier digit
1348
1349 identifier-nondigit:
1350 nondigit
1351 universal-character-name
1352 any other implementation-defined characters
1353
1354 nondigit:
1355 _
1356 [a-zA-Z] /* regular expression */
1357
1358 digit:
1359 [0-9] /* regular expression */
1360
1361 1.4) Universal character names
1362
1363 universal-character-name:
1364 \u hex-quad
1365 \U hex-quad hex-quad
1366
1367 hex-quad:
1368 hexadecimal-digit hexadecimal-digit hexadecimal-digit hexadecimal-digit
1369
1370 1.5) Constants
1371
1372 constant:
1373 integer-constant
1374 enumeration-constant
1375 character-constant
1376
1377 integer-constant:
1378 decimal-constant integer-suffix-opt
1379 octal-constant integer-suffix-opt
1380 hexadecimal-constant integer-suffix-opt
1381
1382 decimal-constant:
1383 nonzero-digit
1384 decimal-constant digit
1385
1386 octal-constant:
1387 0
1388 octal-constant octal-digit
1389
1390 hexadecimal-constant:
1391 hexadecimal-prefix hexadecimal-digit
1392 hexadecimal-constant hexadecimal-digit
1393
1394 hexadecimal-prefix:
1395 0x
1396 0X
1397
1398 nonzero-digit:
1399 [1-9]
1400
1401 integer-suffix:
1402 unsigned-suffix long-suffix-opt
1403 unsigned-suffix long-long-suffix
1404 long-suffix unsigned-suffix-opt
1405 long-long-suffix unsigned-suffix-opt
1406
1407 unsigned-suffix:
1408 u
1409 U
1410
1411 long-suffix:
1412 l
1413 L
1414
1415 long-long-suffix:
1416 ll
1417 LL
1418
1419 enumeration-constant:
1420 identifier
1421 string-literal
1422
1423 character-constant:
1424 ' c-char-sequence '
1425 L' c-char-sequence '
1426
1427 c-char-sequence:
1428 c-char
1429 c-char-sequence c-char
1430
1431 c-char:
1432 any member of source charset except single-quote ('), backslash
1433 (\), or new-line character.
1434 escape-sequence
1435
1436 escape-sequence:
1437 simple-escape-sequence
1438 octal-escape-sequence
1439 hexadecimal-escape-sequence
1440 universal-character-name
1441
1442 simple-escape-sequence: one of
1443 \' \" \? \\ \a \b \f \n \r \t \v
1444
1445 octal-escape-sequence:
1446 \ octal-digit
1447 \ octal-digit octal-digit
1448 \ octal-digit octal-digit octal-digit
1449
1450 hexadecimal-escape-sequence:
1451 \x hexadecimal-digit
1452 hexadecimal-escape-sequence hexadecimal-digit
1453
1454 1.6) String literals
1455
1456 string-literal:
1457 " s-char-sequence-opt "
1458 L" s-char-sequence-opt "
1459
1460 s-char-sequence:
1461 s-char
1462 s-char-sequence s-char
1463
1464 s-char:
1465 any member of source charset except double-quote ("), backslash
1466 (\), or new-line character.
1467 escape-sequence
1468
1469 1.7) Punctuators
1470
1471 punctuator: one of
1472 [ ] ( ) { } . -> * + - < > : ; ... = ,
1473
1474
1475 2) Phrase structure grammar
1476
1477 primary-expression:
1478 identifier
1479 constant
1480 string-literal
1481 ( unary-expression )
1482
1483 postfix-expression:
1484 primary-expression
1485 postfix-expression [ unary-expression ]
1486 postfix-expression . identifier
1487 postfix-expressoin -> identifier
1488
1489 unary-expression:
1490 postfix-expression
1491 unary-operator postfix-expression
1492
1493 unary-operator: one of
1494 + -
1495
1496 assignment-operator:
1497 =
1498
1499 type-assignment-operator:
1500 :=
1501
1502 constant-expression-range:
1503 unary-expression ... unary-expression
1504
1505 2.2) Declarations:
1506
1507 declaration:
1508 declaration-specifiers declarator-list-opt ;
1509 ctf-specifier ;
1510
1511 declaration-specifiers:
1512 storage-class-specifier declaration-specifiers-opt
1513 type-specifier declaration-specifiers-opt
1514 type-qualifier declaration-specifiers-opt
1515
1516 declarator-list:
1517 declarator
1518 declarator-list , declarator
1519
1520 abstract-declarator-list:
1521 abstract-declarator
1522 abstract-declarator-list , abstract-declarator
1523
1524 storage-class-specifier:
1525 typedef
1526
1527 type-specifier:
1528 void
1529 char
1530 short
1531 int
1532 long
1533 float
1534 double
1535 signed
1536 unsigned
1537 _Bool
1538 _Complex
1539 _Imaginary
1540 struct-specifier
1541 variant-specifier
1542 enum-specifier
1543 typedef-name
1544 ctf-type-specifier
1545
1546 align-attribute:
1547 align ( unary-expression )
1548
1549 struct-specifier:
1550 struct identifier-opt { struct-or-variant-declaration-list-opt } align-attribute-opt
1551 struct identifier align-attribute-opt
1552
1553 struct-or-variant-declaration-list:
1554 struct-or-variant-declaration
1555 struct-or-variant-declaration-list struct-or-variant-declaration
1556
1557 struct-or-variant-declaration:
1558 specifier-qualifier-list struct-or-variant-declarator-list ;
1559 declaration-specifiers-opt storage-class-specifier declaration-specifiers-opt declarator-list ;
1560 typealias declaration-specifiers abstract-declarator-list type-assignment-operator declaration-specifiers abstract-declarator-list ;
1561 typealias declaration-specifiers abstract-declarator-list type-assignment-operator declarator-list ;
1562
1563 specifier-qualifier-list:
1564 type-specifier specifier-qualifier-list-opt
1565 type-qualifier specifier-qualifier-list-opt
1566
1567 struct-or-variant-declarator-list:
1568 struct-or-variant-declarator
1569 struct-or-variant-declarator-list , struct-or-variant-declarator
1570
1571 struct-or-variant-declarator:
1572 declarator
1573 declarator-opt : unary-expression
1574
1575 variant-specifier:
1576 variant identifier-opt variant-tag-opt { struct-or-variant-declaration-list }
1577 variant identifier variant-tag
1578
1579 variant-tag:
1580 < unary-expression >
1581
1582 enum-specifier:
1583 enum identifier-opt { enumerator-list }
1584 enum identifier-opt { enumerator-list , }
1585 enum identifier
1586 enum identifier-opt : declaration-specifiers { enumerator-list }
1587 enum identifier-opt : declaration-specifiers { enumerator-list , }
1588
1589 enumerator-list:
1590 enumerator
1591 enumerator-list , enumerator
1592
1593 enumerator:
1594 enumeration-constant
1595 enumeration-constant assignment-operator unary-expression
1596 enumeration-constant assignment-operator constant-expression-range
1597
1598 type-qualifier:
1599 const
1600
1601 declarator:
1602 pointer-opt direct-declarator
1603
1604 direct-declarator:
1605 identifier
1606 ( declarator )
1607 direct-declarator [ unary-expression ]
1608
1609 abstract-declarator:
1610 pointer-opt direct-abstract-declarator
1611
1612 direct-abstract-declarator:
1613 identifier-opt
1614 ( abstract-declarator )
1615 direct-abstract-declarator [ unary-expression ]
1616 direct-abstract-declarator [ ]
1617
1618 pointer:
1619 * type-qualifier-list-opt
1620 * type-qualifier-list-opt pointer
1621
1622 type-qualifier-list:
1623 type-qualifier
1624 type-qualifier-list type-qualifier
1625
1626 typedef-name:
1627 identifier
1628
1629 2.3) CTF-specific declarations
1630
1631 ctf-specifier:
1632 event { ctf-assignment-expression-list-opt }
1633 stream { ctf-assignment-expression-list-opt }
1634 trace { ctf-assignment-expression-list-opt }
1635 typealias declaration-specifiers abstract-declarator-list type-assignment-operator declaration-specifiers abstract-declarator-list
1636 typealias declaration-specifiers abstract-declarator-list type-assignment-operator declarator-list
1637
1638 ctf-type-specifier:
1639 floating_point { ctf-assignment-expression-list-opt }
1640 integer { ctf-assignment-expression-list-opt }
1641 string { ctf-assignment-expression-list-opt }
1642 string
1643
1644 ctf-assignment-expression-list:
1645 ctf-assignment-expression ;
1646 ctf-assignment-expression-list ctf-assignment-expression ;
1647
1648 ctf-assignment-expression:
1649 unary-expression assignment-operator unary-expression
1650 unary-expression type-assignment-operator type-specifier
1651 declaration-specifiers-opt storage-class-specifier declaration-specifiers-opt declarator-list
1652 typealias declaration-specifiers abstract-declarator-list type-assignment-operator declaration-specifiers abstract-declarator-list
1653 typealias declaration-specifiers abstract-declarator-list type-assignment-operator declarator-list
This page took 0.107566 seconds and 4 git commands to generate.