Grammar: remove redundant ;
[ctf.git] / common-trace-format-proposal.txt
CommitLineData
5ba9f198 1
4767a9e7 2RFC: Common Trace Format (CTF) Proposal (pre-v1.7)
5ba9f198
MD
3
4Mathieu Desnoyers, EfficiOS Inc.
5
6The goal of the present document is to propose a trace format that suits the
cc089c3a 7needs of the embedded, telecom, high-performance and kernel communities. It is
5ba9f198 8based on the Common Trace Format Requirements (v1.4) document. It is designed to
cc089c3a 9allow traces to be natively generated by the Linux kernel, Linux user-space
6c7226e9
MD
10applications written in C/C++, and hardware components. One major element of
11CTF is the Trace Stream Description Language (TSDL) which flexibility
12enables description of various binary trace stream layouts.
cc089c3a
MD
13
14The latest version of this document can be found at:
15
16 git tree: git://git.efficios.com/ctf.git
17 gitweb: http://git.efficios.com/?p=ctf.git
5ba9f198
MD
18
19A reference implementation of a library to read and write this trace format is
20being implemented within the BabelTrace project, a converter between trace
21formats. The development tree is available at:
22
23 git tree: git://git.efficios.com/babeltrace.git
24 gitweb: http://git.efficios.com/?p=babeltrace.git
25
26
271. Preliminary definitions
28
3bf79539
MD
29 - Event Trace: An ordered sequence of events.
30 - Event Stream: An ordered sequence of events, containing a subset of the
31 trace event types.
32 - Event Packet: A sequence of physically contiguous events within an event
33 stream.
5ba9f198
MD
34 - Event: This is the basic entry in a trace. (aka: a trace record).
35 - An event identifier (ID) relates to the class (a type) of event within
3bf79539
MD
36 an event stream.
37 e.g. event: irq_entry.
5ba9f198
MD
38 - An event (or event record) relates to a specific instance of an event
39 class.
3bf79539
MD
40 e.g. event: irq_entry, at time X, on CPU Y
41 - Source Architecture: Architecture writing the trace.
42 - Reader Architecture: Architecture reading the trace.
5ba9f198
MD
43
44
452. High-level representation of a trace
46
3bf79539
MD
47A trace is divided into multiple event streams. Each event stream contains a
48subset of the trace event types.
5ba9f198 49
3bf79539
MD
50The final output of the trace, after its generation and optional transport over
51the network, is expected to be either on permanent or temporary storage in a
52virtual file system. Because each event stream is appended to while a trace is
53being recorded, each is associated with a separate file for output. Therefore,
54a stored trace can be represented as a directory containing one file per stream.
5ba9f198 55
6672e9e1
MD
56Meta-data description associated with the trace contains information on
57trace event types expressed in the Trace Stream Description Language
58(TSDL). This language describes:
5ba9f198
MD
59
60- Trace version.
61- Types available.
6672e9e1 62- Per-trace event header description.
3bf79539 63- Per-stream event header description.
6672e9e1 64- Per-stream event context description.
5ba9f198 65- Per-event
3bf79539 66 - Event type to stream mapping.
5ba9f198
MD
67 - Event type to name mapping.
68 - Event type to ID mapping.
6672e9e1 69 - Event context description.
5ba9f198
MD
70 - Event fields description.
71
72
3bf79539 733. Event stream
5ba9f198 74
6672e9e1
MD
75An event stream can be divided into contiguous event packets of variable
76size. These subdivisions have a variable size. An event packet can
77contain a certain amount of padding at the end. The stream header is
78repeated at the beginning of each event packet. The rationale for the
79event stream design choices is explained in Appendix B. Stream Header
80Rationale.
5ba9f198 81
3bf79539
MD
82The event stream header will therefore be referred to as the "event packet
83header" throughout the rest of this document.
5ba9f198
MD
84
85
864. Types
87
1fad7a85
MD
88Types are organized as type classes. Each type class belong to either of two
89kind of types: basic types or compound types.
90
5ba9f198
MD
914.1 Basic types
92
1fad7a85
MD
93A basic type is a scalar type, as described in this section. It includes
94integers, GNU/C bitfields, enumerations, and floating point values.
5ba9f198
MD
95
964.1.1 Type inheritance
97
80fd2569
MD
98Type specifications can be inherited to allow deriving types from a
99type class. For example, see the uint32_t named type derived from the "integer"
100type class below ("Integers" section). Types have a precise binary
101representation in the trace. A type class has methods to read and write these
102types, but must be derived into a type to be usable in an event field.
5ba9f198
MD
103
1044.1.2 Alignment
105
106We define "byte-packed" types as aligned on the byte size, namely 8-bit.
107We define "bit-packed" types as following on the next bit, as defined by the
370eae99 108"Integers" section.
5ba9f198 109
6672e9e1 110Each basic type must specify its alignment, in bits. Examples of
ec4404a7
MD
111possible alignments are: bit-packed (align = 1), byte-packed (align =
1128), or word-aligned (e.g. align = 32 or align = 64). The choice depends
113on the architecture preference and compactness vs performance trade-offs
114of the implementation. Architectures providing fast unaligned write
115byte-packed basic types to save space, aligning each type on byte
116boundaries (8-bit). Architectures with slow unaligned writes align types
117on specific alignment values. If no specific alignment is declared for a
118type, it is assumed to be bit-packed for integers with size not multiple
119of 8 bits and for gcc bitfields. All other basic types are byte-packed
120by default. It is however recommended to always specify the alignment
121explicitly. Alignment values must be power of two. Compound types are
122aligned as specified in their individual specification.
5ba9f198 123
6672e9e1 124TSDL meta-data attribute representation of a specific alignment:
5ba9f198
MD
125
126 align = value; /* value in bits */
127
1284.1.3 Byte order
129
3bf79539
MD
130By default, the native endianness of the source architecture the trace is used.
131Byte order can be overridden for a basic type by specifying a "byte_order"
132attribute. Typical use-case is to specify the network byte order (big endian:
133"be") to save data captured from the network into the trace without conversion.
134If not specified, the byte order is native.
5ba9f198 135
6672e9e1 136TSDL meta-data representation:
5ba9f198
MD
137
138 byte_order = native OR network OR be OR le; /* network and be are aliases */
139
1404.1.4 Size
141
142Type size, in bits, for integers and floats is that returned by "sizeof()" in C
143multiplied by CHAR_BIT.
144We require the size of "char" and "unsigned char" types (CHAR_BIT) to be fixed
145to 8 bits for cross-endianness compatibility.
146
6672e9e1 147TSDL meta-data representation:
5ba9f198
MD
148
149 size = value; (value is in bits)
150
1514.1.5 Integers
152
6672e9e1
MD
153Signed integers are represented in two-complement. Integer alignment,
154size, signedness and byte ordering are defined in the TSDL meta-data.
155Integers aligned on byte size (8-bit) and with length multiple of byte
156size (8-bit) correspond to the C99 standard integers. In addition,
157integers with alignment and/or size that are _not_ a multiple of the
158byte size are permitted; these correspond to the C99 standard bitfields,
159with the added specification that the CTF integer bitfields have a fixed
160binary representation. A MIT-licensed reference implementation of the
161CTF portable bitfields is available at:
5ba9f198
MD
162
163 http://git.efficios.com/?p=babeltrace.git;a=blob;f=include/babeltrace/bitfield.h
164
165Binary representation of integers:
166
167- On little and big endian:
168 - Within a byte, high bits correspond to an integer high bits, and low bits
169 correspond to low bits.
170- On little endian:
171 - Integer across multiple bytes are placed from the less significant to the
172 most significant.
173 - Consecutive integers are placed from lower bits to higher bits (even within
174 a byte).
175- On big endian:
176 - Integer across multiple bytes are placed from the most significant to the
177 less significant.
178 - Consecutive integers are placed from higher bits to lower bits (even within
179 a byte).
180
181This binary representation is derived from the bitfield implementation in GCC
182for little and big endian. However, contrary to what GCC does, integers can
6672e9e1 183cross units boundaries (no padding is required). Padding can be explicitly
5ba9f198
MD
184added (see 4.1.6 GNU/C bitfields) to follow the GCC layout if needed.
185
6672e9e1 186TSDL meta-data representation:
5ba9f198 187
80fd2569 188 integer {
5ba9f198
MD
189 signed = true OR false; /* default false */
190 byte_order = native OR network OR be OR le; /* default native */
191 size = value; /* value in bits, no default */
192 align = value; /* value in bits */
56b77c06
MD
193 /* based used for pretty-printing output, default: decimal. */
194 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
195 OR octal OR oct OR o OR 8 OR binary OR b OR 2;
13441bf5
MD
196 /* character encoding, default: none */
197 encoding = none or UTF8 or ASCII;
2152348f 198 }
5ba9f198 199
80fd2569 200Example of type inheritance (creation of a uint32_t named type):
5ba9f198 201
359894ac 202typealias integer {
9e4e34e9 203 size = 32;
5ba9f198
MD
204 signed = false;
205 align = 32;
38b8da21 206} := uint32_t;
5ba9f198 207
80fd2569 208Definition of a named 5-bit signed bitfield:
5ba9f198 209
359894ac 210typealias integer {
5ba9f198
MD
211 size = 5;
212 signed = true;
213 align = 1;
38b8da21 214} := int5_t;
5ba9f198 215
13441bf5
MD
216The character encoding field can be used to specify that the integer
217must be printed as a text character when read. e.g.:
218
219typealias integer {
220 size = 8;
221 align = 8;
222 signed = false;
223 encoding = UTF8;
224} := utf_char;
225
226
5ba9f198
MD
2274.1.6 GNU/C bitfields
228
229The GNU/C bitfields follow closely the integer representation, with a
230particularity on alignment: if a bitfield cannot fit in the current unit, the
80fd2569
MD
231unit is padded and the bitfield starts at the following unit. The unit size is
232defined by the size of the type "unit_type".
5ba9f198 233
6672e9e1 234TSDL meta-data representation:
80fd2569 235
d674f4b8 236 unit_type name:size;
80fd2569 237
5ba9f198
MD
238As an example, the following structure declared in C compiled by GCC:
239
240struct example {
241 short a:12;
242 short b:5;
243};
244
2152348f
MD
245The example structure is aligned on the largest element (short). The second
246bitfield would be aligned on the next unit boundary, because it would not fit in
247the current unit.
5ba9f198
MD
248
2494.1.7 Floating point
250
6672e9e1 251The floating point values byte ordering is defined in the TSDL meta-data.
5ba9f198
MD
252
253Floating point values follow the IEEE 754-2008 standard interchange formats.
254Description of the floating point values include the exponent and mantissa size
255in bits. Some requirements are imposed on the floating point values:
256
257- FLT_RADIX must be 2.
258- mant_dig is the number of digits represented in the mantissa. It is specified
259 by the ISO C99 standard, section 5.2.4, as FLT_MANT_DIG, DBL_MANT_DIG and
260 LDBL_MANT_DIG as defined by <float.h>.
261- exp_dig is the number of digits represented in the exponent. Given that
262 mant_dig is one bit more than its actual size in bits (leading 1 is not
263 needed) and also given that the sign bit always takes one bit, exp_dig can be
264 specified as:
265
266 - sizeof(float) * CHAR_BIT - FLT_MANT_DIG
267 - sizeof(double) * CHAR_BIT - DBL_MANT_DIG
268 - sizeof(long double) * CHAR_BIT - LDBL_MANT_DIG
269
6672e9e1 270TSDL meta-data representation:
5ba9f198 271
80fd2569 272floating_point {
ec4404a7
MD
273 exp_dig = value;
274 mant_dig = value;
275 byte_order = native OR network OR be OR le;
276 align = value;
2152348f 277}
5ba9f198
MD
278
279Example of type inheritance:
280
359894ac 281typealias floating_point {
5ba9f198
MD
282 exp_dig = 8; /* sizeof(float) * CHAR_BIT - FLT_MANT_DIG */
283 mant_dig = 24; /* FLT_MANT_DIG */
284 byte_order = native;
ec4404a7 285 align = 32;
38b8da21 286} := float;
5ba9f198
MD
287
288TODO: define NaN, +inf, -inf behavior.
289
ec4404a7
MD
290Bit-packed, byte-packed or larger alignments can be used for floating
291point values, similarly to integers.
292
5ba9f198
MD
2934.1.8 Enumerations
294
295Enumerations are a mapping between an integer type and a table of strings. The
296numerical representation of the enumeration follows the integer type specified
6672e9e1
MD
297by the meta-data. The enumeration mapping table is detailed in the enumeration
298description within the meta-data. The mapping table maps inclusive value
299ranges (or single values) to strings. Instead of being limited to simple
3bf79539 300"value -> string" mappings, these enumerations map
80fd2569 301"[ start_value ... end_value ] -> string", which map inclusive ranges of
3bf79539
MD
302values to strings. An enumeration from the C language can be represented in
303this format by having the same start_value and end_value for each element, which
304is in fact a range of size 1. This single-value range is supported without
4767a9e7 305repeating the start and end values with the value = string declaration.
80fd2569 306
a9b83695 307enum name : integer_type {
359894ac 308 somestring = start_value1 ... end_value1,
80fd2569
MD
309 "other string" = start_value2 ... end_value2,
310 yet_another_string, /* will be assigned to end_value2 + 1 */
311 "some other string" = value,
312 ...
313};
314
315If the values are omitted, the enumeration starts at 0 and increment of 1 for
316each entry:
317
a9b83695 318enum name : unsigned int {
80fd2569
MD
319 ZERO,
320 ONE,
321 TWO,
322 TEN = 10,
323 ELEVEN,
3bf79539 324};
5ba9f198 325
80fd2569 326Overlapping ranges within a single enumeration are implementation defined.
5ba9f198 327
2152348f
MD
328A nameless enumeration can be declared as a field type or as part of a typedef:
329
a9b83695 330enum : integer_type {
2152348f
MD
331 ...
332}
333
c2742c56
MD
334Enumerations omitting the container type ": integer_type" use the "int"
335type (for compatibility with C99). The "int" type must be previously
336declared. E.g.:
337
338typealias integer { size = 32; align = 32; signed = true } := int;
339
340enum {
341 ...
342}
343
1fad7a85 344
5ba9f198
MD
3454.2 Compound types
346
1fad7a85
MD
347Compound are aggregation of type declarations. Compound types include
348structures, variant, arrays, sequences, and strings.
349
5ba9f198
MD
3504.2.1 Structures
351
352Structures are aligned on the largest alignment required by basic types
353contained within the structure. (This follows the ISO/C standard for structures)
354
6672e9e1 355TSDL meta-data representation of a named structure:
5ba9f198 356
80fd2569
MD
357struct name {
358 field_type field_name;
359 field_type field_name;
360 ...
361};
5ba9f198
MD
362
363Example:
364
80fd2569
MD
365struct example {
366 integer { /* Nameless type */
367 size = 16;
368 signed = true;
369 align = 16;
370 } first_field_name;
6672e9e1 371 uint64_t second_field_name; /* Named type declared in the meta-data */
3bf79539 372};
5ba9f198
MD
373
374The fields are placed in a sequence next to each other. They each possess a
375field name, which is a unique identifier within the structure.
376
2152348f 377A nameless structure can be declared as a field type or as part of a typedef:
80fd2569
MD
378
379struct {
380 ...
2152348f 381}
80fd2569 382
ec4404a7
MD
383Alignment for a structure compound type can be forced to a minimum value
384by adding an "align" specifier after the declaration of a structure
385body. This attribute is read as: align(value). The value is specified in
386bits. The structure will be aligned on the maximum value between this
387attribute and the alignment required by the basic types contained within
388the structure. e.g.
389
390struct {
391 ...
392} align(32)
393
77a98c82 3944.2.2 Variants (Discriminated/Tagged Unions)
fcba70d4 395
fdf2bb05
MD
396A CTF variant is a selection between different types. A CTF variant must
397always be defined within the scope of a structure or within fields
398contained within a structure (defined recursively). A "tag" enumeration
399field must appear in either the same lexical scope, prior to the variant
1ab22b2a
MD
400field (in field declaration order), in an upper lexical scope (see
401Section 7.3.1), or in an upper dynamic scope (see Section 7.3.2). The
402type selection is indicated by the mapping from the enumeration value to
403the string used as variant type selector. The field to use as tag is
404specified by the "tag_field", specified between "< >" after the
fdf2bb05
MD
405"variant" keyword for unnamed variants, and after "variant name" for
406named variants.
fcba70d4
MD
407
408The alignment of the variant is the alignment of the type as selected by the tag
409value for the specific instance of the variant. The alignment of the type
410containing the variant is independent of the variant alignment. The size of the
411variant is the size as selected by the tag value for the specific instance of
412the variant.
413
414A named variant declaration followed by its definition within a structure
415declaration:
416
417variant name {
418 field_type sel1;
419 field_type sel2;
420 field_type sel3;
421 ...
422};
423
424struct {
a9b83695 425 enum : integer_type { sel1, sel2, sel3, ... } tag_field;
fcba70d4
MD
426 ...
427 variant name <tag_field> v;
428}
429
430An unnamed variant definition within a structure is expressed by the following
6672e9e1 431TSDL meta-data:
fcba70d4
MD
432
433struct {
a9b83695 434 enum : integer_type { sel1, sel2, sel3, ... } tag_field;
fcba70d4
MD
435 ...
436 variant <tag_field> {
437 field_type sel1;
438 field_type sel2;
439 field_type sel3;
440 ...
441 } v;
442}
443
444Example of a named variant within a sequence that refers to a single tag field:
445
446variant example {
447 uint32_t a;
448 uint64_t b;
449 short c;
450};
451
452struct {
a9b83695 453 enum : uint2_t { a, b, c } choice;
1ab22b2a
MD
454 unsigned int seqlen;
455 variant example <choice> v[seqlen];
fcba70d4
MD
456}
457
458Example of an unnamed variant:
459
460struct {
a9b83695 461 enum : uint2_t { a, b, c, d } choice;
fcba70d4
MD
462 /* Unrelated fields can be added between the variant and its tag */
463 int32_t somevalue;
464 variant <choice> {
465 uint32_t a;
466 uint64_t b;
467 short c;
468 struct {
469 unsigned int field1;
470 uint64_t field2;
471 } d;
472 } s;
473}
474
475Example of an unnamed variant within an array:
476
477struct {
a9b83695 478 enum : uint2_t { a, b, c } choice;
fcba70d4
MD
479 variant <choice> {
480 uint32_t a;
481 uint64_t b;
482 short c;
15850440 483 } v[10];
fcba70d4
MD
484}
485
486Example of a variant type definition within a structure, where the defined type
487is then declared within an array of structures. This variant refers to a tag
488located in an upper lexical scope. This example clearly shows that a variant
489type definition referring to the tag "x" uses the closest preceding field from
490the lexical scope of the type definition.
491
492struct {
a9b83695 493 enum : uint2_t { a, b, c, d } x;
fcba70d4
MD
494
495 typedef variant <x> { /*
496 * "x" refers to the preceding "x" enumeration in the
497 * lexical scope of the type definition.
498 */
499 uint32_t a;
500 uint64_t b;
501 short c;
502 } example_variant;
503
504 struct {
a9b83695 505 enum : int { x, y, z } x; /* This enumeration is not used by "v". */
fcba70d4 506 example_variant v; /*
a9b83695 507 * "v" uses the "enum : uint2_t { a, b, c, d }"
fcba70d4
MD
508 * tag.
509 */
510 } a[10];
511}
512
5134.2.3 Arrays
5ba9f198 514
6672e9e1
MD
515Arrays are fixed-length. Their length is declared in the type
516declaration within the meta-data. They contain an array of "inner type"
517elements, which can refer to any type not containing the type of the
518array being declared (no circular dependency). The length is the number
519of elements in an array.
5ba9f198 520
6672e9e1 521TSDL meta-data representation of a named array:
80fd2569
MD
522
523typedef elem_type name[length];
5ba9f198 524
2152348f 525A nameless array can be declared as a field type within a structure, e.g.:
5ba9f198 526
2152348f 527 uint8_t field_name[10];
80fd2569 528
ec4404a7 529Arrays are always aligned on their element alignment requirement.
5ba9f198 530
fcba70d4 5314.2.4 Sequences
5ba9f198 532
1ab22b2a
MD
533Sequences are dynamically-sized arrays. They refer to a a "length"
534unsigned integer field, which must appear in either the same lexical scope,
535prior to the sequence field (in field declaration order), in an upper
536lexical scope (see Section 7.3.1), or in an upper dynamic scope (see
537Section 7.3.2). This length field represents the number of elements in
538the sequence. The sequence per se is an array of "inner type" elements.
5ba9f198 539
1ab22b2a 540TSDL meta-data representation for a sequence type definition:
80fd2569 541
1ab22b2a
MD
542struct {
543 unsigned int length_field;
544 typedef elem_type typename[length_field];
545 typename seq_field_name;
546}
547
548A sequence can also be declared as a field type, e.g.:
80fd2569 549
1ab22b2a
MD
550struct {
551 unsigned int length_field;
552 long seq_field_name[length_field];
553}
80fd2569 554
1ab22b2a
MD
555Multiple sequences can refer to the same length field, and these length
556fields can be in a different upper dynamic scope:
557
558e.g., assuming the stream.event.header defines:
559
560stream {
561 ...
562 id = 1;
563 event.header := struct {
564 uint16_t seq_len;
565 };
566};
567
568event {
569 ...
570 stream_id = 1;
571 fields := struct {
572 long seq_a[stream.event.header.seq_len];
573 char seq_b[stream.event.header.seq_len];
574 };
575};
80fd2569 576
1ab22b2a 577The sequence elements follow the "array" specifications.
5ba9f198 578
fcba70d4 5794.2.5 Strings
5ba9f198
MD
580
581Strings are an array of bytes of variable size and are terminated by a '\0'
6672e9e1
MD
582"NULL" character. Their encoding is described in the TSDL meta-data. In
583absence of encoding attribute information, the default encoding is
584UTF-8.
5ba9f198 585
6672e9e1 586TSDL meta-data representation of a named string type:
80fd2569 587
359894ac 588typealias string {
5ba9f198 589 encoding = UTF8 OR ASCII;
38b8da21 590} := name;
5ba9f198 591
80fd2569
MD
592A nameless string type can be declared as a field type:
593
594string field_name; /* Use default UTF8 encoding */
5ba9f198 595
ec4404a7
MD
596Strings are always aligned on byte size.
597
3bf79539
MD
5985. Event Packet Header
599
6672e9e1
MD
600The event packet header consists of two parts: the "event packet header"
601is the same for all streams of a trace. The second part, the "event
602packet context", is described on a per-stream basis. Both are described
603in the TSDL meta-data. The packets are aligned on architecture-page-sized
604addresses.
3bf79539 605
6672e9e1 606Event packet header (all fields are optional, specified by TSDL meta-data):
3bf79539 607
6672e9e1
MD
608- Magic number (CTF magic number: 0xC1FC1FC1) specifies that this is a
609 CTF packet. This magic number is optional, but when present, it should
610 come at the very beginning of the packet.
611- Trace UUID, used to ensure the event packet match the meta-data used.
612 (note: we cannot use a meta-data checksum in every cases instead of a
613 UUID because meta-data can be appended to while tracing is active)
614 This field is optional.
615- Stream ID, used as reference to stream description in meta-data.
616 This field is optional if there is only one stream description in the
617 meta-data, but becomes required if there are more than one stream in
618 the TSDL meta-data description.
3bf79539 619
6672e9e1 620Event packet context (all fields are optional, specified by TSDL meta-data):
3bf79539
MD
621
622- Event packet content size (in bytes).
623- Event packet size (in bytes, includes padding).
624- Event packet content checksum (optional). Checksum excludes the event packet
625 header.
626- Per-stream event packet sequence count (to deal with UDP packet loss). The
627 number of significant sequence counter bits should also be present, so
b11853af 628 wrap-arounds are dealt with correctly.
6672e9e1 629- Time-stamp at the beginning and time-stamp at the end of the event packet.
3bf79539
MD
630 Both timestamps are written in the packet header, but sampled respectively
631 while (or before) writing the first event and while (or after) writing the
632 last event in the packet. The inclusive range between these timestamps should
633 include all event timestamps assigned to events contained within the packet.
5ba9f198 634- Events discarded count
3bf79539
MD
635 - Snapshot of a per-stream free-running counter, counting the number of
636 events discarded that were supposed to be written in the stream prior to
637 the first event in the event packet.
5ba9f198 638 * Note: producer-consumer buffer full condition should fill the current
3bf79539 639 event packet with padding so we know exactly where events have been
5ba9f198 640 discarded.
3bf79539
MD
641- Lossless compression scheme used for the event packet content. Applied
642 directly to raw data. New types of compression can be added in following
643 versions of the format.
5ba9f198
MD
644 0: no compression scheme
645 1: bzip2
646 2: gzip
3bf79539
MD
647 3: xz
648- Cypher used for the event packet content. Applied after compression.
5ba9f198
MD
649 0: no encryption
650 1: AES
3bf79539 651- Checksum scheme used for the event packet content. Applied after encryption.
5ba9f198
MD
652 0: no checksum
653 1: md5
654 2: sha1
655 3: crc32
656
6672e9e1 6575.1 Event Packet Header Description
3bf79539 658
fc5425db 659The event packet header layout is indicated by the trace packet.header
6672e9e1
MD
660field. Here is a recommended structure type for the packet header with
661the fields typically expected (although these fields are each optional):
fc5425db 662
80fd2569
MD
663struct event_packet_header {
664 uint32_t magic;
3fde5da1 665 uint8_t uuid[16];
3bf79539 666 uint32_t stream_id;
80fd2569 667};
5ba9f198 668
fc5425db
MD
669trace {
670 ...
671 packet.header := struct event_packet_header;
672};
673
6672e9e1
MD
674If the magic number is not present, tools such as "file" will have no
675mean to discover the file type.
676
3fde5da1
MD
677If the uuid is not present, no validation that the meta-data actually
678corresponds to the stream is performed.
37b39e47
MD
679
680If the stream_id packet header field is missing, the trace can only
681contain a single stream. Its "id" field can be left out, and its events
682don't need to declare a "stream_id" field.
683
684
3bf79539
MD
6855.2 Event Packet Context Description
686
687Event packet context example. These are declared within the stream declaration
6672e9e1 688in the meta-data. All these fields are optional. If the packet size field is
6a7c61df 689missing, the whole stream only contains a single packet. If the content
f6b81726
MD
690size field is missing, the packet is filled (no padding). The content
691and packet sizes include all headers.
3bf79539
MD
692
693An example event packet context type:
694
80fd2569 695struct event_packet_context {
3bf79539
MD
696 uint64_t timestamp_begin;
697 uint64_t timestamp_end;
698 uint32_t checksum;
699 uint32_t stream_packet_count;
700 uint32_t events_discarded;
701 uint32_t cpu_id;
702 uint32_t/uint16_t content_size;
703 uint32_t/uint16_t packet_size;
704 uint8_t stream_packet_count_bits; /* Significant counter bits */
705 uint8_t compression_scheme;
706 uint8_t encryption_scheme;
3b0f8e4d 707 uint8_t checksum_scheme;
3bf79539 708};
5ba9f198 709
fcba70d4 710
5ba9f198
MD
7116. Event Structure
712
713The overall structure of an event is:
714
6672e9e1
MD
7151 - Stream Packet Context (as specified by the stream meta-data)
716 2 - Event Header (as specified by the stream meta-data)
717 3 - Stream Event Context (as specified by the stream meta-data)
718 4 - Event Context (as specified by the event meta-data)
719 5 - Event Payload (as specified by the event meta-data)
5ba9f198 720
fdf2bb05 721This structure defines an implicit dynamic scoping, where variants
7d9d7e92
MD
722located in inner structures (those with a higher number in the listing
723above) can refer to the fields of outer structures (with lower number in
6c7226e9 724the listing above). See Section 7.3 TSDL Scopes for more detail.
5ba9f198 725
fdf2bb05 7266.1 Event Header
fcba70d4 727
6672e9e1 728Event headers can be described within the meta-data. We hereby propose, as an
fcba70d4
MD
729example, two types of events headers. Type 1 accommodates streams with less than
73031 event IDs. Type 2 accommodates streams with 31 or more event IDs.
5ba9f198 731
3bf79539
MD
732One major factor can vary between streams: the number of event IDs assigned to
733a stream. Luckily, this information tends to stay relatively constant (modulo
5ba9f198 734event registration while trace is being recorded), so we can specify different
3bf79539 735representations for streams containing few event IDs and streams containing
6672e9e1
MD
736many event IDs, so we end up representing the event ID and time-stamp as
737densely as possible in each case.
5ba9f198 738
fcba70d4
MD
739The header is extended in the rare occasions where the information cannot be
740represented in the ranges available in the standard event header. They are also
3bf79539
MD
741used in the rare occasions where the data required for a field could not be
742collected: the flag corresponding to the missing field within the missing_fields
743array is then set to 1.
5ba9f198 744
6672e9e1
MD
745Types uintX_t represent an X-bit unsigned integer, as declared with
746either:
5ba9f198 747
6672e9e1
MD
748 typealias integer { size = X; align = X; signed = false } := uintX_t;
749
750 or
751
752 typealias integer { size = X; align = 1; signed = false } := uintX_t;
5ba9f198 753
fdf2bb05 7546.1.1 Type 1 - Few event IDs
5ba9f198
MD
755
756 - Aligned on 32-bit (or 8-bit if byte-packed, depending on the architecture
757 preference).
5ba9f198 758 - Native architecture byte ordering.
fcba70d4
MD
759 - For "compact" selection
760 - Fixed size: 32 bits.
761 - For "extended" selection
762 - Size depends on the architecture and variant alignment.
5ba9f198 763
80fd2569 764struct event_header_1 {
fcba70d4
MD
765 /*
766 * id: range: 0 - 30.
767 * id 31 is reserved to indicate an extended header.
768 */
a9b83695 769 enum : uint5_t { compact = 0 ... 30, extended = 31 } id;
fcba70d4
MD
770 variant <id> {
771 struct {
772 uint27_t timestamp;
773 } compact;
774 struct {
775 uint32_t id; /* 32-bit event IDs */
776 uint64_t timestamp; /* 64-bit timestamps */
777 } extended;
778 } v;
cb108fea 779} align(32); /* or align(8) */
5ba9f198 780
5ba9f198 781
fdf2bb05 7826.1.2 Type 2 - Many event IDs
5ba9f198 783
fcba70d4 784 - Aligned on 16-bit (or 8-bit if byte-packed, depending on the architecture
5ba9f198 785 preference).
5ba9f198 786 - Native architecture byte ordering.
fcba70d4
MD
787 - For "compact" selection
788 - Size depends on the architecture and variant alignment.
789 - For "extended" selection
790 - Size depends on the architecture and variant alignment.
5ba9f198 791
80fd2569 792struct event_header_2 {
fcba70d4
MD
793 /*
794 * id: range: 0 - 65534.
795 * id 65535 is reserved to indicate an extended header.
796 */
a9b83695 797 enum : uint16_t { compact = 0 ... 65534, extended = 65535 } id;
fcba70d4
MD
798 variant <id> {
799 struct {
800 uint32_t timestamp;
801 } compact;
802 struct {
803 uint32_t id; /* 32-bit event IDs */
804 uint64_t timestamp; /* 64-bit timestamps */
805 } extended;
806 } v;
cb108fea 807} align(16); /* or align(8) */
5ba9f198 808
5ba9f198
MD
809
8106.2 Event Context
811
6672e9e1
MD
812The event context contains information relative to the current event.
813The choice and meaning of this information is specified by the TSDL
814stream and event meta-data descriptions. The stream context is applied
815to all events within the stream. The stream context structure follows
816the event header. The event context is applied to specific events. Its
817structure follows the stream context structure.
5ba9f198 818
fcba70d4
MD
819An example of stream-level event context is to save the event payload size with
820each event, or to save the current PID with each event. These are declared
6672e9e1 821within the stream declaration within the meta-data:
5ba9f198 822
fcba70d4
MD
823 stream {
824 ...
6672e9e1 825 event.context := struct {
80fd2569
MD
826 uint pid;
827 uint16_t payload_size;
6672e9e1 828 };
fcba70d4
MD
829 };
830
831An example of event-specific event context is to declare a bitmap of missing
832fields, only appended after the stream event context if the extended event
833header is selected. NR_FIELDS is the number of fields within the event (a
834numeric value).
5ba9f198 835
fcba70d4
MD
836 event {
837 context = struct {
838 variant <id> {
839 struct { } compact;
840 struct {
841 uint1_t missing_fields[NR_FIELDS]; /* missing event fields bitmap */
842 } extended;
843 } v;
844 };
845 ...
846 }
5ba9f198
MD
847
8486.3 Event Payload
849
850An event payload contains fields specific to a given event type. The fields
6672e9e1 851belonging to an event type are described in the event-specific meta-data
5ba9f198
MD
852within a structure type.
853
8546.3.1 Padding
855
856No padding at the end of the event payload. This differs from the ISO/C standard
857for structures, but follows the CTF standard for structures. In a trace, even
858though it makes sense to align the beginning of a structure, it really makes no
859sense to add padding at the end of the structure, because structures are usually
860not followed by a structure of the same type.
861
862This trick can be done by adding a zero-length "end" field at the end of the C
863structures, and by using the offset of this field rather than using sizeof()
3bf79539 864when calculating the size of a structure (see Appendix "A. Helper macros").
5ba9f198
MD
865
8666.3.2 Alignment
867
868The event payload is aligned on the largest alignment required by types
869contained within the payload. (This follows the ISO/C standard for structures)
870
871
6c7226e9
MD
8727. Trace Stream Description Language (TSDL)
873
874The Trace Stream Description Language (TSDL) allows expression of the
875binary trace streams layout in a C99-like Domain Specific Language
876(DSL).
877
878
6672e9e1 8797.1 Meta-data
6c7226e9
MD
880
881The trace stream layout description is located in the trace meta-data.
882The meta-data is itself located in a stream identified by its name:
883"metadata".
5ba9f198 884
6672e9e1
MD
885The meta-data description can be expressed in two different formats:
886text-only and packet-based. The text-only description facilitates
887generation of meta-data and provides a convenient way to enter the
888meta-data information by hand. The packet-based meta-data provides the
889CTF stream packet facilities (checksumming, compression, encryption,
890network-readiness) for meta-data stream generated and transported by a
891tracer.
892
893The text-only meta-data file is a plain text TSDL description.
894
895The packet-based meta-data is made of "meta-data packets", which each
896start with a meta-data packet header. The packet-based meta-data
897description is detected by reading the magic number "0x75D11D57" at the
898beginning of the file. This magic number is also used to detect the
899endianness of the architecture by trying to read the CTF magic number
900and its counterpart in reversed endianness. The events within the
901meta-data stream have no event header nor event context. Each event only
5ae2352d
MD
902contains a "sequence" payload, which is a sequence of bits using the
903"trace.packet.header.content_size" field as a placeholder for its
904length. The formatting of this sequence of bits is a plain-text
905representation of the TSDL description. Each meta-data packet start with
906a special packet header, specific to the meta-data stream, which
907contains, exactly:
6672e9e1
MD
908
909struct metadata_packet_header {
2daeaa3a 910 uint32_t magic; /* 0x75D11D57 */
3fde5da1 911 uint8_t uuid[16]; /* Unique Universal Identifier */
6672e9e1
MD
912 uint32_t checksum; /* 0 if unused */
913 uint32_t content_size; /* in bits */
914 uint32_t packet_size; /* in bits */
915 uint8_t compression_scheme; /* 0 if unused */
916 uint8_t encryption_scheme; /* 0 if unused */
917 uint8_t checksum_scheme; /* 0 if unused */
918};
919
920The packet-based meta-data can be converted to a text-only meta-data by
921concatenating all the strings in contains.
4fafe1ad 922
6672e9e1
MD
923In the textual representation of the meta-data, the text contained
924within "/*" and "*/", as well as within "//" and end of line, are
925treated as comments. Boolean values can be represented as true, TRUE,
926or 1 for true, and false, FALSE, or 0 for false. Within the string-based
927meta-data description, the trace UUID is represented as a string of
928hexadecimal digits and dashes "-". In the event packet header, the trace
929UUID is represented as an array of bytes.
fcba70d4 930
fdf2bb05 931
6c7226e9 9327.2 Declaration vs Definition
fdf2bb05
MD
933
934A declaration associates a layout to a type, without specifying where
935this type is located in the event structure hierarchy (see Section 6).
936This therefore includes typedef, typealias, as well as all type
937specifiers. In certain circumstances (typedef, structure field and
938variant field), a declaration is followed by a declarator, which specify
939the newly defined type name (for typedef), or the field name (for
940declarations located within structure and variants). Array and sequence,
941declared with square brackets ("[" "]"), are part of the declarator,
a9b83695 942similarly to C99. The enumeration base type is specified by
6c7226e9 943": enum_base", which is part of the type specifier. The variant tag
a9b83695 944name, specified between "<" ">", is also part of the type specifier.
fdf2bb05
MD
945
946A definition associates a type to a location in the event structure
b9606a77
MD
947hierarchy (see Section 6). This association is denoted by ":=", as shown
948in Section 7.3.
fdf2bb05
MD
949
950
6c7226e9 9517.3 TSDL Scopes
fdf2bb05 952
6c7226e9
MD
953TSDL uses two different types of scoping: a lexical scope is used for
954declarations and type definitions, and a dynamic scope is used for
2e7d7fcb
MD
955variants references to tag fields and for sequence references to length
956fields.
fdf2bb05 957
6c7226e9 9587.3.1 Lexical Scope
fdf2bb05 959
d285084f
MD
960Each of "trace", "stream", "event", "struct" and "variant" have their own
961nestable declaration scope, within which types can be declared using "typedef"
fdf2bb05 962and "typealias". A root declaration scope also contains all declarations
7d9d7e92 963located outside of any of the aforementioned declarations. An inner
fdf2bb05 964declaration scope can refer to type declared within its container
7d9d7e92
MD
965lexical scope prior to the inner declaration scope. Redefinition of a
966typedef or typealias is not valid, although hiding an upper scope
fdf2bb05
MD
967typedef or typealias is allowed within a sub-scope.
968
6c7226e9 9697.3.2 Dynamic Scope
fdf2bb05 970
7d9d7e92
MD
971A dynamic scope consists in the lexical scope augmented with the
972implicit event structure definition hierarchy presented at Section 6.
2e7d7fcb
MD
973The dynamic scope is used for variant tag and sequence length
974definitions. It is used at definition time to look up the location of
975the tag field associated with a variant, and to lookup up the location
976of the length field associated with a sequence.
977
978Therefore, variants (or sequences) in lower levels in the dynamic scope
979(e.g. event context) can refer to a tag (or length) field located in
980upper levels (e.g. in the event header) by specifying, in this case, the
981associated tag with <header.field_name>. This allows, for instance, the
982event context to define a variant referring to the "id" field of the
983event header as selector.
fdf2bb05
MD
984
985The target dynamic scope must be specified explicitly when referring to
986a field outside of the local static scope. The dynamic scope prefixes
987are thus:
988
e0d9e2c7 989 - Trace Packet Header: <trace.packet.header. >,
7d9d7e92
MD
990 - Stream Packet Context: <stream.packet.context. >,
991 - Event Header: <stream.event.header. >,
992 - Stream Event Context: <stream.event.context. >,
993 - Event Context: <event.context. >,
994 - Event Payload: <event.fields. >.
fdf2bb05
MD
995
996Multiple declarations of the same field name within a single scope is
997not valid. It is however valid to re-use the same field name in
998different scopes. There is no possible conflict, because the dynamic
999scope must be specified when a variant refers to a tag field located in
1000a different dynamic scope.
1001
457d8b0a
MD
1002The information available in the dynamic scopes can be thought of as the
1003current tracing context. At trace production, information about the
1004current context is saved into the specified scope field levels. At trace
1005consumption, for each event, the current trace context is therefore
1006readable by accessing the upper dynamic scopes.
1007
fdf2bb05 1008
6c7226e9 10097.4 TSDL Examples
d285084f 1010
6672e9e1 1011The grammar representing the TSDL meta-data is presented in Appendix C.
7df6b93a 1012TSDL Grammar. This section presents a rather lighter reading that
6672e9e1 1013consists in examples of TSDL meta-data, with template values.
969f30c0 1014
350bb633
MD
1015The stream "id" can be left out if there is only one stream in the
1016trace. The event "id" field can be left out if there is only one event
1017in a stream.
1018
5ba9f198 1019trace {
fdf2bb05 1020 major = value; /* Trace format version */
5ba9f198 1021 minor = value;
fdf2bb05 1022 uuid = "aaaaaaaa-aaaa-aaaa-aaaa-aaaaaaaaaaaa"; /* Trace UUID */
58997e9e 1023 byte_order = be OR le; /* Endianness (required) */
fc5425db
MD
1024 packet.header := struct {
1025 uint32_t magic;
3fde5da1 1026 uint8_t uuid[16];
fc5425db
MD
1027 uint32_t stream_id;
1028 };
3bf79539 1029};
5ba9f198 1030
3bf79539
MD
1031stream {
1032 id = stream_id;
fdf2bb05 1033 /* Type 1 - Few event IDs; Type 2 - Many event IDs. See section 6.1. */
4fa992a5
MD
1034 event.header := event_header_1 OR event_header_2;
1035 event.context := struct {
77a98c82 1036 ...
3bf79539 1037 };
4fa992a5 1038 packet.context := struct {
77a98c82 1039 ...
3bf79539
MD
1040 };
1041};
5ba9f198
MD
1042
1043event {
3d13ef1a 1044 name = event_name;
3bf79539 1045 id = value; /* Numeric identifier within the stream */
67f02e24 1046 stream_id = stream_id;
4fa992a5 1047 context := struct {
fcba70d4
MD
1048 ...
1049 };
4fa992a5 1050 fields := struct {
80fd2569
MD
1051 ...
1052 };
3bf79539 1053};
5ba9f198
MD
1054
1055/* More detail on types in section 4. Types */
1056
3d13ef1a
MD
1057/*
1058 * Named types:
1059 *
4fa992a5 1060 * Type declarations behave similarly to the C standard.
3d13ef1a
MD
1061 */
1062
80af8ac6 1063typedef aliased_type_specifiers new_type_declarators;
2152348f 1064
3d13ef1a 1065/* e.g.: typedef struct example new_type_name[10]; */
80fd2569 1066
4fa992a5
MD
1067/*
1068 * typealias
1069 *
1070 * The "typealias" declaration can be used to give a name (including
80af8ac6
MD
1071 * pointer declarator specifier) to a type. It should also be used to
1072 * map basic C types (float, int, unsigned long, ...) to a CTF type.
1073 * Typealias is a superset of "typedef": it also allows assignment of a
38b8da21 1074 * simple variable identifier to a type.
4fa992a5
MD
1075 */
1076
1077typealias type_class {
80fd2569 1078 ...
38b8da21 1079} := type_specifiers type_declarator;
2152348f 1080
3d13ef1a
MD
1081/*
1082 * e.g.:
4fa992a5 1083 * typealias integer {
3d13ef1a
MD
1084 * size = 32;
1085 * align = 32;
1086 * signed = false;
38b8da21 1087 * } := struct page *;
359894ac
MD
1088 *
1089 * typealias integer {
1090 * size = 32;
1091 * align = 32;
1092 * signed = true;
38b8da21 1093 * } := int;
3d13ef1a 1094 */
80fd2569
MD
1095
1096struct name {
3bf79539
MD
1097 ...
1098};
5ba9f198 1099
fcba70d4
MD
1100variant name {
1101 ...
1102};
1103
a9b83695 1104enum name : integer_type {
3bf79539
MD
1105 ...
1106};
1107
2152348f 1108
4fa992a5
MD
1109/*
1110 * Unnamed types, contained within compound type fields, typedef or typealias.
1111 */
2152348f 1112
80fd2569
MD
1113struct {
1114 ...
2152348f 1115}
5ba9f198 1116
ec4404a7
MD
1117struct {
1118 ...
1119} align(value)
1120
fcba70d4
MD
1121variant {
1122 ...
1123}
1124
a9b83695 1125enum : integer_type {
80fd2569 1126 ...
2152348f
MD
1127}
1128
1129typedef type new_type[length];
3bf79539 1130
2152348f
MD
1131struct {
1132 type field_name[length];
1133}
1134
1135typedef type new_type[length_type];
1136
1137struct {
1138 type field_name[length_type];
1139}
1140
1141integer {
80fd2569 1142 ...
2152348f 1143}
3bf79539 1144
2152348f 1145floating_point {
80fd2569 1146 ...
2152348f
MD
1147}
1148
1149struct {
1150 integer_type field_name:size; /* GNU/C bitfield */
1151}
1152
1153struct {
1154 string field_name;
1155}
3bf79539 1156
fcba70d4 1157
3bf79539 1158A. Helper macros
5ba9f198
MD
1159
1160The two following macros keep track of the size of a GNU/C structure without
1161padding at the end by placing HEADER_END as the last field. A one byte end field
1162is used for C90 compatibility (C99 flexible arrays could be used here). Note
1163that this does not affect the effective structure size, which should always be
1164calculated with the header_sizeof() helper.
1165
1166#define HEADER_END char end_field
1167#define header_sizeof(type) offsetof(typeof(type), end_field)
3bf79539
MD
1168
1169
1170B. Stream Header Rationale
1171
1172An event stream is divided in contiguous event packets of variable size. These
1173subdivisions allow the trace analyzer to perform a fast binary search by time
1174within the stream (typically requiring to index only the event packet headers)
1175without reading the whole stream. These subdivisions have a variable size to
1176eliminate the need to transfer the event packet padding when partially filled
1177event packets must be sent when streaming a trace for live viewing/analysis.
1178An event packet can contain a certain amount of padding at the end. Dividing
1179streams into event packets is also useful for network streaming over UDP and
1180flight recorder mode tracing (a whole event packet can be swapped out of the
1181buffer atomically for reading).
1182
1183The stream header is repeated at the beginning of each event packet to allow
1184flexibility in terms of:
1185
1186 - streaming support,
1187 - allowing arbitrary buffers to be discarded without making the trace
1188 unreadable,
1189 - allow UDP packet loss handling by either dealing with missing event packet
1190 or asking for re-transmission.
1191 - transparently support flight recorder mode,
1192 - transparently support crash dump.
1193
6c7226e9
MD
1194
1195C. TSDL Grammar
fcba70d4 1196
4fa992a5 1197/*
6c7226e9 1198 * Common Trace Format (CTF) Trace Stream Description Language (TSDL) Grammar.
4fa992a5
MD
1199 *
1200 * Inspired from the C99 grammar:
1201 * http://www.open-std.org/jtc1/sc22/wg14/www/docs/n1124.pdf (Annex A)
6c7226e9
MD
1202 * and c++1x grammar (draft)
1203 * http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2011/n3291.pdf (Annex A)
4fa992a5
MD
1204 *
1205 * Specialized for CTF needs by including only constant and declarations from
1206 * C99 (excluding function declarations), and by adding support for variants,
6c7226e9
MD
1207 * sequences and CTF-specific specifiers. Enumeration container types
1208 * semantic is inspired from c++1x enum-base.
4fa992a5
MD
1209 */
1210
12111) Lexical grammar
1212
12131.1) Lexical elements
1214
1215token:
1216 keyword
1217 identifier
1218 constant
1219 string-literal
1220 punctuator
1221
12221.2) Keywords
1223
1224keyword: is one of
1225
ec4404a7 1226align
4fa992a5
MD
1227const
1228char
1229double
1230enum
1231event
1232floating_point
1233float
1234integer
1235int
1236long
1237short
1238signed
1239stream
1240string
1241struct
1242trace
3e1e1a78 1243typealias
4fa992a5
MD
1244typedef
1245unsigned
1246variant
1247void
1248_Bool
1249_Complex
1250_Imaginary
1251
1252
12531.3) Identifiers
1254
1255identifier:
1256 identifier-nondigit
1257 identifier identifier-nondigit
1258 identifier digit
1259
1260identifier-nondigit:
1261 nondigit
1262 universal-character-name
1263 any other implementation-defined characters
1264
1265nondigit:
1266 _
1267 [a-zA-Z] /* regular expression */
1268
1269digit:
1270 [0-9] /* regular expression */
1271
12721.4) Universal character names
1273
1274universal-character-name:
1275 \u hex-quad
1276 \U hex-quad hex-quad
1277
1278hex-quad:
1279 hexadecimal-digit hexadecimal-digit hexadecimal-digit hexadecimal-digit
1280
12811.5) Constants
1282
1283constant:
1284 integer-constant
1285 enumeration-constant
1286 character-constant
1287
1288integer-constant:
1289 decimal-constant integer-suffix-opt
1290 octal-constant integer-suffix-opt
1291 hexadecimal-constant integer-suffix-opt
1292
1293decimal-constant:
1294 nonzero-digit
1295 decimal-constant digit
1296
1297octal-constant:
1298 0
1299 octal-constant octal-digit
1300
1301hexadecimal-constant:
1302 hexadecimal-prefix hexadecimal-digit
1303 hexadecimal-constant hexadecimal-digit
1304
1305hexadecimal-prefix:
1306 0x
1307 0X
1308
1309nonzero-digit:
1310 [1-9]
1311
1312integer-suffix:
1313 unsigned-suffix long-suffix-opt
1314 unsigned-suffix long-long-suffix
1315 long-suffix unsigned-suffix-opt
1316 long-long-suffix unsigned-suffix-opt
1317
1318unsigned-suffix:
1319 u
1320 U
1321
1322long-suffix:
1323 l
1324 L
1325
1326long-long-suffix:
1327 ll
1328 LL
1329
4fa992a5
MD
1330enumeration-constant:
1331 identifier
1332 string-literal
1333
1334character-constant:
1335 ' c-char-sequence '
1336 L' c-char-sequence '
1337
1338c-char-sequence:
1339 c-char
1340 c-char-sequence c-char
1341
1342c-char:
1343 any member of source charset except single-quote ('), backslash
1344 (\), or new-line character.
1345 escape-sequence
1346
1347escape-sequence:
1348 simple-escape-sequence
1349 octal-escape-sequence
1350 hexadecimal-escape-sequence
1351 universal-character-name
1352
1353simple-escape-sequence: one of
1354 \' \" \? \\ \a \b \f \n \r \t \v
1355
1356octal-escape-sequence:
1357 \ octal-digit
1358 \ octal-digit octal-digit
1359 \ octal-digit octal-digit octal-digit
1360
1361hexadecimal-escape-sequence:
1362 \x hexadecimal-digit
1363 hexadecimal-escape-sequence hexadecimal-digit
1364
13651.6) String literals
1366
1367string-literal:
1368 " s-char-sequence-opt "
1369 L" s-char-sequence-opt "
1370
1371s-char-sequence:
1372 s-char
1373 s-char-sequence s-char
1374
1375s-char:
1376 any member of source charset except double-quote ("), backslash
1377 (\), or new-line character.
1378 escape-sequence
1379
13801.7) Punctuators
1381
1382punctuator: one of
1383 [ ] ( ) { } . -> * + - < > : ; ... = ,
1384
1385
13862) Phrase structure grammar
1387
1388primary-expression:
1389 identifier
1390 constant
1391 string-literal
1392 ( unary-expression )
1393
1394postfix-expression:
1395 primary-expression
1396 postfix-expression [ unary-expression ]
1397 postfix-expression . identifier
1398 postfix-expressoin -> identifier
1399
1400unary-expression:
1401 postfix-expression
1402 unary-operator postfix-expression
1403
1404unary-operator: one of
1405 + -
1406
4fa992a5
MD
1407assignment-operator:
1408 =
1409
b9606a77
MD
1410type-assignment-operator:
1411 :=
1412
4fa992a5 1413constant-expression-range:
73d61ac3 1414 unary-expression ... unary-expression
4fa992a5
MD
1415
14162.2) Declarations:
1417
1418declaration:
689e04b4 1419 declaration-specifiers declarator-list-opt ;
4fa992a5
MD
1420 ctf-specifier ;
1421
1422declaration-specifiers:
689e04b4 1423 storage-class-specifier declaration-specifiers-opt
4fa992a5
MD
1424 type-specifier declaration-specifiers-opt
1425 type-qualifier declaration-specifiers-opt
1426
1427declarator-list:
1428 declarator
1429 declarator-list , declarator
1430
d285084f
MD
1431abstract-declarator-list:
1432 abstract-declarator
1433 abstract-declarator-list , abstract-declarator
1434
4fa992a5
MD
1435storage-class-specifier:
1436 typedef
1437
1438type-specifier:
1439 void
1440 char
1441 short
1442 int
1443 long
1444 float
1445 double
1446 signed
1447 unsigned
1448 _Bool
1449 _Complex
cfdd51ec 1450 _Imaginary
9dfcfc0f
MD
1451 struct-specifier
1452 variant-specifier
4fa992a5
MD
1453 enum-specifier
1454 typedef-name
1455 ctf-type-specifier
1456
ec4404a7 1457align-attribute:
73d61ac3 1458 align ( unary-expression )
ec4404a7 1459
4fa992a5 1460struct-specifier:
ec4404a7
MD
1461 struct identifier-opt { struct-or-variant-declaration-list-opt } align-attribute-opt
1462 struct identifier align-attribute-opt
4fa992a5
MD
1463
1464struct-or-variant-declaration-list:
1465 struct-or-variant-declaration
1466 struct-or-variant-declaration-list struct-or-variant-declaration
1467
1468struct-or-variant-declaration:
1469 specifier-qualifier-list struct-or-variant-declarator-list ;
eacb16d1 1470 declaration-specifiers-opt storage-class-specifier declaration-specifiers-opt declarator-list ;
6143bab7
MD
1471 typealias declaration-specifiers abstract-declarator-list type-assignment-operator declaration-specifiers abstract-declarator-list ;
1472 typealias declaration-specifiers abstract-declarator-list type-assignment-operator declarator-list ;
4fa992a5
MD
1473
1474specifier-qualifier-list:
1475 type-specifier specifier-qualifier-list-opt
1476 type-qualifier specifier-qualifier-list-opt
1477
1478struct-or-variant-declarator-list:
1479 struct-or-variant-declarator
1480 struct-or-variant-declarator-list , struct-or-variant-declarator
1481
1482struct-or-variant-declarator:
1483 declarator
73d61ac3 1484 declarator-opt : unary-expression
4fa992a5
MD
1485
1486variant-specifier:
1487 variant identifier-opt variant-tag-opt { struct-or-variant-declaration-list }
1488 variant identifier variant-tag
1489
1490variant-tag:
1491 < identifier >
1492
1493enum-specifier:
1494 enum identifier-opt { enumerator-list }
1495 enum identifier-opt { enumerator-list , }
1496 enum identifier
a9b83695
MD
1497 enum identifier-opt : declaration-specifiers { enumerator-list }
1498 enum identifier-opt : declaration-specifiers { enumerator-list , }
4fa992a5
MD
1499
1500enumerator-list:
1501 enumerator
1502 enumerator-list , enumerator
1503
1504enumerator:
1505 enumeration-constant
8d2d41f7
MD
1506 enumeration-constant assignment-operator unary-expression
1507 enumeration-constant assignment-operator constant-expression-range
4fa992a5
MD
1508
1509type-qualifier:
1510 const
1511
1512declarator:
1513 pointer-opt direct-declarator
1514
1515direct-declarator:
1516 identifier
1517 ( declarator )
1ab22b2a 1518 direct-declarator [ unary-expression ]
4fa992a5 1519
d285084f
MD
1520abstract-declarator:
1521 pointer-opt direct-abstract-declarator
1522
1523direct-abstract-declarator:
1524 identifier-opt
1525 ( abstract-declarator )
1ab22b2a 1526 direct-abstract-declarator [ unary-expression ]
d285084f
MD
1527 direct-abstract-declarator [ ]
1528
4fa992a5 1529pointer:
3b0f8e4d
MD
1530 * type-qualifier-list-opt
1531 * type-qualifier-list-opt pointer
4fa992a5
MD
1532
1533type-qualifier-list:
1534 type-qualifier
1535 type-qualifier-list type-qualifier
1536
4fa992a5
MD
1537typedef-name:
1538 identifier
1539
15402.3) CTF-specific declarations
1541
1542ctf-specifier:
1543 event { ctf-assignment-expression-list-opt }
1544 stream { ctf-assignment-expression-list-opt }
1545 trace { ctf-assignment-expression-list-opt }
b12919a5
MD
1546 typealias declaration-specifiers abstract-declarator-list type-assignment-operator declaration-specifiers abstract-declarator-list
1547 typealias declaration-specifiers abstract-declarator-list type-assignment-operator declarator-list
4fa992a5
MD
1548
1549ctf-type-specifier:
1550 floating_point { ctf-assignment-expression-list-opt }
1551 integer { ctf-assignment-expression-list-opt }
1552 string { ctf-assignment-expression-list-opt }
7609d3c7 1553 string
4fa992a5
MD
1554
1555ctf-assignment-expression-list:
13441bf5
MD
1556 ctf-assignment-expression ;
1557 ctf-assignment-expression-list ctf-assignment-expression ;
4fa992a5
MD
1558
1559ctf-assignment-expression:
1560 unary-expression assignment-operator unary-expression
1561 unary-expression type-assignment-operator type-specifier
eacb16d1 1562 declaration-specifiers-opt storage-class-specifier declaration-specifiers-opt declarator-list
6143bab7
MD
1563 typealias declaration-specifiers abstract-declarator-list type-assignment-operator declaration-specifiers abstract-declarator-list
1564 typealias declaration-specifiers abstract-declarator-list type-assignment-operator declarator-list
This page took 0.182678 seconds and 4 git commands to generate.