Clarify declaration/definition and scope
[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
MD
9allow traces to be natively generated by the Linux kernel, Linux user-space
10applications written in C/C++, and hardware components.
11
12The latest version of this document can be found at:
13
14 git tree: git://git.efficios.com/ctf.git
15 gitweb: http://git.efficios.com/?p=ctf.git
5ba9f198
MD
16
17A reference implementation of a library to read and write this trace format is
18being implemented within the BabelTrace project, a converter between trace
19formats. The development tree is available at:
20
21 git tree: git://git.efficios.com/babeltrace.git
22 gitweb: http://git.efficios.com/?p=babeltrace.git
23
24
251. Preliminary definitions
26
3bf79539
MD
27 - Event Trace: An ordered sequence of events.
28 - Event Stream: An ordered sequence of events, containing a subset of the
29 trace event types.
30 - Event Packet: A sequence of physically contiguous events within an event
31 stream.
5ba9f198
MD
32 - Event: This is the basic entry in a trace. (aka: a trace record).
33 - An event identifier (ID) relates to the class (a type) of event within
3bf79539
MD
34 an event stream.
35 e.g. event: irq_entry.
5ba9f198
MD
36 - An event (or event record) relates to a specific instance of an event
37 class.
3bf79539
MD
38 e.g. event: irq_entry, at time X, on CPU Y
39 - Source Architecture: Architecture writing the trace.
40 - Reader Architecture: Architecture reading the trace.
5ba9f198
MD
41
42
432. High-level representation of a trace
44
3bf79539
MD
45A trace is divided into multiple event streams. Each event stream contains a
46subset of the trace event types.
5ba9f198 47
3bf79539
MD
48The final output of the trace, after its generation and optional transport over
49the network, is expected to be either on permanent or temporary storage in a
50virtual file system. Because each event stream is appended to while a trace is
51being recorded, each is associated with a separate file for output. Therefore,
52a stored trace can be represented as a directory containing one file per stream.
5ba9f198 53
3bf79539 54A metadata event stream contains information on trace event types. It describes:
5ba9f198
MD
55
56- Trace version.
57- Types available.
3bf79539
MD
58- Per-stream event header description.
59- Per-stream event header selection.
60- Per-stream event context fields.
5ba9f198 61- Per-event
3bf79539 62 - Event type to stream mapping.
5ba9f198
MD
63 - Event type to name mapping.
64 - Event type to ID mapping.
65 - Event fields description.
66
67
3bf79539 683. Event stream
5ba9f198 69
3bf79539
MD
70An event stream is divided in contiguous event packets of variable size. These
71subdivisions have a variable size. An event packet can contain a certain amount
72of padding at the end. The rationale for the event stream design choices is
73explained in Appendix B. Stream Header Rationale.
5ba9f198 74
3bf79539
MD
75An event stream is divided in contiguous event packets of variable size. These
76subdivisions have a variable size. An event packet can contain a certain amount
77of padding at the end. The stream header is repeated at the beginning of each
78event packet.
5ba9f198 79
3bf79539
MD
80The event stream header will therefore be referred to as the "event packet
81header" throughout the rest of this document.
5ba9f198
MD
82
83
844. Types
85
1fad7a85
MD
86Types are organized as type classes. Each type class belong to either of two
87kind of types: basic types or compound types.
88
5ba9f198
MD
894.1 Basic types
90
1fad7a85
MD
91A basic type is a scalar type, as described in this section. It includes
92integers, GNU/C bitfields, enumerations, and floating point values.
5ba9f198
MD
93
944.1.1 Type inheritance
95
80fd2569
MD
96Type specifications can be inherited to allow deriving types from a
97type class. For example, see the uint32_t named type derived from the "integer"
98type class below ("Integers" section). Types have a precise binary
99representation in the trace. A type class has methods to read and write these
100types, but must be derived into a type to be usable in an event field.
5ba9f198
MD
101
1024.1.2 Alignment
103
104We define "byte-packed" types as aligned on the byte size, namely 8-bit.
105We define "bit-packed" types as following on the next bit, as defined by the
106"bitfields" section.
5ba9f198 107
3bf79539
MD
108All basic types, except bitfields, are either aligned on an architecture-defined
109specific alignment or byte-packed, depending on the architecture preference.
110Architectures providing fast unaligned write byte-packed basic types to save
5ba9f198 111space, aligning each type on byte boundaries (8-bit). Architectures with slow
3bf79539
MD
112unaligned writes align types on specific alignment values. If no specific
113alignment is declared for a type nor its parents, it is assumed to be bit-packed
114for bitfields and byte-packed for other types.
5ba9f198 115
3bf79539 116Metadata attribute representation of a specific alignment:
5ba9f198
MD
117
118 align = value; /* value in bits */
119
1204.1.3 Byte order
121
3bf79539
MD
122By default, the native endianness of the source architecture the trace is used.
123Byte order can be overridden for a basic type by specifying a "byte_order"
124attribute. Typical use-case is to specify the network byte order (big endian:
125"be") to save data captured from the network into the trace without conversion.
126If not specified, the byte order is native.
5ba9f198
MD
127
128Metadata representation:
129
130 byte_order = native OR network OR be OR le; /* network and be are aliases */
131
1324.1.4 Size
133
134Type size, in bits, for integers and floats is that returned by "sizeof()" in C
135multiplied by CHAR_BIT.
136We require the size of "char" and "unsigned char" types (CHAR_BIT) to be fixed
137to 8 bits for cross-endianness compatibility.
138
139Metadata representation:
140
141 size = value; (value is in bits)
142
1434.1.5 Integers
144
145Signed integers are represented in two-complement. Integer alignment, size,
146signedness and byte ordering are defined in the metadata. Integers aligned on
147byte size (8-bit) and with length multiple of byte size (8-bit) correspond to
148the C99 standard integers. In addition, integers with alignment and/or size that
149are _not_ a multiple of the byte size are permitted; these correspond to the C99
150standard bitfields, with the added specification that the CTF integer bitfields
151have a fixed binary representation. A MIT-licensed reference implementation of
152the CTF portable bitfields is available at:
153
154 http://git.efficios.com/?p=babeltrace.git;a=blob;f=include/babeltrace/bitfield.h
155
156Binary representation of integers:
157
158- On little and big endian:
159 - Within a byte, high bits correspond to an integer high bits, and low bits
160 correspond to low bits.
161- On little endian:
162 - Integer across multiple bytes are placed from the less significant to the
163 most significant.
164 - Consecutive integers are placed from lower bits to higher bits (even within
165 a byte).
166- On big endian:
167 - Integer across multiple bytes are placed from the most significant to the
168 less significant.
169 - Consecutive integers are placed from higher bits to lower bits (even within
170 a byte).
171
172This binary representation is derived from the bitfield implementation in GCC
173for little and big endian. However, contrary to what GCC does, integers can
174cross units boundaries (no padding is required). Padding can be explicitely
175added (see 4.1.6 GNU/C bitfields) to follow the GCC layout if needed.
176
177Metadata representation:
178
80fd2569 179 integer {
5ba9f198
MD
180 signed = true OR false; /* default false */
181 byte_order = native OR network OR be OR le; /* default native */
182 size = value; /* value in bits, no default */
183 align = value; /* value in bits */
2152348f 184 }
5ba9f198 185
80fd2569 186Example of type inheritance (creation of a uint32_t named type):
5ba9f198 187
359894ac 188typealias integer {
9e4e34e9 189 size = 32;
5ba9f198
MD
190 signed = false;
191 align = 32;
359894ac 192} : uint32_t;
5ba9f198 193
80fd2569 194Definition of a named 5-bit signed bitfield:
5ba9f198 195
359894ac 196typealias integer {
5ba9f198
MD
197 size = 5;
198 signed = true;
199 align = 1;
359894ac 200} : int5_t;
5ba9f198
MD
201
2024.1.6 GNU/C bitfields
203
204The GNU/C bitfields follow closely the integer representation, with a
205particularity on alignment: if a bitfield cannot fit in the current unit, the
80fd2569
MD
206unit is padded and the bitfield starts at the following unit. The unit size is
207defined by the size of the type "unit_type".
5ba9f198 208
2152348f 209Metadata representation:
80fd2569
MD
210
211 unit_type name:size:
212
5ba9f198
MD
213As an example, the following structure declared in C compiled by GCC:
214
215struct example {
216 short a:12;
217 short b:5;
218};
219
2152348f
MD
220The example structure is aligned on the largest element (short). The second
221bitfield would be aligned on the next unit boundary, because it would not fit in
222the current unit.
5ba9f198
MD
223
2244.1.7 Floating point
225
226The floating point values byte ordering is defined in the metadata.
227
228Floating point values follow the IEEE 754-2008 standard interchange formats.
229Description of the floating point values include the exponent and mantissa size
230in bits. Some requirements are imposed on the floating point values:
231
232- FLT_RADIX must be 2.
233- mant_dig is the number of digits represented in the mantissa. It is specified
234 by the ISO C99 standard, section 5.2.4, as FLT_MANT_DIG, DBL_MANT_DIG and
235 LDBL_MANT_DIG as defined by <float.h>.
236- exp_dig is the number of digits represented in the exponent. Given that
237 mant_dig is one bit more than its actual size in bits (leading 1 is not
238 needed) and also given that the sign bit always takes one bit, exp_dig can be
239 specified as:
240
241 - sizeof(float) * CHAR_BIT - FLT_MANT_DIG
242 - sizeof(double) * CHAR_BIT - DBL_MANT_DIG
243 - sizeof(long double) * CHAR_BIT - LDBL_MANT_DIG
244
245Metadata representation:
246
80fd2569 247floating_point {
5ba9f198
MD
248 exp_dig = value;
249 mant_dig = value;
250 byte_order = native OR network OR be OR le;
2152348f 251}
5ba9f198
MD
252
253Example of type inheritance:
254
359894ac 255typealias floating_point {
5ba9f198
MD
256 exp_dig = 8; /* sizeof(float) * CHAR_BIT - FLT_MANT_DIG */
257 mant_dig = 24; /* FLT_MANT_DIG */
258 byte_order = native;
359894ac 259} : float;
5ba9f198
MD
260
261TODO: define NaN, +inf, -inf behavior.
262
2634.1.8 Enumerations
264
265Enumerations are a mapping between an integer type and a table of strings. The
266numerical representation of the enumeration follows the integer type specified
267by the metadata. The enumeration mapping table is detailed in the enumeration
3bf79539
MD
268description within the metadata. The mapping table maps inclusive value ranges
269(or single values) to strings. Instead of being limited to simple
270"value -> string" mappings, these enumerations map
80fd2569 271"[ start_value ... end_value ] -> string", which map inclusive ranges of
3bf79539
MD
272values to strings. An enumeration from the C language can be represented in
273this format by having the same start_value and end_value for each element, which
274is in fact a range of size 1. This single-value range is supported without
4767a9e7 275repeating the start and end values with the value = string declaration.
80fd2569 276
4767a9e7
MD
277If a numeric value is encountered between < >, it represents the integer type
278size used to hold the enumeration, in bits.
279
cfc73fdc 280enum name <integer_type OR size> {
359894ac 281 somestring = start_value1 ... end_value1,
80fd2569
MD
282 "other string" = start_value2 ... end_value2,
283 yet_another_string, /* will be assigned to end_value2 + 1 */
284 "some other string" = value,
285 ...
286};
287
288If the values are omitted, the enumeration starts at 0 and increment of 1 for
289each entry:
290
cfc73fdc 291enum name <32> {
80fd2569
MD
292 ZERO,
293 ONE,
294 TWO,
295 TEN = 10,
296 ELEVEN,
3bf79539 297};
5ba9f198 298
80fd2569 299Overlapping ranges within a single enumeration are implementation defined.
5ba9f198 300
2152348f
MD
301A nameless enumeration can be declared as a field type or as part of a typedef:
302
303enum <integer_type> {
304 ...
305}
306
1fad7a85 307
5ba9f198
MD
3084.2 Compound types
309
1fad7a85
MD
310Compound are aggregation of type declarations. Compound types include
311structures, variant, arrays, sequences, and strings.
312
5ba9f198
MD
3134.2.1 Structures
314
315Structures are aligned on the largest alignment required by basic types
316contained within the structure. (This follows the ISO/C standard for structures)
317
80fd2569 318Metadata representation of a named structure:
5ba9f198 319
80fd2569
MD
320struct name {
321 field_type field_name;
322 field_type field_name;
323 ...
324};
5ba9f198
MD
325
326Example:
327
80fd2569
MD
328struct example {
329 integer { /* Nameless type */
330 size = 16;
331 signed = true;
332 align = 16;
333 } first_field_name;
334 uint64_t second_field_name; /* Named type declared in the metadata */
3bf79539 335};
5ba9f198
MD
336
337The fields are placed in a sequence next to each other. They each possess a
338field name, which is a unique identifier within the structure.
339
2152348f 340A nameless structure can be declared as a field type or as part of a typedef:
80fd2569
MD
341
342struct {
343 ...
2152348f 344}
80fd2569 345
77a98c82 3464.2.2 Variants (Discriminated/Tagged Unions)
fcba70d4 347
fdf2bb05
MD
348A CTF variant is a selection between different types. A CTF variant must
349always be defined within the scope of a structure or within fields
350contained within a structure (defined recursively). A "tag" enumeration
351field must appear in either the same lexical scope, prior to the variant
352field (in field declaration order), in an uppermost lexical scope (see
353Section 7.2.1), or in an uppermost dynamic scope (see Section 7.2.2).
354The type selection is indicated by the mapping from the enumeration
355value to the string used as variant type selector. The field to use as
356tag is specified by the "tag_field", specified between "< >" after the
357"variant" keyword for unnamed variants, and after "variant name" for
358named variants.
fcba70d4
MD
359
360The alignment of the variant is the alignment of the type as selected by the tag
361value for the specific instance of the variant. The alignment of the type
362containing the variant is independent of the variant alignment. The size of the
363variant is the size as selected by the tag value for the specific instance of
364the variant.
365
366A named variant declaration followed by its definition within a structure
367declaration:
368
369variant name {
370 field_type sel1;
371 field_type sel2;
372 field_type sel3;
373 ...
374};
375
376struct {
377 enum <integer_type or size> { sel1, sel2, sel3, ... } tag_field;
378 ...
379 variant name <tag_field> v;
380}
381
382An unnamed variant definition within a structure is expressed by the following
383metadata:
384
385struct {
386 enum <integer_type or size> { sel1, sel2, sel3, ... } tag_field;
387 ...
388 variant <tag_field> {
389 field_type sel1;
390 field_type sel2;
391 field_type sel3;
392 ...
393 } v;
394}
395
396Example of a named variant within a sequence that refers to a single tag field:
397
398variant example {
399 uint32_t a;
400 uint64_t b;
401 short c;
402};
403
404struct {
405 enum <uint2_t> { a, b, c } choice;
15850440 406 variant example <choice> v[unsigned int];
fcba70d4
MD
407}
408
409Example of an unnamed variant:
410
411struct {
412 enum <uint2_t> { a, b, c, d } choice;
413 /* Unrelated fields can be added between the variant and its tag */
414 int32_t somevalue;
415 variant <choice> {
416 uint32_t a;
417 uint64_t b;
418 short c;
419 struct {
420 unsigned int field1;
421 uint64_t field2;
422 } d;
423 } s;
424}
425
426Example of an unnamed variant within an array:
427
428struct {
429 enum <uint2_t> { a, b, c } choice;
430 variant <choice> {
431 uint32_t a;
432 uint64_t b;
433 short c;
15850440 434 } v[10];
fcba70d4
MD
435}
436
437Example of a variant type definition within a structure, where the defined type
438is then declared within an array of structures. This variant refers to a tag
439located in an upper lexical scope. This example clearly shows that a variant
440type definition referring to the tag "x" uses the closest preceding field from
441the lexical scope of the type definition.
442
443struct {
444 enum <uint2_t> { a, b, c, d } x;
445
446 typedef variant <x> { /*
447 * "x" refers to the preceding "x" enumeration in the
448 * lexical scope of the type definition.
449 */
450 uint32_t a;
451 uint64_t b;
452 short c;
453 } example_variant;
454
455 struct {
456 enum <int> { x, y, z } x; /* This enumeration is not used by "v". */
457 example_variant v; /*
458 * "v" uses the "enum <uint2_t> { a, b, c, d }"
459 * tag.
460 */
461 } a[10];
462}
463
4644.2.3 Arrays
5ba9f198
MD
465
466Arrays are fixed-length. Their length is declared in the type declaration within
467the metadata. They contain an array of "inner type" elements, which can refer to
468any type not containing the type of the array being declared (no circular
3bf79539 469dependency). The length is the number of elements in an array.
5ba9f198 470
2152348f 471Metadata representation of a named array:
80fd2569
MD
472
473typedef elem_type name[length];
5ba9f198 474
2152348f 475A nameless array can be declared as a field type within a structure, e.g.:
5ba9f198 476
2152348f 477 uint8_t field_name[10];
80fd2569 478
5ba9f198 479
fcba70d4 4804.2.4 Sequences
5ba9f198
MD
481
482Sequences are dynamically-sized arrays. They start with an integer that specify
483the length of the sequence, followed by an array of "inner type" elements.
3bf79539 484The length is the number of elements in the sequence.
5ba9f198 485
2152348f 486Metadata representation for a named sequence:
80fd2569
MD
487
488typedef elem_type name[length_type];
489
490A nameless sequence can be declared as a field type, e.g.:
491
80fd2569
MD
492long field_name[int];
493
494The length type follows the integer types specifications, and the sequence
5ba9f198
MD
495elements follow the "array" specifications.
496
fcba70d4 4974.2.5 Strings
5ba9f198
MD
498
499Strings are an array of bytes of variable size and are terminated by a '\0'
500"NULL" character. Their encoding is described in the metadata. In absence of
501encoding attribute information, the default encoding is UTF-8.
502
80fd2569
MD
503Metadata representation of a named string type:
504
359894ac 505typealias string {
5ba9f198 506 encoding = UTF8 OR ASCII;
359894ac 507} : name;
5ba9f198 508
80fd2569
MD
509A nameless string type can be declared as a field type:
510
511string field_name; /* Use default UTF8 encoding */
5ba9f198 512
3bf79539
MD
5135. Event Packet Header
514
515The event packet header consists of two part: one is mandatory and have a fixed
516layout. The second part, the "event packet context", has its layout described in
517the metadata.
5ba9f198 518
3bf79539
MD
519- Aligned on page size. Fixed size. Fields either aligned or packed (depending
520 on the architecture preference).
521 No padding at the end of the event packet header. Native architecture byte
5ba9f198 522 ordering.
3bf79539
MD
523
524Fixed layout (event packet header):
525
5ba9f198
MD
526- Magic number (CTF magic numbers: 0xC1FC1FC1 and its reverse endianness
527 representation: 0xC11FFCC1) It needs to have a non-symmetric bytewise
528 representation. Used to distinguish between big and little endian traces (this
529 information is determined by knowing the endianness of the architecture
530 reading the trace and comparing the magic number against its value and the
531 reverse, 0xC11FFCC1). This magic number specifies that we use the CTF metadata
532 description language described in this document. Different magic numbers
533 should be used for other metadata description languages.
3bf79539 534- Trace UUID, used to ensure the event packet match the metadata used.
5ba9f198
MD
535 (note: we cannot use a metadata checksum because metadata can be appended to
536 while tracing is active)
3bf79539
MD
537- Stream ID, used as reference to stream description in metadata.
538
539Metadata-defined layout (event packet context):
540
541- Event packet content size (in bytes).
542- Event packet size (in bytes, includes padding).
543- Event packet content checksum (optional). Checksum excludes the event packet
544 header.
545- Per-stream event packet sequence count (to deal with UDP packet loss). The
546 number of significant sequence counter bits should also be present, so
547 wrap-arounds are deal with correctly.
548- Timestamp at the beginning and timestamp at the end of the event packet.
549 Both timestamps are written in the packet header, but sampled respectively
550 while (or before) writing the first event and while (or after) writing the
551 last event in the packet. The inclusive range between these timestamps should
552 include all event timestamps assigned to events contained within the packet.
5ba9f198 553- Events discarded count
3bf79539
MD
554 - Snapshot of a per-stream free-running counter, counting the number of
555 events discarded that were supposed to be written in the stream prior to
556 the first event in the event packet.
5ba9f198 557 * Note: producer-consumer buffer full condition should fill the current
3bf79539 558 event packet with padding so we know exactly where events have been
5ba9f198 559 discarded.
3bf79539
MD
560- Lossless compression scheme used for the event packet content. Applied
561 directly to raw data. New types of compression can be added in following
562 versions of the format.
5ba9f198
MD
563 0: no compression scheme
564 1: bzip2
565 2: gzip
3bf79539
MD
566 3: xz
567- Cypher used for the event packet content. Applied after compression.
5ba9f198
MD
568 0: no encryption
569 1: AES
3bf79539 570- Checksum scheme used for the event packet content. Applied after encryption.
5ba9f198
MD
571 0: no checksum
572 1: md5
573 2: sha1
574 3: crc32
575
3bf79539
MD
5765.1 Event Packet Header Fixed Layout Description
577
80fd2569
MD
578struct event_packet_header {
579 uint32_t magic;
580 uint8_t trace_uuid[16];
3bf79539 581 uint32_t stream_id;
80fd2569 582};
5ba9f198 583
3bf79539
MD
5845.2 Event Packet Context Description
585
586Event packet context example. These are declared within the stream declaration
587in the metadata. All these fields are optional except for "content_size" and
588"packet_size", which must be present in the context.
589
590An example event packet context type:
591
80fd2569 592struct event_packet_context {
3bf79539
MD
593 uint64_t timestamp_begin;
594 uint64_t timestamp_end;
595 uint32_t checksum;
596 uint32_t stream_packet_count;
597 uint32_t events_discarded;
598 uint32_t cpu_id;
599 uint32_t/uint16_t content_size;
600 uint32_t/uint16_t packet_size;
601 uint8_t stream_packet_count_bits; /* Significant counter bits */
602 uint8_t compression_scheme;
603 uint8_t encryption_scheme;
3b0f8e4d 604 uint8_t checksum_scheme;
3bf79539 605};
5ba9f198 606
fcba70d4 607
5ba9f198
MD
6086. Event Structure
609
610The overall structure of an event is:
611
fcba70d4 6121 - Stream Packet Context (as specified by the stream metadata)
fdf2bb05
MD
613 2 - Event Header (as specified by the stream metadata)
614 3 - Stream Event Context (as specified by the stream metadata)
615 4 - Event Context (as specified by the event metadata)
616 5 - Event Payload (as specified by the event metadata)
5ba9f198 617
fdf2bb05
MD
618This structure defines an implicit dynamic scoping, where variants
619located in structures with higher number can refer to the fields of
620structures with lower number. See Section 7.2 Metadata Scopes for more
621detail.
5ba9f198 622
fdf2bb05 6236.1 Event Header
fcba70d4
MD
624
625Event headers can be described within the metadata. We hereby propose, as an
626example, two types of events headers. Type 1 accommodates streams with less than
62731 event IDs. Type 2 accommodates streams with 31 or more event IDs.
5ba9f198 628
3bf79539
MD
629One major factor can vary between streams: the number of event IDs assigned to
630a stream. Luckily, this information tends to stay relatively constant (modulo
5ba9f198 631event registration while trace is being recorded), so we can specify different
3bf79539 632representations for streams containing few event IDs and streams containing
5ba9f198
MD
633many event IDs, so we end up representing the event ID and timestamp as densely
634as possible in each case.
635
fcba70d4
MD
636The header is extended in the rare occasions where the information cannot be
637represented in the ranges available in the standard event header. They are also
3bf79539
MD
638used in the rare occasions where the data required for a field could not be
639collected: the flag corresponding to the missing field within the missing_fields
640array is then set to 1.
5ba9f198
MD
641
642Types uintX_t represent an X-bit unsigned integer.
643
644
fdf2bb05 6456.1.1 Type 1 - Few event IDs
5ba9f198
MD
646
647 - Aligned on 32-bit (or 8-bit if byte-packed, depending on the architecture
648 preference).
5ba9f198 649 - Native architecture byte ordering.
fcba70d4
MD
650 - For "compact" selection
651 - Fixed size: 32 bits.
652 - For "extended" selection
653 - Size depends on the architecture and variant alignment.
5ba9f198 654
80fd2569 655struct event_header_1 {
fcba70d4
MD
656 /*
657 * id: range: 0 - 30.
658 * id 31 is reserved to indicate an extended header.
659 */
660 enum <uint5_t> { compact = 0 ... 30, extended = 31 } id;
661 variant <id> {
662 struct {
663 uint27_t timestamp;
664 } compact;
665 struct {
666 uint32_t id; /* 32-bit event IDs */
667 uint64_t timestamp; /* 64-bit timestamps */
668 } extended;
669 } v;
5ba9f198
MD
670};
671
5ba9f198 672
fdf2bb05 6736.1.2 Type 2 - Many event IDs
5ba9f198 674
fcba70d4 675 - Aligned on 16-bit (or 8-bit if byte-packed, depending on the architecture
5ba9f198 676 preference).
5ba9f198 677 - Native architecture byte ordering.
fcba70d4
MD
678 - For "compact" selection
679 - Size depends on the architecture and variant alignment.
680 - For "extended" selection
681 - Size depends on the architecture and variant alignment.
5ba9f198 682
80fd2569 683struct event_header_2 {
fcba70d4
MD
684 /*
685 * id: range: 0 - 65534.
686 * id 65535 is reserved to indicate an extended header.
687 */
688 enum <uint16_t> { compact = 0 ... 65534, extended = 65535 } id;
689 variant <id> {
690 struct {
691 uint32_t timestamp;
692 } compact;
693 struct {
694 uint32_t id; /* 32-bit event IDs */
695 uint64_t timestamp; /* 64-bit timestamps */
696 } extended;
697 } v;
5ba9f198
MD
698};
699
5ba9f198
MD
700
7016.2 Event Context
702
703The event context contains information relative to the current event. The choice
fcba70d4
MD
704and meaning of this information is specified by the metadata "stream" and
705"event" information. The "stream" context is applied to all events within the
706stream. The "stream" context structure follows the event header. The "event"
707context is applied to specific events. Its structure follows the "stream"
708context stucture.
5ba9f198 709
fcba70d4
MD
710An example of stream-level event context is to save the event payload size with
711each event, or to save the current PID with each event. These are declared
712within the stream declaration within the metadata:
5ba9f198 713
fcba70d4
MD
714 stream {
715 ...
716 event {
717 ...
4fa992a5 718 context := struct {
80fd2569
MD
719 uint pid;
720 uint16_t payload_size;
3bf79539 721 };
fcba70d4
MD
722 }
723 };
724
725An example of event-specific event context is to declare a bitmap of missing
726fields, only appended after the stream event context if the extended event
727header is selected. NR_FIELDS is the number of fields within the event (a
728numeric value).
5ba9f198 729
fcba70d4
MD
730 event {
731 context = struct {
732 variant <id> {
733 struct { } compact;
734 struct {
735 uint1_t missing_fields[NR_FIELDS]; /* missing event fields bitmap */
736 } extended;
737 } v;
738 };
739 ...
740 }
5ba9f198
MD
741
7426.3 Event Payload
743
744An event payload contains fields specific to a given event type. The fields
745belonging to an event type are described in the event-specific metadata
746within a structure type.
747
7486.3.1 Padding
749
750No padding at the end of the event payload. This differs from the ISO/C standard
751for structures, but follows the CTF standard for structures. In a trace, even
752though it makes sense to align the beginning of a structure, it really makes no
753sense to add padding at the end of the structure, because structures are usually
754not followed by a structure of the same type.
755
756This trick can be done by adding a zero-length "end" field at the end of the C
757structures, and by using the offset of this field rather than using sizeof()
3bf79539 758when calculating the size of a structure (see Appendix "A. Helper macros").
5ba9f198
MD
759
7606.3.2 Alignment
761
762The event payload is aligned on the largest alignment required by types
763contained within the payload. (This follows the ISO/C standard for structures)
764
765
5ba9f198
MD
7667. Metadata
767
3bf79539
MD
768The meta-data is located in a stream named "metadata". It is made of "event
769packets", which each start with an event packet header. The event type within
770the metadata stream have no event header nor event context. Each event only
5ba9f198 771contains a null-terminated "string" payload, which is a metadata description
3bf79539
MD
772entry. The events are packed one next to another. Each event packet start with
773an event packet header, which contains, amongst other fields, the magic number
fdf2bb05
MD
774and trace UUID. The trace UUID is represented as a string of hexadecimal digits
775and dashes "-".
5ba9f198
MD
776
777The metadata can be parsed by reading through the metadata strings, skipping
fcba70d4
MD
778newlines and null-characters. Type names are made of a single identifier, and
779can be surrounded by prefix/postfix. Text contained within "/*" and "*/", as
c6d7abc5
MD
780well as within "//" and end of line, are treated as comments. Boolean values can
781be represented as true, TRUE, or 1 for true, and false, FALSE, or 0 for false.
fcba70d4 782
fdf2bb05
MD
783
7847.1 Declaration vs Definition
785
786A declaration associates a layout to a type, without specifying where
787this type is located in the event structure hierarchy (see Section 6).
788This therefore includes typedef, typealias, as well as all type
789specifiers. In certain circumstances (typedef, structure field and
790variant field), a declaration is followed by a declarator, which specify
791the newly defined type name (for typedef), or the field name (for
792declarations located within structure and variants). Array and sequence,
793declared with square brackets ("[" "]"), are part of the declarator,
794similarly to C99.
795
796A definition associates a type to a location in the event structure
797hierarchy (see Section 6).
798
799
8007.2 Metadata Scopes
801
802CTF metadata uses two different types of scoping: a lexical scope is
803used for declarations and type definitions, and a dynamic scope is used
804for variants references to tag fields.
805
8067.2.1 Lexical Scope
807
d285084f
MD
808Each of "trace", "stream", "event", "struct" and "variant" have their own
809nestable declaration scope, within which types can be declared using "typedef"
fdf2bb05
MD
810and "typealias". A root declaration scope also contains all declarations
811located outside of any of the aforementioned declarations. An innermost
812declaration scope can refer to type declared within its container
813lexical scope prior to the innermost declaration scope. Redefinition of
814a typedef or typealias is not valid, although hiding an uppermost scope
815typedef or typealias is allowed within a sub-scope.
816
8177.2.2 Dynamic Scope
818
819For variant tag definition only, the dynamic scope used to look up the
820location of the associated tag field consists in the lexical scope of
821the structures where the variant is declared, extended with the implicit
822dynamic scope specified by the event structure hierarchy presented at
823the beginning of Section 6. Therefore, lower levels in the dynamic scope
824(e.g. event context) can refer to a tag field located in upper levels
825(e.g. in the event header) by specifying, in this case,
826"header.field_name" as tag identifier. This allows, for instance, the
827event context to define a variant referring to the "id" field of the
828event header as selector.
829
830The target dynamic scope must be specified explicitly when referring to
831a field outside of the local static scope. The dynamic scope prefixes
832are thus:
833
834 - Stream Packet Context: "stream.packet.context.",
835 - Event Header: "stream.event.header.",
836 - Stream Event Context: "stream.event.context.",
837 - Event Context: "event.context.",
838 - Event Payload: "event.fields.".
839
840Multiple declarations of the same field name within a single scope is
841not valid. It is however valid to re-use the same field name in
842different scopes. There is no possible conflict, because the dynamic
843scope must be specified when a variant refers to a tag field located in
844a different dynamic scope.
845
846
8477.2 Metadata Examples
d285084f 848
fcba70d4 849The grammar representing the CTF metadata is presented in
fdf2bb05
MD
850Appendix C. CTF Metadata Grammar. This section presents a rather ligher
851reading that consists in examples of CTF metadata, with template values:
5ba9f198
MD
852
853trace {
fdf2bb05 854 major = value; /* Trace format version */
5ba9f198 855 minor = value;
fdf2bb05 856 uuid = "aaaaaaaa-aaaa-aaaa-aaaa-aaaaaaaaaaaa"; /* Trace UUID */
3bf79539
MD
857 word_size = value;
858};
5ba9f198 859
3bf79539
MD
860stream {
861 id = stream_id;
fdf2bb05 862 /* Type 1 - Few event IDs; Type 2 - Many event IDs. See section 6.1. */
4fa992a5
MD
863 event.header := event_header_1 OR event_header_2;
864 event.context := struct {
77a98c82 865 ...
3bf79539 866 };
4fa992a5 867 packet.context := struct {
77a98c82 868 ...
3bf79539
MD
869 };
870};
5ba9f198
MD
871
872event {
3d13ef1a 873 name = event_name;
3bf79539
MD
874 id = value; /* Numeric identifier within the stream */
875 stream = stream_id;
4fa992a5 876 context := struct {
fcba70d4
MD
877 ...
878 };
4fa992a5 879 fields := struct {
80fd2569
MD
880 ...
881 };
3bf79539 882};
5ba9f198
MD
883
884/* More detail on types in section 4. Types */
885
3d13ef1a
MD
886/*
887 * Named types:
888 *
4fa992a5 889 * Type declarations behave similarly to the C standard.
3d13ef1a
MD
890 */
891
892typedef aliased_type_prefix aliased_type new_type aliased_type_postfix;
2152348f 893
3d13ef1a 894/* e.g.: typedef struct example new_type_name[10]; */
80fd2569 895
4fa992a5
MD
896/*
897 * typealias
898 *
899 * The "typealias" declaration can be used to give a name (including
359894ac
MD
900 * prefix/postfix) to a type. It should also be used to map basic C types
901 * (float, int, unsigned long, ...) to a CTF type. Typealias is a superset of
902 * "typedef": it also allows assignment of a simple variable identifier to a
903 * type.
4fa992a5
MD
904 */
905
906typealias type_class {
80fd2569 907 ...
fcba70d4 908} : new_type_prefix new_type new_type_postfix;
2152348f 909
3d13ef1a
MD
910/*
911 * e.g.:
4fa992a5 912 * typealias integer {
3d13ef1a
MD
913 * size = 32;
914 * align = 32;
915 * signed = false;
fcba70d4 916 * } : struct page *;
359894ac
MD
917 *
918 * typealias integer {
919 * size = 32;
920 * align = 32;
921 * signed = true;
922 * } : int;
3d13ef1a 923 */
80fd2569
MD
924
925struct name {
3bf79539
MD
926 ...
927};
5ba9f198 928
fcba70d4
MD
929variant name {
930 ...
931};
932
cfc73fdc 933enum name <integer_type or size> {
3bf79539
MD
934 ...
935};
936
2152348f 937
4fa992a5
MD
938/*
939 * Unnamed types, contained within compound type fields, typedef or typealias.
940 */
2152348f 941
80fd2569
MD
942struct {
943 ...
2152348f 944}
5ba9f198 945
fcba70d4
MD
946variant {
947 ...
948}
949
4767a9e7 950enum <integer_type or size> {
80fd2569 951 ...
2152348f
MD
952}
953
954typedef type new_type[length];
3bf79539 955
2152348f
MD
956struct {
957 type field_name[length];
958}
959
960typedef type new_type[length_type];
961
962struct {
963 type field_name[length_type];
964}
965
966integer {
80fd2569 967 ...
2152348f 968}
3bf79539 969
2152348f 970floating_point {
80fd2569 971 ...
2152348f
MD
972}
973
974struct {
975 integer_type field_name:size; /* GNU/C bitfield */
976}
977
978struct {
979 string field_name;
980}
3bf79539 981
fcba70d4 982
3bf79539 983A. Helper macros
5ba9f198
MD
984
985The two following macros keep track of the size of a GNU/C structure without
986padding at the end by placing HEADER_END as the last field. A one byte end field
987is used for C90 compatibility (C99 flexible arrays could be used here). Note
988that this does not affect the effective structure size, which should always be
989calculated with the header_sizeof() helper.
990
991#define HEADER_END char end_field
992#define header_sizeof(type) offsetof(typeof(type), end_field)
3bf79539
MD
993
994
995B. Stream Header Rationale
996
997An event stream is divided in contiguous event packets of variable size. These
998subdivisions allow the trace analyzer to perform a fast binary search by time
999within the stream (typically requiring to index only the event packet headers)
1000without reading the whole stream. These subdivisions have a variable size to
1001eliminate the need to transfer the event packet padding when partially filled
1002event packets must be sent when streaming a trace for live viewing/analysis.
1003An event packet can contain a certain amount of padding at the end. Dividing
1004streams into event packets is also useful for network streaming over UDP and
1005flight recorder mode tracing (a whole event packet can be swapped out of the
1006buffer atomically for reading).
1007
1008The stream header is repeated at the beginning of each event packet to allow
1009flexibility in terms of:
1010
1011 - streaming support,
1012 - allowing arbitrary buffers to be discarded without making the trace
1013 unreadable,
1014 - allow UDP packet loss handling by either dealing with missing event packet
1015 or asking for re-transmission.
1016 - transparently support flight recorder mode,
1017 - transparently support crash dump.
1018
1019The event stream header will therefore be referred to as the "event packet
1020header" throughout the rest of this document.
fcba70d4
MD
1021
1022C. CTF Metadata Grammar
1023
4fa992a5
MD
1024/*
1025 * Common Trace Format (CTF) Metadata Grammar.
1026 *
1027 * Inspired from the C99 grammar:
1028 * http://www.open-std.org/jtc1/sc22/wg14/www/docs/n1124.pdf (Annex A)
1029 *
1030 * Specialized for CTF needs by including only constant and declarations from
1031 * C99 (excluding function declarations), and by adding support for variants,
1032 * sequences and CTF-specific specifiers.
1033 */
1034
10351) Lexical grammar
1036
10371.1) Lexical elements
1038
1039token:
1040 keyword
1041 identifier
1042 constant
1043 string-literal
1044 punctuator
1045
10461.2) Keywords
1047
1048keyword: is one of
1049
1050const
1051char
1052double
1053enum
1054event
1055floating_point
1056float
1057integer
1058int
1059long
1060short
1061signed
1062stream
1063string
1064struct
1065trace
3e1e1a78 1066typealias
4fa992a5
MD
1067typedef
1068unsigned
1069variant
1070void
1071_Bool
1072_Complex
1073_Imaginary
1074
1075
10761.3) Identifiers
1077
1078identifier:
1079 identifier-nondigit
1080 identifier identifier-nondigit
1081 identifier digit
1082
1083identifier-nondigit:
1084 nondigit
1085 universal-character-name
1086 any other implementation-defined characters
1087
1088nondigit:
1089 _
1090 [a-zA-Z] /* regular expression */
1091
1092digit:
1093 [0-9] /* regular expression */
1094
10951.4) Universal character names
1096
1097universal-character-name:
1098 \u hex-quad
1099 \U hex-quad hex-quad
1100
1101hex-quad:
1102 hexadecimal-digit hexadecimal-digit hexadecimal-digit hexadecimal-digit
1103
11041.5) Constants
1105
1106constant:
1107 integer-constant
1108 enumeration-constant
1109 character-constant
1110
1111integer-constant:
1112 decimal-constant integer-suffix-opt
1113 octal-constant integer-suffix-opt
1114 hexadecimal-constant integer-suffix-opt
1115
1116decimal-constant:
1117 nonzero-digit
1118 decimal-constant digit
1119
1120octal-constant:
1121 0
1122 octal-constant octal-digit
1123
1124hexadecimal-constant:
1125 hexadecimal-prefix hexadecimal-digit
1126 hexadecimal-constant hexadecimal-digit
1127
1128hexadecimal-prefix:
1129 0x
1130 0X
1131
1132nonzero-digit:
1133 [1-9]
1134
1135integer-suffix:
1136 unsigned-suffix long-suffix-opt
1137 unsigned-suffix long-long-suffix
1138 long-suffix unsigned-suffix-opt
1139 long-long-suffix unsigned-suffix-opt
1140
1141unsigned-suffix:
1142 u
1143 U
1144
1145long-suffix:
1146 l
1147 L
1148
1149long-long-suffix:
1150 ll
1151 LL
1152
1153digit-sequence:
1154 digit
1155 digit-sequence digit
1156
1157hexadecimal-digit-sequence:
1158 hexadecimal-digit
1159 hexadecimal-digit-sequence hexadecimal-digit
1160
1161enumeration-constant:
1162 identifier
1163 string-literal
1164
1165character-constant:
1166 ' c-char-sequence '
1167 L' c-char-sequence '
1168
1169c-char-sequence:
1170 c-char
1171 c-char-sequence c-char
1172
1173c-char:
1174 any member of source charset except single-quote ('), backslash
1175 (\), or new-line character.
1176 escape-sequence
1177
1178escape-sequence:
1179 simple-escape-sequence
1180 octal-escape-sequence
1181 hexadecimal-escape-sequence
1182 universal-character-name
1183
1184simple-escape-sequence: one of
1185 \' \" \? \\ \a \b \f \n \r \t \v
1186
1187octal-escape-sequence:
1188 \ octal-digit
1189 \ octal-digit octal-digit
1190 \ octal-digit octal-digit octal-digit
1191
1192hexadecimal-escape-sequence:
1193 \x hexadecimal-digit
1194 hexadecimal-escape-sequence hexadecimal-digit
1195
11961.6) String literals
1197
1198string-literal:
1199 " s-char-sequence-opt "
1200 L" s-char-sequence-opt "
1201
1202s-char-sequence:
1203 s-char
1204 s-char-sequence s-char
1205
1206s-char:
1207 any member of source charset except double-quote ("), backslash
1208 (\), or new-line character.
1209 escape-sequence
1210
12111.7) Punctuators
1212
1213punctuator: one of
1214 [ ] ( ) { } . -> * + - < > : ; ... = ,
1215
1216
12172) Phrase structure grammar
1218
1219primary-expression:
1220 identifier
1221 constant
1222 string-literal
1223 ( unary-expression )
1224
1225postfix-expression:
1226 primary-expression
1227 postfix-expression [ unary-expression ]
1228 postfix-expression . identifier
1229 postfix-expressoin -> identifier
1230
1231unary-expression:
1232 postfix-expression
1233 unary-operator postfix-expression
1234
1235unary-operator: one of
1236 + -
1237
4fa992a5
MD
1238assignment-operator:
1239 =
1240
1241constant-expression:
1242 unary-expression
1243
1244constant-expression-range:
1245 constant-expression ... constant-expression
1246
12472.2) Declarations:
1248
1249declaration:
3b0f8e4d
MD
1250 declaration-specifiers ;
1251 declaration-specifiers storage-class-specifier declaration-specifiers declarator-list ;
4fa992a5
MD
1252 ctf-specifier ;
1253
1254declaration-specifiers:
4fa992a5
MD
1255 type-specifier declaration-specifiers-opt
1256 type-qualifier declaration-specifiers-opt
1257
1258declarator-list:
1259 declarator
1260 declarator-list , declarator
1261
d285084f
MD
1262abstract-declarator-list:
1263 abstract-declarator
1264 abstract-declarator-list , abstract-declarator
1265
4fa992a5
MD
1266storage-class-specifier:
1267 typedef
1268
1269type-specifier:
1270 void
1271 char
1272 short
1273 int
1274 long
1275 float
1276 double
1277 signed
1278 unsigned
1279 _Bool
1280 _Complex
9dfcfc0f
MD
1281 struct-specifier
1282 variant-specifier
4fa992a5
MD
1283 enum-specifier
1284 typedef-name
1285 ctf-type-specifier
1286
1287struct-specifier:
3b0f8e4d 1288 struct identifier-opt { struct-or-variant-declaration-list-opt }
4fa992a5
MD
1289 struct identifier
1290
1291struct-or-variant-declaration-list:
1292 struct-or-variant-declaration
1293 struct-or-variant-declaration-list struct-or-variant-declaration
1294
1295struct-or-variant-declaration:
1296 specifier-qualifier-list struct-or-variant-declarator-list ;
550aca33 1297 declaration-specifiers storage-class-specifier declaration-specifiers declarator-list ;
d285084f
MD
1298 typealias declaration-specifiers abstract-declarator-list : declaration-specifiers abstract-declarator-list ;
1299 typealias declaration-specifiers abstract-declarator-list : declarator-list ;
4fa992a5
MD
1300
1301specifier-qualifier-list:
1302 type-specifier specifier-qualifier-list-opt
1303 type-qualifier specifier-qualifier-list-opt
1304
1305struct-or-variant-declarator-list:
1306 struct-or-variant-declarator
1307 struct-or-variant-declarator-list , struct-or-variant-declarator
1308
1309struct-or-variant-declarator:
1310 declarator
1311 declarator-opt : constant-expression
1312
1313variant-specifier:
1314 variant identifier-opt variant-tag-opt { struct-or-variant-declaration-list }
1315 variant identifier variant-tag
1316
1317variant-tag:
1318 < identifier >
1319
1320enum-specifier:
1321 enum identifier-opt { enumerator-list }
1322 enum identifier-opt { enumerator-list , }
1323 enum identifier
359894ac
MD
1324 enum identifier-opt < declaration-specifiers > { enumerator-list }
1325 enum identifier-opt < declaration-specifiers > { enumerator-list , }
1326 enum identifier < declaration-specifiers >
4fa992a5
MD
1327 enum identifier-opt < integer-constant > { enumerator-list }
1328 enum identifier-opt < integer-constant > { enumerator-list , }
1329 enum identifier < integer-constant >
1330
1331enumerator-list:
1332 enumerator
1333 enumerator-list , enumerator
1334
1335enumerator:
1336 enumeration-constant
1337 enumeration-constant = constant-expression
1338 enumeration-constant = constant-expression-range
1339
1340type-qualifier:
1341 const
1342
1343declarator:
1344 pointer-opt direct-declarator
1345
1346direct-declarator:
1347 identifier
1348 ( declarator )
1349 direct-declarator [ type-specifier ]
1350 direct-declarator [ constant-expression ]
1351
d285084f
MD
1352abstract-declarator:
1353 pointer-opt direct-abstract-declarator
1354
1355direct-abstract-declarator:
1356 identifier-opt
1357 ( abstract-declarator )
1358 direct-abstract-declarator [ type-specifier ]
1359 direct-abstract-declarator [ constant-expression ]
1360 direct-abstract-declarator [ ]
1361
4fa992a5 1362pointer:
3b0f8e4d
MD
1363 * type-qualifier-list-opt
1364 * type-qualifier-list-opt pointer
4fa992a5
MD
1365
1366type-qualifier-list:
1367 type-qualifier
1368 type-qualifier-list type-qualifier
1369
4fa992a5
MD
1370typedef-name:
1371 identifier
1372
13732.3) CTF-specific declarations
1374
1375ctf-specifier:
1376 event { ctf-assignment-expression-list-opt }
1377 stream { ctf-assignment-expression-list-opt }
1378 trace { ctf-assignment-expression-list-opt }
d285084f
MD
1379 typealias declaration-specifiers abstract-declarator-list : declaration-specifiers abstract-declarator-list ;
1380 typealias declaration-specifiers abstract-declarator-list : declarator-list ;
4fa992a5
MD
1381
1382ctf-type-specifier:
1383 floating_point { ctf-assignment-expression-list-opt }
1384 integer { ctf-assignment-expression-list-opt }
1385 string { ctf-assignment-expression-list-opt }
1386
1387ctf-assignment-expression-list:
1388 ctf-assignment-expression
1389 ctf-assignment-expression-list ; ctf-assignment-expression
1390
1391ctf-assignment-expression:
1392 unary-expression assignment-operator unary-expression
1393 unary-expression type-assignment-operator type-specifier
550aca33 1394 declaration-specifiers storage-class-specifier declaration-specifiers declarator-list
d285084f
MD
1395 typealias declaration-specifiers abstract-declarator-list : declaration-specifiers abstract-declarator-list
1396 typealias declaration-specifiers abstract-declarator-list : declarator-list
This page took 0.140609 seconds and 4 git commands to generate.