1 Common Trace Format (CTF) Specification (v1.8.2)
3 Mathieu Desnoyers, EfficiOS Inc.
5 The goal of the present document is to specify a trace format that suits the
6 needs of the embedded, telecom, high-performance and kernel communities. It is
7 based on the Common Trace Format Requirements (v1.4) document. It is designed to
8 allow traces to be natively generated by the Linux kernel, Linux user-space
9 applications written in C/C++, and hardware components. One major element of
10 CTF is the Trace Stream Description Language (TSDL) which flexibility
11 enables description of various binary trace stream layouts.
13 The latest version of this document can be found at:
15 git tree: git://git.efficios.com/ctf.git
16 gitweb: http://git.efficios.com/?p=ctf.git
18 A reference implementation of a library to read and write this trace format is
19 being implemented within the BabelTrace project, a converter between trace
20 formats. The development tree is available at:
22 git tree: git://git.efficios.com/babeltrace.git
23 gitweb: http://git.efficios.com/?p=babeltrace.git
25 The CE Workgroup of the Linux Foundation, Ericsson, and EfficiOS have
31 1. Preliminary definitions
32 2. High-level representation of a trace
36 4.1.1 Type inheritance
46 4.2.2 Variants (Discriminated/Tagged Unions)
50 5. Event Packet Header
51 5.1 Event Packet Header Description
52 5.2 Event Packet Context Description
55 6.1.1 Type 1 - Few event IDs
56 6.1.2 Type 2 - Many event IDs
57 6.2 Stream Event Context and Event Context
61 7. Trace Stream Description Language (TSDL)
63 7.2 Declaration vs Definition
66 7.3.2 Static and Dynamic Scopes
71 1. Preliminary definitions
73 - Event Trace: An ordered sequence of events.
74 - Event Stream: An ordered sequence of events, containing a subset of the
76 - Event Packet: A sequence of physically contiguous events within an event
78 - Event: This is the basic entry in a trace. (aka: a trace record).
79 - An event identifier (ID) relates to the class (a type) of event within
81 e.g. event: irq_entry.
82 - An event (or event record) relates to a specific instance of an event
84 e.g. event: irq_entry, at time X, on CPU Y
85 - Source Architecture: Architecture writing the trace.
86 - Reader Architecture: Architecture reading the trace.
89 2. High-level representation of a trace
91 A trace is divided into multiple event streams. Each event stream contains a
92 subset of the trace event types.
94 The final output of the trace, after its generation and optional transport over
95 the network, is expected to be either on permanent or temporary storage in a
96 virtual file system. Because each event stream is appended to while a trace is
97 being recorded, each is associated with a distinct set of files for
98 output. Therefore, a stored trace can be represented as a directory
99 containing zero, one or more files per stream.
101 Meta-data description associated with the trace contains information on
102 trace event types expressed in the Trace Stream Description Language
103 (TSDL). This language describes:
107 - Per-trace event header description.
108 - Per-stream event header description.
109 - Per-stream event context description.
111 - Event type to stream mapping.
112 - Event type to name mapping.
113 - Event type to ID mapping.
114 - Event context description.
115 - Event fields description.
120 An event stream can be divided into contiguous event packets of variable
121 size. An event packet can contain a certain amount of padding at the
122 end. The stream header is repeated at the beginning of each event
123 packet. The rationale for the event stream design choices is explained
124 in Appendix B. Stream Header Rationale.
126 The event stream header will therefore be referred to as the "event packet
127 header" throughout the rest of this document.
132 Types are organized as type classes. Each type class belong to either of two
133 kind of types: basic types or compound types.
137 A basic type is a scalar type, as described in this section. It includes
138 integers, GNU/C bitfields, enumerations, and floating point values.
140 4.1.1 Type inheritance
142 Type specifications can be inherited to allow deriving types from a
143 type class. For example, see the uint32_t named type derived from the "integer"
144 type class below ("Integers" section). Types have a precise binary
145 representation in the trace. A type class has methods to read and write these
146 types, but must be derived into a type to be usable in an event field.
150 We define "byte-packed" types as aligned on the byte size, namely 8-bit.
151 We define "bit-packed" types as following on the next bit, as defined by the
154 Each basic type must specify its alignment, in bits. Examples of
155 possible alignments are: bit-packed (align = 1), byte-packed (align =
156 8), or word-aligned (e.g. align = 32 or align = 64). The choice depends
157 on the architecture preference and compactness vs performance trade-offs
158 of the implementation. Architectures providing fast unaligned write
159 byte-packed basic types to save space, aligning each type on byte
160 boundaries (8-bit). Architectures with slow unaligned writes align types
161 on specific alignment values. If no specific alignment is declared for a
162 type, it is assumed to be bit-packed for integers with size not multiple
163 of 8 bits and for gcc bitfields. All other basic types are byte-packed
164 by default. It is however recommended to always specify the alignment
165 explicitly. Alignment values must be power of two. Compound types are
166 aligned as specified in their individual specification.
168 The base offset used for field alignment is the start of the packet
169 containing the field. For instance, a field aligned on 32-bit needs to
170 be at an offset multiple of 32-bit from the start of the packet that
173 TSDL meta-data attribute representation of a specific alignment:
175 align = value; /* value in bits */
179 By default, byte order of a basic type is the byte order described in
180 the trace description. It can be overridden by specifying a
181 "byte_order" attribute for a basic type. Typical use-case is to specify
182 the network byte order (big endian: "be") to save data captured from the
183 network into the trace without conversion.
185 TSDL meta-data representation:
187 byte_order = native OR network OR be OR le; /* network and be are aliases */
189 The "native" keyword selects the byte order described in the trace
190 description. The "network" byte order is an alias for big endian.
192 Even though the trace description section is not per se a type, for sake
193 of clarity, it should be noted that "native" and "network" byte orders
194 are only allowed within type declaration. The byte_order specified in
195 the trace description section only accepts "be" or "le" values.
199 Type size, in bits, for integers and floats is that returned by "sizeof()" in C
200 multiplied by CHAR_BIT.
201 We require the size of "char" and "unsigned char" types (CHAR_BIT) to be fixed
202 to 8 bits for cross-endianness compatibility.
204 TSDL meta-data representation:
206 size = value; (value is in bits)
210 Signed integers are represented in two-complement. Integer alignment,
211 size, signedness and byte ordering are defined in the TSDL meta-data.
212 Integers aligned on byte size (8-bit) and with length multiple of byte
213 size (8-bit) correspond to the C99 standard integers. In addition,
214 integers with alignment and/or size that are _not_ a multiple of the
215 byte size are permitted; these correspond to the C99 standard bitfields,
216 with the added specification that the CTF integer bitfields have a fixed
217 binary representation. Integer size needs to be a positive integer.
218 Integers of size 0 are forbidden. A MIT-licensed reference
219 implementation of the CTF portable bitfields is available at:
221 http://git.efficios.com/?p=babeltrace.git;a=blob;f=include/babeltrace/bitfield.h
223 Binary representation of integers:
225 - On little and big endian:
226 - Within a byte, high bits correspond to an integer high bits, and low bits
227 correspond to low bits.
229 - Integer across multiple bytes are placed from the less significant to the
231 - Consecutive integers are placed from lower bits to higher bits (even within
234 - Integer across multiple bytes are placed from the most significant to the
236 - Consecutive integers are placed from higher bits to lower bits (even within
239 This binary representation is derived from the bitfield implementation in GCC
240 for little and big endian. However, contrary to what GCC does, integers can
241 cross units boundaries (no padding is required). Padding can be explicitly
242 added (see 4.1.6 GNU/C bitfields) to follow the GCC layout if needed.
244 TSDL meta-data representation:
247 signed = true OR false; /* default false */
248 byte_order = native OR network OR be OR le; /* default native */
249 size = value; /* value in bits, no default */
250 align = value; /* value in bits */
251 /* based used for pretty-printing output, default: decimal. */
252 base = decimal OR dec OR d OR i OR u OR 10 OR hexadecimal OR hex OR x OR X OR p OR 16
253 OR octal OR oct OR o OR 8 OR binary OR b OR 2;
254 /* character encoding, default: none */
255 encoding = none or UTF8 or ASCII;
258 Example of type inheritance (creation of a uint32_t named type):
266 Definition of a named 5-bit signed bitfield:
274 The character encoding field can be used to specify that the integer
275 must be printed as a text character when read. e.g.:
285 4.1.6 GNU/C bitfields
287 The GNU/C bitfields follow closely the integer representation, with a
288 particularity on alignment: if a bitfield cannot fit in the current unit, the
289 unit is padded and the bitfield starts at the following unit. The unit size is
290 defined by the size of the type "unit_type".
292 TSDL meta-data representation:
296 As an example, the following structure declared in C compiled by GCC:
303 The example structure is aligned on the largest element (short). The second
304 bitfield would be aligned on the next unit boundary, because it would not fit in
309 The floating point values byte ordering is defined in the TSDL meta-data.
311 Floating point values follow the IEEE 754-2008 standard interchange formats.
312 Description of the floating point values include the exponent and mantissa size
313 in bits. Some requirements are imposed on the floating point values:
315 - FLT_RADIX must be 2.
316 - mant_dig is the number of digits represented in the mantissa. It is specified
317 by the ISO C99 standard, section 5.2.4, as FLT_MANT_DIG, DBL_MANT_DIG and
318 LDBL_MANT_DIG as defined by <float.h>.
319 - exp_dig is the number of digits represented in the exponent. Given that
320 mant_dig is one bit more than its actual size in bits (leading 1 is not
321 needed) and also given that the sign bit always takes one bit, exp_dig can be
324 - sizeof(float) * CHAR_BIT - FLT_MANT_DIG
325 - sizeof(double) * CHAR_BIT - DBL_MANT_DIG
326 - sizeof(long double) * CHAR_BIT - LDBL_MANT_DIG
328 TSDL meta-data representation:
333 byte_order = native OR network OR be OR le;
337 Example of type inheritance:
339 typealias floating_point {
340 exp_dig = 8; /* sizeof(float) * CHAR_BIT - FLT_MANT_DIG */
341 mant_dig = 24; /* FLT_MANT_DIG */
346 TODO: define NaN, +inf, -inf behavior.
348 Bit-packed, byte-packed or larger alignments can be used for floating
349 point values, similarly to integers.
353 Enumerations are a mapping between an integer type and a table of strings. The
354 numerical representation of the enumeration follows the integer type specified
355 by the meta-data. The enumeration mapping table is detailed in the enumeration
356 description within the meta-data. The mapping table maps inclusive value
357 ranges (or single values) to strings. Instead of being limited to simple
358 "value -> string" mappings, these enumerations map
359 "[ start_value ... end_value ] -> string", which map inclusive ranges of
360 values to strings. An enumeration from the C language can be represented in
361 this format by having the same start_value and end_value for each element, which
362 is in fact a range of size 1. This single-value range is supported without
363 repeating the start and end values with the value = string declaration.
364 Enumerations need to contain at least one entry.
366 enum name : integer_type {
367 somestring = start_value1 ... end_value1,
368 "other string" = start_value2 ... end_value2,
369 yet_another_string, /* will be assigned to end_value2 + 1 */
370 "some other string" = value,
374 If the values are omitted, the enumeration starts at 0 and increment of 1 for
375 each entry. An entry with omitted value that follows a range entry takes
376 as value the end_value of the previous range + 1:
378 enum name : unsigned int {
386 Overlapping ranges within a single enumeration are implementation defined.
388 A nameless enumeration can be declared as a field type or as part of a typedef:
390 enum : integer_type {
394 Enumerations omitting the container type ": integer_type" use the "int"
395 type (for compatibility with C99). The "int" type must be previously
398 typealias integer { size = 32; align = 32; signed = true; } := int;
407 Compound are aggregation of type declarations. Compound types include
408 structures, variant, arrays, sequences, and strings.
412 Structures are aligned on the largest alignment required by basic types
413 contained within the structure. (This follows the ISO/C standard for structures)
415 TSDL meta-data representation of a named structure:
418 field_type field_name;
419 field_type field_name;
426 integer { /* Nameless type */
431 uint64_t second_field_name; /* Named type declared in the meta-data */
434 The fields are placed in a sequence next to each other. They each
435 possess a field name, which is a unique identifier within the structure.
436 The identifier is not allowed to use any reserved keyword
437 (see Section C.1.2). Replacing reserved keywords with
438 underscore-prefixed field names is recommended. Fields starting with an
439 underscore should have their leading underscore removed by the CTF trace
442 A nameless structure can be declared as a field type or as part of a typedef:
448 Alignment for a structure compound type can be forced to a minimum value
449 by adding an "align" specifier after the declaration of a structure
450 body. This attribute is read as: align(value). The value is specified in
451 bits. The structure will be aligned on the maximum value between this
452 attribute and the alignment required by the basic types contained within
459 4.2.2 Variants (Discriminated/Tagged Unions)
461 A CTF variant is a selection between different types. A CTF variant must
462 always be defined within the scope of a structure or within fields
463 contained within a structure (defined recursively). A "tag" enumeration
464 field must appear in either the same static scope, prior to the variant
465 field (in field declaration order), in an upper static scope , or in an
466 upper dynamic scope (see Section 7.3.2). The type selection is indicated
467 by the mapping from the enumeration value to the string used as variant
468 type selector. The field to use as tag is specified by the "tag_field",
469 specified between "< >" after the "variant" keyword for unnamed
470 variants, and after "variant name" for named variants.
472 The alignment of the variant is the alignment of the type as selected by
473 the tag value for the specific instance of the variant. The size of the
474 variant is the size as selected by the tag value for the specific
475 instance of the variant.
477 The alignment of the type containing the variant is independent of the
478 variant alignment. For instance, if a structure contains two fields, a
479 32-bit integer, aligned on 32 bits, and a variant, which contains two
480 choices: either a 32-bit field, aligned on 32 bits, or a 64-bit field,
481 aligned on 64 bits, the alignment of the outmost structure will be
482 32-bit (the alignment of its largest field, disregarding the alignment
483 of the variant). The alignment of the variant will depend on the
484 selector: if the variant's 32-bit field is selected, its alignment will
485 be 32-bit, or 64-bit otherwise. It is important to note that variants
486 are specifically tailored for compactness in a stream. Therefore, the
487 relative offsets of compound type fields can vary depending on
488 the offset at which the compound type starts if it contains a variant
489 that itself contains a type with alignment larger than the largest field
490 contained within the compound type. This is caused by the fact that the
491 compound type may contain the enumeration that select the variant's
492 choice, and therefore the alignment to be applied to the compound type
493 cannot be determined before encountering the enumeration.
495 Each variant type selector possess a field name, which is a unique
496 identifier within the variant. The identifier is not allowed to use any
497 reserved keyword (see Section C.1.2). Replacing reserved keywords with
498 underscore-prefixed field names is recommended. Fields starting with an
499 underscore should have their leading underscore removed by the CTF trace
503 A named variant declaration followed by its definition within a structure
514 enum : integer_type { sel1, sel2, sel3, ... } tag_field;
516 variant name <tag_field> v;
519 An unnamed variant definition within a structure is expressed by the following
523 enum : integer_type { sel1, sel2, sel3, ... } tag_field;
525 variant <tag_field> {
533 Example of a named variant within a sequence that refers to a single tag field:
542 enum : uint2_t { a, b, c } choice;
544 variant example <choice> v[seqlen];
547 Example of an unnamed variant:
550 enum : uint2_t { a, b, c, d } choice;
551 /* Unrelated fields can be added between the variant and its tag */
564 Example of an unnamed variant within an array:
567 enum : uint2_t { a, b, c } choice;
575 Example of a variant type definition within a structure, where the defined type
576 is then declared within an array of structures. This variant refers to a tag
577 located in an upper static scope. This example clearly shows that a variant
578 type definition referring to the tag "x" uses the closest preceding field from
579 the static scope of the type definition.
582 enum : uint2_t { a, b, c, d } x;
584 typedef variant <x> { /*
585 * "x" refers to the preceding "x" enumeration in the
586 * static scope of the type definition.
594 enum : int { x, y, z } x; /* This enumeration is not used by "v". */
595 example_variant v; /*
596 * "v" uses the "enum : uint2_t { a, b, c, d }"
604 Arrays are fixed-length. Their length is declared in the type
605 declaration within the meta-data. They contain an array of "inner type"
606 elements, which can refer to any type not containing the type of the
607 array being declared (no circular dependency). The length is the number
608 of elements in an array.
610 TSDL meta-data representation of a named array:
612 typedef elem_type name[length];
614 A nameless array can be declared as a field type within a structure, e.g.:
616 uint8_t field_name[10];
618 Arrays are always aligned on their element alignment requirement.
622 Sequences are dynamically-sized arrays. They refer to a "length"
623 unsigned integer field, which must appear in either the same static scope,
624 prior to the sequence field (in field declaration order), in an upper
625 static scope, or in an upper dynamic scope (see Section 7.3.2). This
626 length field represents the number of elements in the sequence. The
627 sequence per se is an array of "inner type" elements.
629 TSDL meta-data representation for a sequence type definition:
632 unsigned int length_field;
633 typedef elem_type typename[length_field];
634 typename seq_field_name;
637 A sequence can also be declared as a field type, e.g.:
640 unsigned int length_field;
641 long seq_field_name[length_field];
644 Multiple sequences can refer to the same length field, and these length
645 fields can be in a different upper dynamic scope:
647 e.g., assuming the stream.event.header defines:
652 event.header := struct {
661 long seq_a[stream.event.header.seq_len];
662 char seq_b[stream.event.header.seq_len];
666 The sequence elements follow the "array" specifications.
670 Strings are an array of bytes of variable size and are terminated by a '\0'
671 "NULL" character. Their encoding is described in the TSDL meta-data. In
672 absence of encoding attribute information, the default encoding is
675 TSDL meta-data representation of a named string type:
678 encoding = UTF8 OR ASCII;
681 A nameless string type can be declared as a field type:
683 string field_name; /* Use default UTF8 encoding */
685 Strings are always aligned on byte size.
687 5. Event Packet Header
689 The event packet header consists of two parts: the "event packet header"
690 is the same for all streams of a trace. The second part, the "event
691 packet context", is described on a per-stream basis. Both are described
692 in the TSDL meta-data.
694 Event packet header (all fields are optional, specified by TSDL meta-data):
696 - Magic number (CTF magic number: 0xC1FC1FC1) specifies that this is a
697 CTF packet. This magic number is optional, but when present, it should
698 come at the very beginning of the packet.
699 - Trace UUID, used to ensure the event packet match the meta-data used.
700 (note: we cannot use a meta-data checksum in every cases instead of a
701 UUID because meta-data can be appended to while tracing is active)
702 This field is optional.
703 - Stream ID, used as reference to stream description in meta-data.
704 This field is optional if there is only one stream description in the
705 meta-data, but becomes required if there are more than one stream in
706 the TSDL meta-data description.
708 Event packet context (all fields are optional, specified by TSDL meta-data):
710 - Event packet content size (in bits).
711 - Event packet size (in bits, includes padding).
712 - Event packet content checksum. Checksum excludes the event packet
714 - Per-stream event packet sequence count (to deal with UDP packet loss). The
715 number of significant sequence counter bits should also be present, so
716 wrap-arounds are dealt with correctly.
717 - Time-stamp at the beginning and time-stamp at the end of the event packet.
718 Both timestamps are written in the packet header, but sampled respectively
719 while (or before) writing the first event and while (or after) writing the
720 last event in the packet. The inclusive range between these timestamps should
721 include all event timestamps assigned to events contained within the packet.
722 The timestamp at the beginning of an event packet is guaranteed to be
723 below or equal the timestamp at the end of that event packet.
724 The timestamp at the end of an event packet is guaranteed to be below
725 or equal the timestamps at the end of any following packet within the
726 same stream. See Section 8. Clocks for more detail.
727 - Events discarded count
728 - Snapshot of a per-stream free-running counter, counting the number of
729 events discarded that were supposed to be written in the stream after
730 the last event in the event packet.
731 * Note: producer-consumer buffer full condition can fill the current
732 event packet with padding so we know exactly where events have been
733 discarded. However, if the buffer full condition chooses not
734 to fill the current event packet with padding, all we know
735 about the timestamp range in which the events have been
736 discarded is that it is somewhere between the beginning and
737 the end of the packet.
738 - Lossless compression scheme used for the event packet content. Applied
739 directly to raw data. New types of compression can be added in following
740 versions of the format.
741 0: no compression scheme
745 - Cypher used for the event packet content. Applied after compression.
748 - Checksum scheme used for the event packet content. Applied after encryption.
754 5.1 Event Packet Header Description
756 The event packet header layout is indicated by the trace packet.header
757 field. Here is a recommended structure type for the packet header with
758 the fields typically expected (although these fields are each optional):
760 struct event_packet_header {
768 packet.header := struct event_packet_header;
771 If the magic number is not present, tools such as "file" will have no
772 mean to discover the file type.
774 If the uuid is not present, no validation that the meta-data actually
775 corresponds to the stream is performed.
777 If the stream_id packet header field is missing, the trace can only
778 contain a single stream. Its "id" field can be left out, and its events
779 don't need to declare a "stream_id" field.
782 5.2 Event Packet Context Description
784 Event packet context example. These are declared within the stream declaration
785 in the meta-data. All these fields are optional. If the packet size field is
786 missing, the whole stream only contains a single packet. If the content
787 size field is missing, the packet is filled (no padding). The content
788 and packet sizes include all headers.
790 An example event packet context type:
792 struct event_packet_context {
793 uint64_t timestamp_begin;
794 uint64_t timestamp_end;
796 uint32_t stream_packet_count;
797 uint32_t events_discarded;
799 uint64_t/uint32_t/uint16_t content_size;
800 uint64_t/uint32_t/uint16_t packet_size;
801 uint8_t compression_scheme;
802 uint8_t encryption_scheme;
803 uint8_t checksum_scheme;
809 The overall structure of an event is:
811 1 - Event Header (as specified by the stream meta-data)
812 2 - Stream Event Context (as specified by the stream meta-data)
813 3 - Event Context (as specified by the event meta-data)
814 4 - Event Payload (as specified by the event meta-data)
816 This structure defines an implicit dynamic scoping, where variants
817 located in inner structures (those with a higher number in the listing
818 above) can refer to the fields of outer structures (with lower number in
819 the listing above). See Section 7.3 TSDL Scopes for more detail.
821 The total length of an event is defined as the difference between the
822 end of its Event Payload and the end of the previous event's Event
823 Payload. Therefore, it includes the event header alignment padding, and
824 all its fields and their respective alignment padding. Events of length
829 Event headers can be described within the meta-data. We hereby propose, as an
830 example, two types of events headers. Type 1 accommodates streams with less than
831 31 event IDs. Type 2 accommodates streams with 31 or more event IDs.
833 One major factor can vary between streams: the number of event IDs assigned to
834 a stream. Luckily, this information tends to stay relatively constant (modulo
835 event registration while trace is being recorded), so we can specify different
836 representations for streams containing few event IDs and streams containing
837 many event IDs, so we end up representing the event ID and time-stamp as
838 densely as possible in each case.
840 The header is extended in the rare occasions where the information cannot be
841 represented in the ranges available in the standard event header. They are also
842 used in the rare occasions where the data required for a field could not be
843 collected: the flag corresponding to the missing field within the missing_fields
844 array is then set to 1.
846 Types uintX_t represent an X-bit unsigned integer, as declared with
849 typealias integer { size = X; align = X; signed = false; } := uintX_t;
853 typealias integer { size = X; align = 1; signed = false; } := uintX_t;
855 For more information about timestamp fields, see Section 8. Clocks.
857 6.1.1 Type 1 - Few event IDs
859 - Aligned on 32-bit (or 8-bit if byte-packed, depending on the architecture
861 - Native architecture byte ordering.
862 - For "compact" selection
863 - Fixed size: 32 bits.
864 - For "extended" selection
865 - Size depends on the architecture and variant alignment.
867 struct event_header_1 {
870 * id 31 is reserved to indicate an extended header.
872 enum : uint5_t { compact = 0 ... 30, extended = 31 } id;
878 uint32_t id; /* 32-bit event IDs */
879 uint64_t timestamp; /* 64-bit timestamps */
882 } align(32); /* or align(8) */
885 6.1.2 Type 2 - Many event IDs
887 - Aligned on 16-bit (or 8-bit if byte-packed, depending on the architecture
889 - Native architecture byte ordering.
890 - For "compact" selection
891 - Size depends on the architecture and variant alignment.
892 - For "extended" selection
893 - Size depends on the architecture and variant alignment.
895 struct event_header_2 {
897 * id: range: 0 - 65534.
898 * id 65535 is reserved to indicate an extended header.
900 enum : uint16_t { compact = 0 ... 65534, extended = 65535 } id;
906 uint32_t id; /* 32-bit event IDs */
907 uint64_t timestamp; /* 64-bit timestamps */
910 } align(16); /* or align(8) */
913 6.2 Stream Event Context and Event Context
915 The event context contains information relative to the current event.
916 The choice and meaning of this information is specified by the TSDL
917 stream and event meta-data descriptions. The stream context is applied
918 to all events within the stream. The stream context structure follows
919 the event header. The event context is applied to specific events. Its
920 structure follows the stream context structure.
922 An example of stream-level event context is to save the event payload size with
923 each event, or to save the current PID with each event. These are declared
924 within the stream declaration within the meta-data:
928 event.context := struct {
930 uint16_t payload_size;
934 An example of event-specific event context is to declare a bitmap of missing
935 fields, only appended after the stream event context if the extended event
936 header is selected. NR_FIELDS is the number of fields within the event (a
944 uint1_t missing_fields[NR_FIELDS]; /* missing event fields bitmap */
953 An event payload contains fields specific to a given event type. The fields
954 belonging to an event type are described in the event-specific meta-data
955 within a structure type.
959 No padding at the end of the event payload. This differs from the ISO/C standard
960 for structures, but follows the CTF standard for structures. In a trace, even
961 though it makes sense to align the beginning of a structure, it really makes no
962 sense to add padding at the end of the structure, because structures are usually
963 not followed by a structure of the same type.
965 This trick can be done by adding a zero-length "end" field at the end of the C
966 structures, and by using the offset of this field rather than using sizeof()
967 when calculating the size of a structure (see Appendix "A. Helper macros").
971 The event payload is aligned on the largest alignment required by types
972 contained within the payload. (This follows the ISO/C standard for structures)
975 7. Trace Stream Description Language (TSDL)
977 The Trace Stream Description Language (TSDL) allows expression of the
978 binary trace streams layout in a C99-like Domain Specific Language
984 The trace stream layout description is located in the trace meta-data.
985 The meta-data is itself located in a stream identified by its name:
988 The meta-data description can be expressed in two different formats:
989 text-only and packet-based. The text-only description facilitates
990 generation of meta-data and provides a convenient way to enter the
991 meta-data information by hand. The packet-based meta-data provides the
992 CTF stream packet facilities (checksumming, compression, encryption,
993 network-readiness) for meta-data stream generated and transported by a
996 The text-only meta-data file is a plain-text TSDL description. This file
997 must begin with the following characters to identify the file as a CTF
998 TSDL text-based metadata file (without the double-quotes) :
1002 It must be followed by a space, and the version of the specification
1003 followed by the CTF trace, e.g.:
1007 These characters allow automated discovery of file type and CTF
1008 specification version. They are interpreted as a the beginning of a
1009 comment by the TSDL metadata parser. The comment can be continued to
1010 contain extra commented characters before it is closed.
1012 The packet-based meta-data is made of "meta-data packets", which each
1013 start with a meta-data packet header. The packet-based meta-data
1014 description is detected by reading the magic number "0x75D11D57" at the
1015 beginning of the file. This magic number is also used to detect the
1016 endianness of the architecture by trying to read the CTF magic number
1017 and its counterpart in reversed endianness. The events within the
1018 meta-data stream have no event header nor event context. Each event only
1019 contains a special "sequence" payload, which is a sequence of bits which
1020 length is implicitly calculated by using the
1021 "trace.packet.header.content_size" field, minus the packet header size.
1022 The formatting of this sequence of bits is a plain-text representation
1023 of the TSDL description. Each meta-data packet start with a special
1024 packet header, specific to the meta-data stream, which contains,
1027 struct metadata_packet_header {
1028 uint32_t magic; /* 0x75D11D57 */
1029 uint8_t uuid[16]; /* Unique Universal Identifier */
1030 uint32_t checksum; /* 0 if unused */
1031 uint32_t content_size; /* in bits */
1032 uint32_t packet_size; /* in bits */
1033 uint8_t compression_scheme; /* 0 if unused */
1034 uint8_t encryption_scheme; /* 0 if unused */
1035 uint8_t checksum_scheme; /* 0 if unused */
1036 uint8_t major; /* CTF spec version major number */
1037 uint8_t minor; /* CTF spec version minor number */
1040 The packet-based meta-data can be converted to a text-only meta-data by
1041 concatenating all the strings it contains.
1043 In the textual representation of the meta-data, the text contained
1044 within "/*" and "*/", as well as within "//" and end of line, are
1045 treated as comments. Boolean values can be represented as true, TRUE,
1046 or 1 for true, and false, FALSE, or 0 for false. Within the string-based
1047 meta-data description, the trace UUID is represented as a string of
1048 hexadecimal digits and dashes "-". In the event packet header, the trace
1049 UUID is represented as an array of bytes.
1052 7.2 Declaration vs Definition
1054 A declaration associates a layout to a type, without specifying where
1055 this type is located in the event structure hierarchy (see Section 6).
1056 This therefore includes typedef, typealias, as well as all type
1057 specifiers. In certain circumstances (typedef, structure field and
1058 variant field), a declaration is followed by a declarator, which specify
1059 the newly defined type name (for typedef), or the field name (for
1060 declarations located within structure and variants). Array and sequence,
1061 declared with square brackets ("[" "]"), are part of the declarator,
1062 similarly to C99. The enumeration base type is specified by
1063 ": enum_base", which is part of the type specifier. The variant tag
1064 name, specified between "<" ">", is also part of the type specifier.
1066 A definition associates a type to a location in the event structure
1067 hierarchy (see Section 6). This association is denoted by ":=", as shown
1073 TSDL uses three different types of scoping: a lexical scope is used for
1074 declarations and type definitions, and static and dynamic scopes are
1075 used for variants references to tag fields (with relative and absolute
1076 path lookups) and for sequence references to length fields.
1080 Each of "trace", "env", "stream", "event", "struct" and "variant" have
1081 their own nestable declaration scope, within which types can be declared
1082 using "typedef" and "typealias". A root declaration scope also contains
1083 all declarations located outside of any of the aforementioned
1084 declarations. An inner declaration scope can refer to type declared
1085 within its container lexical scope prior to the inner declaration scope.
1086 Redefinition of a typedef or typealias is not valid, although hiding an
1087 upper scope typedef or typealias is allowed within a sub-scope.
1089 7.3.2 Static and Dynamic Scopes
1091 A local static scope consists in the scope generated by the declaration
1092 of fields within a compound type. A static scope is a local static scope
1093 augmented with the nested sub-static-scopes it contains.
1095 A dynamic scope consists in the static scope augmented with the
1096 implicit event structure definition hierarchy presented at Section 6.
1098 Multiple declarations of the same field name within a local static scope
1099 is not valid. It is however valid to re-use the same field name in
1100 different local scopes.
1102 Nested static and dynamic scopes form lookup paths. These are used for
1103 variant tag and sequence length references. They are used at the variant
1104 and sequence definition site to look up the location of the tag field
1105 associated with a variant, and to lookup up the location of the length
1106 field associated with a sequence.
1108 Variants and sequences can refer to a tag field either using a relative
1109 path or an absolute path. The relative path is relative to the scope in
1110 which the variant or sequence performing the lookup is located.
1111 Relative paths are only allowed to lookup within the same static scope,
1112 which includes its nested static scopes. Lookups targeting parent static
1113 scopes need to be performed with an absolute path.
1115 Absolute path lookups use the full path including the dynamic scope
1116 followed by a "." and then the static scope. Therefore, variants (or
1117 sequences) in lower levels in the dynamic scope (e.g. event context) can
1118 refer to a tag (or length) field located in upper levels (e.g. in the
1119 event header) by specifying, in this case, the associated tag with
1120 <stream.event.header.field_name>. This allows, for instance, the event
1121 context to define a variant referring to the "id" field of the event
1124 The dynamic scope prefixes are thus:
1126 - Trace Environment: <env. >,
1127 - Trace Packet Header: <trace.packet.header. >,
1128 - Stream Packet Context: <stream.packet.context. >,
1129 - Event Header: <stream.event.header. >,
1130 - Stream Event Context: <stream.event.context. >,
1131 - Event Context: <event.context. >,
1132 - Event Payload: <event.fields. >.
1135 The target dynamic scope must be specified explicitly when referring to
1136 a field outside of the static scope (absolute scope reference). No
1137 conflict can occur between relative and dynamic paths, because the
1138 keywords "trace", "stream", and "event" are reserved, and thus
1139 not permitted as field names. It is recommended that field names
1140 clashing with CTF and C99 reserved keywords use an underscore prefix to
1141 eliminate the risk of generating a description containing an invalid
1142 field name. Consequently, fields starting with an underscore should have
1143 their leading underscore removed by the CTF trace readers.
1146 The information available in the dynamic scopes can be thought of as the
1147 current tracing context. At trace production, information about the
1148 current context is saved into the specified scope field levels. At trace
1149 consumption, for each event, the current trace context is therefore
1150 readable by accessing the upper dynamic scopes.
1155 The grammar representing the TSDL meta-data is presented in Appendix C.
1156 TSDL Grammar. This section presents a rather lighter reading that
1157 consists in examples of TSDL meta-data, with template values.
1159 The stream "id" can be left out if there is only one stream in the
1160 trace. The event "id" field can be left out if there is only one event
1164 major = value; /* CTF spec version major number */
1165 minor = value; /* CTF spec version minor number */
1166 uuid = "aaaaaaaa-aaaa-aaaa-aaaa-aaaaaaaaaaaa"; /* Trace UUID */
1167 byte_order = be OR le; /* Endianness (required) */
1168 packet.header := struct {
1176 * The "env" (environment) scope contains assignment expressions. The
1177 * field names and content are implementation-defined.
1180 pid = value; /* example */
1181 proc_name = "name"; /* example */
1187 /* Type 1 - Few event IDs; Type 2 - Many event IDs. See section 6.1. */
1188 event.header := event_header_1 OR event_header_2;
1189 event.context := struct {
1192 packet.context := struct {
1198 name = "event_name";
1199 id = value; /* Numeric identifier within the stream */
1200 stream_id = stream_id;
1202 model.emf.uri = "string";
1212 name = "event_name";
1219 /* More detail on types in section 4. Types */
1224 * Type declarations behave similarly to the C standard.
1227 typedef aliased_type_specifiers new_type_declarators;
1229 /* e.g.: typedef struct example new_type_name[10]; */
1234 * The "typealias" declaration can be used to give a name (including
1235 * pointer declarator specifier) to a type. It should also be used to
1236 * map basic C types (float, int, unsigned long, ...) to a CTF type.
1237 * Typealias is a superset of "typedef": it also allows assignment of a
1238 * simple variable identifier to a type.
1241 typealias type_class {
1243 } := type_specifiers type_declarator;
1247 * typealias integer {
1251 * } := struct page *;
1253 * typealias integer {
1268 enum name : integer_type {
1274 * Unnamed types, contained within compound type fields, typedef or typealias.
1289 enum : integer_type {
1293 typedef type new_type[length];
1296 type field_name[length];
1299 typedef type new_type[length_type];
1302 type field_name[length_type];
1314 integer_type field_name:size; /* GNU/C bitfield */
1324 Clock metadata allows to describe the clock topology of the system, as
1325 well as to detail each clock parameter. In absence of clock description,
1326 it is assumed that all fields named "timestamp" use the same clock
1327 source, which increments once per nanosecond.
1329 Describing a clock and how it is used by streams is threefold: first,
1330 the clock and clock topology should be described in a "clock"
1331 description block, e.g.:
1334 name = cycle_counter_sync;
1335 uuid = "62189bee-96dc-11e0-91a8-cfa3d89f3923";
1336 description = "Cycle counter synchronized across CPUs";
1337 freq = 1000000000; /* frequency, in Hz */
1338 /* precision in seconds is: 1000 * (1/freq) */
1341 * clock value offset from Epoch is:
1342 * offset_s + (offset * (1/freq))
1344 offset_s = 1326476837;
1349 The mandatory "name" field specifies the name of the clock identifier,
1350 which can later be used as a reference. The optional field "uuid" is the
1351 unique identifier of the clock. It can be used to correlate different
1352 traces that use the same clock. An optional textual description string
1353 can be added with the "description" field. The "freq" field is the
1354 initial frequency of the clock, in Hz. If the "freq" field is not
1355 present, the frequency is assumed to be 1000000000 (providing clock
1356 increment of 1 ns). The optional "precision" field details the
1357 uncertainty on the clock measurements, in (1/freq) units. The "offset_s"
1358 and "offset" fields indicate the offset from POSIX.1 Epoch, 1970-01-01
1359 00:00:00 +0000 (UTC), to the zero of value of the clock. The "offset_s"
1360 field is in seconds. The "offset" field is in (1/freq) units. If any of
1361 the "offset_s" or "offset" field is not present, it is assigned the 0
1362 value. The field "absolute" is TRUE if the clock is a global reference
1363 across different clock uuid (e.g. NTP time). Otherwise, "absolute" is
1364 FALSE, and the clock can be considered as synchronized only with other
1365 clocks that have the same uuid.
1368 Secondly, a reference to this clock should be added within an integer
1372 size = 64; align = 1; signed = false;
1373 map = clock.cycle_counter_sync.value;
1376 Thirdly, stream declarations can reference the clock they use as a
1379 struct packet_context {
1380 uint64_ccnt_t ccnt_begin;
1381 uint64_ccnt_t ccnt_end;
1387 event.header := struct {
1388 uint64_ccnt_t timestamp;
1391 packet.context := struct packet_context;
1394 For a N-bit integer type referring to a clock, if the integer overflows
1395 compared to the N low order bits of the clock prior value found in the
1396 same stream, then it is assumed that one, and only one, overflow
1397 occurred. It is therefore important that events encoding time on a small
1398 number of bits happen frequently enough to detect when more than one
1399 N-bit overflow occurs.
1401 In a packet context, clock field names ending with "_begin" and "_end"
1402 have a special meaning: this refers to the time-stamps at, respectively,
1403 the beginning and the end of each packet.
1408 The two following macros keep track of the size of a GNU/C structure without
1409 padding at the end by placing HEADER_END as the last field. A one byte end field
1410 is used for C90 compatibility (C99 flexible arrays could be used here). Note
1411 that this does not affect the effective structure size, which should always be
1412 calculated with the header_sizeof() helper.
1414 #define HEADER_END char end_field
1415 #define header_sizeof(type) offsetof(typeof(type), end_field)
1418 B. Stream Header Rationale
1420 An event stream is divided in contiguous event packets of variable size. These
1421 subdivisions allow the trace analyzer to perform a fast binary search by time
1422 within the stream (typically requiring to index only the event packet headers)
1423 without reading the whole stream. These subdivisions have a variable size to
1424 eliminate the need to transfer the event packet padding when partially filled
1425 event packets must be sent when streaming a trace for live viewing/analysis.
1426 An event packet can contain a certain amount of padding at the end. Dividing
1427 streams into event packets is also useful for network streaming over UDP and
1428 flight recorder mode tracing (a whole event packet can be swapped out of the
1429 buffer atomically for reading).
1431 The stream header is repeated at the beginning of each event packet to allow
1432 flexibility in terms of:
1434 - streaming support,
1435 - allowing arbitrary buffers to be discarded without making the trace
1437 - allow UDP packet loss handling by either dealing with missing event packet
1438 or asking for re-transmission.
1439 - transparently support flight recorder mode,
1440 - transparently support crash dump.
1446 * Common Trace Format (CTF) Trace Stream Description Language (TSDL) Grammar.
1448 * Inspired from the C99 grammar:
1449 * http://www.open-std.org/jtc1/sc22/wg14/www/docs/n1124.pdf (Annex A)
1450 * and c++1x grammar (draft)
1451 * http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2011/n3291.pdf (Annex A)
1453 * Specialized for CTF needs by including only constant and declarations from
1454 * C99 (excluding function declarations), and by adding support for variants,
1455 * sequences and CTF-specific specifiers. Enumeration container types
1456 * semantic is inspired from c++1x enum-base.
1461 1.1) Lexical elements
1508 identifier identifier-nondigit
1511 identifier-nondigit:
1513 universal-character-name
1514 any other implementation-defined characters
1518 [a-zA-Z] /* regular expression */
1521 [0-9] /* regular expression */
1523 1.4) Universal character names
1525 universal-character-name:
1527 \U hex-quad hex-quad
1530 hexadecimal-digit hexadecimal-digit hexadecimal-digit hexadecimal-digit
1536 enumeration-constant
1540 decimal-constant integer-suffix-opt
1541 octal-constant integer-suffix-opt
1542 hexadecimal-constant integer-suffix-opt
1546 decimal-constant digit
1550 octal-constant octal-digit
1552 hexadecimal-constant:
1553 hexadecimal-prefix hexadecimal-digit
1554 hexadecimal-constant hexadecimal-digit
1564 unsigned-suffix long-suffix-opt
1565 unsigned-suffix long-long-suffix
1566 long-suffix unsigned-suffix-opt
1567 long-long-suffix unsigned-suffix-opt
1581 enumeration-constant:
1587 L' c-char-sequence '
1591 c-char-sequence c-char
1594 any member of source charset except single-quote ('), backslash
1595 (\), or new-line character.
1599 simple-escape-sequence
1600 octal-escape-sequence
1601 hexadecimal-escape-sequence
1602 universal-character-name
1604 simple-escape-sequence: one of
1605 \' \" \? \\ \a \b \f \n \r \t \v
1607 octal-escape-sequence:
1609 \ octal-digit octal-digit
1610 \ octal-digit octal-digit octal-digit
1612 hexadecimal-escape-sequence:
1613 \x hexadecimal-digit
1614 hexadecimal-escape-sequence hexadecimal-digit
1616 1.6) String literals
1619 " s-char-sequence-opt "
1620 L" s-char-sequence-opt "
1624 s-char-sequence s-char
1627 any member of source charset except double-quote ("), backslash
1628 (\), or new-line character.
1634 [ ] ( ) { } . -> * + - < > : ; ... = ,
1637 2) Phrase structure grammar
1643 ( unary-expression )
1647 postfix-expression [ unary-expression ]
1648 postfix-expression . identifier
1649 postfix-expressoin -> identifier
1653 unary-operator postfix-expression
1655 unary-operator: one of
1658 assignment-operator:
1661 type-assignment-operator:
1664 constant-expression-range:
1665 unary-expression ... unary-expression
1670 declaration-specifiers declarator-list-opt ;
1673 declaration-specifiers:
1674 storage-class-specifier declaration-specifiers-opt
1675 type-specifier declaration-specifiers-opt
1676 type-qualifier declaration-specifiers-opt
1680 declarator-list , declarator
1682 abstract-declarator-list:
1684 abstract-declarator-list , abstract-declarator
1686 storage-class-specifier:
1709 align ( unary-expression )
1712 struct identifier-opt { struct-or-variant-declaration-list-opt } align-attribute-opt
1713 struct identifier align-attribute-opt
1715 struct-or-variant-declaration-list:
1716 struct-or-variant-declaration
1717 struct-or-variant-declaration-list struct-or-variant-declaration
1719 struct-or-variant-declaration:
1720 specifier-qualifier-list struct-or-variant-declarator-list ;
1721 declaration-specifiers-opt storage-class-specifier declaration-specifiers-opt declarator-list ;
1722 typealias declaration-specifiers abstract-declarator-list type-assignment-operator declaration-specifiers abstract-declarator-list ;
1723 typealias declaration-specifiers abstract-declarator-list type-assignment-operator declarator-list ;
1725 specifier-qualifier-list:
1726 type-specifier specifier-qualifier-list-opt
1727 type-qualifier specifier-qualifier-list-opt
1729 struct-or-variant-declarator-list:
1730 struct-or-variant-declarator
1731 struct-or-variant-declarator-list , struct-or-variant-declarator
1733 struct-or-variant-declarator:
1735 declarator-opt : unary-expression
1738 variant identifier-opt variant-tag-opt { struct-or-variant-declaration-list }
1739 variant identifier variant-tag
1742 < unary-expression >
1745 enum identifier-opt { enumerator-list }
1746 enum identifier-opt { enumerator-list , }
1748 enum identifier-opt : declaration-specifiers { enumerator-list }
1749 enum identifier-opt : declaration-specifiers { enumerator-list , }
1753 enumerator-list , enumerator
1756 enumeration-constant
1757 enumeration-constant assignment-operator unary-expression
1758 enumeration-constant assignment-operator constant-expression-range
1764 pointer-opt direct-declarator
1769 direct-declarator [ unary-expression ]
1771 abstract-declarator:
1772 pointer-opt direct-abstract-declarator
1774 direct-abstract-declarator:
1776 ( abstract-declarator )
1777 direct-abstract-declarator [ unary-expression ]
1778 direct-abstract-declarator [ ]
1781 * type-qualifier-list-opt
1782 * type-qualifier-list-opt pointer
1784 type-qualifier-list:
1786 type-qualifier-list type-qualifier
1791 2.3) CTF-specific declarations
1794 clock { ctf-assignment-expression-list-opt }
1795 event { ctf-assignment-expression-list-opt }
1796 stream { ctf-assignment-expression-list-opt }
1797 env { ctf-assignment-expression-list-opt }
1798 trace { ctf-assignment-expression-list-opt }
1799 callsite { ctf-assignment-expression-list-opt }
1800 typealias declaration-specifiers abstract-declarator-list type-assignment-operator declaration-specifiers abstract-declarator-list
1801 typealias declaration-specifiers abstract-declarator-list type-assignment-operator declarator-list
1804 floating_point { ctf-assignment-expression-list-opt }
1805 integer { ctf-assignment-expression-list-opt }
1806 string { ctf-assignment-expression-list-opt }
1809 ctf-assignment-expression-list:
1810 ctf-assignment-expression ;
1811 ctf-assignment-expression-list ctf-assignment-expression ;
1813 ctf-assignment-expression:
1814 unary-expression assignment-operator unary-expression
1815 unary-expression type-assignment-operator type-specifier
1816 declaration-specifiers-opt storage-class-specifier declaration-specifiers-opt declarator-list
1817 typealias declaration-specifiers abstract-declarator-list type-assignment-operator declaration-specifiers abstract-declarator-list
1818 typealias declaration-specifiers abstract-declarator-list type-assignment-operator declarator-list