[ctf.git] / common-trace-format-linux-proposal.txt

RFC: Common Trace Format Proposal for Linux (v1.5)

Mathieu Desnoyers, EfficiOS Inc.

The goal of the present document is to propose a trace format that suits the
needs of the embedded, telecom, high-performance and kernel communities.  It is
based on the Common Trace Format Requirements (v1.4) document. It is designed to
allow tracing that is natively generated by the Linux kernel and Linux
user-space applications written in C/C++.

A reference implementation of a library to read and write this trace format is
being implemented within the BabelTrace project, a converter between trace
formats. The development tree is available at:

  git tree:   git://git.efficios.com/babeltrace.git
  gitweb:     http://git.efficios.com/?p=babeltrace.git


1. Preliminary definitions

  - Event Trace: An ordered sequence of events.
  - Event Stream: An ordered sequence of events, containing a subset of the
                  trace event types.
  - Event Packet: A sequence of physically contiguous events within an event
                  stream.
  - Event: This is the basic entry in a trace. (aka: a trace record).
    - An event identifier (ID) relates to the class (a type) of event within
      an event stream.
        e.g. event: irq_entry.
    - An event (or event record) relates to a specific instance of an event
      class.
        e.g. event: irq_entry, at time X, on CPU Y
  - Source Architecture: Architecture writing the trace.
  - Reader Architecture: Architecture reading the trace.


2. High-level representation of a trace

A trace is divided into multiple event streams. Each event stream contains a
subset of the trace event types.

The final output of the trace, after its generation and optional transport over
the network, is expected to be either on permanent or temporary storage in a
virtual file system. Because each event stream is appended to while a trace is
being recorded, each is associated with a separate file for output.  Therefore,
a stored trace can be represented as a directory containing one file per stream.

A metadata event stream contains information on trace event types. It describes:

- Trace version.
- Types available.
- Per-stream event header description.
- Per-stream event header selection.
- Per-stream event context fields.
- Per-event
  - Event type to stream mapping.
  - Event type to name mapping.
  - Event type to ID mapping.
  - Event fields description.


3. Event stream

An event stream is divided in contiguous event packets of variable size. These
subdivisions have a variable size. An event packet can contain a certain amount
of padding at the end. The rationale for the event stream design choices is
explained in Appendix B. Stream Header Rationale.

An event stream is divided in contiguous event packets of variable size. These
subdivisions have a variable size. An event packet can contain a certain amount
of padding at the end.  The stream header is repeated at the beginning of each
event packet.

The event stream header will therefore be referred to as the "event packet
header" throughout the rest of this document.


4. Types

4.1 Basic types

A basic type is a scalar type, as described in this section.

4.1.1 Type inheritance

Type specifications can be inherited to allow deriving concrete types from an
abstract type. For example, see the uint32_t type derived from the "integer"
abstract type below ("Integers" section). Concrete types have a precise binary
representation in the trace. Abstract types have methods to read and write these
types, but must be derived into a concrete type to be usable in an event field.

Concrete types inherit from abstract types. Abstract types can inherit from
other abstract types.

4.1.2 Alignment

We define "byte-packed" types as aligned on the byte size, namely 8-bit.
We define "bit-packed" types as following on the next bit, as defined by the
"bitfields" section.

All basic types, except bitfields, are either aligned on an architecture-defined
specific alignment or byte-packed, depending on the architecture preference.
Architectures providing fast unaligned write byte-packed basic types to save
space, aligning each type on byte boundaries (8-bit). Architectures with slow
unaligned writes align types on specific alignment values. If no specific
alignment is declared for a type nor its parents, it is assumed to be bit-packed
for bitfields and byte-packed for other types.

Metadata attribute representation of a specific alignment:

  align = value;                                /* value in bits */

4.1.3 Byte order

By default, the native endianness of the source architecture the trace is used.
Byte order can be overridden for a basic type by specifying a "byte_order"
attribute. Typical use-case is to specify the network byte order (big endian:
"be") to save data captured from the network into the trace without conversion.
If not specified, the byte order is native.

Metadata representation:

  byte_order = native OR network OR be OR le;   /* network and be are aliases */

4.1.4 Size

Type size, in bits, for integers and floats is that returned by "sizeof()" in C
multiplied by CHAR_BIT.
We require the size of "char" and "unsigned char" types (CHAR_BIT) to be fixed
to 8 bits for cross-endianness compatibility.

Metadata representation:

  size = value;    (value is in bits)

4.1.5 Integers

Signed integers are represented in two-complement. Integer alignment, size,
signedness and byte ordering are defined in the metadata. Integers aligned on
byte size (8-bit) and with length multiple of byte size (8-bit) correspond to
the C99 standard integers. In addition, integers with alignment and/or size that
are _not_ a multiple of the byte size are permitted; these correspond to the C99
standard bitfields, with the added specification that the CTF integer bitfields
have a fixed binary representation. A MIT-licensed reference implementation of
the CTF portable bitfields is available at:

  http://git.efficios.com/?p=babeltrace.git;a=blob;f=include/babeltrace/bitfield.h

Binary representation of integers:

- On little and big endian:
  - Within a byte, high bits correspond to an integer high bits, and low bits
    correspond to low bits.
- On little endian:
  - Integer across multiple bytes are placed from the less significant to the
    most significant.
  - Consecutive integers are placed from lower bits to higher bits (even within
    a byte).
- On big endian:
  - Integer across multiple bytes are placed from the most significant to the
    less significant.
  - Consecutive integers are placed from higher bits to lower bits (even within
    a byte).

This binary representation is derived from the bitfield implementation in GCC
for little and big endian. However, contrary to what GCC does, integers can
cross units boundaries (no padding is required). Padding can be explicitely
added (see 4.1.6 GNU/C bitfields) to follow the GCC layout if needed.

Metadata representation:

  abstract_type integer {
    signed = true OR false;                     /* default false */
    byte_order = native OR network OR be OR le; /* default native */
    size = value;                               /* value in bits, no default */
    align = value;                              /* value in bits */
  };

Example of type inheritance (creation of a concrete type uint32_t):

type uint32_t {
  parent = integer;
  size = 8;
  signed = false;
  align = 32;
};

Definition of a 5-bit signed bitfield:

type int5_t {
  parent = integer;
  size = 5;
  signed = true;
  align = 1;
};

4.1.6 GNU/C bitfields

The GNU/C bitfields follow closely the integer representation, with a
particularity on alignment: if a bitfield cannot fit in the current unit, the
unit is padded and the bitfield starts at the following unit. We therefore need
to express the extra "unit size" information.

Metadata representation:

abstract_type gcc_bitfield {
  parent = integer;
  unit_size = value;
};

As an example, the following structure declared in C compiled by GCC:

struct example {
  short a:12;
  short b:5;
};

Would correspond to the following structure, aligned on the largest element
(short). The second bitfield would be aligned on the next unit boundary, because
it would not fit in the current unit.

type struct_example {
  parent = struct;
  fields = {
    type {
      parent = gcc_bitfield;
      unit_size = 16;                           /* sizeof(short) */
      size = 12;
      signed = true;
      align = 1;
    } a;
    type {
      parent = gcc_bitfield;
      unit_size = 16;                           /* sizeof(short) */
      size = 5;
      signed = true;
      align = 1;
    } b;
  };
};

4.1.7 Floating point

The floating point values byte ordering is defined in the metadata.

Floating point values follow the IEEE 754-2008 standard interchange formats.
Description of the floating point values include the exponent and mantissa size
in bits. Some requirements are imposed on the floating point values:

- FLT_RADIX must be 2.
- mant_dig is the number of digits represented in the mantissa. It is specified
  by the ISO C99 standard, section 5.2.4, as FLT_MANT_DIG, DBL_MANT_DIG and
  LDBL_MANT_DIG as defined by <float.h>.
- exp_dig is the number of digits represented in the exponent. Given that
  mant_dig is one bit more than its actual size in bits (leading 1 is not
  needed) and also given that the sign bit always takes one bit, exp_dig can be
  specified as:

  - sizeof(float) * CHAR_BIT - FLT_MANT_DIG
  - sizeof(double) * CHAR_BIT - DBL_MANT_DIG
  - sizeof(long double) * CHAR_BIT - LDBL_MANT_DIG

Metadata representation:

abstract_type floating_point {
   exp_dig = value;
   mant_dig = value;
   byte_order = native OR network OR be OR le;
};

Example of type inheritance:

type float {
  exp_dig = 8;         /* sizeof(float) * CHAR_BIT - FLT_MANT_DIG */
  mant_dig = 24;       /* FLT_MANT_DIG */
  byte_order = native;
};

TODO: define NaN, +inf, -inf behavior.

4.1.8 Enumerations

Enumerations are a mapping between an integer type and a table of strings. The
numerical representation of the enumeration follows the integer type specified
by the metadata. The enumeration mapping table is detailed in the enumeration
description within the metadata. The mapping table maps inclusive value ranges
(or single values) to strings. Instead of being limited to simple
"value -> string" mappings, these enumerations map
"[ start_value .. end_value ] -> string", which map inclusive ranges of
values to strings.  An enumeration from the C language can be represented in
this format by having the same start_value and end_value for each element, which
is in fact a range of size 1. This single-value range is supported without
repeating the start and end values with the { value, string } declaration.

abstract_type enum {
  parent = integer;
  map = {
    { { start_value, end_value }, string },
    { { start_value, end_value }, string },
    { { start_value, end_value }, string },
    { value, string },
    { value, string },
    ...
  };
};


4.2 Compound types

4.2.1 Structures

Structures are aligned on the largest alignment required by basic types
contained within the structure. (This follows the ISO/C standard for structures)

Metadata representation:

abstract_type struct {
  fields = {
    field_type field_name;
    field_type field_name;
    ...
  };
}  

Example:

type struct_example {
  parent = struct;
  fields = {
    type {                       /* Nameless type */
      parent = integer;
      size = 16;
      signed = true;
      align = 16;
    } first_field_name;
    uint64_t second_field_name;  /* Named type declared in the metadata */
  };
};

The fields are placed in a sequence next to each other. They each possess a
field name, which is a unique identifier within the structure.

4.2.2 Arrays

Arrays are fixed-length. Their length is declared in the type declaration within
the metadata. They contain an array of "inner type" elements, which can refer to
any type not containing the type of the array being declared (no circular
dependency). The length is the number of elements in an array.

Metadata representation:

abstract_type array {
  length = value;
  elem_type = type;
};

E.g.:

type example_array {
  parent = array;
  length = 10;
  elem_type = uint32_t;
};

4.2.3 Sequences

Sequences are dynamically-sized arrays. They start with an integer that specify
the length of the sequence, followed by an array of "inner type" elements.
The length is the number of elements in the sequence.

abstract_type sequence {
  length_type = type;   /* Inheriting from integer */
  elem_type = type;
};

The integer type follows the integer types specifications, and the sequence
elements follow the "array" specifications.

4.2.4 Strings

Strings are an array of bytes of variable size and are terminated by a '\0'
"NULL" character.  Their encoding is described in the metadata. In absence of
encoding attribute information, the default encoding is UTF-8.

abstract_type string {
  encoding = UTF8 OR ASCII;
};


5. Event Packet Header

The event packet header consists of two part: one is mandatory and have a fixed
layout. The second part, the "event packet context", has its layout described in
the metadata.

- Aligned on page size. Fixed size. Fields either aligned or packed (depending
  on the architecture preference).
  No padding at the end of the event packet header. Native architecture byte
  ordering.

Fixed layout (event packet header):

- Magic number (CTF magic numbers: 0xC1FC1FC1 and its reverse endianness
  representation: 0xC11FFCC1) It needs to have a non-symmetric bytewise
  representation. Used to distinguish between big and little endian traces (this
  information is determined by knowing the endianness of the architecture
  reading the trace and comparing the magic number against its value and the
  reverse, 0xC11FFCC1). This magic number specifies that we use the CTF metadata
  description language described in this document. Different magic numbers
  should be used for other metadata description languages.
- Trace UUID, used to ensure the event packet match the metadata used.
  (note: we cannot use a metadata checksum because metadata can be appended to
   while tracing is active)
- Stream ID, used as reference to stream description in metadata.

Metadata-defined layout (event packet context):

- Event packet content size (in bytes).
- Event packet size (in bytes, includes padding).
- Event packet content checksum (optional). Checksum excludes the event packet
  header.
- Per-stream event packet sequence count (to deal with UDP packet loss). The
  number of significant sequence counter bits should also be present, so
  wrap-arounds are deal with correctly.
- Timestamp at the beginning and timestamp at the end of the event packet.
  Both timestamps are written in the packet header, but sampled respectively
  while (or before) writing the first event and while (or after) writing the
  last event in the packet. The inclusive range between these timestamps should
  include all event timestamps assigned to events contained within the packet.
- Events discarded count
  - Snapshot of a per-stream free-running counter, counting the number of
    events discarded that were supposed to be written in the stream prior to
    the first event in the event packet.
    * Note: producer-consumer buffer full condition should fill the current
            event packet with padding so we know exactly where events have been
            discarded.
- Lossless compression scheme used for the event packet content. Applied
  directly to raw data. New types of compression can be added in following
  versions of the format.
  0: no compression scheme
  1: bzip2
  2: gzip
  3: xz
- Cypher used for the event packet content. Applied after compression.
  0: no encryption
  1: AES
- Checksum scheme used for the event packet content. Applied after encryption.
  0: no checksum
  1: md5
  2: sha1
  3: crc32

5.1 Event Packet Header Fixed Layout Description

type event_packet_header {
  parent = struct;
  fields = {
    uint32_t magic;
    type {
      parent = array; 
      length = 16; 
      elem_type = uint8_t;  
    } trace_uuid;
  uint32_t stream_id;
}

5.2 Event Packet Context Description

Event packet context example. These are declared within the stream declaration
in the metadata. All these fields are optional except for "content_size" and
"packet_size", which must be present in the context.

An example event packet context type:

type event_packet_context {
  uint64_t timestamp_begin;
  uint64_t timestamp_end;
  uint32_t checksum;
  uint32_t stream_packet_count;
  uint32_t events_discarded;
  uint32_t cpu_id;
  uint32_t/uint16_t content_size;
  uint32_t/uint16_t packet_size;
  uint8_t  stream_packet_count_bits;    /* Significant counter bits */
  uint8_t  compression_scheme;
  uint8_t  encryption_scheme;
  uint8_t  checksum;
};

6. Event Structure

The overall structure of an event is:

  - Event Header (as specifed by the stream metadata)
  - Extended Event Header (as specified by the event header)
  - Event Context (as specified by the stream metadata)
  - Event Payload (as specified by the event metadata)


6.1 Event Header

One major factor can vary between streams: the number of event IDs assigned to
a stream. Luckily, this information tends to stay relatively constant (modulo
event registration while trace is being recorded), so we can specify different
representations for streams containing few event IDs and streams containing
many event IDs, so we end up representing the event ID and timestamp as densely
as possible in each case.

We therefore provide two types of events headers. Type 1 accommodates streams
with less than 31 event IDs. Type 2 accommodates streams with 31 or more event
IDs.

The "extended headers" are used in the rare occasions where the information
cannot be represented in the ranges available in the event header. They are also
used in the rare occasions where the data required for a field could not be
collected: the flag corresponding to the missing field within the missing_fields
array is then set to 1.

Types uintX_t represent an X-bit unsigned integer.


6.1.1 Type 1 - Few event IDs

  - Aligned on 32-bit (or 8-bit if byte-packed, depending on the architecture
    preference).
  - Fixed size: 32 bits.
  - Native architecture byte ordering.

type event_header_1 {
  parent = struct;
  fields = {
    uint5_t id;         /*
                         * id: range: 0 - 30.
                         * id 31 is reserved to indicate a following
                         * extended header.
                         */
    uint27_t timestamp;
  };
};

The end of a type 1 header is aligned on a 32-bit boundary (or packed).


6.1.2 Extended Type 1 Event Header

  - Follows struct event_header_1, which is aligned on 32-bit, so no need to
    realign.
  - Variable size (depends on the number of fields per event).
  - Native architecture byte ordering.

type event_header_1_ext {
  parent = struct;
  fields = {
    uint32_t id;                /* 32-bit event IDs */
    uint64_t timestamp;         /* 64-bit timestamps */
    type {
      parent = array;
      length = NR_FIELDS;       /* Number of fields within the event */
      elem_type = uint1_t;      /* 1-bit bitfield */
    } missing_fields;
  };
};


6.1.3 Type 2 - Many event IDs

  - Aligned on 32-bit (or 8-bit if byte-packed, depending on the architecture
    preference).
  - Fixed size: 48 bits.
  - Native architecture byte ordering.

type event_header_2 {
  parent = struct;
  fields = {
    uint32_t timestamp;
    uint16_t id;        /*
                         * id: range: 0 - 65534.
                         * id 65535 is reserved to indicate a following
                         * extended header.
                         */
  };
};

The end of a type 2 header is aligned on a 16-bit boundary (or 8-bit if
byte-packed).


6.1.4 Extended Type 2 Event Header

  - Follows struct event_header_2, which alignment end on a 16-bit boundary, so
    we need to align on 64-bit integer architecture alignment (or 8-bit if
    byte-packed).
  - Variable size (depends on the number of fields per event).
  - Native architecture byte ordering.

type event_header_2_ext {
  parent = struct;
  fields = {
    uint64_t timestamp;         /* 64-bit timestamps */ 
    uint32_t id;                /* 32-bit event IDs */
    type {
      parent = array;
      length = NR_FIELDS;       /* Number of fields within the event */
      elem_type = uint1_t;      /* 1-bit bitfield */
    } missing_fields;
  };
};


6.2 Event Context

The event context contains information relative to the current event. The choice
and meaning of this information is specified by the metadata "stream"
information. For this trace format, event context is usually empty, except when
the metadata "stream" information specifies otherwise by declaring a non-empty
structure for the event context. An example of event context is to save the
event payload size with each event, or to save the current PID with each event.
These are declared within the stream declaration within the metadata.

An example event context type:

      type event_context {
        parent = struct;
        fields = {
          uint pid;
          uint16_t payload_size;
        };
      };


6.3 Event Payload

An event payload contains fields specific to a given event type. The fields
belonging to an event type are described in the event-specific metadata
within a structure type.

6.3.1 Padding

No padding at the end of the event payload. This differs from the ISO/C standard
for structures, but follows the CTF standard for structures. In a trace, even
though it makes sense to align the beginning of a structure, it really makes no
sense to add padding at the end of the structure, because structures are usually
not followed by a structure of the same type.

This trick can be done by adding a zero-length "end" field at the end of the C
structures, and by using the offset of this field rather than using sizeof()
when calculating the size of a structure (see Appendix "A. Helper macros").

6.3.2 Alignment

The event payload is aligned on the largest alignment required by types
contained within the payload. (This follows the ISO/C standard for structures)



7. Metadata

The meta-data is located in a stream named "metadata". It is made of "event
packets", which each start with an event packet header. The event type within
the metadata stream have no event header nor event context. Each event only
contains a null-terminated "string" payload, which is a metadata description
entry. The events are packed one next to another. Each event packet start with
an event packet header, which contains, amongst other fields, the magic number
and trace UUID.

The metadata can be parsed by reading through the metadata strings, skipping
newlines and null-characters. Type names may contain spaces.

trace {
  major = value;        /* Trace format version */
  minor = value;
  uuid = value;         /* Trace UUID */
  word_size = value;
};

stream {
  id = stream_id;
  event {
    /* Type 1 - Few event IDs; Type 2 - Many event IDs. See section 6.1. */
    header_type = event_header_1 OR event_header_2;
    /*
     * Extended event header type. Only present if specified in event header
     * on a per-event basis.
     */
    header_type_ext = event_header_1_ext OR event_header_2_ext;
    context_type = type inheriting from "struct" abstract type;
  };
  packet {
    context_type = type inheriting from "struct" abstract type;
  };
};

event {
  name = eventname;
  id = value;                   /* Numeric identifier within the stream */
  stream = stream_id;
  fields = type inheriting from "struct" abstract type;
};

/* More detail on types in section 4. Types */

/* Named types */
type {
  name = typename;
  ...
};

/* Unnamed types, contained within compound type fields or type assignments. */
type {
  ...
};


Structure types used for fields and context_type implicitly inherit from
"struct" and require no "type" identifier before the braces. E.g.:

  fields = {
    typeA name1;
    typeB name2;
    ...
  };


A. Helper macros

The two following macros keep track of the size of a GNU/C structure without
padding at the end by placing HEADER_END as the last field. A one byte end field
is used for C90 compatibility (C99 flexible arrays could be used here). Note
that this does not affect the effective structure size, which should always be
calculated with the header_sizeof() helper.

#define HEADER_END              char end_field
#define header_sizeof(type)     offsetof(typeof(type), end_field)


B. Stream Header Rationale

An event stream is divided in contiguous event packets of variable size. These
subdivisions allow the trace analyzer to perform a fast binary search by time
within the stream (typically requiring to index only the event packet headers)
without reading the whole stream. These subdivisions have a variable size to
eliminate the need to transfer the event packet padding when partially filled
event packets must be sent when streaming a trace for live viewing/analysis.
An event packet can contain a certain amount of padding at the end. Dividing
streams into event packets is also useful for network streaming over UDP and
flight recorder mode tracing (a whole event packet can be swapped out of the
buffer atomically for reading).

The stream header is repeated at the beginning of each event packet to allow
flexibility in terms of:

  - streaming support,
  - allowing arbitrary buffers to be discarded without making the trace
    unreadable,
  - allow UDP packet loss handling by either dealing with missing event packet
    or asking for re-transmission.
  - transparently support flight recorder mode,
  - transparently support crash dump.

The event stream header will therefore be referred to as the "event packet
header" throughout the rest of this document.