Philippe Proulx [Mon, 1 Jun 2020 15:17:27 +0000 (11:17 -0400)]
Rename `msg`/`ctx` properties 👉 `message`/`context`
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 29 May 2020 22:12:19 +0000 (18:12 -0400)]
barectf: use package names (do not import "from")
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 29 May 2020 21:58:26 +0000 (17:58 -0400)]
cli.py: fix PEP 8 errors, as reported by `flake8`
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 29 May 2020 21:56:49 +0000 (17:56 -0400)]
cli.py: format
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 29 May 2020 21:56:00 +0000 (17:56 -0400)]
cli.py: standardize `exc` as exception name
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 29 May 2020 21:54:05 +0000 (17:54 -0400)]
cli.py: replace `'...'.format(...)` with f-strings
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 29 May 2020 20:53:13 +0000 (16:53 -0400)]
config_parse.py: replace `'...'.format()` with f-strings
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 29 May 2020 20:52:21 +0000 (16:52 -0400)]
pyproject.toml: require Python 3.6+
I want to use f-strings.
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 29 May 2020 19:57:17 +0000 (15:57 -0400)]
config_parse.py: fix PEP 8 errors, as reported by `flake8`
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 29 May 2020 19:56:33 +0000 (15:56 -0400)]
tox.ini: make `flake8` ignore `templates.py`
`templates.py` is only a bunch of strings; no application code there.
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 29 May 2020 19:56:11 +0000 (15:56 -0400)]
tox.ini: run `flake8` for the `barectf` directory only
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 29 May 2020 19:47:55 +0000 (15:47 -0400)]
config_parse.py: comment `_ConfigParseErrorCtx` and `_ConfigParseError`
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 29 May 2020 19:35:31 +0000 (15:35 -0400)]
_YamlConfigParser: rename field type creation methods (more explicit)
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 29 May 2020 19:33:18 +0000 (15:33 -0400)]
_YamlConfigParser: rename `_root_yaml_path` 👉 `_root_path`
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 29 May 2020 19:30:52 +0000 (15:30 -0400)]
config_parse.py: use a `_YamlConfigParser` object a single time
This patch changes `_YamlConfigParser` so that you can only use it once:
the object parses the file on construction, and then makes the
configuration object available through its `config` property.
Because you can't parse another file with the same parser, we don't need
to care about any stale state.
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 29 May 2020 19:21:49 +0000 (15:21 -0400)]
config_parse.py: normalize exception contexts and messages
This patch also changes the terminology of the messages to use "type"
and "field type" as much as possible, for example "event type",
"stream type", and "integer field type".
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 29 May 2020 16:48:49 +0000 (12:48 -0400)]
config_parse.py: update/improve comments
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 29 May 2020 15:29:03 +0000 (11:29 -0400)]
barectf: rename `ConfigParseError` 👉 `_ConfigParseError`
This is not a class which the API user can instantiate.
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Thu, 28 May 2020 20:32:52 +0000 (16:32 -0400)]
barectf: use JSON schema to validate special root field type fields
This patch removes almost everything from the
`_MetadataSpecialFieldsValidator` class, moving those checks to the
`schemas/2/config/config.yaml` JSON schema instead.
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Thu, 28 May 2020 18:18:48 +0000 (14:18 -0400)]
barectf: schemas: use conditionals intead of `oneOf` when possible
Using `if`/`then`/`else` instead of `oneOf` makes the `jsonschema`
validator create much more readable errors.
For example:
Before:
Error:
Configuration: Cannot create configuration from YAML file
`config.yaml`
Configuration object:
`metadata` property:
`clocks` property:
`some_clock` property:
`offset` property:
`seconds` property: -2 is not valid under any of the given
schemas: -2 is less than the minimum of 0; -2 is not of type
'null' (from schema `2/config/config`)
Now:
Error:
Configuration: Cannot create configuration from YAML file
`config.yaml`
Configuration object:
`metadata` property:
`clocks` property:
`some_clock` property:
`offset` property:
`seconds` property: -2 is less than the minimum of 0
(from schema `2/config/config`)
This is because, with conditionals, we assume that the user intended
something with some valid schema, and add more constraints. The example
above is for an integer having a minimum value of 0, or a null value. As
soon as we know that the value is an integer, then we know its minimum
must be 0; we decide to not care about the fact that the whole value
could also be null (which the old message indicated). In my opinion,
this is a better UX.
The example above is simple. For more complex schemas, the UX gain is
even more obvious.
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Thu, 28 May 2020 18:16:47 +0000 (14:16 -0400)]
_SchemaValidator.validate(): use "element" prefix for array elements
Otherwise the context's object name would be:
`2` property
instead of
Element 2
for example.
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Wed, 27 May 2020 03:04:48 +0000 (23:04 -0400)]
LICENSE: reflow licence header for 72 columns
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Wed, 27 May 2020 03:04:32 +0000 (23:04 -0400)]
LICENSE: update copyright years
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Wed, 27 May 2020 03:02:30 +0000 (23:02 -0400)]
barectf: reflow licence headers for 72 columns
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Wed, 27 May 2020 03:00:16 +0000 (23:00 -0400)]
barectf: update copyright years
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Wed, 27 May 2020 02:58:10 +0000 (22:58 -0400)]
gen.py, templates.py: replace double quotes with backticks (generated)
Style preference.
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Wed, 27 May 2020 02:54:36 +0000 (22:54 -0400)]
cli.py, config_parse.py: replace double quotes with backticks
Style preference.
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 22 May 2020 15:06:31 +0000 (11:06 -0400)]
Use JSON schemas to validate the layout and types of the YAML config.
This patch changes `config_parse.py` so as to use JSON schemas
(<https://json-schema.org/>) to validate many aspects of a barectf YAML
configuration file instead of having redundant, manual checks.
JSON schemas are found in the `barectf` package's `schemas` directory by
an instance of the new `_SchemaValidator` class. Such an object finds
all the YAML files in the `schemas` directory and, for each of them:
1. Loads it (YAML to Python dictionary).
2. Gets its ID (`$id` property).
3. Adds the schema (1.) to a local schema store using its ID (2.).
Then _SchemaValidator.validate() uses this schema store to build a JSON
schema reference resolver and validator from the `jsonschema` package
(<https://pypi.org/project/jsonschema/>).
The `jsonschema` dependency is added to `pyproject.toml` and
`poetry.lock`. We need a version which supports JSON Schema draft 7
because we need the conditional keywords (`if`/`then`).
Because the barectf YAML configuration file format supports features
which do not exist natively in YAML (inclusions, field type aliases,
field type inheritance, and log level aliases), we can't have a single
JSON schema for the "raw" configuration file. There are actually five
validation steps, each one having its JSON schema:
1. Make sure the configuration object is minimally valid, that is, it's
an object and has a `version` property with a supported value.
Schema: `config/config-min.yaml`.
2. Make sure the configuration object is valid for the inclusion phase.
Those schemas only validate that the metadata/stream/event objects
are objects and that any `$include` property is valid.
Each time the YAML configuration parser loads a partial YAML file for
inclusion, it validates the resulting object using the corresponding
schema depending on what's being included.
Knowing those objects are valid as such is enough to process
inclusions without caring about the resulting configuration object's
validity.
Schemas:
* `2/config/clock-pre-include.yaml`
* `2/config/config-pre-include.yaml`
* `2/config/event-pre-include.yaml`
* `2/config/include-prop.yaml`
* `2/config/metadata-pre-include.yaml`
* `2/config/stream-pre-include.yaml`
* `2/config/trace-pre-include.yaml`
3. Make sure the configuration object is valid for the field type
expansion phase.
This schema digs into compound field types recursively to make sure
any field type is either a string (field type alias) or an object
with either a `class` property or an `$inherit`/`inherit` (mutually
exclusive) property.
Knowing the configuration object is valid as such is enough to expand
field types without caring about the resulting configuration object's
validity. So, for example, a resulting, expanded field type could be:
class: string
size: 16
value-type:
class: array
meow: mix
We just don't care at this point, as long as field types are
"complete".
Schema: `2/config/config-pre-field-type-expansion.yaml`.
4. Make sure the configuration object is valid for the log level
expansion phase.
This schema validates the `$log-levels` property of the metadata
object as well as the `log-level` property of any event object.
Knowing the configuration object is valid as such is enough to expand
log levels, that is, to replace log level strings with their numeric
value, without caring about the resulting configuration object's
validity.
Schema: `2/config/config-pre-log-level-expansion.yaml`.
5. Make sure the configuration object is valid.
This validates the final, effective configuration object which, at
this point, do not contain any:
* `$include` properties.
* Strings as field types.
* `$inherit`/`inherit` properties (field type objects).
* `$log-levels` property (metadata object).
* Strings for the `log-level` properties (event objects).
Also, this is the configuration object which the `--dump-config`
option prints now.
* `2/config/byte-order-prop.yaml`
* `2/config/config.yaml`
* `2/config/field-type.yaml`
* `2/config/uuid-prop.yaml`
After step 5., there's not much to validate in `config_parse.py` itself:
* Make sure referred clock type objects (in `property-mappings` properties
of integer field type objects) exist.
* Make sure identifiers are valid (the schemas do not always validate
that they exclude CTF keywords for technical reasons).
* Make sure alignment values are valid (powers of two).
* Make sure there's only one default stream type name and that it
exists.
* Make sure the values of enumeration field type members are within the
field type's range depending on the value (integer) field type's size.
* Everything that remains in `_BarectfMetadataValidator`.
* Everything in `_MetadataSpecialFieldsValidator`, which is untouched
because that's not something we validate with JSON schemas, although
we could possibly, but it might stretch the use case.
_SchemaValidator.validate() catches a `jsonschema` exception and
converts it to a barectf `ConfigParseError` exception to avoid leaking
`jsonschema` objects (implementation detail) from barectf calls. I made
an effort to make the error object as readable as possible, for example
converting the instance path to context object names, but there's room
for improvement here.
No functional change intended, except for the modified raised
`ConfigParseError` objects. Tests are not changed and pass.
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Wed, 20 May 2020 18:58:16 +0000 (14:58 -0400)]
config_parse.py: add _append_error_ctx() to remove redundant code
_append_error_ctx() appends context to a given `ConfigParseError`
exception and then reraises it.
Using
catch ConfigParseError as exc:
_append_error_ctx(exc, 'My object', 'My message')
instead of
catch ConfigParseError as exc:
exc.append_ctx('My object', 'My message')
raise
makes sure we don't forget to reraise.
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 15 May 2020 21:12:39 +0000 (17:12 -0400)]
Make configuration and metadata objects immutable
This patch changes how configuration and metadata objects are built.
Before this patch, the YAML configuration parser in `config.py` created
default metadata objects and then set their properties as it parsed the
YAML node.
Having such mutable objects means they can change at anytime once they
exist.
This patch changes this so that all configuration and metadata objects
accept all they need at construction time and do not offer setters.
`config.py` now only contains the `Config` and `ConfigOptions` classes
as well as the from_file() function. All the YAML parsing and
configuration/metadata construction is in `config_parse.py`.
In `config_parse.py`, there are mutable "pseudo" classes which
correspond to the immutable versions. They are only used during the
parsing stage and, at the end, converted to their public, immutable
equivalents.
The configuration validation is done at the YAML/pseudo level, which
means the public configuration/metadata classes do not validate anything
(narrow contract). Their constructors _should_ at least assert that
everything looks okay, but this work is not part of this patch.
No functional changes intended by this patch at the CLI level, which is
why all tests still pass without modification.
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 15 May 2020 19:55:14 +0000 (15:55 -0400)]
Capitalize exception messages and what the CLI prints
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 15 May 2020 19:35:00 +0000 (15:35 -0400)]
test.bash: pass all `.bats` files to `bats`
This makes `bats` accumulate tests and run all of them with a global
progress, which is much better.
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 15 May 2020 19:31:59 +0000 (15:31 -0400)]
barectf_config_check_fail(): check that failure is caused by cfg. error
This patch makes barectf_config_check_fail() in `config/common.bash`
check that the output of `barectf` contains the string
configuration: cannot create configuration
so as to ensure that the failure is _at least_ caused by a configuration
error, even if we don't know if the real failure cause matches what the
test intends to verify.
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 15 May 2020 19:31:33 +0000 (15:31 -0400)]
tests/config/pass/everything/pass.bats: quote `$CC`
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 15 May 2020 19:31:11 +0000 (15:31 -0400)]
tests/config/pass/everything/pass.bats: make test directory the CWD
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 15 May 2020 19:30:34 +0000 (15:30 -0400)]
tests/config/fail/yaml/invalid.yaml: fix invalid YAML (list/neg. value)
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 15 May 2020 19:29:46 +0000 (15:29 -0400)]
cli: print traceback for unknown exceptions
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 15 May 2020 19:28:59 +0000 (15:28 -0400)]
Append context to a `ConfigError` instead of wrapping another exception
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 15 May 2020 01:42:48 +0000 (21:42 -0400)]
Remove everything related to CTF sequences and variants
Those CTF types are not supported by barectf currently, yet some code
was written to support them at the configuration level. This is some
rather complex (dead) code to resolve field paths and validate those
types, but without anything to test it. In fact, the code exists to
parse a variant type, for example, but barectf fails with an error (not
supported) when you do so.
The code is still in this repository's history; let's add it back when
we can really test it.
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Thu, 14 May 2020 19:22:09 +0000 (15:22 -0400)]
Convert project to a Poetry project
Poetry <https://python-poetry.org/> handles the build creation and
`setup.py` and `PKG-INFO` generation.
Poetry also handles dependencies and can create a corresponding virtual
environment having them installed.
Poetry can also publish.
`poetry.lock` was generated by `poetry install` and ensures this
specific version of barectf is tested with those specific versions of
the dependencies.
Setuptools is a run-time dependency for its `pkg_resources` package. I
use `*` for its version constraint because somehow Poetry needs
Setuptools itself so it skips this dependency (see
<https://github.com/python-poetry/poetry/issues/1584>). The `setuptools`
dependency still ends up in `setup.py`.
`tox.ini` is calibrated so that tox doesn't manage any dependency;
everything is done through Poetry (except `flake8` which doesn't need
barectf to be installed).
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Thu, 14 May 2020 16:47:24 +0000 (12:47 -0400)]
barectf/__init__.py: build version string instead of parsing it
This is more natural and less complicated.
Leaving get_version_tuple() as some modules use it.
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Michael Jeanson [Tue, 4 Jun 2019 15:52:19 +0000 (11:52 -0400)]
Add git-review config
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Change-Id: I12ee3f57fc2e9cafb74cd489add4f8dbb89b2ca9
Philippe Proulx [Mon, 25 Mar 2019 20:28:03 +0000 (16:28 -0400)]
barectf v2.3.1
Signed-off-by: Philippe Proulx <pproulx@efficios.com>
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Mon, 25 Mar 2019 20:05:13 +0000 (16:05 -0400)]
Fix: do not sample clock when closing/opening a packet for an event not fitting
This patch adds members to the generated barectf context to save the
current (becomes last) event's timestamp and use it when closing and
opening packets when the event does not fit the current packet instead
of sampling the clock again.
Without this patch, with this scenario:
1. Application calls a tracing function.
2. Tracing function samples the clock.
3. Event does not fit the current packet size: current packet is closed,
sampling the clock, and new packet is opened, also sampling the
clock.
4. Event is serialized to the new packet.
The new packet's timestamp is greater than its first event's timestamp,
because this event's timestamp is the result of sampling the clock
before sampling it to close the current packet and open a new one.
Babeltrace 1 accepts and reads the resulting stream, but Babeltrace 2's
muxer is more strict and refuses a message which is in the past.
Signed-off-by: Philippe Proulx <pproulx@efficios.com>
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Fri, 11 Jan 2019 21:52:35 +0000 (16:52 -0500)]
Fix: use the same actions to compute event size and to serialize
Issue
=====
In `gen.py`, the code to generate the event size computation C code and
the event serialization C code are not the same. There are optimizations
to avoid generating useless alignment statements in one which are
missing in the other, and those optimizations have a bug. This bug can
lead to the space reservation phase of a tracing function passing
(enough space for a given event to serialize), but the serialization
phase writing passed the end of the packet buffer.
Solution
========
Make barectf use the exact same actions to generate the event size
computation C code and the event serialization C code.
In `gen.py`, `_SerializationActions` is a new class of which you can
use an instance to append a root scope field type
(_SerializationActions.append_root_scope_type()) and to retrieve a list
of resulting serialization actions. The two types of serialization
actions are:
`_AlignSerializationAction`:
Force the alignment of the current bit position within the packet.
`_SerializeSerializationAction`:
Serialize a field from a given type.
Each action object contains:
* An optional field type which is responsible for the action.
* An offset within the current byte when this action needs
to be executed (to generate the appropriate bt_bitfield_write_*()
calls).
* A list of names which identify this action (complete field name/path).
_SerializationActions.append_root_scope_type() performs the necessary
optimizations (does not align when specific conditions are met to avoid
a useless alignment statement). Then those optimizations apply to both
the event size computation and event serialization C code, the only
difference being that the event size computation code adds to a partial
sum variable instead of writing to the packet buffer.
The alignment optimization technique is straightforward at this point
and could be improved: if the alignment requirement did not change since
the last alignment, and if there's no padding between a field and the
previous field, then do not align for this field; otherwise, align.
Known drawbacks
===============
Because the alignment optimization technique is changed to be simpler,
some barectf configurations could make barectf generate more C code
(which means more compiled code and slower execution eventually).
However, the previously generated C code was incorrect anyway.
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Tue, 17 Oct 2017 21:40:57 +0000 (17:40 -0400)]
barectf v2.3.0
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Tue, 17 Oct 2017 21:30:40 +0000 (17:30 -0400)]
tox.ini: no `pep8` env by default, fix testenv:pep8 env
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Tue, 17 Oct 2017 21:30:18 +0000 (17:30 -0400)]
barectf/config.py: fix whitespaces (PEP 8)
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Michael Jeanson [Tue, 17 Oct 2017 20:43:21 +0000 (16:43 -0400)]
tox: Add usedevelop
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Michael Jeanson [Tue, 17 Oct 2017 20:36:59 +0000 (16:36 -0400)]
Fix: fail tests succeed on env failure
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Philippe Proulx [Tue, 17 Oct 2017 19:25:25 +0000 (15:25 -0400)]
tests/common.bash: use $status, not $?, when using `run`
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Tue, 17 Oct 2017 19:11:40 +0000 (15:11 -0400)]
Merge branch 'tox' of https://github.com/mjeanson/barectf
Michael Jeanson [Tue, 17 Oct 2017 18:45:42 +0000 (14:45 -0400)]
Add tox support
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Philippe Proulx [Tue, 17 Oct 2017 18:53:10 +0000 (14:53 -0400)]
Merge branch 'master' of https://github.com/mjeanson/barectf
Philippe Proulx [Tue, 17 Oct 2017 18:52:16 +0000 (14:52 -0400)]
Fix: templates.py: call bt_piecewise_rshift() with _vtype (BE)
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Michael Jeanson [Tue, 17 Oct 2017 18:50:48 +0000 (14:50 -0400)]
Fix: return proper exit code on test failure
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Philippe Proulx [Sat, 19 Nov 2016 19:23:18 +0000 (14:23 -0500)]
linux-fs-simple example Makefile: make platform depend on barectf.h
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Sat, 19 Nov 2016 19:00:14 +0000 (14:00 -0500)]
Add tracing activity and tracing section API
The new barectf_is_tracing_enabled() function allows the instrumented
program to enable or disable tracing for a specific barectf context.
This is a thread-safe API, in that it only sets a member of the context
to 0 or 1. Disabling the tracing activity does not guarantee, however,
that the current packet or back-end is not currently being altered. See
the next paragraph about this.
The new barectf_is_in_tracing_section() function allows the instrumented
program to know if its tracing thread is currently in a barectf section
or not. barectf_is_in_tracing_section_ptr() returns a volatile pointer
to this same information.
Both functions can be used together to implement a snapshot action for a
given platform, something like:
1. The instrumented thread calls the barectf tracing functions during
its execution. When a packet is closed, the platform appends it to a
ring buffer, overwriting the oldest recorded one if there's no space
left. This is called a flight recorder mode.
2. The controller (another thread of the instrumented program, or an
external host) recognizes a trigger where we would only need the
latest data to analyze the situation. Therefore tracing must be
stopped, because we don't want the current ring buffer packets to be
overwritten anymore.
3. The controller disables tracing. It is safe to call
barectf_is_tracing_enabled(ctx, 0) from another thread.
4. When barectf_is_tracing_enabled(ctx, 0) is called, the barectf
tracing function currently being executed still needs to finish. At
this point it could still alter the current packet. Therefore the
controller waits until barectf_is_in_tracing_section() is 0.
5. Once barectf_is_in_tracing_section() returns 0, it won't return 1
until barectf_is_tracing_enabled(ctx, 1) is called again, so it is
safe to read the ring buffer, and also to read the current packet.
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Tue, 15 Nov 2016 18:08:59 +0000 (13:08 -0500)]
README.md: fix links
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Tue, 15 Nov 2016 18:00:46 +0000 (13:00 -0500)]
barectf v2.2.1
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Tue, 15 Nov 2016 17:55:48 +0000 (12:55 -0500)]
README.md: update content
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Tue, 15 Nov 2016 17:13:32 +0000 (12:13 -0500)]
doc/examples: use -ansi flag to ensure ANSI C compatibility
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Thu, 10 Nov 2016 22:33:19 +0000 (17:33 -0500)]
barectf(1): write short option's argument too
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Thu, 10 Nov 2016 22:19:43 +0000 (17:19 -0500)]
Claim to generate ANSI C code
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Thu, 10 Nov 2016 22:17:52 +0000 (17:17 -0500)]
Remove <stdint.h> include in generated bitfield.h
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Thu, 10 Nov 2016 21:44:36 +0000 (16:44 -0500)]
Remove unused macros from generated bitfield header
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Thu, 10 Nov 2016 21:43:39 +0000 (16:43 -0500)]
gen.py: put union variable at the top of a nested block
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Thu, 10 Nov 2016 18:16:59 +0000 (13:16 -0500)]
Remove dependency on GNU C extensions
__typeof__() dependency is removed by passing the type to the
macro explicitly.
Statement expression dependency is removed by using an in/out
macro parameter instead.
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Wed, 9 Nov 2016 22:53:22 +0000 (17:53 -0500)]
Support C++ build of generated and linux-fs platform files
Mostly void * casting stuff.
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Wed, 9 Nov 2016 17:36:56 +0000 (12:36 -0500)]
barectf-platform-linux-fs.h: add __cplusplus check
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Wed, 9 Nov 2016 17:32:12 +0000 (12:32 -0500)]
barectf.c: remove unused _packet_is_full()
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Wed, 9 Nov 2016 17:29:06 +0000 (12:29 -0500)]
barectf.h: add __cplusplus check
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Michael Jeanson [Thu, 8 Sep 2016 20:43:35 +0000 (16:43 -0400)]
Merge pull request #11 from mjeanson/master
Update CI link in README.md
Michael Jeanson [Mon, 29 Aug 2016 15:46:12 +0000 (11:46 -0400)]
Update CI link in README.md
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Philippe Proulx [Wed, 13 Jul 2016 15:41:38 +0000 (11:41 -0400)]
Refer to official website barectf.org
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Wed, 13 Jul 2016 15:35:03 +0000 (11:35 -0400)]
Add MIT license header to generated files
Fixes: #8
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philipp Wagner [Tue, 12 Jul 2016 15:47:44 +0000 (17:47 +0200)]
Fix package name in Ubuntu install instructions
Philippe Proulx [Thu, 14 Apr 2016 00:52:47 +0000 (20:52 -0400)]
Make sure all source files have a MIT license header
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Thu, 14 Apr 2016 00:18:09 +0000 (20:18 -0400)]
examples/barectf-tracepoint: fix comment
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Thu, 14 Apr 2016 00:12:08 +0000 (20:12 -0400)]
doc/man: update man page's date and version
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Wed, 13 Apr 2016 21:00:09 +0000 (17:00 -0400)]
barectf v2.2.0
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Wed, 13 Apr 2016 20:49:41 +0000 (16:49 -0400)]
docs/examples/.gitignore: ignore "metadata"
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Wed, 13 Apr 2016 19:52:48 +0000 (15:52 -0400)]
barectf-tracepoint: add QEMU ARM target example
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Tue, 12 Apr 2016 21:20:38 +0000 (17:20 -0400)]
barectf-tracepoint ex: rename "barectf" -> "barectf-linux-fs"
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Tue, 12 Apr 2016 21:02:19 +0000 (17:02 -0400)]
barectf-tracepoint ex: put platform-specific stuff in separate files
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Thu, 7 Apr 2016 04:05:50 +0000 (00:05 -0400)]
doc/examples/linux-fs-simple: use $(BARECTF), no view target
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Thu, 7 Apr 2016 03:46:45 +0000 (23:46 -0400)]
Add doc/examples/barectf-tracepoint
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Thu, 7 Apr 2016 03:46:08 +0000 (23:46 -0400)]
Add extra/barectf-tracepoint.h (tracepoint() macro wrapper)
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Thu, 7 Apr 2016 01:53:26 +0000 (21:53 -0400)]
tests/config/fail/stream: add tests for "$default" prop
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Thu, 7 Apr 2016 01:50:15 +0000 (21:50 -0400)]
tests/config/fail/metadata: add tests for default stream name
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Thu, 7 Apr 2016 01:49:32 +0000 (21:49 -0400)]
tests/config/fail: remove .c/.h files
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Thu, 7 Apr 2016 01:41:54 +0000 (21:41 -0400)]
tests/config/pass/everything: fix gen-default-stream-def
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Thu, 7 Apr 2016 01:40:52 +0000 (21:40 -0400)]
doc/examples/linux-fs-simple: generate prefix/def stream name defs
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Thu, 7 Apr 2016 01:39:54 +0000 (21:39 -0400)]
tests/config/fail/config: remove options-gen-default-stream-def-unknown-stream-name
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Thu, 7 Apr 2016 01:37:10 +0000 (21:37 -0400)]
Make config option "gen-default-stream-def" a boolean
Now that there's a way to specify a default stream, we only
need to know if we want or not to generate this definition.
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Wed, 6 Apr 2016 23:33:03 +0000 (19:33 -0400)]
doc/examples/linux-fs-simple: use 2.2 config with default stream
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Wed, 6 Apr 2016 23:32:43 +0000 (19:32 -0400)]
gen, templates: generate default stream trace definitions
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Wed, 6 Apr 2016 23:17:34 +0000 (19:17 -0400)]
config: specify default stream in metadata/stream objects
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Wed, 6 Apr 2016 16:50:50 +0000 (12:50 -0400)]
gen, templates: generate prefix/default stream definitions
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
Philippe Proulx [Wed, 6 Apr 2016 16:41:26 +0000 (12:41 -0400)]
tests/config/pass/everything: add config options to test
Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>
This page took 0.040978 seconds and 4 git commands to generate.