ddtrace 0.51.1 → 0.54.1

data/docs/GettingStarted.md

@@ -28,8 +28,10 @@ To contribute, check out the [contribution guidelines][contribution docs] and [d
  - [Integration instrumentation](#integration-instrumentation)
      - [Action Cable](#action-cable)
      - [Action View](#action-view)
+     - [Action Mailer](#action-mailer)
      - [Active Model Serializers](#active-model-serializers)
      - [Action Pack](#action-pack)
+     - [Active Job](#active-job)
      - [Active Record](#active-record)
      - [Active Support](#active-support)
      - [AWS](#aws)
@@ -44,12 +46,12 @@ To contribute, check out the [contribution guidelines][contribution docs] and [d
      - [Grape](#grape)
      - [GraphQL](#graphql)
      - [gRPC](#grpc)
-     - [http.rb](#http-rb)
+     - [http.rb](#httprb)
      - [httpclient](#httpclient)
      - [httpx](#httpx)
      - [MongoDB](#mongodb)
      - [MySQL2](#mysql2)
-     - [Net/HTTP](#net-http)
+     - [Net/HTTP](#nethttp)
      - [Presto](#presto)
      - [Qless](#qless)
      - [Que](#que)
@@ -86,6 +88,8 @@ To contribute, check out the [contribution guidelines][contribution docs] and [d
          - [For application runtime](#for-application-runtime)
      - [OpenTracing](#opentracing)
      - [Profiling](#profiling)
+         - [Troubleshooting](#troubleshooting)
+         - [Profiling Resque jobs](#profiling-resque-jobs)
  - [Known issues and suggested configurations](#known-issues-and-suggested-configurations)
     - [Payload too large](#payload-too-large)
     - [Stack level too deep](#stack-level-too-deep)
@@ -131,13 +135,18 @@ To contribute, check out the [contribution guidelines][contribution docs] and [d
 
 *EOL* indicates support is no longer provided.
 
+### Microsoft Windows support
+
+Using `ddtrace ` on Microsoft Windows is currently unsupported. We'll still accept community contributions and issues,
+but will consider them as having low priority.
+
 ## Installation
 
 The following steps will help you quickly start tracing your Ruby application.
 
 ### Configure the Datadog Agent for APM
 
-Before downloading tracing on your application, install the Datadog Agent. The Ruby APM tracer sends trace data through the Datadog Agent.
+Before downloading tracing on your application, [install the Datadog Agent on the host](https://docs.datadoghq.com/agent/). The Ruby APM tracer sends trace data through the Datadog Agent.
 
 Install and configure the Datadog Agent to receive traces from your now instrumented application. By default the Datadog Agent is enabled in your `datadog.yaml` file under `apm_enabled: true` and listens for trace traffic at `localhost:8126`. For containerized environments, follow the steps below to enable trace collection within the Datadog Agent.
 
@@ -163,7 +172,7 @@ Install and configure the Datadog Agent to receive traces from your now instrume
 
 2. Install the gem with `bundle install`
 
-3. You can configure, override, or disable any specific integration settings by also adding a [Rails Manual Configuration](#rails-manual-configuration) file.
+3. You can configure, override, or disable any specific integration settings by also adding a Rails manual instrumentation configuration file (next).
 
 #### Manual instrumentation
 
@@ -203,7 +212,7 @@ Install and configure the Datadog Agent to receive traces from your now instrume
     require 'ddtrace/auto_instrument'
     ```
 
-    You can configure, override, or disable any specific integration settings by also adding a [Ruby Manual Configuration Block](#ruby-manual-configuration).
+    You can configure, override, or disable any specific integration settings by also adding a Ruby manual configuration block (next).
 
 #### Manual instrumentation
 
@@ -386,14 +395,16 @@ For a list of available integrations, and their configuration options, please re
 | Name                     | Key                        | Versions Supported: MRI  | Versions Supported: JRuby | How to configure                    | Gem source                                                                     |
 | ------------------------ | -------------------------- | ------------------------ | --------------------------| ----------------------------------- | ------------------------------------------------------------------------------ |
 | Action Cable             | `action_cable`             | `>= 5.0`                 | `>= 5.0`                  | *[Link](#action-cable)*             | *[Link](https://github.com/rails/rails/tree/master/actioncable)*               |
+| Action Mailer            | `action_mailer`            | `>= 5.0`                 | `>= 5.0`                  | *[Link](#action-mailer)*            | *[Link](https://github.com/rails/rails/tree/master/actionmailer)*              |
 | Action View              | `action_view`              | `>= 3.0`                 | `>= 3.0`                  | *[Link](#action-view)*              | *[Link](https://github.com/rails/rails/tree/master/actionview)*                |
 | Active Model Serializers | `active_model_serializers` | `>= 0.9`                 | `>= 0.9`                  | *[Link](#active-model-serializers)* | *[Link](https://github.com/rails-api/active_model_serializers)*                |
 | Action Pack              | `action_pack`              | `>= 3.0`                 | `>= 3.0`                  | *[Link](#action-pack)*              | *[Link](https://github.com/rails/rails/tree/master/actionpack)*                |
+| Active Job               | `active_job`               | `>= 4.2`                 | `>= 4.2`                  | *[Link](#active-job)*               | *[Link](https://github.com/rails/rails/tree/master/activejob)*             |
 | Active Record            | `active_record`            | `>= 3.0`                 | `>= 3.0`                  | *[Link](#active-record)*            | *[Link](https://github.com/rails/rails/tree/master/activerecord)*              |
 | Active Support           | `active_support`           | `>= 3.0`                 | `>= 3.0`                  | *[Link](#active-support)*           | *[Link](https://github.com/rails/rails/tree/master/activesupport)*             |
 | AWS                      | `aws`                      | `>= 2.0`                 | `>= 2.0`                  | *[Link](#aws)*                      | *[Link](https://github.com/aws/aws-sdk-ruby)*                                  |
 | Concurrent Ruby          | `concurrent_ruby`          | `>= 0.9`                 | `>= 0.9`                  | *[Link](#concurrent-ruby)*          | *[Link](https://github.com/ruby-concurrency/concurrent-ruby)*                  |
-| Cucumber                 | `cucumber`                 | `>= 3.0`                 | `>= 1.7.16`               | *[Link](#cucumber)*                | *[Link](https://github.com/cucumber/cucumber-ruby)*                            |
+| Cucumber                 | `cucumber`                 | `>= 3.0`                 | `>= 1.7.16`               | *[Link](#cucumber)*                 | *[Link](https://github.com/cucumber/cucumber-ruby)*                            |
 | Dalli                    | `dalli`                    | `>= 2.0`                 | `>= 2.0`                  | *[Link](#dalli)*                    | *[Link](https://github.com/petergoldstein/dalli)*                              |
 | DelayedJob               | `delayed_job`              | `>= 4.1`                 | `>= 4.1`                  | *[Link](#delayedjob)*               | *[Link](https://github.com/collectiveidea/delayed_job)*                        |
 | Elasticsearch            | `elasticsearch`            | `>= 1.0`                 | `>= 1.0`                  | *[Link](#elasticsearch)*            | *[Link](https://github.com/elastic/elasticsearch-ruby)*                        |
@@ -403,7 +414,7 @@ For a list of available integrations, and their configuration options, please re
 | Grape                    | `grape`                    | `>= 1.0`                 | `>= 1.0`                  | *[Link](#grape)*                    | *[Link](https://github.com/ruby-grape/grape)*                                  |
 | GraphQL                  | `graphql`                  | `>= 1.7.9`               | `>= 1.7.9`                | *[Link](#graphql)*                  | *[Link](https://github.com/rmosolgo/graphql-ruby)*                             |
 | gRPC                     | `grpc`                     | `>= 1.7`                 | *gem not available*       | *[Link](#grpc)*                     | *[Link](https://github.com/grpc/grpc/tree/master/src/rubyc)*                   |
-| http.rb                  | `httprb`                   | `>= 2.0`                 | `>= 2.0`                  | *[Link](#http-rb)*                  | *[Link](https://github.com/httprb/http)*                                       |
+| http.rb                  | `httprb`                   | `>= 2.0`                 | `>= 2.0`                  | *[Link](#httprb)*                  | *[Link](https://github.com/httprb/http)*                                       |
 | httpclient                | `httpclient`              | `>= 2.2`                 | `>= 2.2`                  | *[Link](#httpclient)*               | *[Link](https://github.com/nahi/httpclient)*                                     |
 | httpx                     | `httpx`                   | `>= 0.11`                | `>= 0.11`                 | *[Link](#httpx)*                    | *[Link](https://gitlab.com/honeyryderchuck/httpx)*                             |
 | Kafka                    | `ruby-kafka`               | `>= 0.7.10`              | `>= 0.7.10`               | *[Link](#kafka)*                    | *[Link](https://github.com/zendesk/ruby-kafka)*                                |
@@ -469,6 +480,27 @@ Where `options` is an optional `Hash` that accepts the following parameters:
 | `service_name` | Service name used for rendering instrumentation. | `action_view` |
 | `template_base_path` | Used when the template name is parsed. If you don't store your templates in the `views/` folder, you may need to change this value | `'views/'` |
 
+### Action Mailer
+
+The Action Mailer integration provides tracing for Rails 5 ActionMailer actions.
+
+You can enable it through `Datadog.configure`:
+
+```ruby
+require 'ddtrace'
+ Datadog.configure do |c|
+  c.use :action_mailer, options
+end
+```
+
+Where `options` is an optional `Hash` that accepts the following parameters:
+
+| Key | Description | Default |
+| --- | ----------- | ------- |
+| `analytics_enabled` | Enable analytics for spans produced by this integration. `true` for on, `nil` to defer to global setting, `false` for off. | `false` |
+| `service_name` | Service name used for `action_mailer` instrumentation | `'action_mailer'` |
+| `email_data` | Whether or not to append additional email payload metadata to `action_mailer.deliver` spans. Fields include `['subject', 'to', 'from', 'bcc', 'cc', 'date', 'perform_deliveries']`. | `false` |
+
 ### Active Model Serializers
 
 The Active Model Serializers integration traces the `serialize` event for version 0.9+ and the `render` event for version 0.10+.
@@ -508,6 +540,28 @@ Where `options` is an optional `Hash` that accepts the following parameters:
 | ---| --- | --- |
 | `service_name` | Service name used for rendering instrumentation. | `action_pack` |
 
+### Active Job
+
+Most of the time, Active Job is set up as part of Rails, but it can be activated separately:
+
+```ruby
+require 'active_job'
+require 'ddtrace'
+
+Datadog.configure do |c|
+  c.use :active_job, options
+end
+
+ExampleJob.perform_later
+```
+
+Where `options` is an optional `Hash` that accepts the following parameters:
+
+| Key | Description | Default |
+| --- | ----------- | ------- |
+| `service_name` | Service name used for `active_job` instrumentation | `'active_job'` |
+| `log_injection` | Automatically enables injection [Trace Correlation](#trace-correlation) information, such as `dd.trace_id`, into Active Job logs. Supports the default logger (`ActiveSupport::TaggedLogging`) and `Lograge`. Details on the format of Trace Correlation information can be found in the [Trace Correlation](#trace-correlation) section. | `false` |
+
 ### Active Record
 
 Most of the time, Active Record is set up as part of a web framework (Rails, Sinatra...) however, it can be set up alone:
@@ -1028,6 +1082,7 @@ Where `options` is an optional `Hash` that accepts the following parameters:
 | Key | Description | Default |
 | --- | ----------- | ------- |
 | `service_name` | Service name used for `grpc` instrumentation | `'grpc'` |
+| `error_handler` | Custom error handler invoked when a request is an error. A `Proc` that accepts `span` and `error` parameters. Sets error on the span by default. | `proc { |span, error| span.set_error(error) unless span.nil? }` |
 
 **Configuring clients to use different settings**
 
@@ -1167,6 +1222,37 @@ Where `options` is an optional `Hash` that accepts the following parameters:
 | `quantize` | Hash containing options for quantization. May include `:show` with an Array of keys to not quantize (or `:all` to skip quantization), or `:exclude` with Array of keys to exclude entirely. | `{ show: [:collection, :database, :operation] }` |
 | `service_name` | Service name used for `mongo` instrumentation | `'mongodb'` |
 
+**Configuring trace settings per connection**
+
+You can configure trace settings per connection by using the `describes` option:
+
+```ruby
+# Provide a `:describes` option with a connection key.
+# Any of the following keys are acceptable and equivalent to one another.
+# If a block is provided, it yields a Settings object that
+# accepts any of the configuration options listed above.
+
+Datadog.configure do |c|
+  # Network connection string
+  c.use :mongo, describes: '127.0.0.1:27017', service_name: 'mongo-primary'
+
+  # Network connection regular expression
+  c.use :mongo, describes: /localhost.*/, service_name: 'mongo-secondary'
+end
+
+client = Mongo::Client.new([ '127.0.0.1:27017' ], :database => 'artists')
+collection = client[:people]
+collection.insert_one({ name: 'Steve' })
+# Traced call will belong to `mongo-primary` service
+
+client = Mongo::Client.new([ 'localhost:27017' ], :database => 'artists')
+collection = client[:people]
+collection.insert_one({ name: 'Steve' })
+# Traced call will belong to `mongo-secondary` service
+```
+
+When multiple `describes` configurations match a connection, the latest configured rule that matches will be applied.
+
 ### MySQL2
 
 The MySQL2 integration traces any SQL command sent through `mysql2` gem.
@@ -1421,6 +1507,7 @@ Where `options` is an optional `Hash` that accepts the following parameters:
 | `database_service` | Database service name used when tracing database activity | `'<app_name>-<adapter_name>'` |
 | `distributed_tracing` | Enables [distributed tracing](#distributed-tracing) so that this service trace is connected with a trace of another service if tracing headers are received | `true` |
 | `exception_controller` | Class or Module which identifies a custom exception controller class. Tracer provides improved error behavior when it can identify custom exception controllers. By default, without this option, it 'guesses' what a custom exception controller looks like. Providing this option aids this identification. | `nil` |
+| `job_service` | Service name used when tracing ActiveJob activity. | `<app_name>-active_job` |
 | `middleware` | Add the trace middleware to the Rails application. Set to `false` if you don't want the middleware to load. | `true` |
 | `middleware_names` | Enables any short-circuited middleware requests to display the middleware name as a resource for the trace. | `false` |
 | `service_name` | Service name used when tracing application requests (on the `rack` level) | `'<app_name>'` (inferred from your Rails application namespace) |
@@ -1431,7 +1518,6 @@ Where `options` is an optional `Hash` that accepts the following parameters:
 
 | MRI Versions  | JRuby Versions | Rails Versions |
 | ------------- | -------------- | -------------- |
-|  2.0          |                |  3.0 - 3.2     |
 |  2.1          |                |  3.0 - 4.2     |
 |  2.2 - 2.3    |                |  3.0 - 5.2     |
 |  2.4          |                |  4.2.8 - 5.2   |
@@ -2122,7 +2208,7 @@ For more details on how to activate distributed tracing for integrations, see th
 - [Rack](#rack)
 - [Rails](#rails)
 - [Sinatra](#sinatra)
-- [http.rb](#http-rb)
+- [http.rb](#httprb)
 - [httpclient](#httpclient)
 - [httpx](#httpx)
 
@@ -2340,7 +2426,9 @@ Datadog.tracer.trace('my.operation') { logger.warn('This is a traced operation.'
 
 ### Configuring the transport layer
 
-By default, the tracer submits trace data using `Net::HTTP` to `127.0.0.1:8126`, the default location for the Datadog trace agent process. However, the tracer can be configured to send its trace data to alternative destinations, or by alternative protocols.
+By default, the tracer submits trace data using the Unix socket `/var/run/datadog/apm.socket`, if one is created by the Agent. Otherwise, it connects via HTTP to `127.0.0.1:8126`, the default TCP location the Agent listens on.
+
+However, the tracer can be configured to send its trace data to alternative destinations, or by alternative protocols.
 
 Some basic settings, such as hostname and port, can be configured using [tracer settings](#tracer-settings).
 
@@ -2366,7 +2454,7 @@ To use, first configure your trace agent to listen by Unix socket, then configur
 ```ruby
 Datadog.configure do |c|
   c.tracer.transport_options = proc { |t|
-    # Provide filepath to trace agent Unix socket
+    # Provide local path to trace agent Unix socket
     t.adapter :unix, '/tmp/ddagent/trace.sock'
   }
 end
@@ -2410,7 +2498,7 @@ The tracer and its integrations can produce some additional metrics that can pro
 To configure your application for metrics collection:
 
 1. [Configure your Datadog agent for StatsD](https://docs.datadoghq.com/developers/dogstatsd/#setup)
-2. Add `gem 'dogstatsd-ruby', '~> 5.2'` to your Gemfile
+2. Add `gem 'dogstatsd-ruby', '~> 5.3'` to your Gemfile
 
 #### For application runtime
 
@@ -2439,11 +2527,13 @@ See the [Dogstatsd documentation](https://www.rubydoc.info/github/DataDog/dogsta
 
 The stats are VM specific and will include:
 
-| Name                        | Type    | Description                                              |
-| --------------------------  | ------- | -------------------------------------------------------- |
-| `runtime.ruby.class_count`  | `gauge` | Number of classes in memory space.                       |
-| `runtime.ruby.thread_count` | `gauge` | Number of threads.                                       |
-| `runtime.ruby.gc.*`.        | `gauge` | Garbage collection statistics: collected from `GC.stat`. |
+| Name                        | Type    | Description                                              | Available on |
+| --------------------------  | ------- | -------------------------------------------------------- | ------------ |
+| `runtime.ruby.class_count`  | `gauge` | Number of classes in memory space.                       | CRuby        |
+| `runtime.ruby.gc.*`         | `gauge` | Garbage collection statistics: collected from `GC.stat`. | All runtimes |
+| `runtime.ruby.thread_count` | `gauge` | Number of threads.                                       | All runtimes |
+| `runtime.ruby.global_constant_state` | `gauge` | Global constant cache generation.               | CRuby        |
+| `runtime.ruby.global_method_state`   | `gauge` | [Global method cache generation.](https://tenderlovemaking.com/2015/12/23/inline-caching-in-mri.html) | [CRuby < 3.0.0](https://docs.ruby-lang.org/en/3.0.0/NEWS_md.html#label-Implementation+improvements) |
 
 In addition, all metrics include the following tags:
 
@@ -2489,7 +2579,11 @@ However, additional instrumentation provided by Datadog can be activated alongsi
 
 **Setup**
 
-To get started with profiling, follow the [Profiler Getting Started Guide](https://docs.datadoghq.com/tracing/profiler/getting_started/?code-lang=ruby).
+To get started with profiling, follow the [Enabling the Ruby Profiler](https://docs.datadoghq.com/tracing/profiler/enabling/ruby/) guide.
+
+#### Troubleshooting
+
+If you run into issues with profiling, please check the [Profiler Troubleshooting Guide](https://docs.datadoghq.com/tracing/profiler/profiler_troubleshooting/?code-lang=ruby).
 
 #### Profiling Resque jobs
 

data/docs/ProfilingDevelopment.md

@@ -11,7 +11,7 @@ Components below live inside <../lib/ddtrace/profiling>:
 * `Collectors::Stack`: Collects stack trace samples from Ruby threads for both CPU-time (if available) and wall-clock.
   Runs on its own background thread.
 * `Encoding::Profile`: Encodes gathered data into the pprof format.
-* `Events::Stack`, `Events::StackSample`, `Events::StackExceptionSample`: Entity classes used to represent stacks.
+* `Events::Stack`, `Events::StackSample`: Entity classes used to represent stacks.
 * `Ext::CPU`: Monkey patches Ruby's `Thread` with our `Ext::CThread` to enable CPU-time profiling.
 * `Ext::CThread`: Extension used to enable CPU-time profiling via use of Pthread's `getcpuclockid`.
 * `Ext::Forking`: Monkey patches `Kernel#fork`, adding a `Kernel#at_fork` callback mechanism which is used to restore
@@ -20,6 +20,7 @@ Components below live inside <../lib/ddtrace/profiling>:
 * `Tasks::Setup`: Takes care of loading our extensions/monkey patches to handle fork() and CPU profiling.
 * `Transport::*` (in <../lib/ddtrace/profiling/transport>): Implements transmission of profiling payloads to the Datadog agent
   or backend.
+* `TraceIdentifiers::*`: Used to retrieve trace id and span id from tracers, to be used to connect traces to profiles.
 * `BacktraceLocation`: Entity class used to represent an entry in a stack trace.
 * `Buffer`: Bounded buffer used to store profiling events.
 * `Exporter`: Writes profiling data to a given transport.
@@ -41,22 +42,14 @@ flow:
 4. The `Setup` task activates our extensions
     * `Datadog::Profiling::Ext::Forking`
     * `Datadog::Profiling::Ext::CPU`
-5. Still inside `Datadog::Components`, the `build_profiler` method then creates and wires up the Profiler:
-    ```ruby
-          recorder = build_profiler_recorder(settings)
-          collectors = build_profiler_collectors(settings, recorder)
-          exporters = build_profiler_exporters(settings)
-          scheduler = build_profiler_scheduler(settings, recorder, exporters)
-
-          Datadog::Profiler.new(collectors, scheduler)
-    ```
+5. Still inside `Datadog::Components`, the `build_profiler` method then creates and wires up the Profiler as such:
     ```asciiflow
             +------------+
             |  Profiler  |
-            +-+--------+-+
-              |        |
-              v        v
-    +---------+--+  +--+--------+
+            +-+-------+--+
+              |       |
+              v       v
+    +---------+--+  +-+---------+
     | Collectors |  | Scheduler |
     +---------+--+  +-+-------+-+
               |       |       |
@@ -86,3 +79,29 @@ takes care of encoding the data and reporting it to the datadog agent (or to the
 ## How CPU-time profiling works
 
 **TODO**: Document our pthread-based approach to getting CPU-time for threads.
+
+## How linking of traces to profiles works
+
+The [code hotspots feature](https://docs.datadoghq.com/tracing/profiler/connect_traces_and_profiles) allows users to start
+from a trace and then to investigate the profile that corresponds to that trace.
+
+This works in two steps:
+1. Linking a trace to the profile that was gathered while it executed
+2. Enabling the filtering of a profile to contain only the samples relating to a given trace/span
+
+To link a trace to a profile, we must ensure that both have the same `runtime-id` tag.
+This tag is in `Datadog::Runtime::Identity.id` and is automatically added by both the tracer and the profiler to reported
+traces/profiles.
+
+The profiler backend links a trace covering a given time interval to the profiles covering the same time interval,
+whenever they share the same `runtime-id`.
+
+To further enable filtering of a profile to show only samples related to a given trace/span, each sample taken by the
+profiler is tagged with the `local root span id` and `span id` for the given trace/span.
+
+This is done using the `Datadog::Profiling::TraceIdentifiers::Helper` that retrieves a `root_span_id` and `span_id`, if
+available, from the supported tracers. This helper is called by the `Collectors::Stack` during sampling.
+
+Note that if a given trace executes too fast, it's possible that the profiler will not contain any samples for that
+specific trace. Nevertheless, the linking still works and is useful, as it allows users to explore what was going on their
+profile at that time, even if they can't filter down to the specific request.

data/ext/ddtrace_profiling_native_extension/NativeExtensionDesign.md

@@ -0,0 +1,86 @@
+# Profiling Native Extension Design
+
+The profiling native extension is used to implement features which are expensive (in terms of resources) or otherwise
+impossible to implement using Ruby code.
+
+This extension is quite coupled with MRI Ruby ("C Ruby") internals, and is not intended to support other rubies such as
+JRuby or TruffleRuby. When below we say "Ruby", read it as "MRI Ruby".
+
+## Disabling
+
+The profiling native extension can be disabled by setting `DD_PROFILING_NO_EXTENSION=true` when installing or running
+the gem. Setting `DD_PROFILING_NO_EXTENSION` at installation time skips compilation of the extension entirely.
+
+(If you're a customer and needed to use this, please tell us why on <https://github.com/DataDog/dd-trace-rb/issues/new>.)
+
+Currently the profiler can still "limp along" when the native extension is disabled, but the plan is to require it
+in future releases -- e.g. disabling the extension will disable profiling entirely.
+
+## Safety
+
+The profiling native extension is (and must always be) designed to **not cause failures** during gem installation, even
+if some features, Ruby versions, or operating systems are not supported.
+
+E.g. the extension must cleanly build on Ruby 2.1 (or the oldest Ruby version we support at the time) on Windows,
+even if at run time it will effectively do nothing for such a setup.
+
+We have a CI setup to help validate this, but this is really important to keep in mind when adding to or changing the
+existing codebase.
+
+## Usage of private VM headers
+
+To implement some of the features below, we sometimes require access to private Ruby header files (that describe VM
+internal types, structures and functions).
+
+Because these private header files are not included in regular Ruby installations, we have two different workarounds:
+
+1. for Ruby versions >= 2.6 we make use use the Ruby private MJIT header
+2. for Ruby versions < 2.6 (legacy Rubies) we make use of the `debase-ruby_core_source` gem
+
+Functions which make use of these headers are defined in the <private_vm_api_acccess.c> file.
+
+**Important Note**: Our medium/long-term plan is to stop relying on all private Ruby headers, and instead request and
+contribute upstream changes so that they become official public VM APIs.
+
+### Approach 1: Using the Ruby private MJIT header
+
+Ruby versions >= 2.6 introduced a JIT compiler called MJIT. This compiler does not directly generate machine code;
+instead it generates C code and uses the system C compiler to turn it into machine code.
+
+The generated C code `#include`s a private header -- which we reference as "the MJIT header" everywhere.
+The MJIT header gets shipped with all MJIT-enabled Rubies and includes the layout of many internal VM structures;
+and of course the intention is that it is only used by the Ruby MJIT compiler.
+
+This header is placed inside the `include/` directory in a Ruby installation, and is named for that specific Ruby
+version. e.g. `rb_mjit_min_header-2.7.4.h`.
+
+### Approach 2: Using the `debase-ruby_core_source` gem
+
+The [`debase-ruby_core_source`](https://github.com/ruby-debug/debase-ruby_core_source) contains almost no code;
+instead, it just contains per-Ruby-version folders with the private VM headers (`.h`) files for that version.
+
+Thus, even though a regular Ruby installation does not include these files, we can access the copy inside this gem.
+
+## Feature: Getting thread CPU-time clock_ids
+
+* **OS support**: Linux
+* **Ruby support**: 2.6+
+
+To enable CPU-time profiling, we use the `pthread_getcpuclockid(pthread_t thread, clockid_t *clockid)` C function to
+obtain a `clockid_t` that can then be used with the `Process.clock_gettime` method (or directly with the
+`clock_gettime()` C function).
+
+The challenge with using `pthread_getcpuclockid()` is that we need to get the `pthread_t` for a given Ruby `Thread`
+object. We previously did this with a weird combination of monkey patching and `pthread_self()` (effectively patching
+every `Thread` to run `pthread_self()` at initialization time and stash that value somewhere), but this had a number
+of downsides.
+
+The approach we use in the profiling native extension is to reach inside the internal structure of the `Thread` object,
+and extract the `pthread_t` that Ruby itself keeps, but does not expose. This is implemented in the `pthread_id_for()`
+function in `private_vm_api_acccess.c`. Thus, using this trick we can at any point in execution go from a `Thread`
+object into the `clockid_t` that we need.
+
+Note that `pthread_getcpuclockid()` is not available on macOS (nor, obviously, on Windows), and hence this feature
+is currently Linux-specific. Thus, in the <clock_id_from_pthread.c> file we implement the feature for supported Ruby
+setups but if something is missing we instead compile in <clock_id_noop.c> that includes a no-op implementation of the
+feature.

data/ext/ddtrace_profiling_native_extension/clock_id.h

@@ -0,0 +1,4 @@
+#pragma once
+
+void self_test_clock_id();
+VALUE clock_id_for(VALUE self, VALUE thread);

data/ext/ddtrace_profiling_native_extension/clock_id_from_pthread.c

@@ -0,0 +1,52 @@
+#include "extconf.h"
+
+// This file is only compiled on systems where pthread_getcpuclockid() is available;
+// Otherwise we compile clock_id_noop.c
+#ifdef HAVE_PTHREAD_GETCPUCLOCKID
+
+#include <pthread.h>
+#include <time.h>
+#include <errno.h>
+
+#include <ruby.h>
+
+#ifdef RUBY_2_1_WORKAROUND
+#include <thread_native.h>
+#else
+#include <ruby/thread_native.h>
+#endif
+
+#include "private_vm_api_access.h"
+
+#include "clock_id.h"
+
+// Validate that our home-cooked pthread_id_for() matches pthread_self() for the current thread
+void self_test_clock_id() {
+  rb_nativethread_id_t expected_pthread_id = pthread_self();
+  rb_nativethread_id_t actual_pthread_id = pthread_id_for(rb_thread_current());
+
+  if (expected_pthread_id != actual_pthread_id) rb_raise(rb_eRuntimeError, "pthread_id_for() self-test failed");
+}
+
+VALUE clock_id_for(VALUE self, VALUE thread) {
+  rb_nativethread_id_t thread_id = pthread_id_for(thread);
+
+  clockid_t clock_id;
+  int error = pthread_getcpuclockid(thread_id, &clock_id);
+
+  if (error == 0) {
+    return CLOCKID2NUM(clock_id);
+  } else {
+    switch(error) {
+      // The more specific error messages are based on the pthread_getcpuclockid(3) man page
+      case ENOENT:
+        rb_exc_raise(rb_syserr_new(error, "Failed to get clock_id for given thread: Per-thread CPU time clocks are not supported by the system."));
+      case ESRCH:
+        rb_exc_raise(rb_syserr_new(error, "Failed to get clock_id for given thread: No thread could be found."));
+      default:
+        rb_exc_raise(rb_syserr_new(error, "Failed to get clock_id for given thread"));
+    }
+  }
+}
+
+#endif

data/ext/ddtrace_profiling_native_extension/clock_id_noop.c

@@ -0,0 +1,14 @@
+#include "extconf.h"
+
+// This file is the dual of clock_id_from_pthread.c for systems where that info
+// is not available.
+#ifndef HAVE_PTHREAD_GETCPUCLOCKID
+
+#include <ruby.h>
+
+#include "clock_id.h"
+
+void self_test_clock_id() { } // Nothing to check
+VALUE clock_id_for(VALUE self, VALUE thread) { return Qnil; } // Nothing to return
+
+#endif