ddtrace 0.54.0 → 1.0.0

data/lib/datadog/profiling/ext/forking.rb

@@ -0,0 +1,99 @@
+# typed: false
+
+module Datadog
+  module Profiling
+    module Ext
+      # Monkey patches `Kernel#fork`, adding a `Kernel#at_fork` callback mechanism which is used to restore
+      # profiling abilities after the VM forks.
+      module Forking
+        def self.supported?
+          Process.respond_to?(:fork)
+        end
+
+        def self.apply!
+          return false unless supported?
+
+          modules = [::Process, ::Kernel]
+          # TODO: Ruby < 2.3 doesn't support Binding#receiver.
+          #       Remove "else #eval" clause when Ruby < 2.3 support is dropped.
+          # NOTE: Modifying the "main" object as we do here is (as far as I know) irreversible. During tests, this change
+          #       will stick around even if we otherwise stub `Process` and `Kernel`.
+          modules << (TOPLEVEL_BINDING.respond_to?(:receiver) ? TOPLEVEL_BINDING.receiver : TOPLEVEL_BINDING.eval('self'))
+
+          # Patch top-level binding, Kernel, Process.
+          # NOTE: We could instead do Kernel.module_eval { def fork; ... end }
+          #       however, this method rewrite is more invasive and irreversible.
+          #       It could also have collisions with other libraries that patch.
+          #       Opt to modify the inheritance of each relevant target instead.
+          modules.each do |mod|
+            clazz = if mod.class <= Module
+                      mod.singleton_class
+                    else
+                      mod.class
+                    end
+
+            clazz.prepend(Kernel)
+          end
+        end
+
+        # Extensions for kernel
+        module Kernel
+          FORK_STAGES = [:prepare, :parent, :child].freeze
+
+          def fork
+            # If a block is provided, it must be wrapped to trigger callbacks.
+            child_block = if block_given?
+                            proc do
+                              # Trigger :child callback
+                              at_fork_blocks[:child].each(&:call) if at_fork_blocks.key?(:child)
+
+                              # Invoke original block
+                              yield
+                            end
+                          end
+
+            # Trigger :prepare callback
+            at_fork_blocks[:prepare].each(&:call) if at_fork_blocks.key?(:prepare)
+
+            # Start fork
+            # If a block is provided, use the wrapped version.
+            result = child_block.nil? ? super : super(&child_block)
+
+            # Trigger correct callbacks depending on whether we're in the parent or child.
+            # If we're in the fork, result = nil: trigger child callbacks.
+            # If we're in the parent, result = fork PID: trigger parent callbacks.
+            # rubocop:disable Style/IfInsideElse
+            if result.nil?
+              # Trigger :child callback
+              at_fork_blocks[:child].each(&:call) if at_fork_blocks.key?(:child)
+            else
+              # Trigger :parent callback
+              at_fork_blocks[:parent].each(&:call) if at_fork_blocks.key?(:parent)
+            end
+            # rubocop:enable Style/IfInsideElse
+
+            # Return PID from #fork
+            result
+          end
+
+          def at_fork(stage = :prepare, &block)
+            raise ArgumentError, 'Bad \'stage\' for ::at_fork' unless FORK_STAGES.include?(stage)
+
+            at_fork_blocks[stage] = [] unless at_fork_blocks.key?(stage)
+            at_fork_blocks[stage] << block
+          end
+
+          module_function
+
+          def at_fork_blocks
+            # Blocks should be shared across all users of this module,
+            # e.g. Process#fork, Kernel#fork, etc. should all invoke the same callbacks.
+            # rubocop:disable Style/ClassVars
+            @@at_fork_blocks ||= {}
+            # rubocop:enable Style/ClassVars
+          end
+        end
+      end
+    end
+  end
+end

data/lib/datadog/profiling/ext.rb

@@ -0,0 +1,57 @@
+# typed: true
+
+module Datadog
+  module Profiling
+    module Ext
+      ENV_ENABLED = 'DD_PROFILING_ENABLED'.freeze
+      ENV_UPLOAD_TIMEOUT = 'DD_PROFILING_UPLOAD_TIMEOUT'.freeze
+      ENV_MAX_FRAMES = 'DD_PROFILING_MAX_FRAMES'.freeze
+      ENV_AGENTLESS = 'DD_PROFILING_AGENTLESS'.freeze
+      ENV_ENDPOINT_COLLECTION_ENABLED = 'DD_PROFILING_ENDPOINT_COLLECTION_ENABLED'.freeze
+
+      module Pprof
+        LABEL_KEY_LOCAL_ROOT_SPAN_ID = 'local root span id'.freeze
+        LABEL_KEY_SPAN_ID = 'span id'.freeze
+        LABEL_KEY_THREAD_ID = 'thread id'.freeze
+        LABEL_KEY_TRACE_ENDPOINT = 'trace endpoint'.freeze
+        SAMPLE_VALUE_NO_VALUE = 0
+        VALUE_TYPE_CPU = 'cpu-time'.freeze
+        VALUE_TYPE_WALL = 'wall-time'.freeze
+        VALUE_UNIT_NANOSECONDS = 'nanoseconds'.freeze
+      end
+
+      module Transport
+        module HTTP
+          URI_TEMPLATE_DD_API = 'https://intake.profile.%s/'.freeze
+
+          FORM_FIELD_RECORDING_START = 'start'.freeze
+          FORM_FIELD_RECORDING_END = 'end'.freeze
+          FORM_FIELD_FAMILY = 'family'.freeze
+          FORM_FIELD_TAG_ENV = 'env'.freeze
+          FORM_FIELD_TAG_HOST = 'host'.freeze
+          FORM_FIELD_TAG_LANGUAGE = 'language'.freeze
+          FORM_FIELD_TAG_PID = 'pid'.freeze
+          FORM_FIELD_TAG_PROFILER_VERSION = 'profiler_version'.freeze
+          FORM_FIELD_TAG_RUNTIME = 'runtime'.freeze
+          FORM_FIELD_TAG_RUNTIME_ENGINE = 'runtime_engine'.freeze
+          FORM_FIELD_TAG_RUNTIME_ID = 'runtime-id'.freeze
+          FORM_FIELD_TAG_RUNTIME_PLATFORM = 'runtime_platform'.freeze
+          FORM_FIELD_TAG_RUNTIME_VERSION = 'runtime_version'.freeze
+          FORM_FIELD_TAG_SERVICE = 'service'.freeze
+          FORM_FIELD_TAG_VERSION = 'version'.freeze
+          FORM_FIELD_TAGS = 'tags'.freeze
+          FORM_FIELD_INTAKE_VERSION = 'version'.freeze
+
+          HEADER_CONTENT_TYPE = 'Content-Type'.freeze
+          HEADER_CONTENT_TYPE_OCTET_STREAM = 'application/octet-stream'.freeze
+
+          FORM_FIELD_PPROF_DATA = 'data[rubyprofile.pprof]'.freeze
+          PPROF_DEFAULT_FILENAME = 'rubyprofile.pprof.gz'.freeze
+
+          FORM_FIELD_CODE_PROVENANCE_DATA = 'data[code-provenance.json]'.freeze
+          CODE_PROVENANCE_FILENAME = 'code-provenance.json.gz'.freeze
+        end
+      end
+    end
+  end
+end

data/lib/datadog/profiling/flush.rb

@@ -0,0 +1,69 @@
+# typed: false
+
+require 'datadog/core/environment/identity'
+require 'datadog/core/environment/socket'
+
+module Datadog
+  module Profiling
+    # Entity class used to represent metadata for a given profile
+    Flush = Struct.new(
+      :start,
+      :finish,
+      :event_groups,
+      :event_count,
+      :code_provenance,
+      :runtime_id,
+      :service,
+      :env,
+      :version,
+      :host,
+      :language,
+      :runtime_engine,
+      :runtime_platform,
+      :runtime_version,
+      :profiler_version,
+      :tags
+    ) do
+      def initialize(
+        start:,
+        finish:,
+        event_groups:,
+        event_count:,
+        code_provenance:,
+        runtime_id: Core::Environment::Identity.id,
+        service: Datadog.configuration.service,
+        env: Datadog.configuration.env,
+        version: Datadog.configuration.version,
+        host: Core::Environment::Socket.hostname,
+        language: Core::Environment::Identity.lang,
+        runtime_engine: Core::Environment::Identity.lang_engine,
+        runtime_platform: Core::Environment::Identity.lang_platform,
+        runtime_version: Core::Environment::Identity.lang_version,
+        profiler_version: Core::Environment::Identity.tracer_version,
+        tags: Datadog.configuration.tags
+      )
+        super(
+          start,
+          finish,
+          event_groups,
+          event_count,
+          code_provenance,
+          runtime_id,
+          service,
+          env,
+          version,
+          host,
+          language,
+          runtime_engine,
+          runtime_platform,
+          runtime_version,
+          profiler_version,
+          tags,
+        )
+      end
+    end
+
+    # Represents a collection of events of a specific type being flushed.
+    EventGroup = Struct.new(:event_class, :events).freeze
+  end
+end

data/lib/datadog/profiling/native_extension.rb

@@ -0,0 +1,41 @@
+# typed: true
+
+module Datadog
+  module Profiling
+    # This module contains classes and methods which are implemented using native code in the
+    # ext/ddtrace_profiling_native_extension folder, as well as some Ruby-level utilities that don't make sense to
+    # write using C
+    module NativeExtension
+      private_class_method def self.working?
+        native_working?
+      end
+
+      unless singleton_class.private_method_defined?(:native_working?)
+        private_class_method def self.native_working?
+          false
+        end
+      end
+
+      unless singleton_class.method_defined?(:clock_id_for)
+        def self.clock_id_for(_)
+          nil
+        end
+      end
+
+      def self.cpu_time_ns_for(thread)
+        clock_id =
+          begin
+            clock_id_for(thread)
+          rescue Errno::ESRCH
+            nil
+          end
+
+        begin
+          ::Process.clock_gettime(clock_id, :nanosecond) if clock_id
+        rescue Errno::EINVAL
+          nil
+        end
+      end
+    end
+  end
+end

data/lib/datadog/profiling/pprof/builder.rb

@@ -0,0 +1,127 @@
+# frozen_string_literal: true
+
+# typed: true
+
+require 'datadog/profiling/flush'
+require 'datadog/profiling/pprof/message_set'
+require 'datadog/profiling/pprof/string_table'
+require 'datadog/core/utils/time'
+
+module Datadog
+  module Profiling
+    module Pprof
+      # Accumulates profile data and produces a Perftools::Profiles::Profile
+      class Builder
+        DEFAULT_ENCODING = 'UTF-8'
+        DESC_FRAME_OMITTED = 'frame omitted'
+        DESC_FRAMES_OMITTED = 'frames omitted'
+
+        attr_reader \
+          :functions,
+          :locations,
+          :mappings,
+          :sample_types,
+          :samples,
+          :string_table
+
+        def initialize
+          @functions = MessageSet.new(1)
+          @locations = initialize_locations_hash
+          @mappings = MessageSet.new(1)
+          @sample_types = MessageSet.new
+          @samples = []
+          @string_table = StringTable.new
+
+          # Cache this proc, since it's pretty expensive to keep recreating it
+          @build_function = method(:build_function).to_proc
+        end
+
+        # The locations hash maps unique BacktraceLocation instances to their corresponding pprof Location objects;
+        # there's a 1:1 correspondence, since BacktraceLocations were already deduped
+        def initialize_locations_hash
+          sequence = Core::Utils::Sequence.new(1)
+          Hash.new do |locations_hash, backtrace_location|
+            locations_hash[backtrace_location] = build_location(sequence.next, backtrace_location)
+          end
+        end
+
+        def encode_profile(profile)
+          Perftools::Profiles::Profile.encode(profile).force_encoding(DEFAULT_ENCODING)
+        end
+
+        def build_profile(start:, finish:)
+          start_ns = Core::Utils::Time.as_utc_epoch_ns(start)
+          finish_ns = Core::Utils::Time.as_utc_epoch_ns(finish)
+
+          Perftools::Profiles::Profile.new(
+            sample_type: @sample_types.messages,
+            sample: @samples,
+            mapping: @mappings.messages,
+            location: @locations.values,
+            function: @functions.messages,
+            string_table: @string_table.strings,
+            time_nanos: start_ns,
+            duration_nanos: finish_ns - start_ns,
+          )
+        end
+
+        def build_value_type(type, unit)
+          Perftools::Profiles::ValueType.new(
+            type: @string_table.fetch(type),
+            unit: @string_table.fetch(unit)
+          )
+        end
+
+        def build_locations(backtrace_locations, length)
+          locations = backtrace_locations.collect { |backtrace_location| @locations[backtrace_location] }
+
+          omitted = length - backtrace_locations.length
+
+          # Add placeholder stack frame if frames were truncated
+          if omitted > 0
+            desc = omitted == 1 ? DESC_FRAME_OMITTED : DESC_FRAMES_OMITTED
+            locations << @locations[Profiling::BacktraceLocation.new('', 0, "#{omitted} #{desc}")]
+          end
+
+          locations
+        end
+
+        def build_location(id, backtrace_location)
+          Perftools::Profiles::Location.new(
+            id: id,
+            line: [build_line(
+              @functions.fetch(
+                backtrace_location.path,
+                backtrace_location.base_label,
+                &@build_function
+              ).id,
+              backtrace_location.lineno
+            )]
+          )
+        end
+
+        def build_line(function_id, line_number)
+          Perftools::Profiles::Line.new(
+            function_id: function_id,
+            line: line_number
+          )
+        end
+
+        def build_function(id, filename, function_name)
+          Perftools::Profiles::Function.new(
+            id: id,
+            name: @string_table.fetch(function_name),
+            filename: @string_table.fetch(filename)
+          )
+        end
+
+        def build_mapping(id, filename)
+          Perftools::Profiles::Mapping.new(
+            id: id,
+            filename: @string_table.fetch(filename)
+          )
+        end
+      end
+    end
+  end
+end

data/lib/datadog/profiling/pprof/converter.rb

@@ -0,0 +1,104 @@
+# typed: true
+
+require 'datadog/profiling/ext'
+
+module Datadog
+  module Profiling
+    module Pprof
+      # Base class for converters that convert profiling events to pprof
+      class Converter
+        attr_reader \
+          :builder
+
+        # Override in child class to define sample types
+        # this converter uses when building samples.
+        def self.sample_value_types
+          raise NotImplementedError
+        end
+
+        def initialize(builder, sample_type_mappings)
+          @builder = builder
+          @sample_type_mappings = sample_type_mappings
+        end
+
+        def group_events(events)
+          # Event grouping in format:
+          # [key, EventGroup]
+          event_groups = {}
+
+          # Aggregate each event into a group
+          # with identical properties, but different values.
+          events.each do |event|
+            key = yield(event)
+            values = build_event_values(event)
+
+            unless key.nil?
+              if event_groups.key?(key)
+                # Update existing group from event
+                update_group(event_groups[key], event, values)
+              else
+                # Add new group
+                event_groups[key] = new_group(event, values)
+              end
+            end
+          end
+
+          event_groups
+        end
+
+        def add_events!(events)
+          raise NotImplementedError
+        end
+
+        def sample_value_index(type)
+          index = @sample_type_mappings[type]
+          raise UnknownSampleTypeMappingError, type unless index
+
+          index
+        end
+
+        def build_event_values(event)
+          # Build a value array that matches the length of the sample types
+          # Populate all values with "no value" by default
+          Array.new(@sample_type_mappings.length, Profiling::Ext::Pprof::SAMPLE_VALUE_NO_VALUE)
+        end
+
+        def debug_statistics
+          # Empty; can be used by subclasses to report a string containing debug statistics to be logged
+        end
+
+        # Represents a grouped event
+        # 'sample' is an example event object from the group.
+        # 'values' is the summation of the group's sample values
+        EventGroup = Struct.new(:sample, :values)
+
+        # Error when the mapping of a sample type to value index is unknown
+        class UnknownSampleTypeMappingError < StandardError
+          attr_reader :type
+
+          def initialize(type)
+            @type = type
+          end
+
+          def message
+            "Mapping for sample value type '#{type}' to index is unknown."
+          end
+        end
+
+        protected
+
+        def new_group(event, values)
+          EventGroup.new(event, values)
+        end
+
+        def update_group(event_group, event, values)
+          # Update values for group
+          group_values = event_group.values
+          group_values.each_with_index do |group_value, i|
+            group_values[i] = group_value + values[i]
+          end
+        end
+      end
+    end
+  end
+end

data/lib/datadog/profiling/pprof/message_set.rb

@@ -0,0 +1,16 @@
+# typed: true
+
+require 'datadog/core/utils/object_set'
+
+module Datadog
+  module Profiling
+    module Pprof
+      # Acts as a unique dictionary of protobuf messages
+      class MessageSet < Core::Utils::ObjectSet
+        def messages
+          objects
+        end
+      end
+    end
+  end
+end

data/lib/datadog/profiling/pprof/payload.rb

@@ -0,0 +1,20 @@
+# typed: true
+
+module Datadog
+  module Profiling
+    module Pprof
+      # Pprof output data.
+      # Includes encoded data and list of types.
+      Payload = Struct.new(:data, :types) do
+        def initialize(data, types)
+          super
+          self.types = types || []
+        end
+
+        def to_s
+          data
+        end
+      end
+    end
+  end
+end

data/lib/datadog/profiling/pprof/pprof_pb.rb

@@ -0,0 +1,83 @@
+# typed: false
+
+# Generated by the protocol buffer compiler.  DO NOT EDIT!
+# source: lib/datadog/profiling/pprof/pprof.proto
+
+require 'google/protobuf'
+
+Google::Protobuf::DescriptorPool.generated_pool.build do
+  add_message "perftools.profiles.Profile" do
+    repeated :sample_type, :message, 1, "perftools.profiles.ValueType"
+    repeated :sample, :message, 2, "perftools.profiles.Sample"
+    repeated :mapping, :message, 3, "perftools.profiles.Mapping"
+    repeated :location, :message, 4, "perftools.profiles.Location"
+    repeated :function, :message, 5, "perftools.profiles.Function"
+    repeated :string_table, :string, 6
+    optional :drop_frames, :int64, 7
+    optional :keep_frames, :int64, 8
+    optional :time_nanos, :int64, 9
+    optional :duration_nanos, :int64, 10
+    optional :period_type, :message, 11, "perftools.profiles.ValueType"
+    optional :period, :int64, 12
+    repeated :comment, :int64, 13
+    optional :default_sample_type, :int64, 14
+  end
+  add_message "perftools.profiles.ValueType" do
+    optional :type, :int64, 1
+    optional :unit, :int64, 2
+  end
+  add_message "perftools.profiles.Sample" do
+    repeated :location_id, :uint64, 1
+    repeated :value, :int64, 2
+    repeated :label, :message, 3, "perftools.profiles.Label"
+  end
+  add_message "perftools.profiles.Label" do
+    optional :key, :int64, 1
+    optional :str, :int64, 2
+    optional :num, :int64, 3
+    optional :num_unit, :int64, 4
+  end
+  add_message "perftools.profiles.Mapping" do
+    optional :id, :uint64, 1
+    optional :memory_start, :uint64, 2
+    optional :memory_limit, :uint64, 3
+    optional :file_offset, :uint64, 4
+    optional :filename, :int64, 5
+    optional :build_id, :int64, 6
+    optional :has_functions, :bool, 7
+    optional :has_filenames, :bool, 8
+    optional :has_line_numbers, :bool, 9
+    optional :has_inline_frames, :bool, 10
+  end
+  add_message "perftools.profiles.Location" do
+    optional :id, :uint64, 1
+    optional :mapping_id, :uint64, 2
+    optional :address, :uint64, 3
+    repeated :line, :message, 4, "perftools.profiles.Line"
+    optional :is_folded, :bool, 5
+  end
+  add_message "perftools.profiles.Line" do
+    optional :function_id, :uint64, 1
+    optional :line, :int64, 2
+  end
+  add_message "perftools.profiles.Function" do
+    optional :id, :uint64, 1
+    optional :name, :int64, 2
+    optional :system_name, :int64, 3
+    optional :filename, :int64, 4
+    optional :start_line, :int64, 5
+  end
+end
+
+module Perftools
+  module Profiles
+    Profile = Google::Protobuf::DescriptorPool.generated_pool.lookup("perftools.profiles.Profile").msgclass
+    ValueType = Google::Protobuf::DescriptorPool.generated_pool.lookup("perftools.profiles.ValueType").msgclass
+    Sample = Google::Protobuf::DescriptorPool.generated_pool.lookup("perftools.profiles.Sample").msgclass
+    Label = Google::Protobuf::DescriptorPool.generated_pool.lookup("perftools.profiles.Label").msgclass
+    Mapping = Google::Protobuf::DescriptorPool.generated_pool.lookup("perftools.profiles.Mapping").msgclass
+    Location = Google::Protobuf::DescriptorPool.generated_pool.lookup("perftools.profiles.Location").msgclass
+    Line = Google::Protobuf::DescriptorPool.generated_pool.lookup("perftools.profiles.Line").msgclass
+    Function = Google::Protobuf::DescriptorPool.generated_pool.lookup("perftools.profiles.Function").msgclass
+  end
+end