Load target build files in parallel using Python multiprocessing.

pylib/gyp/input.py

Index: pylib/gyp/input.py
diff --git a/pylib/gyp/input.py b/pylib/gyp/input.py
index 2678bab2c4233f79cfb6fc71b34401e42871d055..c33e55437e41e9728cf7d9fa14bee05cf3389c7a 100644
--- a/pylib/gyp/input.py
+++ b/pylib/gyp/input.py
@@ -12,12 +12,14 @@ from compiler.ast import Stmt
 import compiler
 import copy
 import gyp.common
+import multiprocessing
 import optparse
 import os.path
 import re
 import shlex
 import subprocess
 import sys
+import time
 
 
 # A list of types that are treated as linkable.
@@ -329,7 +331,7 @@ def ProcessToolsetsInDict(data):
 # a build file that contains targets and is expected to provide a targets dict
 # that contains the targets...
 def LoadTargetBuildFile(build_file_path, data, aux_data, variables, includes,
-                        depth, check):
+                        depth, check, load_dependencies):
   # If depth is set, predefine the DEPTH variable to be a relative path from
   # this build file's directory to the directory identified by depth.
   if depth:
@@ -348,7 +350,7 @@ def LoadTargetBuildFile(build_file_path, data, aux_data, variables, includes,
 
   if build_file_path in data['target_build_files']:
     # Already loaded.
-    return
+    return False
   data['target_build_files'].add(build_file_path)
 
   gyp.DebugOutput(gyp.DEBUG_INCLUDES,
@@ -418,22 +420,118 @@ def LoadTargetBuildFile(build_file_path, data, aux_data, variables, includes,
   # in other words, you can't put a "dependencies" section inside a "post"
   # conditional within a target.
 
+  dependencies = []
   if 'targets' in build_file_data:
     for target_dict in build_file_data['targets']:
       if 'dependencies' not in target_dict:
         continue
       for dependency in target_dict['dependencies']:
-        other_build_file = \
-            gyp.common.ResolveTarget(build_file_path, dependency, None)[0]
-        try:
-          LoadTargetBuildFile(other_build_file, data, aux_data, variables,
-                              includes, depth, check)
-        except Exception, e:
-          gyp.common.ExceptionAppend(
-            e, 'while loading dependencies of %s' % build_file_path)
-          raise
+        dependencies.append(
+            gyp.common.ResolveTarget(build_file_path, dependency, None)[0])
 
-  return data
+  if load_dependencies:
+    for dependency in dependencies:
+      try:
+        LoadTargetBuildFile(dependency, data, aux_data, variables,
+                            includes, depth, check, load_dependencies)
+      except Exception, e:
+        gyp.common.ExceptionAppend(
+          e, 'while loading dependencies of %s' % build_file_path)
+        raise
+  else:
+    return (build_file_path, dependencies)
+
+
+# Wrapper around LoadTargetBuildFile used when LoadTargetBuildFile
+# is executed in a worker process.
+def CallLoadTargetBuildFile(global_flags,
+                            build_file_path, data,
+                            aux_data, variables,
+                            includes, depth, check):
+  # Apply globals so that the worker process behaves the same.
+  for key, value in global_flags.iteritems():
+    globals()[key] = value
+
+  # Save the keys so we can return data that changed.
+  data_keys = set(data)
+  aux_data_keys = set(aux_data)
+
+  result = LoadTargetBuildFile(build_file_path, data,
+                               aux_data, variables,
+                               includes, depth, check, False)
+  if not result:
+    return result
+
+  (build_file_path, dependencies) = result
+
+  data_out = {}
+  for key in data:
+    if key == 'target_build_files':
+      continue
+    if key not in data_keys:
+      data_out[key] = data[key]
+  aux_data_out = {}
+  for key in aux_data:
+    if key not in aux_data_keys:
+      aux_data_out[key] = aux_data[key]
+
+  # This gets serialized and sent back to the main process via a pipe.
+  # It's handled in LoadTargetBuildFileCallback.
+  return (build_file_path,
+          data_out,
+          aux_data_out,
+          dependencies)
+
+
+# Handle the results of LoadTargetBuildFile that executed in a separate
+# process.
+def LoadTargetBuildFileCallback(result):
+  (build_file_path0, data0, aux_data0, dependencies0) = result
+  gyp.data['target_build_files'].add(build_file_path0)

[M-A Ruel, 2012/09/18 19:42:18]

Can you tell me where the "gyp" variable is defined?

[dmazzoni, 2012/09/19 21:14:52]

gyp is the module object.

I cleaned this up by creating a single global called parallel_state and
documenting all of its members.

+  for key in data0:
+    gyp.data[key] = data0[key]
+  for key in aux_data0:
+    gyp.aux_data[key] = aux_data0[key]
+  for new_dependency in dependencies0:
+    if new_dependency not in gyp.scheduled:
+      gyp.scheduled.add(new_dependency)
+      gyp.dependencies.append(new_dependency)
+  gyp.pending -= 1
+
+
+def LoadTargetBuildFileParallel(build_file_path, data, aux_data,
+                                variables, includes, depth, check):
+  gyp.dependencies = [build_file_path]

[M-A Ruel, 2012/09/18 19:42:18]

I find the code a bit hard to read, the "gyp" variable looks magical to me, but
I'm not super familiar with the gyp code.

+  gyp.scheduled = set([build_file_path])
+  gyp.pending = 0
+  gyp.data = data
+  gyp.aux_data = aux_data
+  while gyp.dependencies or gyp.pending:
+    if not gyp.dependencies:
+      time.sleep(0.003)
+      continue
+
+    dependency = gyp.dependencies.pop()
+
+    gyp.pending += 1
+    data_in = {}
+    data_in['target_build_files'] = data['target_build_files']
+    aux_data_in = {}
+    global_flags = {
+      'path_sections': globals()['path_sections'],
+      'non_configuration_keys': globals()['non_configuration_keys'],
+      'absolute_build_file_paths': globals()['absolute_build_file_paths'],
+      'multiple_toolsets': globals()['multiple_toolsets']}
+
+    if 'pool' not in dir(gyp):
+      gyp.pool = multiprocessing.Pool(8)
+    gyp.pool.apply_async(
+        CallLoadTargetBuildFile,
+        args = (global_flags, dependency,
+                data_in, aux_data_in,
+                variables, includes, depth, check),
+        callback = LoadTargetBuildFileCallback)
+    time.sleep(0.003)
 
 
 # Look for the bracket that matches the first bracket seen in a
@@ -2333,7 +2431,7 @@ def VerifyNoCollidingTargets(targets):
 
 
 def Load(build_files, variables, includes, depth, generator_input_info, check,
-         circular_check):
+         circular_check, parallel):
   # Set up path_sections and non_configuration_keys with the default data plus
   # the generator-specifc data.
   global path_sections
@@ -2374,8 +2472,13 @@ def Load(build_files, variables, includes, depth, generator_input_info, check,
     # used as keys to the data dict and for references between input files.
     build_file = os.path.normpath(build_file)
     try:
-      LoadTargetBuildFile(build_file, data, aux_data, variables, includes,
-                          depth, check)
+      if parallel:
+        print >>sys.stderr, 'Using parallel processing (experimental).'
+        LoadTargetBuildFileParallel(build_file, data, aux_data,
+                                    variables, includes, depth, check)
+      else:
+        LoadTargetBuildFile(build_file, data, aux_data,
+                            variables, includes, depth, check, True)
     except Exception, e:
       gyp.common.ExceptionAppend(e, 'while trying to load %s' % build_file)
       raise