Add support for including GNU Gold linker along with --section-orderi…

…ng-file

GNU Gold linker will be auto-detected on Linux systems.
If available the pre-generated section ordering file will be used while
linking the node binary.
This will help improve performance through reduction of iTLB misses
as the most frequently used functions in the runtime will be packed
together.

Add documentation to generate the section reorderding file
Add an option to override this configure option if
LD environment variable is defined.

common.gypi

@@ -42,6 +42,9 @@
     # Don't use ICU data file (icudtl.dat) from V8, we use our own.
     'icu_use_data_file_flag%': 0,
 
+    # use exported value from the environment 
+    'gold_section_ordering_file%': '',
+
     'conditions': [
       ['GENERATOR=="ninja"', {
         'OBJ_DIR': '<(PRODUCT_DIR)/obj',
@@ -431,6 +434,14 @@
         'ldflags': [
           '-Wl,--export-dynamic',
         ],
+      }],
+      ['gold_section_ordering_file!=""', {
+        'cflags': [ '-fuse-ld=gold',
+                    '-ffunction-sections'],
+        'ldflags': [
+          '-fuse-ld=gold',
+          '-Wl,--section-ordering-file=<(gold_section_ordering_file)',
+        ],
       }]
     ],
   }

configure

@@ -38,6 +38,9 @@ import string
 # If not run from node/, cd to node/.
 os.chdir(os.path.dirname(__file__) or '.')
 
+# Set by configure_node(). 
+cross_compiling = False
+
 # gcc and g++ as defaults matches what GYP's Makefile generator does,
 # except on OS X.
 CC = os.environ.get('CC', 'cc' if sys.platform == 'darwin' else 'gcc')
@@ -150,6 +153,12 @@ parser.add_option("--enable-vtune-profiling",
          "JavaScript code executed in nodejs. This feature is only available "
          "for x32, x86, and x64 architectures.")
 
+parser.add_option("--enable-section-ordering",
+   action="store",
+   dest="section_ordering",
+   help="Use GNU Gold linker capability to reorder sections in a binary."
+        "This field will be passed to Gold linker's --section-ordering-file"
+        "Provide a section ordering file to override the default")
 
 parser.add_option("--link-module",
     action="append",
@@ -706,7 +715,6 @@ def check_compiler(o):
   else:
     o['variables']['gas_version'] = get_gas_version(CC)
 
-
 def cc_macros(cc=None):
   """Checks predefined macros using the C compiler command."""
 
@@ -865,6 +873,7 @@ def configure_node(o):
   o['variables']['target_arch'] = target_arch
   o['variables']['node_byteorder'] = sys.byteorder
 
+  global cross_compiling
   cross_compiling = (options.cross_compiling
                      if options.cross_compiling is not None
                      else target_arch != host_arch)
@@ -1099,6 +1108,25 @@ def configure_static(o):
         o['libraries'] += ['-static-libasan']
 
 
+def configure_gold(o):
+  try:
+    proc = subprocess.Popen(['ld.gold'] + ['-v'], stdin=subprocess.PIPE, 
+                            stderr=subprocess.PIPE, stdout=subprocess.PIPE)
+  except OSError:
+    return 0
+  match = re.match(r"(GNU gold) .* ([0-9]\.[0-9]+)",
+                   proc.communicate()[0])
+  if match:
+    gold_version = match.group(2)
+    if gold_version > '1.1':
+      if options.section_ordering:
+        o['variables']['gold_section_ordering_file'] = options.section_ordering
+      else:
+        o['variables']['gold_section_ordering_file'] = \
+        os.path.realpath('tools/gold_linker_section_reordering.txt') 
+  else:
+    return 0
+
 def write(filename, data):
   filename = filename
   print('creating %s' % filename)
@@ -1425,6 +1453,7 @@ output = {
   'libraries': [],
   'defines': [],
   'cflags': [],
+  'ldflags': [],
 }
 
 # Print a warning when the compiler is too old.
@@ -1450,6 +1479,8 @@ configure_v8(output)
 configure_openssl(output)
 configure_intl(output)
 configure_static(output)
+if not cross_compiling and not os.environ.get('LD'):
+  configure_gold(output)
 configure_inspector(output)
 
 # variables should be a root level element,

tools/config_gold_section_ordering.md

@@ -0,0 +1,32 @@
+Once the workload, production server has reached a steady state/ post warm-up phase
+
+- [] Collect the perf profile for the workload:
+```
+     perf record -ag -e cycles:pp  -o <perf_profile_cycles_pp> --  sleep 30 
+```
+`:pp` offers a more precise dist of cycles accounting
+
+- [] Use perf script command to decode the collected profile
+```
+      perf script -i perf_profile_inst -f comm,ip -c node | gzip -c > perf_profile_decoded.pds.gz
+```
+- [] Use nm to dump the binary's symbol information
+```
+        nm -S node > node.nm
+```
+- [] Run the hfsort program to determine the function ordering 
+     for the node binary ideal for this workload.
+     https://github.com/facebook/hhvm/tree/master/hphp/tools/hfsort
+```
+          Usage: hfsort [-p] [-e <EDGCNT_FILE>] <SYMBOL_FILE> <PERF_DATA_FILE>
+
+           -p,   use pettis-hansen algorithm for code layout  (optional)
+
+           -e    <EDGCNT_FILE>, use edge profile result to build the call graph
+
+        hfsort -p node.nm perf_profile_decoded.pds.gz
+```
+This application will create 2 files hotfuncs.txt and result-hfsort.txt
+
+hotfuncs.txt is what is used here. hfsort is one way to generate it.
+hfsort generates the minimal working order set required to optimally run the program.