Change sym_check to filter non-stdlib symbols.

Currently sym_check almost all names found in the binary, including those
which are defined in other libraries. This makes our ABI lists harder to maintain.

This patch adds a --only-stdlib-symbols option to sym_check which removes
all symbols which aren't possibly provided by libc++. It also re-generates
the linux ABI list after making this change.


git-svn-id: https://llvm.org/svn/llvm-project/libcxx/trunk@287294 91177308-0d34-0410-b5e6-96231b3b80d8

4 files changed, 59 insertions, 2 deletions

diff --git a/utils/sym_check/sym_check/extract.py b/utils/sym_check/sym_check/extract.py
index 7bafd8ec4..d132e2282 100644
--- a/utils/sym_check/sym_check/extract.py
+++ b/utils/sym_check/sym_check/extract.py
@@ -12,9 +12,11 @@ extract - A set of function that extract symbol lists from shared libraries.
 """
 import distutils.spawn
 import sys
+import re
 
 from sym_check import util
 
+extract_ignore_names = ['_init', '_fini']
 
 class NMExtractor(object):
     """
@@ -65,7 +67,8 @@ class NMExtractor(object):
             return None
         new_sym = {
             'name': bits[0],
-            'type': bits[1]
+            'type': bits[1],
+            'is_defined': (bits[1].lower() != 'u')
         }
         new_sym['name'] = new_sym['name'].replace('@@', '@')
         new_sym = self._transform_sym_type(new_sym)
@@ -81,6 +84,8 @@ class NMExtractor(object):
         """
         if sym is None or len(sym) < 2:
             return False
+        if sym['name'] in extract_ignore_names:
+            return False
         bad_types = ['t', 'b', 'r', 'd', 'w']
         return (sym['type'] not in bad_types
                 and sym['name'] not in ['__bss_start', '_end', '_edata'])
@@ -148,8 +153,11 @@ class ReadElfExtractor(object):
                 'name': parts[7],
                 'size': int(parts[2]),
                 'type': parts[3],
+                'is_defined': (parts[6] != 'UND')
             }
             assert new_sym['type'] in ['OBJECT', 'FUNC', 'NOTYPE']
+            if new_sym['name'] in extract_ignore_names:
+                continue
             if new_sym['type'] == 'NOTYPE':
                 continue
             if new_sym['type'] == 'FUNC':
diff --git a/utils/sym_check/sym_check/util.py b/utils/sym_check/sym_check/util.py
index fd1b0b1e1..84f8d523c 100644
--- a/utils/sym_check/sym_check/util.py
+++ b/utils/sym_check/sym_check/util.py
@@ -12,7 +12,7 @@ import distutils.spawn
 import signal
 import subprocess
 import sys
-
+import re
 
 def execute_command(cmd, input_str=None):
     """
@@ -135,3 +135,40 @@ def extract_or_load(filename):
     if is_library_file(filename):
         return sym_check.extract.extract_symbols(filename)
     return read_syms_from_file(filename)
+
+def adjust_mangled_name(name):
+    if not name.startswith('__Z'):
+        return name
+    return name[1:]
+
+new_delete_std_symbols = [
+    '_Znam',
+    '_Znwm',
+    '_ZdaPv',
+    '_ZdaPvm',
+    '_ZdlPv',
+    '_ZdlPvm'
+]
+
+def is_stdlib_symbol_name(name):
+    name = adjust_mangled_name(name)
+    if name in new_delete_std_symbols:
+        return True
+    if re.search("@GLIBC|@GCC", name):
+        return False
+    if re.search('(St[0-9])|(__cxa)|(__cxxabi)', name):
+        return True
+    return True
+
+def filter_stdlib_symbols(syms):
+    stdlib_symbols = []
+    other_symbols = []
+    for s in syms:
+        canon_name = adjust_mangled_name(s['name'])
+        if not is_stdlib_symbol_name(canon_name):
+            assert not s['is_defined'] and \
+                   'have non-stdlib symbol defined in stdlib'
+            other_symbols += [s]
+        else:
+            stdlib_symbols += [s]
+    return stdlib_symbols, other_symbols
diff --git a/utils/sym_check/sym_diff.py b/utils/sym_check/sym_diff.py
index 0d88a2b87..842e908dd 100755
--- a/utils/sym_check/sym_diff.py
+++ b/utils/sym_check/sym_diff.py
@@ -27,6 +27,9 @@ def main():
         '--removed-only', dest='removed_only',
         help='Only print removed symbols',
         action='store_true', default=False)
+    parser.add_argument('--only-stdlib-symbols', dest='only_stdlib',
+                        help="Filter all symbols not related to the stdlib",
+                        action='store_true', default=False)
     parser.add_argument(
         '-o', '--output', dest='output',
         help='The output file. stdout is used if not given',
@@ -44,6 +47,10 @@ def main():
     old_syms_list = util.extract_or_load(args.old_syms)
     new_syms_list = util.extract_or_load(args.new_syms)
 
+    if args.only_stdlib:
+        old_syms_list, _ = util.filter_stdlib_symbols(old_syms_list)
+        new_syms_list, _ = util.filter_stdlib_symbols(new_syms_list)
+
     added, removed, changed = diff.diff(old_syms_list, new_syms_list)
     if args.removed_only:
         added = {}
diff --git a/utils/sym_check/sym_extract.py b/utils/sym_check/sym_extract.py
index a0fbb3e63..27765679e 100755
--- a/utils/sym_check/sym_extract.py
+++ b/utils/sym_check/sym_extract.py
@@ -25,11 +25,16 @@ def main():
     parser.add_argument('--names-only', dest='names_only',
                         help='Output only the name of the symbol',
                         action='store_true', default=False)
+    parser.add_argument('--only-stdlib-symbols', dest='only_stdlib',
+                        help="Filter all symbols not related to the stdlib",
+                        action='store_true', default=False)
     args = parser.parse_args()
     if args.output is not None:
         print('Extracting symbols from %s to %s.'
               % (args.library, args.output))
     syms = extract.extract_symbols(args.library)
+    if args.only_stdlib:
+        syms, other_syms = util.filter_stdlib_symbols(syms)
     util.write_syms(syms, out=args.output, names_only=args.names_only)