Convert linknamespace tests from Perl to Python.

This patch replaces conform/linknamespace.pl with a new
conform/linknamespace.py, so continuing the consolidation on Python
instead of Perl for miscellaneous scripts used in building and testing
glibc.  The new script follows the same logic as the old one; as a
recently-added script, there were no major cleanups to be made in the
course of the language conversion.

Tested for x86_64, and with build-many-glibcs.py.  For x86_64 I also
tested that if the Perl and Python scripts were made to print all the
symbols in seen_where and the paths of symbol references by which
those symbols were linked in, even when those symbols were OK,
identical symbol lists appeared in the output with both versions of
the script (the differences in linknamespace.out files were only in
paths to temporary files in diagnostics for e.g. deprecated functions,
and error output for the expected compilation failures when testing
ndbm.h and varargs.h).

	* conform/linknamespace.py: New file.
	* conform/linknamespace.pl: Remove file.
	* conform/Makefile ($(linknamespace-header-tests)): Use
	linknamespace.py instead of linknamespace.pl.  Do not use --tmpdir
	option.

1 files changed, 217 insertions, 0 deletions

diff --git a/conform/linknamespace.py b/conform/linknamespace.py
new file mode 100644
index 0000000000..07a775499d
--- /dev/null
+++ b/conform/linknamespace.py
@@ -0,0 +1,217 @@
+#!/usr/bin/python
+# Check that use of symbols declared in a given header does not result
+# in any symbols being brought in that are not reserved with external
+# linkage for the given standard.
+# Copyright (C) 2014-2018 Free Software Foundation, Inc.
+# This file is part of the GNU C Library.
+#
+# The GNU C Library is free software; you can redistribute it and/or
+# modify it under the terms of the GNU Lesser General Public
+# License as published by the Free Software Foundation; either
+# version 2.1 of the License, or (at your option) any later version.
+#
+# The GNU C Library is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+# Lesser General Public License for more details.
+#
+# You should have received a copy of the GNU Lesser General Public
+# License along with the GNU C Library; if not, see
+# <http://www.gnu.org/licenses/>.
+
+import argparse
+from collections import defaultdict
+import os.path
+import re
+import subprocess
+import sys
+import tempfile
+
+import glibcconform
+
+# The following whitelisted symbols are also allowed for now.
+#
+# * Bug 17576: stdin, stdout, stderr only reserved with external
+# linkage when stdio.h included (and possibly not then), not
+# generally.
+#
+# * Bug 18442: re_syntax_options wrongly brought in by regcomp and
+# used by re_comp.
+#
+WHITELIST = {'stdin', 'stdout', 'stderr', 're_syntax_options'}
+
+
+def list_syms(filename):
+    """Return information about GLOBAL and WEAK symbols listed in readelf
+    -s output."""
+    ret = []
+    cur_file = filename
+    with open(filename, 'r') as syms_file:
+        for line in syms_file:
+            line = line.rstrip()
+            if line.startswith('File: '):
+                cur_file = line[len('File: '):]
+                cur_file = cur_file.split('/')[-1]
+                continue
+            # Architecture-specific st_other bits appear inside [] and
+            # disrupt the format of readelf output.
+            line = re.sub(r'\[.*?\]', '', line)
+            fields = line.split()
+            if len(fields) < 8:
+                continue
+            bind = fields[4]
+            ndx = fields[6]
+            sym = fields[7]
+            if bind not in ('GLOBAL', 'WEAK'):
+                continue
+            if not re.fullmatch('[A-Za-z0-9_]+', sym):
+                continue
+            ret.append((cur_file, sym, bind, ndx != 'UND'))
+    return ret
+
+
+def main():
+    """The main entry point."""
+    parser = argparse.ArgumentParser(description='Check link-time namespace.')
+    parser.add_argument('--header', metavar='HEADER',
+                        help='name of header')
+    parser.add_argument('--standard', metavar='STD',
+                        help='standard to use when processing header')
+    parser.add_argument('--cc', metavar='CC',
+                        help='C compiler to use')
+    parser.add_argument('--flags', metavar='CFLAGS',
+                        help='Compiler flags to use with CC')
+    parser.add_argument('--stdsyms', metavar='FILE',
+                        help='File with list of standard symbols')
+    parser.add_argument('--libsyms', metavar='FILE',
+                        help='File with symbol information from libraries')
+    parser.add_argument('--readelf', metavar='READELF',
+                        help='readelf program to use')
+    args = parser.parse_args()
+
+    # Load the list of symbols that are OK.
+    stdsyms = set()
+    with open(args.stdsyms, 'r') as stdsyms_file:
+        for line in stdsyms_file:
+            stdsyms.add(line.rstrip())
+    stdsyms |= WHITELIST
+
+    # Load information about GLOBAL and WEAK symbols defined or used
+    # in the standard libraries.
+    # Symbols from a given object, except for weak defined symbols.
+    seen_syms = defaultdict(list)
+    # Strong undefined symbols from a given object.
+    strong_undef_syms = defaultdict(list)
+    # Objects defining a given symbol (strongly or weakly).
+    sym_objs = defaultdict(list)
+    for file, name, bind, defined in list_syms(args.libsyms):
+        if defined:
+            sym_objs[name].append(file)
+        if bind == 'GLOBAL' or not defined:
+            seen_syms[file].append(name)
+        if bind == 'GLOBAL' and not defined:
+            strong_undef_syms[file].append(name)
+
+    # Determine what ELF-level symbols are brought in by use of C-level
+    # symbols declared in the given header.
+    #
+    # The rules followed are heuristic and so may produce false
+    # positives and false negatives.
+    #
+    # * All undefined symbols are considered of signficance, but it is
+    # possible that (a) any standard library definition is weak, so
+    # can be overridden by the user's definition, and (b) the symbol
+    # is only used conditionally and not if the program is limited to
+    # standard functionality.
+    #
+    # * If a symbol reference is only brought in by the user using a
+    # data symbol rather than a function from the standard library,
+    # this will not be detected.
+    #
+    # * If a symbol reference is only brought in by crt*.o or libgcc,
+    # this will not be detected.
+    #
+    # * If a symbol reference is only brought in through __builtin_foo
+    # in a standard macro being compiled to call foo, this will not be
+    # detected.
+    #
+    # * Header inclusions should be compiled several times with
+    # different options such as -O2, -D_FORTIFY_SOURCE and
+    # -D_FILE_OFFSET_BITS=64 to find out what symbols are undefined
+    # from such a compilation; this is not yet implemented.
+    #
+    # * This script finds symbols referenced through use of macros on
+    # the basis that if a macro calls an internal function, that
+    # function must also be declared in the header.  However, the
+    # header might also declare implementation-namespace functions
+    # that are not called by any standard macro in the header,
+    # resulting in false positives for any symbols brought in only
+    # through use of those implementation-namespace functions.
+    #
+    # * Namespace issues can apply for dynamic linking as well as
+    # static linking, when a call is from one shared library to
+    # another or uses a PLT entry for a call within a shared library;
+    # such issues are only detected by this script if the same
+    # namespace issue applies for static linking.
+    seen_where = {}
+    files_seen = set()
+    all_undef = {}
+    current_undef = {}
+    compiler = '%s %s' % (args.cc, args.flags)
+    c_syms = glibcconform.list_exported_functions(compiler, args.standard,
+                                                  args.header)
+    with tempfile.TemporaryDirectory() as temp_dir:
+        cincfile_name = os.path.join(temp_dir, 'undef.c')
+        cincfile_o_name = os.path.join(temp_dir, 'undef.o')
+        cincfile_sym_name = os.path.join(temp_dir, 'undef.sym')
+        cincfile_text = ('#include <%s>\n%s\n'
+                         % (args.header,
+                            '\n'.join('void *__glibc_test_%s = (void *) &%s;'
+                                      % (sym, sym) for sym in sorted(c_syms))))
+        with open(cincfile_name, 'w') as cincfile:
+            cincfile.write(cincfile_text)
+        cmd = ('%s %s -D_ISOMAC %s -c %s -o %s'
+               % (args.cc, args.flags, glibcconform.CFLAGS[args.standard],
+                  cincfile_name, cincfile_o_name))
+        subprocess.check_call(cmd, shell=True)
+        cmd = ('LC_ALL=C %s -W -s %s > %s'
+               % (args.readelf, cincfile_o_name, cincfile_sym_name))
+        subprocess.check_call(cmd, shell=True)
+        for file, name, bind, defined in list_syms(cincfile_sym_name):
+            if bind == 'GLOBAL' and not defined:
+                sym_text = '[initial] %s' % name
+                seen_where[name] = sym_text
+                all_undef[name] = sym_text
+                current_undef[name] = sym_text
+
+    while current_undef:
+        new_undef = {}
+        for sym, cu_sym in sorted(current_undef.items()):
+            for file in sym_objs[sym]:
+                if file in files_seen:
+                    continue
+                files_seen.add(file)
+                for ssym in seen_syms[file]:
+                    if ssym not in seen_where:
+                        seen_where[ssym] = ('%s -> [%s] %s'
+                                            % (cu_sym, file, ssym))
+                for usym in strong_undef_syms[file]:
+                    if usym not in all_undef:
+                        usym_text = '%s -> [%s] %s' % (cu_sym, file, usym)
+                        all_undef[usym] = usym_text
+                        new_undef[usym] = usym_text
+        current_undef = new_undef
+
+    ret = 0
+    for sym in sorted(seen_where):
+        if sym.startswith('_'):
+            continue
+        if sym in stdsyms:
+            continue
+        print(seen_where[sym])
+        ret = 1
+    sys.exit(ret)
+
+
+if __name__ == '__main__':
+    main()