1 files changed, 89 insertions, 0 deletions
diff --git a/scripts/libc_early_init_name.py b/scripts/libc_early_init_name.py
new file mode 100644
index 0000000000..a56c2008f3
--- /dev/null
+++ b/scripts/libc_early_init_name.py
@@ -0,0 +1,89 @@
+#!/usr/bin/python3
+# Compute the hash-based name of the __libc_early_init function.
+# Copyright (C) 2022 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
+# <https://www.gnu.org/licenses/>.
+
+"""Compute the name of the __libc_early_init function, which is used
+as a protocol version marker between ld.so and libc.so.
+
+The name contains a hash suffix, and the hash changes if certain key
+files in the source tree change.  Distributions can also configure
+with --with-extra-version-id, to make the computed hash dependent on
+the package version.
+
+"""
+
+import argparse
+import hashlib
+import os
+import string
+import sys
+
+def gnu_hash(s):
+    """Computes the GNU hash of the string."""
+    h = 5381
+    for ch in s:
+        if type(ch) is not int:
+            ch = ord(ch)
+        h = (h * 33 + ch) & 0xffffffff
+    return h
+
+# Parse the command line.
+parser = argparse.ArgumentParser(description=__doc__)
+parser.add_argument('--output', metavar='PATH',
+                    help='path to header file this tool generates')
+parser.add_argument('--extra-version-id', metavar='ID',
+                    help='extra string to influence hash computation')
+parser.add_argument('inputs', metavar='PATH', nargs='*',
+                    help='files whose contents influences the generated hash')
+opts = parser.parse_args()
+
+# Obtain the blobs that affect the generated hash.
+blobs = [(opts.extra_version_id or '').encode('UTF-8')]
+for path in opts.inputs:
+    with open(path, 'rb') as inp:
+        blobs.append(inp.read())
+
+# Hash the file boundaries.
+md = hashlib.sha256()
+md.update(repr([len(blob) for blob in blobs]).encode('UTF-8'))
+
+# And then hash the file contents.  Do not hash the paths, to avoid
+# impacting reproducibility.
+for blob in blobs:
+    md.update(blob)
+
+# These are the bits used to compute the suffix.
+derived_bits = int.from_bytes(md.digest(), byteorder='big', signed=False)
+
+# These digits are used in the suffix (should result in base-62 encoding).
+# They must be valid in C identifiers.
+digits = string.digits + string.ascii_letters
+
+# Generate eight digits as a suffix.  They should provide enough
+# uniqueness (47.6 bits).
+name = '__libc_early_init_'
+for n in range(8):
+    name += digits[derived_bits % len(digits)]
+    derived_bits //= len(digits)
+
+# Write the output file.
+with open(opts.output, 'w') if opts.output else sys.stdout as out:
+    out.write('#define LIBC_EARLY_INIT_NAME {}\n'.format(name))
+    out.write('#define LIBC_EARLY_INIT_NAME_STRING "{}"\n'.format(name))
+    out.write('#define LIBC_EARLY_INIT_GNU_HASH {}\n'.format(
+        gnu_hash(name)))