x86/kprobes: Fix crash when probing CS CALL instructionsgrafted

When using eBPF to probe CS CALL instructions within a function,
a crash can be triggered.

The eBPF tool probes offset 257 of the __hrtimer_run_queues()
function:

<__hrtimer_run_queues+249>:  nopl   0x0(%rax,%rax,1)
<__hrtimer_run_queues+254>:  mov    %r14,%rdi
<__hrtimer_run_queues+257>:  cs call <__x86_indirect_thunk_r12>
<__hrtimer_run_queues+263>:  mov    %eax,%r12d
<__hrtimer_run_queues+266>:  xchg   %ax,%ax
<__hrtimer_run_queues+268>:  mov    %r13,%rdi

Which triggers this crash:

  BUG: unable to handle page fault for address: 00000000000f41c9
  #PF: supervisor write access in kernel mode
  #PF: error_code(0x0002) - not-present page
  PGD 0 P4D 0
  Oops: 0002 [#1] SMP NOPTI
  CPU: 1 PID: 0 Comm: swapper/1 Kdump: loaded Tainted: P
  RIP: 0010:__hrtimer_run_queues+0x106/0x230

Note that __hrtimer_run_queues+0x106 is __hrtimer_run_queues+262, which is
at the 6th byte of the above CS CALL instruction. Since the CS CALL
instruction occupies 6 bytes, the exception occurred in the middle of that
call instruction.

The root cause is that when using eBPF tools to probe in the middle of a
function, a kprobe with INT3 is used as the underlying implementation.

During single-step emulation of the original CALL instruction,
int3_emulate_call() assumes that the probed CALL instruction is 5 bytes
long. However, the actual CS-prefixed CALL instruction occupies 6 bytes,
so it constructs an incorrect exception return address. When the CPU
returns from the kprobe handler, the next instruction to be executed is at
the address of the last byte of that CS CALL instruction. Coincidentally,
starting from that address, the CPU fetches and decodes a completely
different instruction, which ultimately triggers a kernel crash.

Fix the issue by using the actual instruction length obtained from
the instruction decoder when constructing the exception return
address, rather than relying on the hardcoded CALL_INSN_SIZE macro.

[ mingo: Refined the changelog ]

Fixes: 6256e668b7af ("x86/kprobes: Use int3 instead of debug trap for single-step")
Suggested-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Jinke Han <jinkehan@didiglobal.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Acked-by: Yafang Shao <laoar.shao@gmail.com>
Acked-by: Borislav Petkov <bp@alien8.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://patch.msgid.link/20260908073742.GA10517@didi-ThinkCentre-M920t-N000

3 files changed, 581 insertions, 0 deletions

diff --git a/scripts/crypto/gen-aead-testvecs.py b/scripts/crypto/gen-aead-testvecs.py
new file mode 100755
index 000000000..d77d646f7
--- /dev/null
+++ b/scripts/crypto/gen-aead-testvecs.py
@@ -0,0 +1,69 @@
+#!/usr/bin/env python3
+# SPDX-License-Identifier: GPL-2.0-or-later
+#
+# Script that generates known-good data used in the AEAD tests.
+#
+# Requires that python-cryptography be installed.
+#
+# Copyright 2026 Google LLC
+
+import hashlib
+import sys
+import cryptography.hazmat.primitives.ciphers.aead
+
+
+# Deterministically generate 'length' random bytes.
+def rand_bytes(length):
+    seed = length
+    out = []
+    for _ in range(length):
+        seed = (seed * 25214903917 + 11) % 2**48
+        out.append((seed >> 16) % 256)
+    return bytes(out)
+
+
+# Deterministically generate many different AEAD inputs using exactly the same
+# method that the test uses; encrypt them using an independent implementation of
+# the algorithm; compute the checksum of all the resulting (ciphertext, authtag)
+# pairs concatenated to each other; and print the checksum as a C struct.
+def gen_monte_carlo_checksum(alg):
+    blake2s = hashlib.blake2s()
+    for data_len in range(1025):
+        ad_len = data_len % 293
+        pt = rand_bytes(data_len)
+        ad = rand_bytes(ad_len)
+        if alg == "aes-ccm":
+            key_len = [16, 24, 32][data_len % 3]
+            key = rand_bytes(key_len)
+            nonce = rand_bytes([7, 8, 9, 10, 11, 12, 13][data_len % 7])
+            tag_len = [4, 6, 8, 10, 12, 14, 16][data_len % 7]
+            ccm = cryptography.hazmat.primitives.ciphers.aead.AESCCM(
+                key, tag_length=tag_len
+            )
+            ct_and_tag = ccm.encrypt(nonce, pt, ad)
+        elif alg == "aes-gcm":
+            key_len = [16, 24, 32][data_len % 3]
+            key = rand_bytes(key_len)
+            nonce = rand_bytes(12)
+            tag_len = [4, 8, 12, 13, 14, 15, 16][data_len % 7]
+            gcm = cryptography.hazmat.primitives.ciphers.aead.AESGCM(key)
+            # python-cryptography supports only 16-byte GCM tags.  However, in
+            # GCM, shorter tags are simply truncated.  Do that below.
+            ct_and_tag = gcm.encrypt(nonce, pt, ad)[: data_len + tag_len]
+
+        blake2s.update(ct_and_tag)
+
+    name = f"{alg.replace('-', '_')}_monte_carlo_checksum"
+    value = blake2s.digest()
+    print(f"static const u8 {name}[BLAKE2S_HASH_SIZE] = {{")
+    for i in range(0, len(value), 11):
+        line = "\t" + "".join(f"0x{b:02x}, " for b in value[i : i + 11])
+        print(f"{line.rstrip()}")
+    print("};")
+
+
+if len(sys.argv) != 2 or sys.argv[1] not in ("aes-ccm", "aes-gcm"):
+    sys.stderr.write("Usage: gen-aead-testvecs.py [aes-ccm|aes-gcm]\n")
+    sys.exit(1)
+
+gen_monte_carlo_checksum(sys.argv[1])
diff --git a/scripts/crypto/gen-fips-testvecs.py b/scripts/crypto/gen-fips-testvecs.py
new file mode 100755
index 000000000..b8c8a78cb
--- /dev/null
+++ b/scripts/crypto/gen-fips-testvecs.py
@@ -0,0 +1,148 @@
+#!/usr/bin/env python3
+# SPDX-License-Identifier: GPL-2.0-or-later
+#
+# Script that generates lib/crypto/fips-aes.h and lib/crypto/fips-sha.h
+#
+# Requires that python-cryptography be installed.
+#
+# Copyright 2025 Google LLC
+
+import cryptography.hazmat.primitives.ciphers
+import cryptography.hazmat.primitives.ciphers.aead
+import cryptography.hazmat.primitives.cmac
+import hashlib
+import hmac
+
+
+def print_static_u8_array_definition(file, name, value):
+    print("", file=file)
+    print(f"static const u8 {name}[] __initconst __maybe_unused = {{", file=file)
+    for i in range(0, len(value), 8):
+        line = "\t" + "".join(f"0x{b:02x}, " for b in value[i : i + 8])
+        print(f"{line.rstrip()}", file=file)
+    print("};", file=file)
+
+
+def print_header(file):
+    print("/* SPDX-License-Identifier: GPL-2.0-or-later */", file=file)
+    print("/* This file was generated by: gen-fips-testvecs.py */", file=file)
+    print("/* clang-format off */", file=file)
+    print("", file=file)
+    print("#include <linux/fips.h>", file=file)
+
+
+def gen_aes_test_data(file):
+    fips_test_data = b"fips test data\0\0"
+    fips_test_ad = b"fips test ad\0\0\0\0"
+    fips_test_iv = b"fips test iv\0\0\0\0"
+    fips_test_key = b"fips test key\0\0\0"
+    fips_test_xts_key = b"key1" + (b"\0" * 12) + b"key2" + (b"\0" * 12)
+
+    print_header(file)
+    print_static_u8_array_definition(file, "fips_test_data", fips_test_data)
+    print_static_u8_array_definition(file, "fips_test_ad", fips_test_ad)
+    print_static_u8_array_definition(file, "fips_test_iv", fips_test_iv)
+    print_static_u8_array_definition(file, "fips_test_key", fips_test_key)
+    print_static_u8_array_definition(file, "fips_test_xts_key", fips_test_xts_key)
+
+    aes = cryptography.hazmat.primitives.ciphers.algorithms.AES(fips_test_key)
+
+    # AES-CMAC
+    aes_cmac = cryptography.hazmat.primitives.cmac.CMAC(aes)
+    aes_cmac.update(fips_test_data)
+    print_static_u8_array_definition(
+        file, "fips_test_aes_cmac_value", aes_cmac.finalize()
+    )
+
+    # AES-ECB
+    cipher = cryptography.hazmat.primitives.ciphers.Cipher(
+        aes, cryptography.hazmat.primitives.ciphers.modes.ECB()
+    )
+    encryptor = cipher.encryptor()
+    ctext = encryptor.update(fips_test_data) + encryptor.finalize()
+    print_static_u8_array_definition(file, "fips_test_aes_ecb_ctext", ctext)
+
+    # AES-CBC
+    cipher = cryptography.hazmat.primitives.ciphers.Cipher(
+        aes, cryptography.hazmat.primitives.ciphers.modes.CBC(fips_test_iv)
+    )
+    encryptor = cipher.encryptor()
+    ctext = encryptor.update(fips_test_data) + encryptor.finalize()
+    print_static_u8_array_definition(file, "fips_test_aes_cbc_ctext", ctext)
+
+    # AES-CBC-CTS
+    cipher = cryptography.hazmat.primitives.ciphers.Cipher(
+        aes, cryptography.hazmat.primitives.ciphers.modes.CBC(fips_test_iv)
+    )
+    encryptor = cipher.encryptor()
+    ctext = encryptor.update(fips_test_data * 2) + encryptor.finalize()
+    ctext = ctext[16:32] + ctext[0:16]
+    print_static_u8_array_definition(file, "fips_test_aes_cbc_cts_ctext", ctext)
+
+    # AES-CTR
+    cipher = cryptography.hazmat.primitives.ciphers.Cipher(
+        aes, cryptography.hazmat.primitives.ciphers.modes.CTR(fips_test_iv)
+    )
+    encryptor = cipher.encryptor()
+    ctext = encryptor.update(fips_test_data) + encryptor.finalize()
+    print_static_u8_array_definition(file, "fips_test_aes_ctr_ctext", ctext)
+
+    # AES-XTS
+    cipher = cryptography.hazmat.primitives.ciphers.Cipher(
+        cryptography.hazmat.primitives.ciphers.algorithms.AES(fips_test_xts_key),
+        cryptography.hazmat.primitives.ciphers.modes.XTS(fips_test_iv),
+    )
+    encryptor = cipher.encryptor()
+    ctext = encryptor.update(fips_test_data) + encryptor.finalize()
+    print_static_u8_array_definition(file, "fips_test_aes_xts_ctext", ctext)
+
+    # AES-GCM
+    cipher = cryptography.hazmat.primitives.ciphers.aead.AESGCM(fips_test_key)
+    ct_and_tag = cipher.encrypt(
+        nonce=fips_test_iv[:12], data=fips_test_data, associated_data=fips_test_ad
+    )
+    print_static_u8_array_definition(
+        file, "fips_test_aes_gcm_ctext_and_tag", ct_and_tag
+    )
+
+    # AES-CCM
+    cipher = cryptography.hazmat.primitives.ciphers.aead.AESCCM(
+        fips_test_key, tag_length=16
+    )
+    ct_and_tag = cipher.encrypt(
+        nonce=fips_test_iv[:13], data=fips_test_data, associated_data=fips_test_ad
+    )
+    print_static_u8_array_definition(
+        file, "fips_test_aes_ccm_ctext_and_tag", ct_and_tag
+    )
+
+
+def gen_sha_test_data(file):
+    fips_test_data = b"fips test data\0\0"
+    fips_test_key = b"fips test key\0\0\0"
+
+    print_header(file)
+    print_static_u8_array_definition(file, "fips_test_data", fips_test_data)
+    print_static_u8_array_definition(file, "fips_test_key", fips_test_key)
+
+    for alg in "sha1", "sha256", "sha512":
+        ctx = hmac.new(fips_test_key, digestmod=alg)
+        ctx.update(fips_test_data)
+        print_static_u8_array_definition(
+            file, f"fips_test_hmac_{alg}_value", ctx.digest()
+        )
+
+    print_static_u8_array_definition(
+        file, "fips_test_sha3_256_value", hashlib.sha3_256(fips_test_data).digest()
+    )
+
+
+filename = "lib/crypto/fips-aes.h"
+with open(filename, "w") as file:
+    print(f"Generating {filename}")
+    gen_aes_test_data(file)
+
+filename = "lib/crypto/fips-sha.h"
+with open(filename, "w") as file:
+    print(f"Generating {filename}")
+    gen_sha_test_data(file)
diff --git a/scripts/crypto/gen-hash-testvecs.py b/scripts/crypto/gen-hash-testvecs.py
new file mode 100755
index 000000000..f356f87e1
--- /dev/null
+++ b/scripts/crypto/gen-hash-testvecs.py
@@ -0,0 +1,364 @@
+#!/usr/bin/env python3
+# SPDX-License-Identifier: GPL-2.0-or-later
+#
+# Script that generates test vectors for the given hash function.
+#
+# Requires that python-cryptography be installed.
+#
+# Copyright 2025 Google LLC
+
+import cryptography.hazmat.primitives.ciphers
+import cryptography.hazmat.primitives.cmac
+import hashlib
+import hmac
+import sys
+
+DATA_LENS = [0, 1, 2, 3, 16, 32, 48, 49, 63, 64, 65, 127, 128, 129, 256, 511,
+             513, 1000, 3333, 4096, 4128, 4160, 4224, 16384]
+
+# Generate the given number of random bytes, using the length itself as the seed
+# for a simple linear congruential generator (LCG).  The C test code uses the
+# same LCG with the same seeding strategy to reconstruct the data, ensuring
+# reproducibility without explicitly storing the data in the test vectors.
+def rand_bytes(length):
+    seed = length
+    out = []
+    for _ in range(length):
+        seed = (seed * 25214903917 + 11) % 2**48
+        out.append((seed >> 16) % 256)
+    return bytes(out)
+
+AES_256_KEY_SIZE = 32
+
+# AES-CMAC.  Just wraps the implementation from python-cryptography.
+class AesCmac:
+    def __init__(self, key):
+        aes = cryptography.hazmat.primitives.ciphers.algorithms.AES(key)
+        self.cmac = cryptography.hazmat.primitives.cmac.CMAC(aes)
+
+    def update(self, data):
+        self.cmac.update(data)
+
+    def digest(self):
+        return self.cmac.finalize()
+
+POLY1305_KEY_SIZE = 32
+
+# A straightforward, unoptimized implementation of Poly1305.
+# Reference: https://cr.yp.to/mac/poly1305-20050329.pdf
+class Poly1305:
+    def __init__(self, key):
+        assert len(key) == POLY1305_KEY_SIZE
+        self.h = 0
+        rclamp = 0x0ffffffc0ffffffc0ffffffc0fffffff
+        self.r = int.from_bytes(key[:16], byteorder='little') & rclamp
+        self.s = int.from_bytes(key[16:], byteorder='little')
+
+    # Note: this supports partial blocks only at the end.
+    def update(self, data):
+        for i in range(0, len(data), 16):
+            chunk = data[i:i+16]
+            c = int.from_bytes(chunk, byteorder='little') + 2**(8 * len(chunk))
+            self.h = ((self.h + c) * self.r) % (2**130 - 5)
+        return self
+
+    # Note: gen_additional_poly1305_testvecs() relies on this being
+    # nondestructive, i.e. not changing any field of self.
+    def digest(self):
+        m = (self.h + self.s) % 2**128
+        return m.to_bytes(16, byteorder='little')
+
+GHASH_POLY = sum((1 << i) for i in [128, 7, 2, 1, 0])
+GHASH_BLOCK_SIZE = 16
+
+# A straightforward, unoptimized implementation of GHASH.
+class Ghash:
+
+    @staticmethod
+    def reflect_bits_in_bytes(v):
+        res = 0
+        for offs in range(0, 128, 8):
+            for bit in range(8):
+                if (v & (1 << (offs + bit))) != 0:
+                    res ^= 1 << (offs + 7 - bit)
+        return res
+
+    @staticmethod
+    def bytes_to_poly(data):
+        return Ghash.reflect_bits_in_bytes(int.from_bytes(data, byteorder='little'))
+
+    @staticmethod
+    def poly_to_bytes(poly):
+        return Ghash.reflect_bits_in_bytes(poly).to_bytes(16, byteorder='little')
+
+    def __init__(self, key):
+        assert len(key) == 16
+        self.h = Ghash.bytes_to_poly(key)
+        self.acc = 0
+
+    # Note: this supports partial blocks only at the end.
+    def update(self, data):
+        for i in range(0, len(data), 16):
+            # acc += block
+            self.acc ^= Ghash.bytes_to_poly(data[i:i+16])
+            # acc = (acc * h) mod GHASH_POLY
+            product = 0
+            for j in range(127, -1, -1):
+                if (self.h & (1 << j)) != 0:
+                    product ^= self.acc << j
+                if (product & (1 << (128 + j))) != 0:
+                    product ^= GHASH_POLY << j
+            self.acc = product
+        return self
+
+    def digest(self):
+        return Ghash.poly_to_bytes(self.acc)
+
+POLYVAL_POLY = sum((1 << i) for i in [128, 127, 126, 121, 0])
+POLYVAL_BLOCK_SIZE = 16
+
+# A straightforward, unoptimized implementation of POLYVAL.
+# Reference: https://datatracker.ietf.org/doc/html/rfc8452
+class Polyval:
+    def __init__(self, key):
+        assert len(key) == 16
+        self.h = int.from_bytes(key, byteorder='little')
+        self.acc = 0
+
+    # Note: this supports partial blocks only at the end.
+    def update(self, data):
+        for i in range(0, len(data), 16):
+            # acc += block
+            self.acc ^= int.from_bytes(data[i:i+16], byteorder='little')
+            # acc = (acc * h * x^-128) mod POLYVAL_POLY
+            product = 0
+            for j in range(128):
+                if (self.h & (1 << j)) != 0:
+                    product ^= self.acc << j
+                if (product & (1 << j)) != 0:
+                    product ^= POLYVAL_POLY << j
+            self.acc = product >> 128
+        return self
+
+    def digest(self):
+        return self.acc.to_bytes(16, byteorder='little')
+
+def hash_init(alg):
+    # The keyed hash functions are assigned a fixed random key here, to present
+    # them as unkeyed hash functions.  This allows all the test cases for
+    # unkeyed hash functions to work on them.
+    if alg == 'aes-cmac':
+        return AesCmac(rand_bytes(AES_256_KEY_SIZE))
+    if alg == 'ghash':
+        return Ghash(rand_bytes(GHASH_BLOCK_SIZE))
+    if alg == 'poly1305':
+        return Poly1305(rand_bytes(POLY1305_KEY_SIZE))
+    if alg == 'polyval':
+        return Polyval(rand_bytes(POLYVAL_BLOCK_SIZE))
+    return hashlib.new(alg)
+
+def hash_update(ctx, data):
+    ctx.update(data)
+
+def hash_final(ctx):
+    return ctx.digest()
+
+def compute_hash(alg, data):
+    ctx = hash_init(alg)
+    hash_update(ctx, data)
+    return hash_final(ctx)
+
+def print_bytes(prefix, value, bytes_per_line):
+    for i in range(0, len(value), bytes_per_line):
+        line = prefix + ''.join(f'0x{b:02x}, ' for b in value[i:i+bytes_per_line])
+        print(f'{line.rstrip()}')
+
+def print_static_u8_array_definition(name, value):
+    print('')
+    print(f'static const u8 {name} = {{')
+    print_bytes('\t', value, 8)
+    print('};')
+
+def print_c_struct_u8_array_field(name, value):
+    print(f'\t\t.{name} = {{')
+    print_bytes('\t\t\t', value, 8)
+    print('\t\t},')
+
+def alg_digest_size_const(alg):
+    if alg == 'aes-cmac':
+        return 'AES_BLOCK_SIZE'
+    if alg.startswith('blake2'):
+        return f'{alg.upper()}_HASH_SIZE'
+    return f"{alg.upper().replace('-', '_')}_DIGEST_SIZE"
+
+def gen_unkeyed_testvecs(alg):
+    print('')
+    print('static const struct {')
+    print('\tsize_t data_len;')
+    print(f'\tu8 digest[{alg_digest_size_const(alg)}];')
+    print('} hash_testvecs[] = {')
+    for data_len in DATA_LENS:
+        data = rand_bytes(data_len)
+        print('\t{')
+        print(f'\t\t.data_len = {data_len},')
+        print_c_struct_u8_array_field('digest', compute_hash(alg, data))
+        print('\t},')
+    print('};')
+
+    data = rand_bytes(4096)
+    ctx = hash_init(alg)
+    for data_len in range(len(data) + 1):
+        hash_update(ctx, compute_hash(alg, data[:data_len]))
+    print_static_u8_array_definition(
+            f'hash_testvec_consolidated[{alg_digest_size_const(alg)}]',
+            hash_final(ctx))
+
+def gen_additional_sha3_testvecs():
+    max_len = 4096
+    in_data = rand_bytes(max_len)
+    for alg in ['shake128', 'shake256']:
+        ctx = hashlib.new('sha3-256')
+        for in_len in range(max_len + 1):
+            out_len = (in_len * 293) % (max_len + 1)
+            out = hashlib.new(alg, data=in_data[:in_len]).digest(out_len)
+            ctx.update(out)
+        print_static_u8_array_definition(f'{alg}_testvec_consolidated[SHA3_256_DIGEST_SIZE]',
+                                         ctx.digest())
+
+def gen_hmac_testvecs(alg):
+    ctx = hmac.new(rand_bytes(32), digestmod=alg)
+    data = rand_bytes(4096)
+    for data_len in range(len(data) + 1):
+        ctx.update(data[:data_len])
+        key_len = data_len % 293
+        key = rand_bytes(key_len)
+        mac = hmac.digest(key, data[:data_len], alg)
+        ctx.update(mac)
+    print_static_u8_array_definition(
+            f'hmac_testvec_consolidated[{alg.upper()}_DIGEST_SIZE]',
+            ctx.digest())
+
+def gen_additional_blake2_testvecs(alg):
+    if alg == 'blake2s':
+        (max_key_size, max_hash_size) = (32, 32)
+    elif alg == 'blake2b':
+        (max_key_size, max_hash_size) = (64, 64)
+    else:
+        raise ValueError(f'Unsupported alg: {alg}')
+    hashes = b''
+    for key_len in range(max_key_size + 1):
+        for out_len in range(1, max_hash_size + 1):
+            h = hashlib.new(alg, digest_size=out_len, key=rand_bytes(key_len))
+            h.update(rand_bytes(100))
+            hashes += h.digest()
+    print_static_u8_array_definition(
+            f'{alg}_keyed_testvec_consolidated[{alg_digest_size_const(alg)}]',
+            compute_hash(alg, hashes))
+
+def nh_extract_int(bytestr, pos, length):
+    assert pos % 8 == 0 and length % 8 == 0
+    return int.from_bytes(bytestr[pos//8 : pos//8 + length//8], byteorder='little')
+
+# The NH "almost-universal hash function" used in Adiantum.  This is a
+# straightforward translation of the pseudocode from Section 6.3 of the Adiantum
+# paper (https://eprint.iacr.org/2018/720.pdf), except the outer loop is omitted
+# because we assume len(msg) <= 1024.  (The kernel's nh() function is only
+# expected to handle up to 1024 bytes; it's just called repeatedly as needed.)
+def nh(key, msg):
+    (w, s, r, u) = (32, 2, 4, 8192)
+    l = 8 * len(msg)
+    assert l <= u
+    assert l % (2*s*w) == 0
+    h = bytes()
+    for i in range(0, 2*s*w*r, 2*s*w):
+        p = 0
+        for j in range(0, l, 2*s*w):
+            for k in range(0, w*s, w):
+                a0 = nh_extract_int(key, i + j + k, w)
+                a1 = nh_extract_int(key, i + j + k + s*w, w)
+                b0 = nh_extract_int(msg, j + k, w)
+                b1 = nh_extract_int(msg, j + k + s*w, w)
+                p += ((a0 + b0) % 2**w) * ((a1 + b1) % 2**w)
+        h += (p % 2**64).to_bytes(8, byteorder='little')
+    return h
+
+def gen_nh_testvecs():
+    NH_KEY_BYTES = 1072
+    NH_MESSAGE_BYTES = 1024
+    key = rand_bytes(NH_KEY_BYTES)
+    msg = rand_bytes(NH_MESSAGE_BYTES)
+    print_static_u8_array_definition('nh_test_key[NH_KEY_BYTES]', key)
+    print_static_u8_array_definition('nh_test_msg[NH_MESSAGE_BYTES]', msg)
+    for length in [16, 96, 256, 1024]:
+        print_static_u8_array_definition(f'nh_test_val{length}[NH_HASH_BYTES]',
+                                         nh(key, msg[:length]))
+
+def gen_additional_poly1305_testvecs():
+    key = b'\xff' * POLY1305_KEY_SIZE
+    data = b''
+    ctx = Poly1305(key)
+    for _ in range(32):
+        for j in range(0, 4097, 16):
+            ctx.update(b'\xff' * j)
+            data += ctx.digest()
+    print_static_u8_array_definition(
+            'poly1305_allones_macofmacs[POLY1305_DIGEST_SIZE]',
+            Poly1305(key).update(data).digest())
+
+def gen_additional_ghash_testvecs():
+    key = b'\xff' * GHASH_BLOCK_SIZE
+    hashes = b''
+    for data_len in range(0, 4097, 16):
+        hashes += Ghash(key).update(b'\xff' * data_len).digest()
+    print_static_u8_array_definition(
+            'ghash_allones_hashofhashes[GHASH_DIGEST_SIZE]',
+            Ghash(key).update(hashes).digest())
+
+def gen_additional_polyval_testvecs():
+    key = b'\xff' * POLYVAL_BLOCK_SIZE
+    hashes = b''
+    for data_len in range(0, 4097, 16):
+        hashes += Polyval(key).update(b'\xff' * data_len).digest()
+    print_static_u8_array_definition(
+            'polyval_allones_hashofhashes[POLYVAL_DIGEST_SIZE]',
+            Polyval(key).update(hashes).digest())
+
+if len(sys.argv) != 2:
+    sys.stderr.write('Usage: gen-hash-testvecs.py ALGORITHM\n')
+    sys.stderr.write('ALGORITHM may be any supported by Python hashlib;\n')
+    sys.stderr.write('  or aes-cmac, ghash, nh, poly1305, polyval, or sha3.\n')
+    sys.stderr.write('Example: gen-hash-testvecs.py sha512\n')
+    sys.exit(1)
+
+alg = sys.argv[1]
+print('/* SPDX-License-Identifier: GPL-2.0-or-later */')
+print(f'/* This file was generated by: {sys.argv[0]} {" ".join(sys.argv[1:])} */')
+if alg == 'aes-cmac':
+    gen_unkeyed_testvecs(alg)
+elif alg.startswith('blake2'):
+    gen_unkeyed_testvecs(alg)
+    gen_additional_blake2_testvecs(alg)
+elif alg == 'ghash':
+    gen_unkeyed_testvecs(alg)
+    gen_additional_ghash_testvecs()
+elif alg == 'nh':
+    gen_nh_testvecs()
+elif alg == 'poly1305':
+    gen_unkeyed_testvecs(alg)
+    gen_additional_poly1305_testvecs()
+elif alg == 'polyval':
+    gen_unkeyed_testvecs(alg)
+    gen_additional_polyval_testvecs()
+elif alg == 'sha3':
+    print()
+    print('/* SHA3-256 test vectors */')
+    gen_unkeyed_testvecs('sha3-256')
+    print()
+    print('/* SHAKE test vectors */')
+    gen_additional_sha3_testvecs()
+elif alg == 'sm3':
+    gen_unkeyed_testvecs(alg)
+    # Kernel doesn't implement HMAC-SM3 library functions yet.
+else:
+    gen_unkeyed_testvecs(alg)
+    gen_hmac_testvecs(alg)