Commit graph

13686 commits

Author SHA1 Message Date
Manuel Pégourié-Gonnard
6d6f8a4b97 Clarify descriptions of constant-flow helpers
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
2020-09-28 09:51:41 +02:00
Manuel Pégourié-Gonnard
822b3729e7 Remove last use of non-bit operations
According to https://www.bearssl.org/ctmul.html even single-precision
multiplication is not constant-time on some older platforms.

An added benefit of the new code is that it removes the somewhat mysterious
constant 0x1ff - which was selected because at that point the maximum value of
padlen was 256. The new code is perhaps a bit more readable for that reason.

Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
2020-09-18 12:11:22 +02:00
Manuel Pégourié-Gonnard
2a59fb45b5 Add explicit cast when truncating values
MSVC complains about it otherwise.

Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
2020-09-18 12:11:22 +02:00
Manuel Pégourié-Gonnard
6e2a9a7faa Factor repeated code in ssl_cf functions
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
2020-09-18 12:11:21 +02:00
Manuel Pégourié-Gonnard
2ddec4306f Use bit operations for constant-flow padding check
The previous code used comparison operators >= and == that are quite likely to
be compiled to branches by some compilers on some architectures (with some
optimisation levels).

For example, take the following function:

void old_update( size_t data_len, size_t *padlen )
{
    *padlen  *= ( data_len >= *padlen + 1 );
}

With Clang 3.8, let's compile it for the Arm v6-M architecture:

% clang --target=arm-none-eabi -march=armv6-m -Os foo.c -S -o - |
    sed -n '/^old_update:$/,/\.size/p'

old_update:
        .fnstart
@ BB#0:
        .save   {r4, lr}
        push    {r4, lr}
        ldr     r2, [r1]
        adds    r4, r2, #1
        movs    r3, #0
        cmp     r4, r0
        bls     .LBB0_2
@ BB#1:
        mov     r2, r3
.LBB0_2:
        str     r2, [r1]
        pop     {r4, pc}
.Lfunc_end0:
        .size   old_update, .Lfunc_end0-old_update

We can see an unbalanced secret-dependant branch, resulting in a total
execution time depends on the value of the secret (here padlen) in a
straightforward way.

The new version, based on bit operations, doesn't have this issue:

new_update:
        .fnstart
@ BB#0:
        ldr     r2, [r1]
        subs    r0, r0, #1
        subs    r0, r0, r2
        asrs    r0, r0, #31
        bics    r2, r0
        str     r2, [r1]
        bx      lr
.Lfunc_end1:
        .size   new_update, .Lfunc_end1-new_update

(As a bonus, it's smaller and uses less stack.)

While there's no formal guarantee that the version based on bit operations in
C won't be translated using branches by the compiler, experiments tend to show
that's the case [1], and it is commonly accepted knowledge in the practical
crypto community that if we want to sick to C, bit operations are the safest
bet [2].

[1] https://github.com/mpg/ct/blob/master/results
[2] https://github.com/veorq/cryptocoding

Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
2020-09-18 12:10:33 +02:00
Janos Follath
523f0554b6
Merge pull request #748 from ARMmbed/mbedtls-2.24.0r0-pr
Prepare Release Candidate for Mbed TLS 2.24.0
2020-08-27 11:31:49 +01:00
Janos Follath
6012f0ee5b Finalize ChangeLog
Fix alignment where necessary and update ChangeLog header.

Signed-off-by: Janos Follath <janos.follath@arm.com>
2020-08-26 16:23:19 +01:00
Janos Follath
17ffc5da8d Bump version to Mbed TLS 2.24.0
Executed "./scripts/bump_version.sh --version 2.24.0"

Signed-off-by: Janos Follath <janos.follath@arm.com>
2020-08-26 16:22:57 +01:00
Janos Follath
c18a7b8466 Assemble ChangeLog
Executed scripts/assemble_changelog.py.

Signed-off-by: Janos Follath <janos.follath@arm.com>
2020-08-26 14:49:16 +01:00
Janos Follath
d2ce916b58 Merge branch 'development-restricted' 2020-08-26 14:15:34 +01:00
Gilles Peskine
d4b9133850
Merge pull request #3611 from gilles-peskine-arm/psa-coverity-cleanups-202008
Minor fixes in PSA code and tests
2020-08-26 13:18:27 +02:00
Gilles Peskine
9e4d4387f0
Merge pull request #3433 from raoulstrackx/raoul/verify_crl_without_time
Always revoke certificate on CRL
2020-08-26 12:56:11 +02:00
Manuel Pégourié-Gonnard
2db7be1cbb
Merge pull request #3612 from gilles-peskine-arm/psa-mac-negative-tests
PSA: add negative MAC tests
2020-08-26 12:19:25 +02:00
Gilles Peskine
a2e518daf5 Fix the documentation of has_even_parity
The documentation had the boolean meaning of the return value inverted.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-08-26 12:14:37 +02:00
Manuel Pégourié-Gonnard
376712217e
Merge pull request #737 from mpg/changelog-for-local-lucky13-dev-restricted
Add a ChangeLog entry for local Lucky13 variant
2020-08-26 11:52:15 +02:00
Gilles Peskine
ed9fbc6443 Clearer function name for parity check
Return a name that more clearly returns nonzero=true=good, 0=bad. We'd
normally expect check_xxx to return 0=pass, nonzero=fail so
check_parity was a bad name.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-08-26 11:16:50 +02:00
Gilles Peskine
6c75152b9f Explain the purpose of check_parity
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-08-26 10:24:26 +02:00
Gilles Peskine
34f063ca47 Add missing cleanup to hash multipart operation tests
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-08-26 10:24:13 +02:00
Manuel Pégourié-Gonnard
8f18d08fae Clarify that the Lucky 13 fix is quite general
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
2020-08-26 10:10:11 +02:00
Gilles Peskine
29c4a6cf9f Add negative tests for MAC verification
Add negative tests for psa_mac_verify_finish: too large, too small, or
a changed byte.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-08-26 00:16:03 +02:00
Gilles Peskine
090e16cb8b Don't destroy the key during a MAC verification operation
An early draft of the PSA crypto specification required multipart
operations to keep working after destroying the key. This is no longer
the case: instead, now, operations are guaranteed to fail. Mbed TLS
does not comply yet, and still allows the operation to keep going.
Stop testing Mbed TLS's non-compliant behavior.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-08-26 00:16:03 +02:00
Gilles Peskine
8b356b5652 Test other output sizes for psa_mac_sign_finish
Test psa_mac_sign_finish with a smaller or larger buffer.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-08-26 00:16:03 +02:00
Gilles Peskine
5e65cec5e8 Simplify output bounds check in mac_sign test
Rely on Asan to detect a potential buffer overflow, instead of doing a
manual check. This makes the code simpler and Asan can detect
underflows as well as overflows.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-08-26 00:16:03 +02:00
Gilles Peskine
3d404d677e Test PSA_MAC_FINAL_SIZE in mac_sign exactly
We expect PSA_MAC_FINAL_SIZE to be exact in this implementation, so
check it here.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-08-26 00:16:03 +02:00
Gilles Peskine
cd65f4ccac Add empty-output-buffer test cases for single-part hash functions
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-08-26 00:11:23 +02:00
Gilles Peskine
e92c68a878 Note that a failure in cleanup is intentional
In the cleanup code for persistent_key_load_key_from_storage(), we
only attempt to reopen the key so that it will be deleted if it exists
at that point. It's intentional that we do nothing if psa_open_key()
fails here.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-08-26 00:11:23 +02:00
Gilles Peskine
64f13ef6ab Add missing cleanup to some multipart operation tests
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-08-26 00:11:23 +02:00
Gilles Peskine
a09713c795 test cleanup: Annotate file removal after a failed creation
Let static analyzers know that it's ok if remove() fails here.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-08-25 22:50:18 +02:00
Gilles Peskine
169ca7f06d psa_crypto_storage: Annotate file removal after a failed creation
Let static analyzers know that it's ok if psa_its_remove() fails here.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-08-25 22:50:06 +02:00
Gilles Peskine
bab1b52048 psa_its: Annotate file removal after a failed creation
Let static analyzers know that it's ok if remove() fails here.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-08-25 22:49:19 +02:00
Gilles Peskine
14613bcd75 Fix parity tests to actually fail the test on error
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-08-25 22:30:31 +02:00
Janos Follath
d4ac4e037b
Merge pull request #736 from mpg/cf-varpos-copy-dev-restricted
Constant-flow copy of HMAC from variable position
2020-08-25 14:35:55 +01:00
Manuel Pégourié-Gonnard
04b7488411 Fix potential use of uninitialised variable
If any of the TEST_ASSERT()s that are before the call to
mbedtls_pk_warp_as_opaque() failed, when reaching the exit label
psa_destroy_key() would be called with an uninitialized argument.

Found by Clang.

Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
2020-08-25 10:45:51 +02:00
Gilles Peskine
ed19762a22
Merge pull request #3574 from makise-homura/e2k_support
Support building on e2k (Elbrus) architecture
2020-08-25 09:46:36 +02:00
makise-homura
af9513bb48 A different approach of signed-to-unsigned comparison
Suggsted by @hanno-arm

Signed-off-by: makise-homura <akemi_homura@kurisa.ch>
2020-08-24 23:42:49 +03:00
Manuel Pégourié-Gonnard
ba6fc9796a Fix a typo in a comment
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
2020-08-24 12:59:55 +02:00
Manuel Pégourié-Gonnard
dd00bfce34 Improve comments on constant-flow testing in config.h
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
2020-08-24 12:58:36 +02:00
Gilles Peskine
0f343ac87f
Merge pull request #3528 from gufe44/helpers-redirect-restore-output
Fix bug in redirection of unit test outputs
2020-08-24 10:45:08 +02:00
Manuel Pégourié-Gonnard
8a79b9b68c Fix "unused function" warning in some configs
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
2020-08-24 10:29:30 +02:00
Manuel Pégourié-Gonnard
6edfe60e0d
Merge pull request #2182 from hanno-arm/key_pwd
Add support for password protected key files to ssl_server2 and ssl_client2
2020-08-24 09:42:38 +02:00
makise-homura
e014fece50 Don't forget to free G, P, Q, ctr_drbg, and entropy
I might be wrong, but lcc's optimizer is curious about this,
and I am too: shouldn't we free allocated stuff correctly
before exiting `dh_genprime` in this certain point of code?

Signed-off-by: makise-homura <akemi_homura@kurisa.ch>
2020-08-22 23:56:46 +03:00
Gilles Peskine
5fe5b823d4
Merge pull request #3590 from mpg/fix-compat.sh-with-ubuntu-16.04-gnutls
Fix compat.sh with ubuntu 16.04 gnutls
2020-08-21 14:18:25 +02:00
Manuel Pégourié-Gonnard
6c77bc6de2 compat.sh: stop using allow_sha1
After the changes of certificates, it's no longer needed.

Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
2020-08-21 12:34:05 +02:00
Manuel Pégourié-Gonnard
499bf4c0c8 compat.sh: quit using SHA-1 certificates
Replace server2.crt with server2-sha256.crt which, as the name implies, is
just the SHA-256 version of the same certificate.

Replace server1.crt with cert_sha256.crt which, as the name doesn't imply, is
associated with the same key and just have a slightly different Subject Name,
which doesn't matter in this instance.

The other certificates used in this script (server5.crt and server6.crt) are
already signed with SHA-256.

This change is motivated by the fact that recent versions of GnuTLS (or older
versions with the Debian patches) reject SHA-1 in certificates by default, as
they should. There are options to still accept it (%VERIFY_ALLOW_BROKEN and
%VERIFY_ALLOW_SIGN_WITH_SHA1) but:

- they're not available in all versions that reject SHA-1-signed certs;
- moving to SHA-2 just seems cleaner anyway.

Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
2020-08-21 12:25:30 +02:00
Manuel Pégourié-Gonnard
244d06637f compat.sh: enable CBC-SHA-2 suites for GnuTLS
Recent GnuTLS packages on Ubuntu 16.04 have them disabled.

From /usr/share/doc/libgnutls30/changelog.Debian.gz:

gnutls28 (3.4.10-4ubuntu1.5) xenial-security; urgency=medium

  * SECURITY UPDATE: Lucky-13 issues
    [...]
    - debian/patches/CVE-2018-1084x-4.patch: hmac-sha384 and sha256
      ciphersuites were removed from defaults in lib/gnutls_priority.c,
      tests/priorities.c.

Since we do want to test the ciphersuites, explicitly re-enable them in the
server's priority string. (This is a no-op with versions of GnuTLS where those
are already enabled by default.)

Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
2020-08-21 12:06:47 +02:00
Manuel Pégourié-Gonnard
1a6af8489e
Merge pull request #3578 from gilles-peskine-arm/md_setup-leak-development
Fix memory leak in mbedtls_md_setup with HMAC
2020-08-21 09:19:12 +02:00
gufe44
067f6e01f1 Fix bug in redirection of unit test outputs
Avoid replacing handle. stdout is defined as a macro on several platforms.

Signed-off-by: gufe44 <gu981@protonmail.com>
2020-08-21 08:32:26 +02:00
Gilles Peskine
22d70a80af
Merge pull request #738 from danh-arm/dh/remaining-lf-copyright
Update remaining copyright notices to use Linux Foundation guidance
2020-08-20 13:26:17 +02:00
Dan Handley
50118144c6 Update remaining copyright notices to use Linux Foundation guidance
Update copyright notices to newly added files since merge of original
PR #3546 "Update copyright notices to use Linux Foundation guidance".
Generated using the same script.

Signed-off-by: Dan Handley <dan.handley@arm.com>
2020-08-20 11:20:12 +01:00
Manuel Pégourié-Gonnard
53d216081c Add a ChangeLog entry for local Lucky13 variant
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
2020-08-20 12:17:05 +02:00