As said by Linus:
A symmetric naming is only helpful if it implies symmetries in use.
Otherwise it's actively misleading.
In "kzalloc()", the z is meaningful and an important part of what the
caller wants.
In "kzfree()", the z is actively detrimental, because maybe in the
future we really _might_ want to use that "memfill(0xdeadbeef)" or
something. The "zero" part of the interface isn't even _relevant_.
The main reason that kzfree() exists is to clear sensitive information
that should not be leaked to other future users of the same memory
objects.
Rename kzfree() to kfree_sensitive() to follow the example of the recently
added kvfree_sensitive() and make the intention of the API more explicit.
In addition, memzero_explicit() is used to clear the memory to make sure
that it won't get optimized away by the compiler.
The renaming is done by using the command sequence:
git grep -w --name-only kzfree |\
xargs sed -i 's/kzfree/kfree_sensitive/'
followed by some editing of the kfree_sensitive() kerneldoc and adding
a kzfree backward compatibility macro in slab.h.
[akpm@linux-foundation.org: fs/crypto/inline_crypt.c needs linux/slab.h]
[akpm@linux-foundation.org: fix fs/crypto/inline_crypt.c some more]
Suggested-by: Joe Perches <joe@perches.com>
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: David Howells <dhowells@redhat.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
Cc: James Morris <jmorris@namei.org>
Cc: "Serge E. Hallyn" <serge@hallyn.com>
Cc: Joe Perches <joe@perches.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Dan Carpenter <dan.carpenter@oracle.com>
Cc: "Jason A . Donenfeld" <Jason@zx2c4.com>
Link: http://lkml.kernel.org/r/20200616154311.12314-3-longman@redhat.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull crypto updates from Herbert Xu:
"API:
- Fix out-of-sync IVs in self-test for IPsec AEAD algorithms
Algorithms:
- Use formally verified implementation of x86/curve25519
Drivers:
- Enhance hwrng support in caam
- Use crypto_engine for skcipher/aead/rsa/hash in caam
- Add Xilinx AES driver
- Add uacce driver
- Register zip engine to uacce in hisilicon
- Add support for OCTEON TX CPT engine in marvell"
* 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6: (162 commits)
crypto: af_alg - bool type cosmetics
crypto: arm[64]/poly1305 - add artifact to .gitignore files
crypto: caam - limit single JD RNG output to maximum of 16 bytes
crypto: caam - enable prediction resistance in HRWNG
bus: fsl-mc: add api to retrieve mc version
crypto: caam - invalidate entropy register during RNG initialization
crypto: caam - check if RNG job failed
crypto: caam - simplify RNG implementation
crypto: caam - drop global context pointer and init_done
crypto: caam - use struct hwrng's .init for initialization
crypto: caam - allocate RNG instantiation descriptor with GFP_DMA
crypto: ccree - remove duplicated include from cc_aead.c
crypto: chelsio - remove set but not used variable 'adap'
crypto: marvell - enable OcteonTX cpt options for build
crypto: marvell - add the Virtual Function driver for CPT
crypto: marvell - add support for OCTEON TX CPT engine
crypto: marvell - create common Kconfig and Makefile for Marvell
crypto: arm/neon - memzero_explicit aes-cbc key
crypto: bcm - Use scnprintf() for avoiding potential buffer overflow
crypto: atmel-i2c - Fix wakeup fail
...
Do test_aead_vs_generic_impl() before test_aead_inauthentic_inputs() so
that any differences with the generic driver are detected before getting
to the inauthentic input tests, which intentionally use only the driver
being tested (so that they run even if a generic driver is unavailable).
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
rfc4543 was missing from the list of algorithms that may treat the end
of the AAD buffer specially.
Also, with rfc4106, rfc4309, rfc4543, and rfc7539esp, the end of the AAD
buffer is actually supposed to contain a second copy of the IV, and
we've concluded that if the IV copies don't match the behavior is
implementation-defined. So, the fuzz tests can't easily test that case.
So, make the fuzz tests only use inputs where the two IV copies match.
Reported-by: Geert Uytterhoeven <geert+renesas@glider.be>
Fixes: 40153b10d9 ("crypto: testmgr - fuzz AEADs against their generic implementation")
Cc: Stephan Mueller <smueller@chronox.de>
Originally-from: Gilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
This patch enables the selftests for the s390 specific protected key
AES (PAES) cipher implementations:
* cbc-paes-s390
* ctr-paes-s390
* ecb-paes-s390
* xts-paes-s390
PAES is an AES cipher but with encrypted ('protected') key
material. However, the paes ciphers are able to derive an protected
key from clear key material with the help of the pkey kernel module.
So this patch now enables the generic AES tests for the paes
ciphers. Under the hood the setkey() functions rearrange the clear key
values as clear key token and so the pkey kernel module is able to
provide protected key blobs from the given clear key values. The
derived protected key blobs are then used within the paes cipers and
should produce the very same results as the generic AES implementation
with the clear key values.
The s390-paes cipher testlist entries are surrounded
by #if IS_ENABLED(CONFIG_CRYPTO_PAES_S390) because they don't
make any sense on non s390 platforms or without the PAES
cipher implementation.
Link: http://lkml.kernel.org/r/20200213083946.zicarnnt3wizl5ty@gondor.apana.org.au
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
The whole point of using an AEAD over length-preserving encryption is
that the data is authenticated. However currently the fuzz tests don't
test any inauthentic inputs to verify that the data is actually being
authenticated. And only two algorithms ("rfc4543(gcm(aes))" and
"ccm(aes)") even have any inauthentic test vectors at all.
Therefore, update the AEAD fuzz tests to sometimes generate inauthentic
test vectors, either by generating a (ciphertext, AAD) pair without
using the key, or by mutating an authentic pair that was generated.
To avoid flakiness, only assume this works reliably if the auth tag is
at least 8 bytes. Also account for the rfc4106, rfc4309, and rfc7539esp
algorithms intentionally ignoring the last 8 AAD bytes, and for some
algorithms doing extra checks that result in EINVAL rather than EBADMSG.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
In preparation for adding inauthentic input fuzz tests, which don't
require that a generic implementation of the algorithm be available,
refactor test_aead_vs_generic_impl() so that instead there's a
higher-level function test_aead_extra() which initializes a struct
aead_extra_tests_ctx and then calls test_aead_vs_generic_impl() with a
pointer to that struct.
As a bonus, this reduces stack usage.
Also switch from crypto_aead_alg(tfm)->maxauthsize to
crypto_aead_maxauthsize(), now that the latter is available in
<crypto/aead.h>.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The alignment bug in ghash_setkey() fixed by commit 5c6bc4dfa5
("crypto: ghash - fix unaligned memory access in ghash_setkey()")
wasn't reliably detected by the crypto self-tests on ARM because the
tests only set the keys directly from the test vectors.
To improve test coverage, update the tests to sometimes pass misaligned
keys to setkey(). This applies to shash, ahash, skcipher, and aead.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
When checking two implementations of the same skcipher algorithm for
consistency, require that the minimum key size be the same, not just the
maximum key size. There's no good reason to allow different minimum key
sizes.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Currently if the comparison fuzz tests encounter an encryption error
when generating an skcipher or AEAD test vector, they will still test
the decryption side (passing it the uninitialized ciphertext buffer)
and expect it to fail with the same error.
This is sort of broken because it's not well-defined usage of the API to
pass an uninitialized buffer, and furthermore in the AEAD case it's
acceptable for the decryption error to be EBADMSG (meaning "inauthentic
input") even if the encryption error was something else like EINVAL.
Fix this for skcipher by explicitly initializing the ciphertext buffer
on error, and for AEAD by skipping the decryption test on error.
Reported-by: Pascal Van Leeuwen <pvanleeuwen@verimatrix.com>
Fixes: d435e10e67 ("crypto: testmgr - fuzz skciphers against their generic implementation")
Fixes: 40153b10d9 ("crypto: testmgr - fuzz AEADs against their generic implementation")
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Due to the removal of the blkcipher and ablkcipher algorithm types,
crypto_skcipher::keysize is now redundant since it always equals
crypto_skcipher_alg(tfm)->max_keysize.
Remove it and update crypto_skcipher_default_keysize() accordingly.
Also rename crypto_skcipher_default_keysize() to
crypto_skcipher_max_keysize() to clarify that it specifically returns
the maximum key size, not some unspecified "default".
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
In preparation of introducing KPP implementations of Curve25519, import
the set of test cases proposed by the Zinc patch set, but converted to
the KPP format.
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Test vectors for blake2b with various digest sizes. As the algorithm is
the same up to the digest calculation, the key and input data length is
distributed in a way that tests all combinanions of the two over the
digest sizes.
Based on the suggestion from Eric, the following input sizes are tested
[0, 1, 7, 15, 64, 247, 256], where blake2b blocksize is 128, so the
padded and the non-padded input buffers are tested.
blake2b-160 blake2b-256 blake2b-384 blake2b-512
---------------------------------------------------
len=0 | klen=0 klen=1 klen=32 klen=64
len=1 | klen=32 klen=64 klen=0 klen=1
len=7 | klen=64 klen=0 klen=1 klen=32
len=15 | klen=1 klen=32 klen=64 klen=0
len=64 | klen=0 klen=1 klen=32 klen=64
len=247 | klen=32 klen=64 klen=0 klen=1
len=256 | klen=64 klen=0 klen=1 klen=32
Where key:
- klen=0: empty key
- klen=1: 1 byte value 0x42, 'B'
- klen=32: first 32 bytes of the default key, sequence 00..1f
- klen=64: default key, sequence 00..3f
The unkeyed vectors are ordered before keyed, as this is required by
testmgr.
CC: Eric Biggers <ebiggers@kernel.org>
Signed-off-by: David Sterba <dsterba@suse.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Added testvectors for the rfc3686(ctr(sm4)) skcipher algorithm
changes since v1:
- nothing
Signed-off-by: Pascal van Leeuwen <pvanleeuwen@verimatrix.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Added testvectors for the ofb(sm4) and cfb(sm4) skcipher algorithms
changes since v1:
- nothing
Signed-off-by: Pascal van Leeuwen <pvanleeuwen@verimatrix.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Added testvectors for the hmac(sm3) ahash authentication algorithm
changes since v1 & v2:
-nothing
Signed-off-by: Pascal van Leeuwen <pvanleeuwen@verimatrix.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Add a test vector for the ESSIV mode that is the most widely used,
i.e., using cbc(aes) and sha256, in both skcipher and AEAD modes
(the latter is used by tcrypt to encapsulate the authenc template
or h/w instantiations of the same)
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Three variants of AEGIS were proposed for the CAESAR competition, and
only one was selected for the final portfolio: AEGIS128.
The other variants, AEGIS128L and AEGIS256, are not likely to ever turn
up in networking protocols or other places where interoperability
between Linux and other systems is a concern, nor are they likely to
be subjected to further cryptanalysis. However, uninformed users may
think that AEGIS128L (which is faster) is equally fit for use.
So let's remove them now, before anyone starts using them and we are
forced to support them forever.
Note that there are no known flaws in the algorithms or in any of these
implementations, but they have simply outlived their usefulness.
Reviewed-by: Ondrej Mosnacek <omosnace@redhat.com>
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
MORUS was not selected as a winner in the CAESAR competition, which
is not surprising since it is considered to be cryptographically
broken [0]. (Note that this is not an implementation defect, but a
flaw in the underlying algorithm). Since it is unlikely to be in use
currently, let's remove it before we're stuck with it.
[0] https://eprint.iacr.org/2019/172.pdf
Reviewed-by: Ondrej Mosnacek <omosnace@redhat.com>
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Crypto test failures in FIPS mode cause an immediate panic, but
on some system the cryptographic boundary extends beyond just
the Linux controlled domain.
Add a simple atomic notification chain to allow interested parties
to register to receive notification prior to us kicking the bucket.
Signed-off-by: Gilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Pull crypto updates from Herbert Xu:
"Here is the crypto update for 5.3:
API:
- Test shash interface directly in testmgr
- cra_driver_name is now mandatory
Algorithms:
- Replace arc4 crypto_cipher with library helper
- Implement 5 way interleave for ECB, CBC and CTR on arm64
- Add xxhash
- Add continuous self-test on noise source to drbg
- Update jitter RNG
Drivers:
- Add support for SHA204A random number generator
- Add support for 7211 in iproc-rng200
- Fix fuzz test failures in inside-secure
- Fix fuzz test failures in talitos
- Fix fuzz test failures in qat"
* 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6: (143 commits)
crypto: stm32/hash - remove interruptible condition for dma
crypto: stm32/hash - Fix hmac issue more than 256 bytes
crypto: stm32/crc32 - rename driver file
crypto: amcc - remove memset after dma_alloc_coherent
crypto: ccp - Switch to SPDX license identifiers
crypto: ccp - Validate the the error value used to index error messages
crypto: doc - Fix formatting of new crypto engine content
crypto: doc - Add parameter documentation
crypto: arm64/aes-ce - implement 5 way interleave for ECB, CBC and CTR
crypto: arm64/aes-ce - add 5 way interleave routines
crypto: talitos - drop icv_ool
crypto: talitos - fix hash on SEC1.
crypto: talitos - move struct talitos_edesc into talitos.h
lib/scatterlist: Fix mapping iterator when sg->offset is greater than PAGE_SIZE
crypto/NX: Set receive window credits to max number of CRBs in RxFIFO
crypto: asymmetric_keys - select CRYPTO_HASH where needed
crypto: serpent - mark __serpent_setkey_sbox noinline
crypto: testmgr - dynamically allocate crypto_shash
crypto: testmgr - dynamically allocate testvec_config
crypto: talitos - eliminate unneeded 'done' functions at build time
...
The largest stack object in this file is now the shash descriptor.
Since there are many other stack variables, this can push it
over the 1024 byte warning limit, in particular with clang and
KASAN:
crypto/testmgr.c:1693:12: error: stack frame size of 1312 bytes in function '__alg_test_hash' [-Werror,-Wframe-larger-than=]
Make test_hash_vs_generic_impl() do the same thing as the
corresponding eaed and skcipher functions by allocating the
descriptor dynamically. We can still do better than this,
but it brings us well below the 1024 byte limit.
Suggested-by: Eric Biggers <ebiggers@kernel.org>
Fixes: 9a8a6b3f09 ("crypto: testmgr - fuzz hashes against their generic implementation")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Eric Biggers <ebiggers@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
On arm32, we get warnings about high stack usage in some of the functions:
crypto/testmgr.c:2269:12: error: stack frame size of 1032 bytes in function 'alg_test_aead' [-Werror,-Wframe-larger-than=]
static int alg_test_aead(const struct alg_test_desc *desc, const char *driver,
^
crypto/testmgr.c:1693:12: error: stack frame size of 1312 bytes in function '__alg_test_hash' [-Werror,-Wframe-larger-than=]
static int __alg_test_hash(const struct hash_testvec *vecs,
^
On of the larger objects on the stack here is struct testvec_config, so
change that to dynamic allocation.
Fixes: 40153b10d9 ("crypto: testmgr - fuzz AEADs against their generic implementation")
Fixes: d435e10e67 ("crypto: testmgr - fuzz skciphers against their generic implementation")
Fixes: 9a8a6b3f09 ("crypto: testmgr - fuzz hashes against their generic implementation")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Eric Biggers <ebiggers@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
There are no remaining users of the cipher implementation, and there
are no meaningful ways in which the arc4 cipher can be combined with
templates other than ECB (and the way we do provide that combination
is highly dubious to begin with).
So let's drop the arc4 cipher altogether, and only keep the ecb(arc4)
skcipher, which is used in various places in the kernel.
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Call cond_resched() after each fuzz test iteration. This avoids stall
warnings if fuzz_iterations is set very high for testing purposes.
While we're at it, also call cond_resched() after finishing testing each
test vector.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Acked-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
xxhash is currently implemented as a self-contained module in /lib.
This patch enables that module to be used as part of the generic kernel
crypto framework. It adds a simple wrapper to the 64bit version.
I've also added test vectors (with help from Nick Terrell). The upstream
xxhash code is tested by running hashing operation on random 222 byte
data with seed values of 0 and a prime number. The upstream test
suite can be found at https://github.com/Cyan4973/xxHash/blob/cf46e0c/xxhsum.c#L664
Essentially hashing is run on data of length 0,1,14,222 with the
aforementioned seed values 0 and prime 2654435761. The particular random
222 byte string was provided to me by Nick Terrell by reading
/dev/random and the checksums were calculated by the upstream xxsum
utility with the following bash script:
dd if=/dev/random of=TEST_VECTOR bs=1 count=222
for a in 0 1; do
for l in 0 1 14 222; do
for s in 0 2654435761; do
echo algo $a length $l seed $s;
head -c $l TEST_VECTOR | ~/projects/kernel/xxHash/xxhsum -H$a -s$s
done
done
done
This produces output as follows:
algo 0 length 0 seed 0
02cc5d05 stdin
algo 0 length 0 seed 2654435761
02cc5d05 stdin
algo 0 length 1 seed 0
25201171 stdin
algo 0 length 1 seed 2654435761
25201171 stdin
algo 0 length 14 seed 0
c1d95975 stdin
algo 0 length 14 seed 2654435761
c1d95975 stdin
algo 0 length 222 seed 0
b38662a6 stdin
algo 0 length 222 seed 2654435761
b38662a6 stdin
algo 1 length 0 seed 0
ef46db3751d8e999 stdin
algo 1 length 0 seed 2654435761
ac75fda2929b17ef stdin
algo 1 length 1 seed 0
27c3f04c2881203a stdin
algo 1 length 1 seed 2654435761
4a15ed26415dfe4d stdin
algo 1 length 14 seed 0
3d33dc700231dfad stdin
algo 1 length 14 seed 2654435761
ea5f7ddef9a64f80 stdin
algo 1 length 222 seed 0
5f3d3c08ec2bef34 stdin
algo 1 length 222 seed 2654435761
6a9df59664c7ed62 stdin
algo 1 is xx64 variant, algo 0 is the 32 bit variant which is currently
not hooked up.
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: Eric Biggers <ebiggers@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
For hash algorithms implemented using the "shash" algorithm type, test
both the ahash and shash APIs, not just the ahash API.
Testing the ahash API already tests the shash API indirectly, which is
normally good enough. However, there have been corner cases where there
have been shash bugs that don't get exposed through the ahash API. So,
update testmgr to test the shash API too.
This would have detected the arm64 SHA-1 and SHA-2 bugs for which fixes
were just sent out (https://patchwork.kernel.org/patch/10964843/ and
https://patchwork.kernel.org/patch/10965089/):
alg: shash: sha1-ce test failed (wrong result) on test vector 0, cfg="init+finup aligned buffer"
alg: shash: sha224-ce test failed (wrong result) on test vector 0, cfg="init+finup aligned buffer"
alg: shash: sha256-ce test failed (wrong result) on test vector 0, cfg="init+finup aligned buffer"
This also would have detected the bugs fixed by commit 307508d107
("crypto: crct10dif-generic - fix use via crypto_shash_digest()") and
commit dec3d0b107
("crypto: x86/crct10dif-pcl - fix use via crypto_shash_digest()").
Signed-off-by: Eric Biggers <ebiggers@google.com>
Acked-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Based on 1 normalized pattern(s):
this program is free software you can redistribute it and or modify
it under the terms of the gnu general public license as published by
the free software foundation either version 2 of the license or at
your option any later version
extracted by the scancode license scanner the SPDX license identifier
GPL-2.0-or-later
has been chosen to replace the boilerplate/reference in 3029 file(s).
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Allison Randal <allison@lohutok.net>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190527070032.746973796@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Mark sm4 and missing aes using protected keys which are indetical to
same algs with no HW protected keys as tested.
Signed-off-by: Gilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The flags field in 'struct shash_desc' never actually does anything.
The only ostensibly supported flag is CRYPTO_TFM_REQ_MAY_SLEEP.
However, no shash algorithm ever sleeps, making this flag a no-op.
With this being the case, inevitably some users who can't sleep wrongly
pass MAY_SLEEP. These would all need to be fixed if any shash algorithm
actually started sleeping. For example, the shash_ahash_*() functions,
which wrap a shash algorithm with the ahash API, pass through MAY_SLEEP
from the ahash API to the shash API. However, the shash functions are
called under kmap_atomic(), so actually they're assumed to never sleep.
Even if it turns out that some users do need preemption points while
hashing large buffers, we could easily provide a helper function
crypto_shash_update_large() which divides the data into smaller chunks
and calls crypto_shash_update() and cond_resched() for each chunk. It's
not necessary to have a flag in 'struct shash_desc', nor is it necessary
to make individual shash algorithms aware of this at all.
Therefore, remove shash_desc::flags, and document that the
crypto_shash_*() functions can be called from any context.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
When the extra crypto self-tests are enabled, test each AEAD algorithm
against its generic implementation when one is available. This
involves: checking the algorithm properties for consistency, then
randomly generating test vectors using the generic implementation and
running them against the implementation under test. Both good and bad
inputs are tested.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
When the extra crypto self-tests are enabled, test each skcipher
algorithm against its generic implementation when one is available.
This involves: checking the algorithm properties for consistency, then
randomly generating test vectors using the generic implementation and
running them against the implementation under test. Both good and bad
inputs are tested.
This has already detected a bug in the skcipher_walk API, a bug in the
LRW template, and an inconsistency in the cts implementations.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
When the extra crypto self-tests are enabled, test each hash algorithm
against its generic implementation when one is available. This
involves: checking the algorithm properties for consistency, then
randomly generating test vectors using the generic implementation and
running them against the implementation under test. Both good and bad
inputs are tested.
This has already detected a bug in the x86 implementation of poly1305,
bugs in crct10dif, and an inconsistency in cbcmac.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Add some helper functions in preparation for fuzz testing algorithms
against their generic implementation.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
In preparation for fuzz testing algorithms against their generic
implementation, make error messages in testmgr identify test vectors by
name rather than index. Built-in test vectors are simply "named" by
their index in testmgr.h, as before. But (in later patches) generated
test vectors will be given more descriptive names to help developers
debug problems detected with them.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Update testmgr to support testing for specific errors from setkey() and
digest() for hashes; setkey() and encrypt()/decrypt() for skciphers and
ciphers; and setkey(), setauthsize(), and encrypt()/decrypt() for AEADs.
This is useful because algorithms usually restrict the lengths or format
of the message, key, and/or authentication tag in some way. And bad
inputs should be tested too, not just good inputs.
As part of this change, remove the ambiguously-named 'fail' flag and
replace it with 'setkey_error = -EINVAL' for the only test vector that
used it -- the DES weak key test vector. Note that this tightens the
test to require -EINVAL rather than any error code, but AFAICS this
won't cause any test failure.
Other than that, these new fields aren't set on any test vectors yet.
Later patches will do so.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Add testmgr test vectors for EC-RDSA algorithm for every of five
supported parameters (curves). Because there are no officially published
test vectors for the curves, the vectors are generated by gost-engine.
Signed-off-by: Vitaly Chikunov <vt@altlinux.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Some public key algorithms (like EC-DSA) keep in parameters field
important data such as digest and curve OIDs (possibly more for
different EC-DSA variants). Thus, just setting a public key (as
for RSA) is not enough.
Append parameters into the key stream for akcipher_set_{pub,priv}_key.
Appended data is: (u32) algo OID, (u32) parameters length, parameters
data.
This does not affect current akcipher API nor RSA ciphers (they could
ignore it). Idea of appending parameters to the key stream is by Herbert
Xu.
Cc: David Howells <dhowells@redhat.com>
Cc: Denis Kenzior <denkenz@gmail.com>
Cc: keyrings@vger.kernel.org
Signed-off-by: Vitaly Chikunov <vt@altlinux.org>
Reviewed-by: Denis Kenzior <denkenz@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Previous akcipher .verify() just `decrypts' (using RSA encrypt which is
using public key) signature to uncover message hash, which was then
compared in upper level public_key_verify_signature() with the expected
hash value, which itself was never passed into verify().
This approach was incompatible with EC-DSA family of algorithms,
because, to verify a signature EC-DSA algorithm also needs a hash value
as input; then it's used (together with a signature divided into halves
`r||s') to produce a witness value, which is then compared with `r' to
determine if the signature is correct. Thus, for EC-DSA, nor
requirements of .verify() itself, nor its output expectations in
public_key_verify_signature() wasn't sufficient.
Make improved .verify() call which gets hash value as input and produce
complete signature check without any output besides status.
Now for the top level verification only crypto_akcipher_verify() needs
to be called and its return value inspected.
Make sure that `digest' is in kmalloc'd memory (in place of `output`) in
{public,tpm}_key_verify_signature() as insisted by Herbert Xu, and will
be changed in the following commit.
Cc: David Howells <dhowells@redhat.com>
Cc: keyrings@vger.kernel.org
Signed-off-by: Vitaly Chikunov <vt@altlinux.org>
Reviewed-by: Denis Kenzior <denkenz@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Add a module parameter cryptomgr.panic_on_fail which causes the kernel
to panic if any crypto self-tests fail.
Use cases:
- More easily detect crypto self-test failures by boot testing,
e.g. on KernelCI.
- Get a bug report if syzkaller manages to use the template system to
instantiate an algorithm that fails its self-tests.
The command-line option "fips=1" already does this, but it also makes
other changes not wanted for general testing, such as disabling
"unapproved" algorithms. panic_on_fail just does what it says.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
All crypto API algorithms are supposed to support the case where they
are called in a context where SIMD instructions are unusable, e.g. IRQ
context on some architectures. However, this isn't tested for by the
self-tests, causing bugs to go undetected.
Now that all algorithms have been converted to use crypto_simd_usable(),
update the self-tests to test the no-SIMD case. First, a bool
testvec_config::nosimd is added. When set, the crypto operation is
executed with preemption disabled and with crypto_simd_usable() mocked
out to return false on the current CPU.
A bool test_sg_division::nosimd is also added. For hash algorithms it's
honored by the corresponding ->update(). By setting just a subset of
these bools, the case where some ->update()s are done in SIMD context
and some are done in no-SIMD context is also tested.
These bools are then randomly set by generate_random_testvec_config().
For now, all no-SIMD testing is limited to the extra crypto self-tests,
because it might be a bit too invasive for the regular self-tests.
But this could be changed later.
This has already found bugs in the arm64 AES-GCM and ChaCha algorithms.
This would have found some past bugs as well.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Acked-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
So that the no-SIMD fallback code can be tested by the crypto
self-tests, add a macro crypto_simd_usable() which wraps may_use_simd(),
but also returns false if the crypto self-tests have set a per-CPU bool
to disable SIMD in crypto code on the current CPU.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Now that all AEAD algorithms (that I have the hardware to test, at
least) have been fixed to not modify the user-provided aead_request,
remove the workaround from testmgr that reset aead_request::tfm after
each AEAD encryption/decryption.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Allow skcipher test vectors to declare the value the IV buffer should be
updated to at the end of the encryption or decryption operation.
(This check actually used to be supported in testmgr, but it was never
used and therefore got removed except for the AES-Keywrap special case.
But it will be used by CBC and CTR now, so re-add it.)
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Check that algorithms do not change the aead_request structure, as users
may rely on submitting the request again (e.g. after copying new data
into the same source buffer) without reinitializing everything.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Check that algorithms do not change the skcipher_request structure, as
users may rely on submitting the request again (e.g. after copying new
data into the same source buffer) without reinitializing everything.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Convert alg_test_hash() to use the new test framework, adding a list of
testvec_configs to test by default. When the extra self-tests are
enabled, randomly generated testvec_configs are tested as well.
This improves hash test coverage mainly because now all algorithms have
a variety of data layouts tested, whereas before each algorithm was
responsible for declaring its own chunked test cases which were often
missing or provided poor test coverage. The new code also tests both
the MAY_SLEEP and !MAY_SLEEP cases and buffers that cross pages.
This already found bugs in the hash walk code and in the arm32 and arm64
implementations of crct10dif.
I removed the hash chunked test vectors that were the same as
non-chunked ones, but left the ones that were unique.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Convert alg_test_aead() to use the new test framework, using the same
list of testvec_configs that skcipher testing uses.
This significantly improves AEAD test coverage mainly because previously
there was only very limited test coverage of the possible data layouts.
Now the data layouts to test are listed in one place for all algorithms
and optionally are also randomly generated. In fact, only one AEAD
algorithm (AES-GCM) even had a chunked test case before.
This already found bugs in all the AEGIS and MORUS implementations, the
x86 AES-GCM implementation, and the arm64 AES-CCM implementation.
I removed the AEAD chunked test vectors that were the same as
non-chunked ones, but left the ones that were unique.
Note: the rewritten test code allocates an aead_request just once per
algorithm rather than once per encryption/decryption, but some AEAD
algorithms incorrectly change the tfm pointer in the request. It's
nontrivial to fix these, so to move forward I'm temporarily working
around it by resetting the tfm pointer. But they'll need to be fixed.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Convert alg_test_skcipher() to use the new test framework, adding a list
of testvec_configs to test by default. When the extra self-tests are
enabled, randomly generated testvec_configs are tested as well.
This improves skcipher test coverage mainly because now all algorithms
have a variety of data layouts tested, whereas before each algorithm was
responsible for declaring its own chunked test cases which were often
missing or provided poor test coverage. The new code also tests both
the MAY_SLEEP and !MAY_SLEEP cases, different IV alignments, and buffers
that cross pages.
This has already found a bug in the arm64 ctr-aes-neonbs algorithm.
It would have easily found many past bugs.
I removed the skcipher chunked test vectors that were the same as
non-chunked ones, but left the ones that were unique.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Add functions that generate a random testvec_config, in preparation for
using it for randomized fuzz tests.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
To achieve more comprehensive crypto test coverage, I'd like to add fuzz
tests that use random data layouts and request flags.
To be most effective these tests should be part of testmgr, so they
automatically run on every algorithm registered with the crypto API.
However, they will take much longer to run than the current tests and
therefore will only really be intended to be run by developers, whereas
the current tests have a wider audience.
Therefore, add a new kconfig option CONFIG_CRYPTO_MANAGER_EXTRA_TESTS
that can be set by developers to enable these extra, expensive tests.
Similar to the regular tests, also add a module parameter
cryptomgr.noextratests to support disabling the tests.
Finally, another module parameter cryptomgr.fuzz_iterations is added to
control how many iterations the fuzz tests do. Note: for now setting
this to 0 will be equivalent to cryptomgr.noextratests=1. But I opted
for separate parameters to provide more flexibility to add other types
of tests under the "extra tests" category in the future.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Crypto algorithms must produce the same output for the same input
regardless of data layout, i.e. how the src and dst scatterlists are
divided into chunks and how each chunk is aligned. Request flags such
as CRYPTO_TFM_REQ_MAY_SLEEP must not affect the result either.
However, testing of this currently has many gaps. For example,
individual algorithms are responsible for providing their own chunked
test vectors. But many don't bother to do this or test only one or two
cases, providing poor test coverage. Also, other things such as
misaligned IVs and CRYPTO_TFM_REQ_MAY_SLEEP are never tested at all.
Test code is also duplicated between the chunked and non-chunked cases,
making it difficult to make other improvements.
To improve the situation, this patch series basically moves the chunk
descriptions into the testmgr itself so that they are shared by all
algorithms. However, it's done in an extensible way via a new struct
'testvec_config', which describes not just the scaled chunk lengths but
also all other aspects of the crypto operation besides the data itself
such as the buffer alignments, the request flags, whether the operation
is in-place or not, the IV alignment, and for hash algorithms when to
do each update() and when to use finup() vs. final() vs. digest().
Then, this patch series makes skcipher, aead, and hash algorithms be
tested against a list of default testvec_configs, replacing the current
test code. This improves overall test coverage, without reducing test
performance too much. Note that the test vectors themselves are not
changed, except for removing the chunk lists.
This series also adds randomized fuzz tests, enabled by a new kconfig
option intended for developer use only, where skcipher, aead, and hash
algorithms are tested against many randomly generated testvec_configs.
This provides much more comprehensive test coverage.
These improved tests have already exposed many bugs.
To start it off, this initial patch adds the testvec_config and various
helper functions that will be used by the skcipher, aead, and hash test
code that will be converted to use the new testvec_config framework.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The CRC32 is not a cryptographic hash algorithm,
so the FIPS restrictions should not apply to it.
(The CRC32C variant is already allowed.)
This CRC32 variant is used for in dm-crypt legacy TrueCrypt
IV implementation (tcw); detected by cryptsetup test suite
failure in FIPS mode.
Signed-off-by: Milan Broz <gmazyland@gmail.com>
Reviewed-by: Stephan Mueller <smueller@chronox.de>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Instantiating "cryptd(crc32c)" causes a crypto self-test failure because
the crypto_alloc_shash() in alg_test_crc32c() fails. This is because
cryptd(crc32c) is an ahash algorithm, not a shash algorithm; so it can
only be accessed through the ahash API, unlike shash algorithms which
can be accessed through both the ahash and shash APIs.
As the test is testing the shash descriptor format which is only
applicable to shash algorithms, skip it for ahash algorithms.
(Note that it's still important to fix crypto self-test failures even
for weird algorithm instantiations like cryptd(crc32c) that no one
would really use; in fips_enabled mode unprivileged users can use them
to panic the kernel, and also they prevent treating a crypto self-test
failure as a bug when fuzzing the kernel.)
Fixes: 8e3ee85e68 ("crypto: crc32c - Test descriptor context format")
Cc: stable@vger.kernel.org
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
CRYPTO_TFM_REQ_WEAK_KEY confuses newcomers to the crypto API because it
sounds like it is requesting a weak key. Actually, it is requesting
that weak keys be forbidden (for algorithms that have the notion of
"weak keys"; currently only DES and XTS do).
Also it is only one letter away from CRYPTO_TFM_RES_WEAK_KEY, with which
it can be easily confused. (This in fact happened in the UX500 driver,
though just in some debugging messages.)
Therefore, make the intent clear by renaming it to
CRYPTO_TFM_REQ_FORBID_WEAK_KEYS.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Currently testmgr has separate encryption and decryption test vectors
for AEADs. That's massively redundant, since usually the decryption
tests are identical to the encryption tests, just with the input/result
swapped. And for some algorithms it was forgotten to add decryption
test vectors, so for them currently only encryption is being tested.
Therefore, eliminate the redundancy by removing the AEAD decryption test
vectors and updating testmgr to test both AEAD encryption and decryption
using what used to be the encryption test vectors. Naming is adjusted
accordingly: each aead_testvec now has a 'ptext' (plaintext), 'plen'
(plaintext length), 'ctext' (ciphertext), and 'clen' (ciphertext length)
instead of an 'input', 'ilen', 'result', and 'rlen'. "Ciphertext" here
refers to the full ciphertext, including the authentication tag.
For now the scatterlist divisions are just given for the plaintext
length, not also the ciphertext length. For decryption, the last
scatterlist element is just extended by the authentication tag length.
In total, this removes over 5000 lines from testmgr.h, with no reduction
in test coverage since prior patches already copied the few unique
decryption test vectors into the encryption test vectors.
The testmgr.h portion of this patch was automatically generated using
the following awk script, except that I also manually updated the
definition of 'struct aead_testvec' and fixed the location of the
comment describing the AEGIS-128 test vectors.
BEGIN { OTHER = 0; ENCVEC = 1; DECVEC = 2; DECVEC_TAIL = 3; mode = OTHER }
/^static const struct aead_testvec.*_enc_/ { sub("_enc", ""); mode = ENCVEC }
/^static const struct aead_testvec.*_dec_/ { mode = DECVEC }
mode == ENCVEC {
sub(/\.input[[:space:]]*=/, ".ptext\t=")
sub(/\.result[[:space:]]*=/, ".ctext\t=")
sub(/\.ilen[[:space:]]*=/, ".plen\t=")
sub(/\.rlen[[:space:]]*=/, ".clen\t=")
print
}
mode == DECVEC_TAIL && /[^[:space:]]/ { mode = OTHER }
mode == OTHER { print }
mode == ENCVEC && /^};/ { mode = OTHER }
mode == DECVEC && /^};/ { mode = DECVEC_TAIL }
Note that git's default diff algorithm gets confused by the testmgr.h
portion of this patch, and reports too many lines added and removed.
It's better viewed with 'git diff --minimal' (or 'git show --minimal'),
which reports "2 files changed, 1235 insertions(+), 6491 deletions(-)".
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
In preparation for unifying the AEAD encryption and decryption test
vectors, skip AEAD test vectors with the 'novrfy' (verification failure
expected) flag set when testing encryption rather than decryption.
These test vectors only make sense for decryption.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The crc32c context is in CPU endianness, whereas the final digest is
little endian. alg_test_crc32c() got this mixed up. Fix it.
The test passes both before and after, but this patch fixes the
following sparse warning:
crypto/testmgr.c:1912:24: warning: cast to restricted __le32
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Before this, if akcipher_testvec have `public_key_vec' set to true
(i.e. having a public key) only sign/encrypt test is performed, but
verify/decrypt test is skipped.
With a public key we could do encrypt and verify, but to sign and decrypt
a private key is required.
This logic is correct for encrypt/decrypt tests (decrypt is skipped if
no private key). But incorrect for sign/verify tests - sign is performed
no matter if there is no private key, but verify is skipped if there is
a public key.
Rework `test_akcipher_one' to arrange tests properly depending on value
of `public_key_vec` and `siggen_sigver_test'.
No tests were missed since there is only one sign/verify test (which
have `siggen_sigver_test' set to true) and it has a private key, but
future tests could benefit from this improvement.
Signed-off-by: Vitaly Chikunov <vt@altlinux.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Add support for the Adiantum encryption mode. Adiantum was designed by
Paul Crowley and is specified by our paper:
Adiantum: length-preserving encryption for entry-level processors
(https://eprint.iacr.org/2018/720.pdf)
See our paper for full details; this patch only provides an overview.
Adiantum is a tweakable, length-preserving encryption mode designed for
fast and secure disk encryption, especially on CPUs without dedicated
crypto instructions. Adiantum encrypts each sector using the XChaCha12
stream cipher, two passes of an ε-almost-∆-universal (εA∆U) hash
function, and an invocation of the AES-256 block cipher on a single
16-byte block. On CPUs without AES instructions, Adiantum is much
faster than AES-XTS; for example, on ARM Cortex-A7, on 4096-byte sectors
Adiantum encryption is about 4 times faster than AES-256-XTS encryption,
and decryption about 5 times faster.
Adiantum is a specialization of the more general HBSH construction. Our
earlier proposal, HPolyC, was also a HBSH specialization, but it used a
different εA∆U hash function, one based on Poly1305 only. Adiantum's
εA∆U hash function, which is based primarily on the "NH" hash function
like that used in UMAC (RFC4418), is about twice as fast as HPolyC's;
consequently, Adiantum is about 20% faster than HPolyC.
This speed comes with no loss of security: Adiantum is provably just as
secure as HPolyC, in fact slightly *more* secure. Like HPolyC,
Adiantum's security is reducible to that of XChaCha12 and AES-256,
subject to a security bound. XChaCha12 itself has a security reduction
to ChaCha12. Therefore, one need not "trust" Adiantum; one need only
trust ChaCha12 and AES-256. Note that the εA∆U hash function is only
used for its proven combinatorical properties so cannot be "broken".
Adiantum is also a true wide-block encryption mode, so flipping any
plaintext bit in the sector scrambles the entire ciphertext, and vice
versa. No other such mode is available in the kernel currently; doing
the same with XTS scrambles only 16 bytes. Adiantum also supports
arbitrary-length tweaks and naturally supports any length input >= 16
bytes without needing "ciphertext stealing".
For the stream cipher, Adiantum uses XChaCha12 rather than XChaCha20 in
order to make encryption feasible on the widest range of devices.
Although the 20-round variant is quite popular, the best known attacks
on ChaCha are on only 7 rounds, so ChaCha12 still has a substantial
security margin; in fact, larger than AES-256's. 12-round Salsa20 is
also the eSTREAM recommendation. For the block cipher, Adiantum uses
AES-256, despite it having a lower security margin than XChaCha12 and
needing table lookups, due to AES's extensive adoption and analysis
making it the obvious first choice. Nevertheless, for flexibility this
patch also permits the "adiantum" template to be instantiated with
XChaCha20 and/or with an alternate block cipher.
We need Adiantum support in the kernel for use in dm-crypt and fscrypt,
where currently the only other suitable options are block cipher modes
such as AES-XTS. A big problem with this is that many low-end mobile
devices (e.g. Android Go phones sold primarily in developing countries,
as well as some smartwatches) still have CPUs that lack AES
instructions, e.g. ARM Cortex-A7. Sadly, AES-XTS encryption is much too
slow to be viable on these devices. We did find that some "lightweight"
block ciphers are fast enough, but these suffer from problems such as
not having much cryptanalysis or being too controversial.
The ChaCha stream cipher has excellent performance but is insecure to
use directly for disk encryption, since each sector's IV is reused each
time it is overwritten. Even restricting the threat model to offline
attacks only isn't enough, since modern flash storage devices don't
guarantee that "overwrites" are really overwrites, due to wear-leveling.
Adiantum avoids this problem by constructing a
"tweakable super-pseudorandom permutation"; this is the strongest
possible security model for length-preserving encryption.
Of course, storing random nonces along with the ciphertext would be the
ideal solution. But doing that with existing hardware and filesystems
runs into major practical problems; in most cases it would require data
journaling (like dm-integrity) which severely degrades performance.
Thus, for now length-preserving encryption is still needed.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Add a generic implementation of NHPoly1305, an ε-almost-∆-universal hash
function used in the Adiantum encryption mode.
CONFIG_NHPOLY1305 is not selectable by itself since there won't be any
real reason to enable it without also enabling Adiantum support.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Acked-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Now that the generic implementation of ChaCha20 has been refactored to
allow varying the number of rounds, add support for XChaCha12, which is
the XSalsa construction applied to ChaCha12. ChaCha12 is one of the
three ciphers specified by the original ChaCha paper
(https://cr.yp.to/chacha/chacha-20080128.pdf: "ChaCha, a variant of
Salsa20"), alongside ChaCha8 and ChaCha20. ChaCha12 is faster than
ChaCha20 but has a lower, but still large, security margin.
We need XChaCha12 support so that it can be used in the Adiantum
encryption mode, which enables disk/file encryption on low-end mobile
devices where AES-XTS is too slow as the CPUs lack AES instructions.
We'd prefer XChaCha20 (the more popular variant), but it's too slow on
some of our target devices, so at least in some cases we do need the
XChaCha12-based version. In more detail, the problem is that Adiantum
is still much slower than we're happy with, and encryption still has a
quite noticeable effect on the feel of low-end devices. Users and
vendors push back hard against encryption that degrades the user
experience, which always risks encryption being disabled entirely. So
we need to choose the fastest option that gives us a solid margin of
security, and here that's XChaCha12. The best known attack on ChaCha
breaks only 7 rounds and has 2^235 time complexity, so ChaCha12's
security margin is still better than AES-256's. Much has been learned
about cryptanalysis of ARX ciphers since Salsa20 was originally designed
in 2005, and it now seems we can be comfortable with a smaller number of
rounds. The eSTREAM project also suggests the 12-round version of
Salsa20 as providing the best balance among the different variants:
combining very good performance with a "comfortable margin of security".
Note that it would be trivial to add vanilla ChaCha12 in addition to
XChaCha12. However, it's unneeded for now and therefore is omitted.
As discussed in the patch that introduced XChaCha20 support, I
considered splitting the code into separate chacha-common, chacha20,
xchacha20, and xchacha12 modules, so that these algorithms could be
enabled/disabled independently. However, since nearly all the code is
shared anyway, I ultimately decided there would have been little benefit
to the added complexity.
Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Acked-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Add support for the XChaCha20 stream cipher. XChaCha20 is the
application of the XSalsa20 construction
(https://cr.yp.to/snuffle/xsalsa-20081128.pdf) to ChaCha20 rather than
to Salsa20. XChaCha20 extends ChaCha20's nonce length from 64 bits (or
96 bits, depending on convention) to 192 bits, while provably retaining
ChaCha20's security. XChaCha20 uses the ChaCha20 permutation to map the
key and first 128 nonce bits to a 256-bit subkey. Then, it does the
ChaCha20 stream cipher with the subkey and remaining 64 bits of nonce.
We need XChaCha support in order to add support for the Adiantum
encryption mode. Note that to meet our performance requirements, we
actually plan to primarily use the variant XChaCha12. But we believe
it's wise to first add XChaCha20 as a baseline with a higher security
margin, in case there are any situations where it can be used.
Supporting both variants is straightforward.
Since XChaCha20's subkey differs for each request, XChaCha20 can't be a
template that wraps ChaCha20; that would require re-keying the
underlying ChaCha20 for every request, which wouldn't be thread-safe.
Instead, we make XChaCha20 its own top-level algorithm which calls the
ChaCha20 streaming implementation internally.
Similar to the existing ChaCha20 implementation, we define the IV to be
the nonce and stream position concatenated together. This allows users
to seek to any position in the stream.
I considered splitting the code into separate chacha20-common, chacha20,
and xchacha20 modules, so that chacha20 and xchacha20 could be
enabled/disabled independently. However, since nearly all the code is
shared anyway, I ultimately decided there would have been little benefit
to the added complexity of separate modules.
Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Acked-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Add testmgr and tcrypt tests and vectors for Streebog hash function
from RFC 6986 and GOST R 34.11-2012, for HMAC-Streebog vectors are
from RFC 7836 and R 50.1.113-2016.
Cc: linux-integrity@vger.kernel.org
Signed-off-by: Vitaly Chikunov <vt@altlinux.org>
Acked-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
As per Sp800-38A addendum from Oct 2010[1], cts(cbc(aes)) is
allowed as a FIPS mode algorithm. Mark it as such.
[1] https://csrc.nist.gov/publications/detail/sp/800-38a/addendum/final
Signed-off-by: Gilad Ben-Yossef <gilad@benyossef.com>
Reviewed-by: Stephan Mueller <smueller@chronox.de>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
After allocation, output and decomp_output both point to memory chunks of
size COMP_BUF_SIZE. Then, only the first bytes are zeroed out using
sizeof(COMP_BUF_SIZE) as parameter to memset(), because
sizeof(COMP_BUF_SIZE) provides the size of the constant and not the size of
allocated memory.
Instead, the whole allocated memory is meant to be zeroed out. Use
COMP_BUF_SIZE as parameter to memset() directly in order to accomplish
this.
Fixes: 336073840a ("crypto: testmgr - Allow different compression results")
Signed-off-by: Michael Schupikov <michael@schupikov.de>
Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Add additional test vectors from "The SM4 Blockcipher Algorithm And Its
Modes Of Operations" draft-ribose-cfrg-sm4-10 and register cipher speed
tests for sm4.
Signed-off-by: Gilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
These are unused, undesired, and have never actually been used by
anybody. The original authors of this code have changed their mind about
its inclusion. While originally proposed for disk encryption on low-end
devices, the idea was discarded [1] in favor of something else before
that could really get going. Therefore, this patch removes Speck.
[1] https://marc.info/?l=linux-crypto-vger&m=153359499015659
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Acked-by: Eric Biggers <ebiggers@google.com>
Cc: stable@vger.kernel.org
Acked-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The testmgr hash tests were testing init, digest, update and final
methods but not the finup method. Add a test for this one too.
While doing this, make sure we only run the partial tests once with
the digest tests and skip them with the final and finup tests since
they are the same.
Signed-off-by: Gilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Remove the original version of the VMAC template that had the nonce
hardcoded to 0 and produced a digest with the wrong endianness. I'm
unsure whether this had users or not (there are no explicit in-kernel
references to it), but given that the hardcoded nonce made it wildly
insecure unless a unique key was used for each message, let's try
removing it and see if anyone complains.
Leave the new "vmac64" template that requires the nonce to be explicitly
specified as the first 16 bytes of data and uses the correct endianness
for the digest.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Currently the VMAC template uses a "nonce" hardcoded to 0, which makes
it insecure unless a unique key is set for every message. Also, the
endianness of the final digest is wrong: the implementation uses little
endian, but the VMAC specification has it as big endian, as do other
VMAC implementations such as the one in Crypto++.
Add a new VMAC template where the nonce is passed as the first 16 bytes
of data (similar to what is done for Poly1305's nonce), and the digest
is big endian. Call it "vmac64", since the old name of simply "vmac"
didn't clarify whether the implementation is of VMAC-64 or of VMAC-128
(which produce 64-bit and 128-bit digests respectively); so we fix the
naming ambiguity too.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Currently testmgr has separate encryption and decryption test vectors
for symmetric ciphers. That's massively redundant, since with few
exceptions (mostly mistakes, apparently), all decryption tests are
identical to the encryption tests, just with the input/result flipped.
Therefore, eliminate the redundancy by removing the decryption test
vectors and updating testmgr to test both encryption and decryption
using what used to be the encryption test vectors. Naming is adjusted
accordingly: each cipher_testvec now has a 'ptext' (plaintext), 'ctext'
(ciphertext), and 'len' instead of an 'input', 'result', 'ilen', and
'rlen'. Note that it was always the case that 'ilen == rlen'.
AES keywrap ("kw(aes)") is special because its IV is generated by the
encryption. Previously this was handled by specifying 'iv_out' for
encryption and 'iv' for decryption. To make it work cleanly with only
one set of test vectors, put the IV in 'iv', remove 'iv_out', and add a
boolean that indicates that the IV is generated by the encryption.
In total, this removes over 10000 lines from testmgr.h, with no
reduction in test coverage since prior patches already copied the few
unique decryption test vectors into the encryption test vectors.
This covers all algorithms that used 'struct cipher_testvec', e.g. any
block cipher in the ECB, CBC, CTR, XTS, LRW, CTS-CBC, PCBC, OFB, or
keywrap modes, and Salsa20 and ChaCha20. No change is made to AEAD
tests, though we probably can eliminate a similar redundancy there too.
The testmgr.h portion of this patch was automatically generated using
the following awk script, with some slight manual fixups on top (updated
'struct cipher_testvec' definition, updated a few comments, and fixed up
the AES keywrap test vectors):
BEGIN { OTHER = 0; ENCVEC = 1; DECVEC = 2; DECVEC_TAIL = 3; mode = OTHER }
/^static const struct cipher_testvec.*_enc_/ { sub("_enc", ""); mode = ENCVEC }
/^static const struct cipher_testvec.*_dec_/ { mode = DECVEC }
mode == ENCVEC && !/\.ilen[[:space:]]*=/ {
sub(/\.input[[:space:]]*=$/, ".ptext =")
sub(/\.input[[:space:]]*=/, ".ptext\t=")
sub(/\.result[[:space:]]*=$/, ".ctext =")
sub(/\.result[[:space:]]*=/, ".ctext\t=")
sub(/\.rlen[[:space:]]*=/, ".len\t=")
print
}
mode == DECVEC_TAIL && /[^[:space:]]/ { mode = OTHER }
mode == OTHER { print }
mode == ENCVEC && /^};/ { mode = OTHER }
mode == DECVEC && /^};/ { mode = DECVEC_TAIL }
Note that git's default diff algorithm gets confused by the testmgr.h
portion of this patch, and reports too many lines added and removed.
It's better viewed with 'git diff --minimal' (or 'git show --minimal'),
which reports "2 files changed, 919 insertions(+), 11723 deletions(-)".
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Since testmgr uses a single tfm for all tests of each hash algorithm,
once a key is set the tfm won't be unkeyed anymore. But with crc32 and
crc32c, the key is really the "default initial state" and is optional;
those algorithms should have both keyed and unkeyed test vectors, to
verify that implementations use the correct default key.
Simply listing the unkeyed test vectors first isn't guaranteed to work
yet because testmgr makes multiple passes through the test vectors.
crc32c does have an unkeyed test vector listed first currently, but it
only works by chance because the last crc32c test vector happens to use
a key that is the same as the default key.
Therefore, teach testmgr to split hash test vectors into unkeyed and
keyed sections, and do all the unkeyed ones before the keyed ones.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The Blackfin CRC driver was removed by commit 9678a8dc53 ("crypto:
bfin_crc - remove blackfin CRC driver"), but it was forgotten to remove
the corresponding "hmac(crc32)" test vectors. I see no point in keeping
them since nothing else appears to implement or use "hmac(crc32)", which
isn't an algorithm that makes sense anyway because HMAC is meant to be
used with a cryptographically secure hash function, which CRC's are not.
Thus, remove the unneeded test vectors.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
This patch adds test vectors for MORUS-640 and MORUS-1280. The test
vectors were generated using the reference implementation from
SUPERCOP (see code comments for more details).
Signed-off-by: Ondrej Mosnacek <omosnacek@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
This patch adds test vectors for the AEGIS family of AEAD algorithms
(AEGIS-128, AEGIS-128L, and AEGIS-256). The test vectors were
generated using the reference implementation from SUPERCOP (see code
comments for more details).
Signed-off-by: Ondrej Mosnacek <omosnacek@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Due to a snafu "paes" testmgr tests were not ordered
lexicographically, which led to boot time warnings.
Reorder the tests as needed.
Fixes: a794d8d ("crypto: ccree - enable support for hardware keys")
Reported-by: Abdul Haleem <abdhalee@linux.vnet.ibm.com>
Signed-off-by: Gilad Ben-Yossef <gilad@benyossef.com>
Tested-by: Abdul Haleem <abdhalee@linux.vnet.ibm.com>
Tested-by: Corentin Labbe <clabbe.montjoie@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Enable CryptoCell support for hardware keys.
Hardware keys are regular AES keys loaded into CryptoCell internal memory
via firmware, often from secure boot ROM or hardware fuses at boot time.
As such, they can be used for enc/dec purposes like any other key but
cannot (read: extremely hard to) be extracted since since they are not
available anywhere in RAM during runtime.
The mechanism has some similarities to s390 secure keys although the keys
are not wrapped or sealed, but simply loaded offline. The interface was
therefore modeled based on the s390 secure keys support.
Signed-off-by: Gilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The following error is triggered by the ThunderX ZIP driver
if the testmanager is enabled:
[ 199.069437] ThunderX-ZIP 0000:03:00.0: Found ZIP device 0 177d:a01a on Node 0
[ 199.073573] alg: comp: Compression test 1 failed for deflate-generic: output len = 37
The reason for this error is the verification of the compression
results. Verifying the compression result only works if all
algorithm parameters are identical, in this case to the software
implementation.
Different compression engines like the ThunderX ZIP coprocessor
might yield different compression results by tuning the
algorithm parameters. In our case the compressed result is
shorter than the test vector.
We should not forbid different compression results but only
check that compression -> decompression yields the same
result. This is done already in the acomp test. Do something
similar for test_comp().
Signed-off-by: Mahipal Challa <mchalla@cavium.com>
Signed-off-by: Balakrishna Bhamidipati <bbhamidipati@cavium.com>
[jglauber@cavium.com: removed unrelated printk changes, rewrote commit msg,
fixed whitespace and unneeded initialization]
Signed-off-by: Jan Glauber <jglauber@cavium.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Adds zstd support to crypto and scompress. Only supports the default
level.
Previously we held off on this patch, since there weren't any users.
Now zram is ready for zstd support, but depends on CONFIG_CRYPTO_ZSTD,
which isn't defined until this patch is in. I also see a patch adding
zstd to pstore [0], which depends on crypto zstd.
[0] lkml.kernel.org/r/9c9416b2dff19f05fb4c35879aaa83d11ff72c92.1521626182.git.geliangtang@gmail.com
Signed-off-by: Nick Terrell <terrelln@fb.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Add test vectors for Speck64-XTS, generated in userspace using C code.
The inputs were borrowed from the AES-XTS test vectors, with key lengths
adjusted.
xts-speck64-neon passes these tests. However, they aren't currently
applicable for the generic XTS template, as that only supports a 128-bit
block size.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Add test vectors for Speck128-XTS, generated in userspace using C code.
The inputs were borrowed from the AES-XTS test vectors.
Both xts(speck128-generic) and xts-speck128-neon pass these tests.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Add a generic implementation of Speck, including the Speck128 and
Speck64 variants. Speck is a lightweight block cipher that can be much
faster than AES on processors that don't have AES instructions.
We are planning to offer Speck-XTS (probably Speck128/256-XTS) as an
option for dm-crypt and fscrypt on Android, for low-end mobile devices
with older CPUs such as ARMv7 which don't have the Cryptography
Extensions. Currently, such devices are unencrypted because AES is not
fast enough, even when the NEON bit-sliced implementation of AES is
used. Other AES alternatives such as Twofish, Threefish, Camellia,
CAST6, and Serpent aren't fast enough either; it seems that only a
modern ARX cipher can provide sufficient performance on these devices.
This is a replacement for our original proposal
(https://patchwork.kernel.org/patch/10101451/) which was to offer
ChaCha20 for these devices. However, the use of a stream cipher for
disk/file encryption with no space to store nonces would have been much
more insecure than we thought initially, given that it would be used on
top of flash storage as well as potentially on top of F2FS, neither of
which is guaranteed to overwrite data in-place.
Speck has been somewhat controversial due to its origin. Nevertheless,
it has a straightforward design (it's an ARX cipher), and it appears to
be the leading software-optimized lightweight block cipher currently,
with the most cryptanalysis. It's also easy to implement without side
channels, unlike AES. Moreover, we only intend Speck to be used when
the status quo is no encryption, due to AES not being fast enough.
We've also considered a novel length-preserving encryption mode based on
ChaCha20 and Poly1305. While theoretically attractive, such a mode
would be a brand new crypto construction and would be more complicated
and difficult to implement efficiently in comparison to Speck-XTS.
There is confusion about the byte and word orders of Speck, since the
original paper doesn't specify them. But we have implemented it using
the orders the authors recommended in a correspondence with them. The
test vectors are taken from the original paper but were mapped to byte
arrays using the recommended byte and word orders.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Async hash operations can use result pointer in final/finup/digest,
but not in init/update/export/import, so test it for misuse.
Signed-off-by: Kamil Konieczny <k.konieczny@partner.samsung.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
When char is signed, storing the values 0xba (186) and 0xad (173) in the
`guard` array produces signed overflow. Change the type of `guard` to
static unsigned char to correct undefined behavior and reduce function
stack usage.
Signed-off-by: Joey Pabalinas <joeypabalinas@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
testmgr is starting async. crypto ops and waiting for them to complete.
Move it over to generic code doing the same.
This also provides a test of the generic crypto async. wait code.
Signed-off-by: Gilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Add testmgr and tcrypt tests and vectors for SM3 secure hash.
Signed-off-by: Gilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The combination of sha1 and aes was disabled in FIPS Mode
accidentally. This patch reenables it.
Fixes: 284a0f6e87 ("crypto: testmgr - Disable fips-allowed for...")
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Acked-by: Stephan Müller <smueller@chronox.de>
The PKCS#1 RSA implementation is provided with a self test with RSA 2048
and SHA-256. This self test implicitly covers other RSA keys and other
hashes. Also, this self test implies that the pkcs1pad(rsa) is FIPS
140-2 compliant.
Signed-off-by: Stephan Mueller <smueller@chronox.de>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Fix inconsistent format and spelling in hash tests error messages.
Signed-off-by: Gilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The test considers a party that already has a private-public
key pair and a party that provides a NULL key. The kernel will
generate the private-public key pair for the latter, computes
the shared secret on both ends and verifies if it's the same.
The explicit private-public key pair was copied from
the previous test vector.
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Add scomp backend for zlib-deflate compression algorithm.
This backend outputs data using the format defined in rfc1950
(raw deflate surrounded by zlib header and footer).
Signed-off-by: Giovanni Cabiddu <giovanni.cabiddu@intel.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The cipher_null is not a real cipher, FIPS mode should not restrict its use.
It is used for several tests (for example in cryptsetup testsuite) and also
temporarily for reencryption of not yet encrypted device in cryptsetup-reencrypt tool.
Problem is easily reproducible with
cryptsetup benchmark -c null
Signed-off-by: Milan Broz <gmazyland@gmail.com>
Acked-by: Stephan Müller <smueller@chronox.de>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Compression implementations might return valid outputs that
do not match what specified in the test vectors.
For this reason, the testmgr might report that a compression
implementation failed the test even if the data produced
by the compressor is correct.
This implements a decompress-and-verify test for acomp
compression tests rather than a known answer test.
Signed-off-by: Giovanni Cabiddu <giovanni.cabiddu@intel.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Waiman Long
Linus Torvalds
Eric Biggers
Harald Freudenberger
Ard Biesheuvel
David Sterba
Pascal van Leeuwen
Ard Biesheuvel
Hannah Pan
Gilad Ben-Yossef
Arnd Bergmann
Nikolay Borisov
Thomas Gleixner
Vitaly Chikunov
Christopher Diaz Riveros
Milan Broz
Dmitry Eremin-Solenikov
Michael Schupikov
Jason A. Donenfeld
Kees Cook
Ondrej Mosnacek
Mahipal Challa
Nick Terrell
Kamil Konieczny
Joey Pabalinas
Herbert Xu
Stephan Mueller
Tudor-Dan Ambarus
Giovanni Cabiddu