Pull crypto update from Herbert Xu:
"API:
- Add support for AEAD in simd
- Add fuzz testing to testmgr
- Add panic_on_fail module parameter to testmgr
- Use per-CPU struct instead multiple variables in scompress
- Change verify API for akcipher
Algorithms:
- Convert x86 AEAD algorithms over to simd
- Forbid 2-key 3DES in FIPS mode
- Add EC-RDSA (GOST 34.10) algorithm
Drivers:
- Set output IV with ctr-aes in crypto4xx
- Set output IV in rockchip
- Fix potential length overflow with hashing in sun4i-ss
- Fix computation error with ctr in vmx
- Add SM4 protected keys support in ccree
- Remove long-broken mxc-scc driver
- Add rfc4106(gcm(aes)) cipher support in cavium/nitrox"
* 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6: (179 commits)
crypto: ccree - use a proper le32 type for le32 val
crypto: ccree - remove set but not used variable 'du_size'
crypto: ccree - Make cc_sec_disable static
crypto: ccree - fix spelling mistake "protedcted" -> "protected"
crypto: caam/qi2 - generate hash keys in-place
crypto: caam/qi2 - fix DMA mapping of stack memory
crypto: caam/qi2 - fix zero-length buffer DMA mapping
crypto: stm32/cryp - update to return iv_out
crypto: stm32/cryp - remove request mutex protection
crypto: stm32/cryp - add weak key check for DES
crypto: atmel - remove set but not used variable 'alg_name'
crypto: picoxcell - Use dev_get_drvdata()
crypto: crypto4xx - get rid of redundant using_sd variable
crypto: crypto4xx - use sync skcipher for fallback
crypto: crypto4xx - fix cfb and ofb "overran dst buffer" issues
crypto: crypto4xx - fix ctr-aes missing output IV
crypto: ecrdsa - select ASN1 and OID_REGISTRY for EC-RDSA
crypto: ux500 - use ccflags-y instead of CFLAGS_<basename>.o
crypto: ccree - handle tee fips error during power management resume
crypto: ccree - add function to handle cryptocell tee fips error
...
If the user-provided IV needs to be aligned to the algorithm's
alignmask, then skcipher_walk_virt() copies the IV into a new aligned
buffer walk.iv. But skcipher_walk_virt() can fail afterwards, and then
if the caller unconditionally accesses walk.iv, it's a use-after-free.
xts-aes-neonbs doesn't set an alignmask, so currently it isn't affected
by this despite unconditionally accessing walk.iv. However this is more
subtle than desired, and unconditionally accessing walk.iv has caused a
real problem in other algorithms. Thus, update xts-aes-neonbs to start
checking the return value of skcipher_walk_virt().
Fixes: 1abee99eaf ("crypto: arm64/aes - reimplement bit-sliced ARM/NEON implementation for arm64")
Cc: <stable@vger.kernel.org> # v4.11+
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
As we will exhaust the first 32 bits of AT_HWCAP let's start
exposing AT_HWCAP2 to userspace to give us up to 64 caps.
Whilst it's possible to use the remaining 32 bits of AT_HWCAP, we
prefer to expand into AT_HWCAP2 in order to provide a consistent
view to userspace between ILP32 and LP64. However internal to the
kernel we prefer to continue to use the full space of elf_hwcap.
To reduce complexity and allow for future expansion, we now
represent hwcaps in the kernel as ordinals and use a
KERNEL_HWCAP_ prefix. This allows us to support automatic feature
based module loading for all our hwcaps.
We introduce cpu_set_feature to set hwcaps which complements the
existing cpu_have_feature helper. These helpers allow us to clean
up existing direct uses of elf_hwcap and reduce any future effort
required to move beyond 64 caps.
For convenience we also introduce cpu_{have,set}_named_feature which
makes use of the cpu_feature macro to allow providing a hwcap name
without a {KERNEL_}HWCAP_ prefix.
Signed-off-by: Andrew Murray <andrew.murray@arm.com>
[will: use const_ilog2() and tweak documentation]
Signed-off-by: Will Deacon <will.deacon@arm.com>
My patches to make testmgr fuzz algorithms against their generic
implementation detected that the arm64 implementations of "cbcmac(aes)"
handle empty messages differently from the cbcmac template. Namely, the
arm64 implementations return the encrypted initial value, but the cbcmac
template returns the initial value directly.
This isn't actually a meaningful case because any user of cbcmac needs
to prepend the message length, as CCM does; otherwise it's insecure.
However, we should keep the behavior consistent; at the very least this
makes testing easier.
Do it the easy way, which is to change the arm64 implementations to have
the same behavior as the cbcmac template.
For what it's worth, ghash does things essentially the same way: it
returns its initial value when given an empty message, even though in
practice ghash is never passed an empty message.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Replace all calls to may_use_simd() in the arm64 crypto code with
crypto_simd_usable(), in order to allow testing the no-SIMD code paths.
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>
The arm64 gcm-aes-ce algorithm is failing the extra crypto self-tests
following my patches to test the !may_use_simd() code paths, which
previously were untested. The problem is that in the !may_use_simd()
case, an odd number of AES blocks can be processed within each step of
the skcipher_walk. However, the skcipher_walk is being done with a
"stride" of 2 blocks and is advanced by an even number of blocks after
each step. This causes the encryption to produce the wrong ciphertext
and authentication tag, and causes the decryption to incorrectly fail.
Fix it by only processing an even number of blocks per step.
Fixes: c2b24c36e0 ("crypto: arm64/aes-gcm-ce - fix scatterwalk API violation")
Fixes: 71e52c278c ("crypto: arm64/aes-ce-gcm - operate on two input blocks at a time")
Cc: <stable@vger.kernel.org> # v4.19+
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>
Pull crypto update from Herbert Xu:
"API:
- Add helper for simple skcipher modes.
- Add helper to register multiple templates.
- Set CRYPTO_TFM_NEED_KEY when setkey fails.
- Require neither or both of export/import in shash.
- AEAD decryption test vectors are now generated from encryption
ones.
- New option CONFIG_CRYPTO_MANAGER_EXTRA_TESTS that includes random
fuzzing.
Algorithms:
- Conversions to skcipher and helper for many templates.
- Add more test vectors for nhpoly1305 and adiantum.
Drivers:
- Add crypto4xx prng support.
- Add xcbc/cmac/ecb support in caam.
- Add AES support for Exynos5433 in s5p.
- Remove sha384/sha512 from artpec7 as hardware cannot do partial
hash"
[ There is a merge of the Freescale SoC tree in order to pull in changes
required by patches to the caam/qi2 driver. ]
* 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6: (174 commits)
crypto: s5p - add AES support for Exynos5433
dt-bindings: crypto: document Exynos5433 SlimSSS
crypto: crypto4xx - add missing of_node_put after of_device_is_available
crypto: cavium/zip - fix collision with generic cra_driver_name
crypto: af_alg - use struct_size() in sock_kfree_s()
crypto: caam - remove redundant likely/unlikely annotation
crypto: s5p - update iv after AES-CBC op end
crypto: x86/poly1305 - Clear key material from stack in SSE2 variant
crypto: caam - generate hash keys in-place
crypto: caam - fix DMA mapping xcbc key twice
crypto: caam - fix hash context DMA unmap size
hwrng: bcm2835 - fix probe as platform device
crypto: s5p-sss - Use AES_BLOCK_SIZE define instead of number
crypto: stm32 - drop pointless static qualifier in stm32_hash_remove()
crypto: chelsio - Fixed Traffic Stall
crypto: marvell - Remove set but not used variable 'ivsize'
crypto: ccp - Update driver messages to remove some confusion
crypto: adiantum - add 1536 and 4096-byte test vectors
crypto: nhpoly1305 - add a test vector with len % 16 != 0
crypto: arm/aes-ce - update IV after partial final CTR block
...
On big endian arm64 kernels, the xchacha20-neon and xchacha12-neon
self-tests fail because hchacha_block_neon() outputs little endian words
but the C code expects native endianness. Fix it to output the words in
native endianness (which also makes it match the arm32 version).
Fixes: cc7cf991e9 ("crypto: arm64/chacha20 - add XChaCha20 support")
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>
The change to encrypt a fifth ChaCha block using scalar instructions
caused the chacha20-neon, xchacha20-neon, and xchacha12-neon self-tests
to start failing on big endian arm64 kernels. The bug is that the
keystream block produced in 32-bit scalar registers is directly XOR'd
with the data words, which are loaded and stored in native endianness.
Thus in big endian mode the data bytes end up XOR'd with the wrong
bytes. Fix it by byte-swapping the keystream words in big endian mode.
Fixes: 2fe55987b2 ("crypto: arm64/chacha - use combined SIMD/ALU routine for more speed")
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>
Make the arm64 ctr-aes-neon and ctr-aes-ce algorithms update the IV
buffer to contain the next counter after processing a partial final
block, rather than leave it as the last counter. This makes these
algorithms pass the updated AES-CTR tests.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The arm64 NEON bit-sliced implementation of AES-CTR fails the improved
skcipher tests because it sometimes produces the wrong ciphertext. The
bug is that the final keystream block isn't returned from the assembly
code when the number of non-final blocks is zero. This can happen if
the input data ends a few bytes after a page boundary. In this case the
last bytes get "encrypted" by XOR'ing them with uninitialized memory.
Fix the assembly code to return the final keystream block when needed.
Fixes: 88a3f582be ("crypto: arm64/aes - don't use IV buffer to return final keystream block")
Cc: <stable@vger.kernel.org> # v4.11+
Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The x86, arm, and arm64 asm implementations of crct10dif are very
difficult to understand partly because many of the comments, labels, and
macros are named incorrectly: the lengths mentioned are usually off by a
factor of two from the actual code. Many other things are unnecessarily
convoluted as well, e.g. there are many more fold constants than
actually needed and some aren't fully reduced.
This series therefore cleans up all these implementations to be much
more maintainable. I also made some small optimizations where I saw
opportunities, resulting in slightly better performance.
This patch cleans up the arm64 version.
Acked-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The arm64 CRC-T10DIF implementation either uses 8-bit or 64-bit
polynomial multiplication instructions, since the latter are
faster but not mandatory in the architecture.
Since that prevents us from testing both implementations on the
same system, let's expose both implementations to the crypto API,
with the priorities reflecting that the P64 version is the
preferred one if available.
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Remove some code that is no longer called now that we make sure never
to invoke the SIMD routine with less than 16 bytes of input.
Reviewed-by: Eric Biggers <ebiggers@kernel.org>
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The SIMD routine ported from x86 used to have a special code path
for inputs < 16 bytes, which got lost somewhere along the way.
Instead, the current glue code aligns the input pointer to 16 bytes,
which is not really necessary on this architecture (although it
could be beneficial to performance to expose aligned data to the
the NEON routine), but this could result in inputs of less than
16 bytes to be passed in. This not only fails the new extended
tests that Eric has implemented, it also results in the code
reading past the end of the input, which could potentially result
in crashes when dealing with less than 16 bytes of input at the
end of a page which is followed by an unmapped page.
So update the glue code to only invoke the NEON routine if the
input is at least 16 bytes.
Reported-by: Eric Biggers <ebiggers@kernel.org>
Reviewed-by: Eric Biggers <ebiggers@kernel.org>
Fixes: 6ef5737f39 ("crypto: arm64/crct10dif - port x86 SSE implementation to arm64")
Cc: <stable@vger.kernel.org> # v4.10+
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The arm64 GHASH implementation either uses 8-bit or 64-bit
polynomial multiplication instructions, since the latter are
faster but not mandatory in the architecture.
Since that prevents us from testing both implementations on the
same system, let's expose both implementations to the crypto API,
with the priorities reflecting that the P64 version is the
preferred one if available.
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
When the AES-CCM code was first added, the NEON register were saved
and restored eagerly, and so the code avoided doing so, and executed
the scatterwalk in atomic context inside the kernel_neon_begin/end
section.
This has been changed in the meantime, so switch to non-atomic
scatterwalks.
Fixes: bd2ad885e3 ("crypto: arm64/aes-ce-ccm - move kernel mode neon ...")
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Commit 5092fcf349 ("crypto: arm64/aes-ce-ccm: add non-SIMD generic
fallback") introduced C fallback code to replace the NEON routines
when invoked from a context where the NEON is not available (i.e.,
from the context of a softirq taken while the NEON is already being
used in kernel process context)
Fix two logical flaws in the MAC calculation of the associated data.
Reported-by: Eric Biggers <ebiggers@kernel.org>
Fixes: 5092fcf349 ("crypto: arm64/aes-ce-ccm: add non-SIMD generic fallback")
Cc: stable@vger.kernel.org
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The NEON MAC calculation routine fails to handle the case correctly
where there is some data in the buffer, and the input fills it up
exactly. In this case, we enter the loop at the end with w8 == 0,
while a negative value is assumed, and so the loop carries on until
the increment of the 32-bit counter wraps around, which is quite
obviously wrong.
So omit the loop altogether in this case, and exit right away.
Reported-by: Eric Biggers <ebiggers@kernel.org>
Fixes: a3fd82105b ("arm64/crypto: AES in CCM mode using ARMv8 Crypto ...")
Cc: stable@vger.kernel.org
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
To some degree, most known AArch64 micro-architectures appear to be
able to issue ALU instructions in parellel to SIMD instructions
without affecting the SIMD throughput. This means we can use the ALU
to process a fifth ChaCha block while the SIMD is processing four
blocks in parallel.
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Update the 4-way NEON ChaCha routine so it can handle input of any
length >64 bytes in its entirety, rather than having to call into
the 1-way routine and/or memcpy()s via temp buffers to handle the
tail of a ChaCha invocation that is not a multiple of 256 bytes.
On inputs that are a multiple of 256 bytes (and thus in tcrypt
benchmarks), performance drops by around 1% on Cortex-A57, while
performance for inputs drawn randomly from the range [64, 1024)
increases by around 30%.
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Now that the ARM64 NEON implementation of ChaCha20 and XChaCha20 has
been refactored to support varying the number of rounds, add support for
XChaCha12. This is identical to XChaCha20 except for the number of
rounds, which is 12 instead of 20. This can be used by Adiantum.
Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
In preparation for adding XChaCha12 support, rename/refactor the ARM64
NEON implementation of ChaCha20 to support different numbers of rounds.
Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Add an XChaCha20 implementation that is hooked up to the ARM64 NEON
implementation of ChaCha20. This can be used by Adiantum.
A NEON implementation of single-block HChaCha20 is also added so that
XChaCha20 can use it rather than the generic implementation. This
required refactoring the ChaCha20 permutation into its own function.
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 an ARM64 NEON implementation of NHPoly1305, an ε-almost-∆-universal
hash function used in the Adiantum encryption mode. For now, only the
NH portion is actually NEON-accelerated; the Poly1305 part is less
performance-critical so is just implemented in C.
Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Tested-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> # big-endian
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
In commit 54a702f705 ("kbuild: mark $(targets) as .SECONDARY and
remove .PRECIOUS markers"), I missed one important feature of the
.SECONDARY target:
.SECONDARY with no prerequisites causes all targets to be
treated as secondary.
... which agrees with the policy of Kbuild.
Let's move it to scripts/Kbuild.include, with no prerequisites.
Note:
If an intermediate file is generated by $(call if_changed,...), you
still need to add it to "targets" so its .*.cmd file is included.
The arm/arm64 crypto files are generated by $(call cmd,shipped),
so they do not need to be added to "targets", but need to be added
to "clean-files" so "make clean" can properly clean them away.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
In preparation for adding XChaCha12 support, rename/refactor
chacha20-generic to support different numbers of rounds. The
justification for needing XChaCha12 support is explained in more detail
in the patch "crypto: chacha - add XChaCha12 support".
The only difference between ChaCha{8,12,20} are the number of rounds
itself; all other parts of the algorithm are the same. Therefore,
remove the "20" from all definitions, structures, functions, files, etc.
that will be shared by all ChaCha versions.
Also make ->setkey() store the round count in the chacha_ctx (previously
chacha20_ctx). The generic code then passes the round count through to
chacha_block(). There will be a ->setkey() function for each explicitly
allowed round count; the encrypt/decrypt functions will be the same. I
decided not to do it the opposite way (same ->setkey() function for all
round counts, with different encrypt/decrypt functions) because that
would have required more boilerplate code in architecture-specific
implementations of ChaCha and XChaCha.
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>
Commit 2e5d2f33d1 ("crypto: arm64/aes-blk - improve XTS mask handling")
optimized away some reloads of the XTS mask vector, but failed to take
into account that calls into the XTS en/decrypt routines will take a
slightly different code path if a single block of input is split across
different buffers. So let's ensure that the first load occurs
unconditionally, and move the reload to the end so it doesn't occur
needlessly.
Fixes: 2e5d2f33d1 ("crypto: arm64/aes-blk - improve XTS mask handling")
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
In the new arm64 CTS-CBC implementation, return an error code rather
than crashing on inputs shorter than AES_BLOCK_SIZE bytes. Also set
cra_blocksize to AES_BLOCK_SIZE (like is done in the cts template) to
indicate the minimum input size.
Fixes: dd597fb33f ("crypto: arm64/aes-blk - add support for CTS-CBC mode")
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>
The Crypto Extension instantiation of the aes-modes.S collection of
skciphers uses only 15 NEON registers for the round key array, whereas
the pure NEON flavor uses 16 NEON registers for the AES S-box.
This means we have a spare register available that we can use to hold
the XTS mask vector, removing the need to reload it at every iteration
of the inner loop.
Since the pure NEON version does not permit this optimization, tweak
the macros so we can factor out this functionality. Also, replace the
literal load with a short sequence to compose the mask vector.
On Cortex-A53, this results in a ~4% speedup.
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Currently, we rely on the generic CTS chaining mode wrapper to
instantiate the cts(cbc(aes)) skcipher. Due to the high performance
of the ARMv8 Crypto Extensions AES instructions (~1 cycles per byte),
any overhead in the chaining mode layers is amplified, and so it pays
off considerably to fold the CTS handling into the SIMD routines.
On Cortex-A53, this results in a ~50% speedup for smaller input sizes.
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The reasoning of commit f10dc56c64 ("crypto: arm64 - revert NEON yield
for fast AEAD implementations") applies equally to skciphers: the walk
API already guarantees that the input size of each call into the NEON
code is bounded to the size of a page, and so there is no need for an
additional TIF_NEED_RESCHED flag check inside the inner loop. So revert
the skcipher changes to aes-modes.S (but retain the mac ones)
This partially reverts commit 0c8f838a52.
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
For some reason, the asmlinkage prototypes of the NEON routines take
u8[] arguments for the round key arrays, while the actual round keys
are arrays of u32, and so passing them into those routines requires
u8* casts at each occurrence. Fix that.
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The arm64 implementation of the CRC-T10DIF algorithm uses the 64x64 bit
polynomial multiplication instructions, which are optional in the
architecture, and if these instructions are not available, we fall back
to the C routine which is slow and inefficient.
So let's reuse the 64x64 bit PMULL alternative from the GHASH driver that
uses a sequence of ~40 instructions involving 8x8 bit PMULL and some
shifting and masking. This is a lot slower than the original, but it is
still twice as fast as the current [unoptimized] C code on Cortex-A53,
and it is time invariant and much easier on the D-cache.
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Reorganize the CRC-T10DIF asm routine so we can easily instantiate an
alternative version based on 8x8 polynomial multiplication in a
subsequent patch.
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Now that the scalar fallbacks have been moved out of this driver into
the core crc32()/crc32c() routines, we are left with a CRC32 crypto API
driver for arm64 that is based only on 64x64 polynomial multiplication,
which is an optional instruction in the ARMv8 architecture, and is less
and less likely to be available on cores that do not also implement the
CRC32 instructions, given that those are mandatory in the architecture
as of ARMv8.1.
Since the scalar instructions do not require the special handling that
SIMD instructions do, and since they turn out to be considerably faster
on some cores (Cortex-A53) as well, there is really no point in keeping
this code around so let's just remove it.
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Replace the literal load of the addend vector with a sequence that
performs each add individually. This sequence is only 2 instructions
longer than the original, and 2% faster on Cortex-A53.
This is an improvement by itself, but also works around a Clang issue,
whose integrated assembler does not implement the GNU ARM asm syntax
completely, and does not support the =literal notation for FP registers
(more info at https://bugs.llvm.org/show_bug.cgi?id=38642)
Cc: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.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>
Commit 71e52c278c ("crypto: arm64/aes-ce-gcm - operate on
two input blocks at a time") modified the granularity at which
the AES/GCM code processes its input to allow subsequent changes
to be applied that improve performance by using aggregation to
process multiple input blocks at once.
For this reason, it doubled the algorithm's 'chunksize' property
to 2 x AES_BLOCK_SIZE, but retained the non-SIMD fallback path that
processes a single block at a time. In some cases, this violates the
skcipher scatterwalk API, by calling skcipher_walk_done() with a
non-zero residue value for a chunk that is expected to be handled
in its entirety. This results in a WARN_ON() to be hit by the TLS
self test code, but is likely to break other user cases as well.
Unfortunately, none of the current test cases exercises this exact
code path at the moment.
Fixes: 71e52c278c ("crypto: arm64/aes-ce-gcm - operate on two ...")
Reported-by: Vakul Garg <vakul.garg@nxp.com>
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Tested-by: Vakul Garg <vakul.garg@nxp.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
ARMv8.2 specifies special instructions for the SM3 cryptographic hash
and the SM4 symmetric cipher. While it is unlikely that a core would
implement one and not the other, we should only use SM4 instructions
if the SM4 CPU feature bit is set, and we currently check the SM3
feature bit instead. So fix that.
Fixes: e99ce921c4 ("crypto: arm64 - add support for SM4...")
Cc: <stable@vger.kernel.org>
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Enhance the GHASH implementation that uses 64-bit polynomial
multiplication by adding support for 4-way aggregation. This
more than doubles the performance, from 2.4 cycles per byte
to 1.1 cpb on Cortex-A53.
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Checking the TIF_NEED_RESCHED flag is disproportionately costly on cores
with fast crypto instructions and comparatively slow memory accesses.
On algorithms such as GHASH, which executes at ~1 cycle per byte on
cores that implement support for 64 bit polynomial multiplication,
there is really no need to check the TIF_NEED_RESCHED particularly
often, and so we can remove the NEON yield check from the assembler
routines.
However, unlike the AEAD or skcipher APIs, the shash/ahash APIs take
arbitrary input lengths, and so there needs to be some sanity check
to ensure that we don't hog the CPU for excessive amounts of time.
So let's simply cap the maximum input size that is processed in one go
to 64 KB.
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Squeeze out another 5% of performance by minimizing the number
of invocations of kernel_neon_begin()/kernel_neon_end() on the
common path, which also allows some reloads of the key schedule
to be optimized away.
The resulting code runs at 2.3 cycles per byte on a Cortex-A53.
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Implement a faster version of the GHASH transform which amortizes
the reduction modulo the characteristic polynomial across two
input blocks at a time.
On a Cortex-A53, the gcm(aes) performance increases 24%, from
3.0 cycles per byte to 2.4 cpb for large input sizes.
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Update the core AES/GCM transform and the associated plumbing to operate
on 2 AES/GHASH blocks at a time. By itself, this is not expected to
result in a noticeable speedup, but it paves the way for reimplementing
the GHASH component using 2-way aggregation.
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
As it turns out, checking the TIF_NEED_RESCHED flag after each
iteration results in a significant performance regression (~10%)
when running fast algorithms (i.e., ones that use special instructions
and operate in the < 4 cycles per byte range) on in-order cores with
comparatively slow memory accesses such as the Cortex-A53.
Given the speed of these ciphers, and the fact that the page based
nature of the AEAD scatterwalk API guarantees that the core NEON
transform is never invoked with more than a single page's worth of
input, we can estimate the worst case duration of any resulting
scheduling blackout: on a 1 GHz Cortex-A53 running with 64k pages,
processing a page's worth of input at 4 cycles per byte results in
a delay of ~250 us, which is a reasonable upper bound.
So let's remove the yield checks from the fused AES-CCM and AES-GCM
routines entirely.
This reverts commit 7b67ae4d5c and
partially reverts commit 7c50136a8a.
Fixes: 7c50136a8a ("crypto: arm64/aes-ghash - yield NEON after every ...")
Fixes: 7b67ae4d5c ("crypto: arm64/aes-ccm - yield NEON after every ...")
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Calling pmull_gcm_encrypt_block() requires kernel_neon_begin() and
kernel_neon_end() to be used since the routine touches the NEON
register file. Add the missing calls.
Also, since NEON register contents are not preserved outside of
a kernel mode NEON region, pass the key schedule array again.
Fixes: 7c50136a8a ("crypto: arm64/aes-ghash - yield NEON after every ...")
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Will Deacon <will.deacon@arm.com>
