This patch converts xts over to the skcipher interface. It also
optimises the implementation to be based on ECB instead of the
underlying cipher. For compatibility the existing naming scheme
of xts(aes) is maintained as opposed to the more obvious one of
xts(ecb(aes)).
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
This patch converts lrw over to the skcipher interface. It also
optimises the implementation to be based on ECB instead of the
underlying cipher. For compatibility the existing naming scheme
of lrw(aes) is maintained as opposed to the more obvious one of
lrw(ecb(aes)).
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
This patch makes use of the new skcipher walk interface instead of
the obsolete blkcipher walk interface.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
This patch adds the skcipher walk interface which replaces both
blkcipher walk and ablkcipher walk. Just like blkcipher walk it
can also be used for AEAD algorithms.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
This integrates both the accelerated scalar and the NEON implementations
of SHA-224/256 as well as SHA-384/512 from the OpenSSL project.
Relative performance compared to the respective generic C versions:
| SHA256-scalar | SHA256-NEON* | SHA512 |
------------+-----------------+--------------+----------+
Cortex-A53 | 1.63x | 1.63x | 2.34x |
Cortex-A57 | 1.43x | 1.59x | 1.95x |
Cortex-A73 | 1.26x | 1.56x | ? |
The core crypto code was authored by Andy Polyakov of the OpenSSL
project, in collaboration with whom the upstream code was adapted so
that this module can be built from the same version of sha512-armv8.pl.
The version in this patch was taken from OpenSSL commit 32bbb62ea634
("sha/asm/sha512-armv8.pl: fix big-endian support in __KERNEL__ case.")
* The core SHA algorithm is fundamentally sequential, but there is a
secondary transformation involved, called the schedule update, which
can be performed independently. The NEON version of SHA-224/SHA-256
only implements this part of the algorithm using NEON instructions,
the sequential part is always done using scalar instructions.
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
As hw_random core calls ->read with max > 32 or more, make it explicit.
Also remove checks involving 'max' being less than 8.
Signed-off-by: PrasannaKumar Muralidharan <prasannatsmkumar@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The CTR DRBG segments the number of random bytes to be generated into
128 byte blocks. The current code misses the advancement of the output
buffer pointer when the requestor asks for more than 128 bytes of data.
In this case, the next 128 byte block of random numbers is copied to
the beginning of the output buffer again. This implies that only the
first 128 bytes of the output buffer would ever be filled.
The patch adds the advancement of the buffer pointer to fill the entire
buffer.
Signed-off-by: Stephan Mueller <smueller@chronox.de>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
First up, clean up the generated .S files properly on a 'make clean'.
Secondly, force re-generation of these files when building for different
endian-ness than what was built previously. Finally, generate the new
files in the build tree, rather than the source tree.
Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Current multi-buffer hash implementations have a restriction on the total
length of a hash job to 512MB. Hashing larger buffers will result in an
incorrect hash. This extends the limit to 2^62 - 1.
Signed-off-by: Greg Tucker <greg.b.tucker@intel.com>
Acked-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
GF(2^128) multiplication tables are typically used for secret
information, so it's a good idea to zero them on free.
Signed-off-by: Alex Cope <alexcope@google.com>
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Shared descriptors used by ahash_final() and ahash_finup()
are identical, thus get rid of one of them (sh_desc_finup).
Signed-off-by: Horia Geantă <horia.geanta@nxp.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The pointer to the descriptor buffer is not touched,
it always points to start of the descriptor buffer.
Thus, make it const.
Signed-off-by: Horia Geantă <horia.geanta@nxp.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
sec4_sg_entry structure is used only by helper functions in sg_sw_sec4.h.
Since SEC HW S/G entries are to be manipulated only indirectly, via these
functions, move sec4_sg_entry to the corresponding header.
Signed-off-by: Horia Geantă <horia.geanta@nxp.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
This reverts commit 66d2e20280.
Quoting from Russell's findings:
https://www.mail-archive.com/linux-crypto@vger.kernel.org/msg21136.html
[quote]
Okay, I've re-tested, using a different way of measuring, because using
openssl speed is impractical for off-loaded engines. I've decided to
use this way to measure the performance:
dd if=/dev/zero bs=1048576 count=128 | /usr/bin/time openssl dgst -md5
For the threaded IRQs case gives:
0.05user 2.74system 0:05.30elapsed 52%CPU (0avgtext+0avgdata 2400maxresident)k
0.06user 2.52system 0:05.18elapsed 49%CPU (0avgtext+0avgdata 2404maxresident)k
0.12user 2.60system 0:05.61elapsed 48%CPU (0avgtext+0avgdata 2460maxresident)k
=> 5.36s => 25.0MB/s
and the tasklet case:
0.08user 2.53system 0:04.83elapsed 54%CPU (0avgtext+0avgdata 2468maxresident)k
0.09user 2.47system 0:05.16elapsed 49%CPU (0avgtext+0avgdata 2368maxresident)k
0.10user 2.51system 0:04.87elapsed 53%CPU (0avgtext+0avgdata 2460maxresident)k
=> 4.95 => 27.1MB/s
which corresponds to an 8% slowdown for the threaded IRQ case. So,
tasklets are indeed faster than threaded IRQs.
[...]
I think I've proven from the above that this patch needs to be reverted
due to the performance regression, and that there _is_ most definitely
a deterimental effect of switching from tasklets to threaded IRQs.
[/quote]
Signed-off-by: Horia Geantă <horia.geanta@nxp.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
alkcipher_edesc_alloc() and ablkcipher_giv_edesc_alloc() don't
free / unmap resources on error path:
- dmap_map_sg() could fail, thus make sure the return value is checked
- unmap DMA mappings in case of error
Signed-off-by: Horia Geantă <horia.geanta@nxp.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
ERRID is a 4-bit field.
Since err_id values are in [0..15] and err_id_list array size is 16,
the condition "err_id < ARRAY_SIZE(err_id_list)" is always true.
Signed-off-by: Horia Geantă <horia.geanta@nxp.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
REG3 no longer needs to be updated, since it's not used after that.
This shared descriptor command is a leftover of the conversion to
AEAD interface.
Fixes: 479bcc7c5b "crypto: caam - Convert authenc to new AEAD interface"
Signed-off-by: Horia Geantă <horia.geanta@nxp.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Fix the following smatch warnings:
drivers/crypto/caam/caamalg.c:2350 aead_edesc_alloc() warn: we tested 'src_nents' before and it was 'true'
drivers/crypto/caam/caamrng.c:351 caam_rng_init() error: no modifiers for allocation.
Signed-off-by: Horia Geantă <horia.geanta@nxp.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Herbert Xu
Jean Delvare
Ard Biesheuvel
PrasannaKumar Muralidharan
Stephan Mueller
Naveen N. Rao
Greg Tucker
Alex Cope
Wei Yongjun
Geliang Tang
Horia Geantă