linux.git/crypto/tcrypt.c, branch v5.14

crypto: tcrypt - enable tests for xxhash and blake2

2021-05-28T07:11:45+00:00

Fill some of the recently freed up slots in tcrypt with xxhash64 and
blake2b/blake2s, so we can easily benchmark their kernel implementations
from user space.

Signed-off-by: Ard Biesheuvel 
Signed-off-by: Herbert Xu

crypto: serpent - get rid of obsolete tnepres variant

2021-02-10T06:55:56+00:00

It is not trivial to trace back why exactly the tnepres variant of
serpent was added ~17 years ago - Google searches come up mostly empty,
but it seems to be related with the 'kerneli' version, which was based
on an incorrect interpretation of the serpent spec.

In other words, nobody is likely to care anymore today, so let's get rid
of it.

Signed-off-by: Ard Biesheuvel 
Signed-off-by: Herbert Xu

crypto: salsa20 - remove Salsa20 stream cipher algorithm

2021-01-29T05:07:04+00:00

Salsa20 is not used anywhere in the kernel, is not suitable for disk
encryption, and widely considered to have been superseded by ChaCha20.
So let's remove it.

Signed-off-by: Ard Biesheuvel 
Acked-by:  Mike Snitzer 
Signed-off-by: Herbert Xu

crypto: tgr192 - remove Tiger 128/160/192 hash algorithms

2021-01-29T05:07:04+00:00

Tiger is never referenced anywhere in the kernel, and unlikely
to be depended upon by userspace via AF_ALG. So let's remove it.

Signed-off-by: Ard Biesheuvel 
Signed-off-by: Herbert Xu

crypto: rmd320 - remove RIPE-MD 320 hash algorithm

2021-01-29T05:07:04+00:00

RIPE-MD 320 is never referenced anywhere in the kernel, and unlikely
to be depended upon by userspace via AF_ALG. So let's remove it

Signed-off-by: Ard Biesheuvel 
Signed-off-by: Herbert Xu

crypto: rmd256 - remove RIPE-MD 256 hash algorithm

2021-01-29T05:07:03+00:00

RIPE-MD 256 is never referenced anywhere in the kernel, and unlikely
to be depended upon by userspace via AF_ALG. So let's remove it

Signed-off-by: Ard Biesheuvel 
Signed-off-by: Herbert Xu

crypto: rmd128 - remove RIPE-MD 128 hash algorithm

2021-01-29T05:07:03+00:00

RIPE-MD 128 is never referenced anywhere in the kernel, and unlikely
to be depended upon by userspace via AF_ALG. So let's remove it.

Signed-off-by: Ard Biesheuvel 
Signed-off-by: Herbert Xu

crypto: tcrypt - avoid signed overflow in byte count

2021-01-02T21:41:34+00:00

The signed long type used for printing the number of bytes processed in
tcrypt benchmarks limits the range to -/+ 2 GiB, which is not sufficient
to cover the performance of common accelerated ciphers such as AES-NI
when benchmarked with sec=1. So switch to u64 instead.

While at it, fix up a missing printk->pr_cont conversion in the AEAD
benchmark.

Signed-off-by: Ard Biesheuvel 
Signed-off-by: Herbert Xu

crypto: tcrypt - include 1420 byte blocks in aead and skcipher benchmarks

2020-11-27T06:13:45+00:00

WireGuard and IPsec both typically operate on input blocks that are
~1420 bytes in size, given the default Ethernet MTU of 1500 bytes and
the overhead of the VPN metadata.

Many aead and sckipher implementations are optimized for power-of-2
block sizes, and whether they perform well when operating on 1420
byte blocks cannot be easily extrapolated from the performance on
power-of-2 block size. So let's add 1420 bytes explicitly, and round
it up to the next blocksize multiple of the algo in question if it
does not support 1420 byte blocks.

Signed-off-by: Ard Biesheuvel 
Signed-off-by: Herbert Xu

crypto: tcrypt - don't initialize at subsys_initcall time

2020-11-27T06:13:44+00:00

Commit c4741b2305979 ("crypto: run initcalls for generic implementations
earlier") converted tcrypt.ko's module_init() to subsys_initcall(), but
this was unintentional: tcrypt.ko currently cannot be built into the core
kernel, and so the subsys_initcall() gets converted into module_init()
under the hood. Given that tcrypt.ko does not implement a generic version
of a crypto algorithm that has to be available early during boot, there
is no point in running the tcrypt init code earlier than implied by
module_init().

However, for crypto development purposes, we will lift the restriction
that tcrypt.ko must be built as a module, and when builtin, it makes sense
for tcrypt.ko (which does its work inside the module init function) to run
as late as possible. So let's switch to late_initcall() instead.

Signed-off-by: Ard Biesheuvel 
Reviewed-by: Eric Biggers 
Signed-off-by: Herbert Xu