Add support for SM3 cipher in CryptoCell 713.
Signed-off-by: NYael Chemla <yael.chemla@foss.arm.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

encapsulate set_cipher_mode call with another api,
preparation for specific hash behavior as needed in later patches
when SM3 introduced.
Signed-off-by: NYael Chemla <yael.chemla@foss.arm.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

Adjust hash length such that it will not be fixed and general for all algs.
Instead make it suitable for certain context information.
This is preparation for SM3 support.
Signed-off-by: NYael Chemla <yael.chemla@foss.arm.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

finup() operation was incorrect, padding was missing.
Fix by setting the ccree HW to enable padding.
Signed-off-by: NHadar Gat <hadar.gat@arm.com>
[ gilad@benyossef.com: refactored for better code sharing ]
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Cc: stable@vger.kernel.org
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

Some ahash algorithms set .cra_type = &crypto_ahash_type.  But this is
redundant with the C structure type ('struct ahash_alg'), and
crypto_register_ahash() already sets the .cra_type automatically.
Apparently the useless assignment has just been copy+pasted around.

So, remove the useless assignment from all the ahash algorithms.

This patch shouldn't change any actual behavior.
Signed-off-by: NEric Biggers <ebiggers@google.com>
Acked-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

Many ahash algorithms set .cra_flags = CRYPTO_ALG_TYPE_AHASH.  But this
is redundant with the C structure type ('struct ahash_alg'), and
crypto_register_ahash() already sets the type flag automatically,
clearing any type flag that was already there.  Apparently the useless
assignment has just been copy+pasted around.

So, remove the useless assignment from all the ahash algorithms.

This patch shouldn't change any actual behavior.
Signed-off-by: NEric Biggers <ebiggers@google.com>
Acked-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

Add support for the legacy CryptoCell 630 and 710 revs.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

Add CryptoCell async. hash and HMAC support.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

These lines are less than 80 characters so we don't need to break them
up into chunks.
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Remove include files not needed for compilation.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The ccree driver source files were using an inconsistent
naming convention stemming from what the company was called
when they were added.

Move to a single consistent naming convention for better
code readability.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Fold common code in hash call into service functions.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Fix indentation of some function params in hash code for
better readability.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

hash_init was mapping DMA memory that were then being unmap in
hash_digest/final/finup callbacks, which is against the Crypto API
usage rules (see discussion at
https://www.mail-archive.com/linux-crypto@vger.kernel.org/msg30077.html)

Fix it by moving all buffer mapping/unmapping or each Crypto API op.

This also properly deals with hash_import() not knowing if
hash_init was called or not as it now no longer matters.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Move to allocating the buffers needed for requests as part of
the request structure instead of malloc'ing each one on it's
own, making for simpler (and more efficient) code.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The ccree hash code is using a double buffer to hold data
for processing but manages the buffers and their associated
data count in two separate fields and uses a predicate to
chose which to use.

Move to using a proper 2 members array for a much cleaner code.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Remove unused struct field.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This reverts commit c5f39d07 ("staging: ccree: fix leak of import()
after init()") and commit aece0902 ("staging: ccree: Uninitialized
return in ssi_ahash_import()").

This is the wrong solution and ends up relying on uninitialized memory,
although it was not obvious to me at the time.

Cc: stable@vger.kernel.org
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Crypto API tfm providers are required to provide a backlog
service, if so indicated, that queues up requests in the case
of the provider being busy and processing them later.

The ccree driver did not provide this facility. Add it now.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The send_request() function was handling both synchronous
and asynchronous invocations, but were not handling
the asynchronous case, which may be called in an atomic
context, properly as it was sleeping.

Start to fix the problem by breaking up the two use
cases to separate functions calling a common internal
service function and return error instead of sleeping
for the asynchronous case.

The next patch will complete the fix by implementing
proper backlog handling.

Fixes: abefd674 ("staging: ccree: introduce CryptoCell HW driver").
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The ccree driver was using a DMA operation to copy larval digest
from the ccree SRAM to RAM. Replace it with a simple memcpy.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The ccree driver was allocating memory using GFP_KERNEL flag
always, ignoring the flags set in the crypto request. Fix it
by choosing gfp flags based on crypto request flags.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Remove bogus GFP_DMA flag from memory allocations. ccree driver
does not operate over an ISA or similar limited bus.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Replace verbatim GPL v2 copy with SPDX tag.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Now that the only thing left in ssi_config is the maximum
DMA mask length we get move that to ssi_driver.h and get
rid of the file.

All of ccree is now runtime configurable or under Kbuild control.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The ccree driver has had a none standard sysfs interface for debugging.
Replace it with a proper debugfs interface.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Unify naming convention by renaming all ssi_ vars/structs/enums
and variables to cc_*
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Unify naming convention by renaming all DX macros to CC.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Unify naming convention by renaming all SSI macros to CC.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Remove macro controlling build of various features. This
needs to happen dynamically in registration time.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Fix function call parameter indentation according to coding
style guide lines.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Func definitions in the hash implementation were did not adhere to
coding style. Fix them for better readability.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The hash files were using a naming convention which was inconsistent
(ssi vs. cc), included a useless prefix (ssi_hash) and often used too
long function names producing monster such as
ssi_ahash_get_initial_digest_len_sram_addr() that made the call site
hard to read.

Make the code more readable by switching to a simpler, consistent naming
convention for the file.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Remove a no longer needed abstraction around ccree hash crypto API
internals that used to allow same ops to be used in synchronous and
asynchronous fashion.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The return value isn't initialized on some success paths.

Fixes: c5f39d07 ("staging: ccree: fix leak of import() after init()")
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The ccree code made a lot of use of likely/unlikely qualifiers without
proven measurements showing any benefits. Remove them all until we
see what is justified and what is not.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The ccree drivers was marking a lot of big functions in C file as
static inline for no good reason. Remove the inline qualifier from
any but the few truly single line functions.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

crypto_ahash_import() may be called either after
crypto_ahash_init() or without such call. Right now
we always internally call init() as part of
import(), thus leaking memory and mappings if the
user has already called init() herself.

Fix this by only calling init() internally if the
state is not already initialized.

Fixes: commit 454527d0 ("staging: ccree: fix hash import/export")
Cc: stable <stable@vger.kernel.org>
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Reviewed-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Remove unneeded wrapper function to simplify code.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Rename the too long  SSI_MAX_HASH_OPAD_TMP_KEYS_SIZE
to SSI_MAX_OPAD_KEYS_SIZE for better code readability.
Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>