The use of __sync functions for atomic memory access is not
supported in the kernel, and can result in a link error depending
on configuration:

ERROR: "__tsan_atomic32_compare_exchange_strong" [drivers/crypto/hisilicon/sec2/hisi_sec2.ko] undefined!
ERROR: "__tsan_atomic64_fetch_add" [drivers/crypto/hisilicon/sec2/hisi_sec2.ko] undefined!

Use the kernel's own atomic interfaces instead. This way the
debugfs interface actually reads the counter atomically.

Fixes: 416d8220 ("crypto: hisilicon - add HiSilicon SEC V2 driver")
Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

Somehow this was forgotten when Zinc was being split into oddly shaped
pieces, resulting in linker errors. The x86_64 glue has a specific key
generation implementation, but the Arm one does not. However, it can
still receive the NEON speedups by calling the ordinary DH function
using the base point.
Signed-off-by: NJason A. Donenfeld <Jason@zx2c4.com>
Acked-by: NArd Biesheuvel <ardb@kernel.org>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

The hisi_acc_create_sgl_pool() function returns error pointers, it never
returns NULL pointers.

Fixes: 416d8220 ("crypto: hisilicon - add HiSilicon SEC V2 driver")
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

The talitos driver needs to select LIB_DES as it needs calls
des_expand_key.

Fixes: 9d574ae8 ("crypto: talitos/des - switch to new...")
Cc: <stable@vger.kernel.org>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Acked-by: NArd Biesheuvel <ardb@kernel.org>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

For glue code that's used by Zinc, the actual Crypto API functions might
not necessarily exist, and don't need to exist either. Before this
patch, there are valid build configurations that lead to a unbuildable
kernel. This fixes it to conditionalize those symbols on the existence
of the proper config entry.
Signed-off-by: NJason A. Donenfeld <Jason@zx2c4.com>
Acked-by: NArd Biesheuvel <ardb@kernel.org>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

In the vmx crypto Makefile we assign to a variable called TARGET and
pass that to the aesp8-ppc.pl and ghashp8-ppc.pl scripts.

The variable is meant to describe what flavour of powerpc we're
building for, eg. either 32 or 64-bit, and big or little endian.

Unfortunately TARGET is a fairly common name for a make variable, and
if it happens that TARGET is specified as a command line parameter to
make, the value specified on the command line will override our value.

In particular this can happen if the kernel Makefile is driven by an
external Makefile that uses TARGET for something.

This leads to weird build failures, eg:
  nonsense  at /build/linux/drivers/crypto/vmx/ghashp8-ppc.pl line 45.
  /linux/drivers/crypto/vmx/Makefile:20: recipe for target 'drivers/crypto/vmx/ghashp8-ppc.S' failed

Which shows that we passed an empty value for $(TARGET) to the perl
script, confirmed with make V=1:

  perl /linux/drivers/crypto/vmx/ghashp8-ppc.pl  > drivers/crypto/vmx/ghashp8-ppc.S

We can avoid this confusion by using override, to tell make that we
don't want anything to override our variable, even a value specified
on the command line. We can also use a less common name, given the
script calls it "flavour", let's use that.
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

Use chacha20_crypt() instead of chacha_crypt(), since it's not really
appropriate for users of the ChaCha library API to be passing the number
of rounds as an argument.
Signed-off-by: NEric Biggers <ebiggers@google.com>
Acked-by: NArd Biesheuvel <ardb@kernel.org>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

It's not valid to call crypto_unregister_skciphers() without a prior
call to crypto_register_skciphers().

Fixes: 84e03fa3 ("crypto: x86/chacha - expose SIMD ChaCha routine as library function")
Signed-off-by: NEric Biggers <ebiggers@google.com>
Acked-by: NArd Biesheuvel <ardb@kernel.org>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

Use chacha20_setkey() and chacha12_setkey() from
<crypto/internal/chacha.h> instead of defining them again in
chacha_generic.c.
Signed-off-by: NEric Biggers <ebiggers@google.com>
Acked-by: NArd Biesheuvel <ardb@kernel.org>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

On big endian kernel, the GXL crypto driver does not works.
This patch do the necessary modification to permit it to work on BE
kernel (removing bitfield and adds some cpu_to_le32).

Fixes: 48fe583f ("crypto: amlogic - Add crypto accelerator for amlogic GXL")
Signed-off-by: NCorentin Labbe <clabbe@baylibre.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

On big endian kernel, the sun8i-ce crypto driver does not works.
This patch do the necessary modification to permit it to work on BE
kernel (setting descriptor entries as __le32 and adding some cpu_to_le32)

Fixes: 06f751b6 ("crypto: allwinner - Add sun8i-ce Crypto Engine")
Signed-off-by: NCorentin Labbe <clabbe.montjoie@gmail.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

Another instance of CRYPTO_BLKCIPHER made it in just after it was
renamed to CRYPTO_SKCIPHER.  Fix it.
Signed-off-by: NEric Biggers <ebiggers@google.com>
Acked-by: NArd Biesheuvel <ardb@kernel.org>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch enables COMPILE_TEST on the ks-sa-rng driver.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Reviewed-by: NAlexander Sverdlin <alexander.sverdlin@nokia.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

kfree has taken null pointer check into account. so it is safe to
remove the unnecessary check.
Signed-off-by: NChen Wandun <chenwandun@huawei.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

In case of in-place decryption, the "lastc" buffer is used to copy
the last ciphertext block before the decryption of the message. It
is later used to update the req->iv of the skcipher_request.

"lastc" variable is not used to interact with the hardware, there
is no restriction to be of type "u32". Change the type of "lastc"
to "u8".
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

The req->iv of the skcipher_request is expected to contain the
last ciphertext block when the {en,de}crypt operation is done.
In case of in-place decryption, copy the ciphertext in an
intermediate buffer before decryption.

This fixes the following tcrypt tests:
alg: skcipher: atmel-cbc-des encryption test failed (wrong output IV) on test vector 0, cfg="in-place"
00000000: fe dc ba 98 76 54 32 10
alg: skcipher: atmel-cbc-tdes encryption test failed (wrong output IV) on test vector 0, cfg="in-place"
00000000: 7d 33 88 93 0f 93 b2 42
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

When testing BigEndian kernel, the sun4i-ss was failling all crypto
tests.
This patch fix endian issues with it.

Fixes: 6298e948 ("crypto: sunxi-ss - Add Allwinner Security System crypto accelerator")
Signed-off-by: NCorentin Labbe <clabbe.montjoie@gmail.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

Drop the "Invalid keylen" message to debug level, it adds no value, and
when CRYPTO_EXTRA_TEST is enabled, it floods the console.
Signed-off-by: NCorentin Labbe <clabbe.montjoie@gmail.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

The size of the digest is different between MD5 and SHA1 so instead of
using the higher value (5 words), let's use crypto_ahash_digestsize().
Signed-off-by: NCorentin Labbe <clabbe.montjoie@gmail.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

The driver now compile without warnings on 64bits, we can remove the
!64BIT condition.
Signed-off-by: NCorentin Labbe <clabbe.montjoie@gmail.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

If you try to compile this driver on a 64-bit platform then you
will get warnings because it mixes size_t with unsigned int which
only works on 32-bit.

This patch fixes all of the warnings on sun4i-ss-hash.c.
Signed-off-by: NCorentin Labbe <clabbe.montjoie@gmail.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

Here adds maintainer information for security engine driver.
Signed-off-by: NZaibo Xu <xuzaibo@huawei.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

The HiSilicon SEC engine driver uses DebugFS
to provide main debug information for user space.
Signed-off-by: NZaibo Xu <xuzaibo@huawei.com>
Signed-off-by: NLongfang Liu <liulongfang@huawei.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This Documentation is for HiSilicon SEC DebugFS.
Signed-off-by: NLongfang Liu <liulongfang@huawei.com>
Signed-off-by: NKai Ye <yekai13@huawei.com>
Reviewed-by: NZaibo Xu <xuzaibo@huawei.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

HiSilicon SEC engine supports PCI SRIOV. This patch enable this feature.
User can enable VFs and pass through them to VM, same SEC driver can work
in VM to provide skcipher algorithms.
Signed-off-by: NZaibo Xu <xuzaibo@huawei.com>
Signed-off-by: NLongfang Liu <liulongfang@huawei.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

SEC driver provides PCIe hardware device initiation with
AES, SM4, and 3DES skcipher algorithms registered to Crypto.
It uses Hisilicon QM as interface to CPU.
Signed-off-by: NZaibo Xu <xuzaibo@huawei.com>
Signed-off-by: NLongfang Liu <liulongfang@huawei.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

If CRYPTO_DEV_ATMEL_AUTHENC is m, CRYPTO_DEV_ATMEL_SHA is m,
but CRYPTO_DEV_ATMEL_AES is y, building will fail:

drivers/crypto/atmel-aes.o: In function `atmel_aes_authenc_init_tfm':
atmel-aes.c:(.text+0x670): undefined reference to `atmel_sha_authenc_get_reqsize'
atmel-aes.c:(.text+0x67a): undefined reference to `atmel_sha_authenc_spawn'
drivers/crypto/atmel-aes.o: In function `atmel_aes_authenc_setkey':
atmel-aes.c:(.text+0x7e5): undefined reference to `atmel_sha_authenc_setkey'

Make CRYPTO_DEV_ATMEL_AUTHENC depend on CRYPTO_DEV_ATMEL_AES,
and select CRYPTO_DEV_ATMEL_SHA and CRYPTO_AUTHENC for it under there.
Reported-by: NHulk Robot <hulkci@huawei.com>
Suggested-by: NHerbert Xu <herbert@gondor.apana.org.au>
Fixes: 89a82ef8 ("crypto: atmel-authenc - add support to...")
Signed-off-by: NYueHaibing <yuehaibing@huawei.com>
Reviewed-by: NTudor Ambarus <tudor.ambarus@microchip.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

dma_request_slave_channel_reason() is:
#define dma_request_slave_channel_reason(dev, name) \
	dma_request_chan(dev, name)
Signed-off-by: NPeter Ujfalusi <peter.ujfalusi@ti.com>
Reviewed-by: NVinod Koul <vkoul@kernel.org>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

Currently, there is no privilege separation of the SEV command; you can
run them all or none of them. This is less than ideal because it means
that a compromise of the code which launches VMs could make permanent
change to the SEV certifcate chain which will affect others.

These commands are required to attest the VM environment:
 - SEV_PDH_CERT_EXPORT
 - SEV_PLATFORM_STATUS
 - SEV_GET_{ID,ID2}

These commands manage the SEV certificate chain:
 - SEV_PEK_CERR_IMPORT
 - SEV_FACTORY_RESET
 - SEV_PEK_GEN
 - SEV_PEK_CSR
 - SEV_PDH_GEN

Lets add the CAP_SYS_ADMIN check for the group of the commands which alters
the SEV certificate chain to provide some level of privilege separation.

Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: Gary Hook <gary.hook@amd.com>
Cc: Erdem Aktas <erdemaktas@google.com>
Cc: Tom Lendacky <Thomas.Lendacky@amd.com>
Tested-by: NDavid Rientjes <rientjes@google.com>
Co-developed-by: NDavid Rientjes <rientjes@google.com>
Signed-off-by: NDavid Rientjes <rientjes@google.com>
Signed-off-by: NBrijesh Singh <brijesh.singh@amd.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

The TFM context can be renamed to a more appropriate name and the local
varaibles as well, using 'tctx' which seems to be more common than
'mctx'.

The _setkey callback was the last one without the blake2b_ prefix,
rename that too.
Signed-off-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

Now that there's only one call to blake2b_update, we can merge it to the
callback and simplify. The empty input check is split and the rest of
code un-indented.
Signed-off-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

The helper is trival and called once, inlining makes things simpler.
There's a comment to tie it back to the idea behind the code.
Signed-off-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

All the code for param block has been inlined, last_node and outlen from
the state are not used or have become redundant due to other code.
Remove it.
Signed-off-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

The keyed init writes the key bytes to the input buffer and does an
update. We can do that in two ways: fill the buffer and update
immediatelly. This is what current blake2b_init_key does. Any other
following _update or _final will continue from the updated state.

The other way is to write the key and set the number of bytes to process
at the next _update or _final, lazy evaluation. Which leads to the the
simplified code in this patch.
Signed-off-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

The call chain from blake2b_init can be simplified because the param
block is effectively zeros, besides the key.

- blake2b_init0 zeroes state and sets IV
- blake2b_init sets up param block with defaults (key and some 1s)
- init with key, write it to the input buffer and recalculate state

So the compact way is to zero out the state and initialize index 0 of
the state directly with the non-zero values and the key.
Signed-off-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

blake2b_final is called only once, merge it to the crypto API callback
and simplify. This avoids the temporary buffer and swaps the bytes of
internal buffer.
Signed-off-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

Use kmemdup rather than duplicating its implementation
Signed-off-by: NYueHaibing <yuehaibing@huawei.com>
Acked-by: NCorentin Labbe <clabbe.montjoie@gmail.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

If you try to compile this driver on a 64-bit platform then you
will get warnings because it mixes size_t with unsigned int which
only works on 32-bit.

This patch fixes all of the warnings.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Acked-by: NCorentin Labbe <clabbe.montjoie@gmail.com>
Tested-by: NCorentin Labbe <clabbe.montjoie@gmail.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

The ccree driver does not use the ablkcipher interface but contains
a rudimentary reference to it in the naming of an unrelated macro.
Let's rename it to avoid confusion.
Acked-by: NGilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: NArd Biesheuvel <ardb@kernel.org>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

Now that all users of the deprecated ablkcipher interface have been
moved to the skcipher interface, ablkcipher is no longer used and
can be removed.
Reviewed-by: NEric Biggers <ebiggers@kernel.org>
Signed-off-by: NArd Biesheuvel <ardb@kernel.org>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>