Set the flag CRYPTO_ALG_ALLOCATES_MEMORY in the crypto drivers that allocate memory. drivers/crypto/allwinner/sun8i-ce/sun8i-ce-core.c: sun8i_ce_cipher drivers/crypto/allwinner/sun8i-ss/sun8i-ss-core.c: sun8i_ss_cipher drivers/crypto/amlogic/amlogic-gxl-core.c: meson_cipher drivers/crypto/axis/artpec6_crypto.c: artpec6_crypto_common_init drivers/crypto/bcm/cipher.c: spu_skcipher_rx_sg_create drivers/crypto/caam/caamalg.c: aead_edesc_alloc drivers/crypto/caam/caamalg_qi.c: aead_edesc_alloc drivers/crypto/caam/caamalg_qi2.c: aead_edesc_alloc drivers/crypto/caam/caamhash.c: hash_digest_key drivers/crypto/cavium/cpt/cptvf_algs.c: process_request drivers/crypto/cavium/nitrox/nitrox_aead.c: nitrox_process_se_request drivers/crypto/cavium/nitrox/nitrox_skcipher.c: nitrox_process_se_request drivers/crypto/ccp/ccp-crypto-aes-cmac.c: ccp_do_cmac_update drivers/crypto/ccp/ccp-crypto-aes-galois.c: ccp_crypto_enqueue_request drivers/crypto/ccp/ccp-crypto-aes-xts.c: ccp_crypto_enqueue_request drivers/crypto/ccp/ccp-crypto-aes.c: ccp_crypto_enqueue_request drivers/crypto/ccp/ccp-crypto-des3.c: ccp_crypto_enqueue_request drivers/crypto/ccp/ccp-crypto-sha.c: ccp_crypto_enqueue_request drivers/crypto/chelsio/chcr_algo.c: create_cipher_wr drivers/crypto/hisilicon/sec/sec_algs.c: sec_alloc_and_fill_hw_sgl drivers/crypto/hisilicon/sec2/sec_crypto.c: sec_alloc_req_id drivers/crypto/inside-secure/safexcel_cipher.c: safexcel_queue_req drivers/crypto/inside-secure/safexcel_hash.c: safexcel_ahash_enqueue drivers/crypto/ixp4xx_crypto.c: ablk_perform drivers/crypto/marvell/cesa/cipher.c: mv_cesa_skcipher_dma_req_init drivers/crypto/marvell/cesa/hash.c: mv_cesa_ahash_dma_req_init drivers/crypto/marvell/octeontx/otx_cptvf_algs.c: create_ctx_hdr drivers/crypto/n2_core.c: n2_compute_chunks drivers/crypto/picoxcell_crypto.c: spacc_sg_to_ddt drivers/crypto/qat/qat_common/qat_algs.c: qat_alg_skcipher_encrypt drivers/crypto/qce/skcipher.c: qce_skcipher_async_req_handle drivers/crypto/talitos.c : talitos_edesc_alloc drivers/crypto/virtio/virtio_crypto_algs.c: __virtio_crypto_skcipher_do_req drivers/crypto/xilinx/zynqmp-aes-gcm.c: zynqmp_aes_aead_cipher Signed-off-by: Mikulas Patocka <mpatocka@redhat.com> [EB: avoid overly-long lines] Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Even though the qce driver implements asynchronous versions of ecb(aes), cbc(aes)and xts(aes), the fallbacks it allocates are required to be synchronous. Given that SIMD based software implementations are usually asynchronous as well, even though they rarely complete asynchronously (this typically only happens in cases where the request was made from softirq context, while SIMD was already in use in the task context that it interrupted), these implementations are disregarded, and either the generic C version or another table based version implemented in assembler is selected instead. Since falling back to synchronous AES is not only a performance issue, but potentially a security issue as well (due to the fact that table based AES is not time invariant), let's fix this, by allocating an ordinary skcipher as the fallback, and invoke it with the completion routine that was given to the outer request. While at it, remove the pointless memset() from qce_skcipher_init(), and remove the call to it qce_skcipher_init_fallback(). Signed-off-by: Ard Biesheuvel <ardb@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Crypto test driver's test_ahash_speed calls crypto_ahash_update and crypto_ahash_final APIs repeatedly for all the available test vector buffer lengths. if we mark the end for scatterlist based on the current vector size then the subsequent vectors might fail if the later buffer lengths are higher. To avoid this, in qce do not mark the end of scatterlist in update API, the qce_ahash_async_req_handle API already takes care of this copying right amount of buffer from the request scatter list. Signed-off-by: Sivaprakash Murugesan <sivaprak@codeaurora.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
crypto testmgr deliberately corrupts the request context while passing vectors to the import. This is to make sure that drivers do not rely on request but they take all the necessary input from io vec passed to it. qce casts the request context from request parameter, since it is corrupted the sub squent hash request fails and qce hangs. To avoid this re-initialize request context on import. The qce import API alreasy takes care of taking the input vectors from passed io vec. Signed-off-by: Sivaprakash Murugesan <sivaprak@codeaurora.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
crypto test module passes zero length vectors as test input to sha-1 and sha-256. To provide correct output for these vectors, hash zero support has been added as in other crypto drivers. Signed-off-by: Sivaprakash Murugesan <sivaprak@codeaurora.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The CONFIG_CRYPTO_DEV_QCE_SOFT_THRESHOLD symbol was renamed during development, but the stringify reference in the parameter description sneaked by unnoticed. Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
QCE hangs when presented with an AES-XTS request whose length is larger than QCE_SECTOR_SIZE (512-bytes), and is not a multiple of it. Let the fallback cipher handle them. Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Process small blocks using the fallback cipher, as a workaround for an observed failure (DMA-related, apparently) when computing the GCM ghash key. This brings a speed gain as well, since it avoids the latency of using the hardware engine to process small blocks. Using software for all 16-byte requests would be enough to make GCM work, but to increase performance, a larger threshold would be better. Measuring the performance of supported ciphers with openssl speed, software matches hardware at around 768-1024 bytes. Considering the 256-bit ciphers, software is 2-3 times faster than qce at 256-bytes, 30% faster at 512, and about even at 768-bytes. With 128-bit keys, the break-even point would be around 1024-bytes. This adds the 'aes_sw_max_len' parameter, to set the largest request length processed by the software fallback. Its default is being set to 512 bytes, a little lower than the break-even point, to balance the cost in CPU usage. Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The qce crypto driver appends an extra entry to the dst sgl, to maintain
private state information.
When the gcm driver sends requests to the ctr skcipher, it passes the
authentication tag after the actual crypto payload, but it must not be
touched.
Commit 1336c2221bee ("crypto: qce - save a sg table slot for result
buf") limited the destination sgl to avoid overwriting the
authentication tag but it assumed the tag would be in a separate sgl
entry.
This is not always the case, so it is better to limit the length of the
destination buffer to req->cryptlen before appending the result buf.
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The CRYPTO_TFM_RES_BAD_KEY_LEN flag was apparently meant as a way to make the ->setkey() functions provide more information about errors. However, no one actually checks for this flag, which makes it pointless. Also, many algorithms fail to set this flag when given a bad length key. Reviewing just the generic implementations, this is the case for aes-fixed-time, cbcmac, echainiv, nhpoly1305, pcrypt, rfc3686, rfc4309, rfc7539, rfc7539esp, salsa20, seqiv, and xcbc. But there are probably many more in arch/*/crypto/ and drivers/crypto/. Some algorithms can even set this flag when the key is the correct length. For example, authenc and authencesn set it when the key payload is malformed in any way (not just a bad length), the atmel-sha and ccree drivers can set it if a memory allocation fails, and the chelsio driver sets it for bad auth tag lengths, not just bad key lengths. So even if someone actually wanted to start checking this flag (which seems unlikely, since it's been unused for a long time), there would be a lot of work needed to get it working correctly. But it would probably be much better to go back to the drawing board and just define different return values, like -EINVAL if the key is invalid for the algorithm vs. -EKEYREJECTED if the key was rejected by a policy like "no weak keys". That would be much simpler, less error-prone, and easier to test. So just remove this flag. Signed-off-by: Eric Biggers <ebiggers@google.com> Reviewed-by: Horia Geantă <horia.geanta@nxp.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>