openbsd/src/lib/libcrypto/evp, branch libressl-v3.0.0

Correct missing test to determine if length is in bytes or in bits.

2019-05-14T15:40:44+00:00

Issue found by Guido Vranken ok tedu@ tb@

Revert the other hunk of r1.36 as well: in the case of CCM, ccm.key is

2019-05-12T15:52:46+00:00

assigned from aesni_ccm_init_key() via CRYPTO_ccm128_init(), so it needs to be copied over... Pointed out by Guido Vranken. ok jsing

Revert part of r1.36: in the case of GCM, gcm.key is assigned from

2019-05-10T18:41:17+00:00

aesni_gcm_init_key() via CRYPTO_gcm128_init(), so it needs to be copied over... Fixes cryptofuzz issue #14352 and likely also #14374. ok beck jsing

Allocate md_data with calloc to avoid use of uninitialised memory.

2019-04-19T17:04:45+00:00

Found by Guido Vranken when fuzzing and trying to use GOST with HMAC. Fix confirmed by Guido; ok tb@

Some more malloc() to calloc() conversions.

2019-04-14T17:27:42+00:00

ok tb@

Remove two pointless chunks of code.

2019-04-14T17:26:27+00:00

This reverts part of OpenSSL c2fd5d79, which added the same code to AES CCM, GCM and XTS. In the case of CCM and GCM nothing assigns {ccm,gcm}.key so there is never going to be anything to update (unlike XTS). ok tb@

Use calloc() when allocating cipher_data.

2019-04-14T17:16:57+00:00

Avoids use of uninitialised memory. ok tb@

Annotate a future improvement.

2019-04-14T16:46:26+00:00

Avoid potential double-frees following EVP_CIPHER_CTX_copy().

2019-04-14T16:43:49+00:00

In the case of a cipher with a custom copy control, if that control fails we may still have pointers that we do not own in the previously copied cipher data. Avoid potential double-frees by zeroing and freeing the copied cipher data in this case. Issue reported by Guido Vranken. ok tb@

Avoid some out of bound accesses in aesni_cbc_hmac_sha1_cipher().

2019-04-03T15:33:37+00:00

The plen variable can be NO_PAYLOAD_LENGTH == (size_t)-1, so doing tls_aad[plen-4] is no good. Also check that the length of the AAD set via the control interface is equal to 13 since the whole file is written with that case in mind. Note that we no longer use this code in LibreSSL/OpenBSD. We eliminated the use of these control interfaces and stitched cipher modes in libssl a while ago. Problem found by Guido Vranken with his cryptofuzz - thanks! input & ok beck, jsing