drbg: fix issue where DRBG_CTR fails if NO_DF is used (2nd attempt)

Since commit 7c226dfc43 a chained DRBG does not add additional
data anymore when reseeding from its parent. The reason is that
the size of the additional data exceeded the allowed size when
no derivation function was used.

This commit provides an alternative fix: instead of adding the
entire DRBG's complete state, we just add the DRBG's address
in memory, thereby providing some distinction between the different
DRBG instances.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/9802)
This commit is contained in:
Dr. Matthias St. Pierre 2019-05-30 18:37:29 +02:00 committed by Matt Caswell
parent 5520695c73
commit 63180182ec

View file

@ -161,7 +161,9 @@ size_t rand_drbg_get_entropy(RAND_DRBG *drbg,
size_t bytes = 0;
/*
* Get random from parent, include our state as additional input.
* Get random data from parent. Include our address as additional input,
* in order to provide some additional distinction between different
* DRBG child instances.
* Our lock is already held, but we need to lock our parent before
* generating bits from it. (Note: taking the lock will be a no-op
* if locking if drbg->parent->lock == NULL.)
@ -170,7 +172,7 @@ size_t rand_drbg_get_entropy(RAND_DRBG *drbg,
if (RAND_DRBG_generate(drbg->parent,
buffer, bytes_needed,
prediction_resistance,
NULL, 0) != 0)
(unsigned char *)&drbg, sizeof(drbg)) != 0)
bytes = bytes_needed;
drbg->reseed_next_counter
= tsan_load(&drbg->parent->reseed_prop_counter);