Derrick Stolee <derrickstolee@xxxxxxxxxx> writes: >> Instead of using srand() and rand() (which would make sense to wrap with >> git_rand() as you propose), we can simplify our lives by using a CSPRNG, >> which only gets initialized once, as is already the case with >> csprng_bytes(). > > So the idea is to use csprng_bytes() everywhere instead of srand()/rand(). > > I can adjust my local patch to still create git_rand(), but base it on > csprng_bytes() and not collide with your patch. Mimicking rand()'s behavior > is a simpler interface to consume. I am still ambivalent about wasting entropy for something that srand() would suffice, so git_rand() interface may be an welcome addition anyway, that serves an extra layer of indirection to insulate the callers from the implementation. Thanks.
