Lesson articulated by @jaekwon on why we need 80 bits
of entropy at least before we can think of cryptographic
safety. math/rand's seed is a max of 64 bits so can never
be cryptographically secure.
Also added some benchmarks for RandBytes
Fixes https://github.com/tendermint/tmlibs/issues/99
Updates https://github.com/tendermint/tendermint/issues/973
Removed usages of math/rand.DefaultSource in favour of our
own source that's seeded with a completely random source
and is safe for use in concurrent in multiple goroutines.
Also extend some functionality that the stdlib exposes such as
* RandPerm
* RandIntn
* RandInt31
* RandInt63
Also added an integration test whose purpose is to be run as
a consistency check to ensure that our results never repeat
hence that our internal PRNG is uniquely seeded each time.
This integration test can be triggered by setting environment variable:
`TENDERMINT_INTEGRATION_TESTS=true`
for example
```shell
TENDERMINT_INTEGRATION_TESTS=true go test
```
```
@melekes
yeah, bool is superfluous
@ethanfrey
If I remember correctly when I was writing test code, if I call Start() on a Service that is already running, it returns (false, nil). Only if I try to legitimately start it, but it fails in startup do I get an error.
The distinction is quite important to make it safe for reentrant calls. The other approach would be to have a special error type like ErrAlreadyStarted, then check for that in your code explicitly. Kind of like if I make a db call in gorm, and get an error, I check if it is a RecordNotFound error, or whether there was a real error with the db query.
@melekes
Ah, I see. Thanks. I must say I like ErrAlreadyStarted approach more (not just in Golang)
```
Emmanuel Odeke
Ethan Buchman
Jae Kwon
Anton Kaliaev
Ethan Frey
Wolf