Re: [RFC] RNG fixes
| From: | Fleshgrinder | Date: | Sun, 19 Jun 2016 15:50:34 +0000 |
| Subject: | Re: [RFC] RNG fixes | ||
| References: | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 | Groups: | php.internals |
| Request: | Send a blank email to internals+get-94122@lists.php.net to get a copy of this message | ||
On 6/17/2016 7:18 PM, Christoph Becker wrote:
> Consequently, we should remove rot13() as well, see
> <http://news.php.net/php.notes/205744>. And we
> shouldn't stop there as
> include(_once), require(_once), file_get_contents() and readfile() bear
> the risk of file inclusion vulnerabilities … ;)
>
> In my opinion, our job when designing the language and the core
> libraries is not to avoid (or remove) features that can be used to
> produce insecure software, but rather to offer additional features that
> make it easier to produce secure software, and to document potential
> issues and hint at better alternatives. random_*() is such an addition,
> and I don't see an urgent need to get rid of (mt_)rand().
>
Reminds me of https://github.com/rust-lang/rust/issues/32670
Again, our (mt_)rand() functions are not portable, not standards
compliant, slow, outdated, and dangerous for crypto. There was not a
single argument why we should keep them.
Fixing = BC
--
Richard "Fleshgrinder" Fussenegger
Attachment: [application/pgp-signature] OpenPGP digital signature signature.asc
Attachment: [application/pgp-signature] OpenPGP digital signature signature.asc