Danger DO NOT USE

README.md

@@ -1,3 +1,6 @@
+# Danger DO NOT USE
+Just found two bugs while finally finishing the SIMD versions and seeing if "bscrypt2" is even worth it to release (see https://infosec.exchange/@sc00bz/111864686852299659 about "bscrypt2" and "has anyone looked at the code" which I now assume is "no" ignoring https://github.com/Sc00bz/bscrypt/issues/4). Feel free to find both bugs (or more?): 6140f531c379e40e556cc9624ca4c32cb38a56c9e503d4e74ad592ca6229a4e4 and 9cb0d6dbda681a3d4d0521ae8cd566cc888726f83a423498cd8dcd09470baa98. It is very obvious which makes me hope that no one is using this. Fix or "bscrypt2" is coming soon. I now need to do what I should of done awhile ago which is write it in multiple languages and compare output to limit errors.
+
 # bscrypt
 A cache hard password hash/KDF