3 ms·

A subset of Bouncycastle primitives (Argon2, Blake2b, DigestRandomGenerator, HKDFBytesGenerator, Poly1305, Salsa20Engine, Scrypt, SHA256Digest, XSalsa20Engine)

A subset of Bouncycastle primitives (Argon2, Blake2b, DigestRandomGenerator, HKDFBytesGenerator, Poly1305, Salsa20Engine, Scrypt, SHA256Digest, XSalsa20Engine) refactored to use DirectByteBuffer instead of primitive arrays, for the purpose of minimizing leaking information through garbage collector.

The_rationalist 5y ago
I wonder if region pinning is relevant here https://openjdk.java.net/jeps/423#:~:text=G1%20permanently%20pins%20humongous%20regions,exceeds%20a%20specified%20liveness%20threshold https://openjdk.java.net/jeps/423#:~:text=G1%20permanently%2.... Looks like a great project Javaman :)
- java-man 5y ago
  Thank you. I would rather see maybe an annotation instructing the compiler and the runtime not to move allocated objects: @UnmovableMemory byte[] bytes = new byte[...];
  - The_rationalist 5y ago
    Yeah like https://doc.rust-lang.org/std/pin/ https://doc.rust-lang.org/std/pin/ Maybe you could propose the idea to the jdk tracker or mailing lists?https://mail.openjdk.java.net/pipermail/jdk-dev/2022-February/006368.html https://mail.openjdk.java.net/pipermail/jdk-dev/2022-Februar...
    - java-man 5y ago
      All my life, large corporations did nothing but ignore me. I basically gave up at this point. You are right - something like Pin in rust. It will be a new language feature, there are backward compatibility aspects, and the fact that it will need to be correctly done by all implementations (or it would fail its stated purpose). DirectByteBuffer provides a certain guarantee as it is. On the other hand, the secrets might still be leaked via paging, so even this solution is a partial one.
skanga 5y ago
Are there any docs besides the unit tests?
- java-man 5y ago
  It's a replacement for Bouncycastle classes, using DirectByteBuffer wrappers to replace primitive arrays. One difference is explicit zero() method via ICryptoZeroable interface which clears the buffers.
msgilligan 5y ago
Any chance you might add ECDSA?
- java-man 5y ago
  I will not turn down a reasonable contract offer ;-)
  - msgilligan 5y ago
    I'll keep that in mind, but not ready to fund it at this point.

The_rationalist 5y ago

I wonder if region pinning is relevant here https://openjdk.java.net/jeps/423#:~:text=G1%20permanently%20pins%20humongous%20regions,exceeds%20a%20specified%20liveness%20threshold https://openjdk.java.net/jeps/423#:~:text=G1%20permanently%2.... Looks like a great project Javaman :)

java-man 5y ago
Thank you. I would rather see maybe an annotation instructing the compiler and the runtime not to move allocated objects: @UnmovableMemory byte[] bytes = new byte[...];
- The_rationalist 5y ago
  Yeah like https://doc.rust-lang.org/std/pin/ https://doc.rust-lang.org/std/pin/ Maybe you could propose the idea to the jdk tracker or mailing lists?https://mail.openjdk.java.net/pipermail/jdk-dev/2022-February/006368.html https://mail.openjdk.java.net/pipermail/jdk-dev/2022-Februar...
  - java-man 5y ago
    All my life, large corporations did nothing but ignore me. I basically gave up at this point. You are right - something like Pin in rust. It will be a new language feature, there are backward compatibility aspects, and the fact that it will need to be correctly done by all implementations (or it would fail its stated purpose). DirectByteBuffer provides a certain guarantee as it is. On the other hand, the secrets might still be leaked via paging, so even this solution is a partial one.

skanga 5y ago

Are there any docs besides the unit tests?

java-man 5y ago
It's a replacement for Bouncycastle classes, using DirectByteBuffer wrappers to replace primitive arrays. One difference is explicit zero() method via ICryptoZeroable interface which clears the buffers.

msgilligan 5y ago

Any chance you might add ECDSA?

java-man 5y ago
I will not turn down a reasonable contract offer ;-)
- msgilligan 5y ago
  I'll keep that in mind, but not ready to fund it at this point.