Official support for Unsafe proposal

[Andrew Haley]

On 1/17/24 10:10, Quân Anh Mai wrote:
 > Regarding C2, I believe there is so much a compiler can do regarding eliminating bound checks. Of course, there are a lot of rooms for improvement, such as a constraint elimination pass (JDK-8307768). But even then a lot of cases are simply too hard. For example, a Brotli decompressor implemented in Rust can achieve a 11% speedup simply by disabling bound checks in its hot spot [1]
 >
 > [1]: https://dropbox.tech/infrastructure/lossless-compression-with-brotli <https://dropbox.tech/infrastructure/lossless-compression-with-brotli>

Thanks for that reference. The feeling I got from reading it is that
decompression can be vastly improved (in the sense of safety and
reliability) by using a decent language at very little cost.

  "The security and safety of Rust in a SECCOMP sandbox without the
  overhead of a garbage collector is a big win. It is worth the minor
  computational overhead of decompressing in a safe language with
  bounds checked arrays and determinism guarantees."

Bravo, I say.

-- 
Andrew Haley  (he/him)
Java Platform Lead Engineer
Red Hat UK Ltd. <https://www.redhat.com>
https://keybase.io/andrewhaley
EAC8 43EB D3EF DB98 CC77 2FAD A5CD 6035 332F A671