Live data from Hacker News

Rust: 128 bit integers preparing to be released

github.com

11–20 of 173 posts

Re: Rust: 128 bit integers preparing to be released

#11
post #6

People still care about Rust? After the quadratic hashes and rumored memory issues, I'd have thought people would've dropped interest in it by now.

Memory issues? And a lot of implementations of hash functions turn quadratic. I seem to remember php having quadratic hash table performance. And the bug in rust's code required a very specific usage pattern.

edit: hash table vuln article (2011): http://www.securityweek.com/hash-table-collision-attacks-cou...

Re: Rust: 128 bit integers preparing to be released

#13
post #9
post #4

Earlier quoted context omitted.

Finance and scientific computing.

Is this for implementing faster high precision floating point? It didn't mention 128 bit float.

For finance probably but it's mostly useful for statistics from what I understand.

I know fortran and Cobol code that banks use and is used for science (tm) often defines numbers this large for certain operations.

One such i can think of is a map reduce of large datasets. Let's say you want to find as a result a huge sum. 2^128 is a bit bigger then 2^64 and that difference may be big enough to provide the computation needed for getting to Mars rather then getting to the moon on 2^8 machines.

Re: Rust: 128 bit integers preparing to be released

#16

This title implies some things that aren't quite right. Let's take a step back. The current release of stable Rust is 1.13, beta is 1.14, nightly is 1.15. New features land on master, hence 1.15. So this means that, if it does land, it will become available on nightly soon. But it's still behind a feature flag. So it won't actually come out in Rust 1.15. I am not on the relevant subteam here, so I am not 100% sure of…

Yea I didn't know how to word it without it becoming more of a sentence and less of a title.

Re: Rust: 128 bit integers preparing to be released

#17

This title implies some things that aren't quite right. Let's take a step back. The current release of stable Rust is 1.13, beta is 1.14, nightly is 1.15. New features land on master, hence 1.15. So this means that, if it does land, it will become available on nightly soon. But it's still behind a feature flag. So it won't actually come out in Rust 1.15. I am not on the relevant subteam here, so I am not 100% sure of…

Yea I didn't know how to word it without it becoming more of a sentence and less of a title.

Totally fair! I'm not sure what I would have said either.

Re: Rust: 128 bit integers preparing to be released

#18
post #6

People still care about Rust? After the quadratic hashes and rumored memory issues, I'd have thought people would've dropped interest in it by now.

The bug depended on iterating through every item in a hashtable and putting it into a new hashtable AND the new hashtable had to not start with a known capacity. That's a pretty specific edge case, IMO. Plus it's been addressed.

"rumored memory issues" just sounds like FUD.

Re: Rust: 128 bit integers preparing to be released

#19
post #6

People still care about Rust? After the quadratic hashes and rumored memory issues, I'd have thought people would've dropped interest in it by now.

> After the quadratic hashes Reference: http://accidentallyquadratic.tumblr.com/post/153545455987/ru... (bugs happen, fix is in the queue, life goes on) > rumored memory issues Rumors are just rumors. If you (or anyone else) has more than FUD here, please email https://www.rust-lang.org/security.html .

Thanks for the link to the explanation of the hash table bug!

> One proposal is that hash tables ought to retain their insertion order...

This seems to be something that a lot of hash table implementations are converging on:

https://morepypy.blogspot.com/2015/01/faster-more-memory-eff...

Do you happen to have a link to the discussion for the Rust proposal?

Post reply on HN