Issue #16487 has been updated by sam.saffron (Sam Saffron).


Could we do something that allows us to install a gem that gives us the spe=
edup by rerouting stuff in MRI? =


Then you could install a pre-compiled binary of ruby if you wish and do `ge=
m install ruby-simd` to get the optimised code paths? =


There is a concern around "docker" style distributions, cause generally peo=
ple package ruby+pre-compiled gems in a single image. I guess a `ruby-simd`=
 type gem could do a quick runtime check at boot to see if it needs to re-c=
ompile stuff if the "compiled" version does not match the current architect=
ure and just pass a bunch of code to gcc/clang? =



----------------------------------------
Misc #16487: Potential for SIMD usage in ruby-core
https://bugs.ruby-lang.org/issues/16487#change-83747

* Author: byroot (Jean Boussier)
* Status: Open
* Priority: Normal
* Assignee: =

----------------------------------------
### Context

There are several ruby core methods that could be optimized with the use of=
 SIMD instructions.

I experimented a bit on `coderange_scan` https://github.com/Shopify/ruby/pu=
ll/2, and Pavel Rosick=FD experimented on `String#strip` https://github.com=
/ruby/ruby/pull/2815.

### Problem

The downside of SIMD instructions is that they are not universally availabl=
e.
So it means maintaining several versions of the same code, and switching th=
em either statically or dynamically.

And since most Ruby users use precompiled binaries from repositories and su=
ch, it would need to be dynamic if we want most users to benefit from it.

So it's not exactly "free speed", as it means a complexified codebase.

### Question

So the question is to know wether ruby-core is open to patches using SIMD i=
nstructions ? And if so under which conditions.

cc @shyouhei





-- =

https://bugs.ruby-lang.org/

Unsubscribe: <mailto:ruby-core-request / ruby-lang.org?subject=3Dunsubscribe>
<http://lists.ruby-lang.org/cgi-bin/mailman/options/ruby-core>