ruby-core

Jeremy is correct, Hash#[] (and Array#[]) have optimized dispatch.

We should delegate to @_emboss_ everytime we need to convert =C3=9F...

I'll try resurrecting an old eventfd proposal and maybe also bare futexes

[email protected] wrote:

Btw, you're already paying significant costs for a case-insensitive FS

[email protected] wrote:

Have you ever gotten to the bottom of this segfault?

Good to know it works for you. Keep in mind TIME_QUANTUM_USEC=1000 is

[email protected] wrote:

Setting SO_KEEPALIVE on the socket will help, but that still takes at

[email protected] wrote:

I think this is worth mentioning on the www.r-l.o so I made a pull request:

[email protected] wrote:

[email protected] wrote:

(2014/07/16 1:38), [email protected] wrote:

I forgot to note this the last publically-visible struct which may be

[email protected] wrote:

(2014/07/17 23:52), [email protected] wrote:

On Mon, Jul 21, 2014 at 10:40 AM, <[email protected]> wrote:

Can you show a performance difference with always_inline on those

(2014/07/23 1:31), [email protected] wrote:

I don't think removing "inline" is necessary, either. It is a hint

[email protected] wrote:

Thank you for the catch and explanation!