Re: wip-cse

[Ludovic Courtès]

Hi!

Andy Wingo <address@hidden> skribis:

> I went ahead and pushed this branch to stable-2.0.  We can roll it out
> again if it causes problems.

Cool, thanks!

> I spent some time poking at vlists and ended up squeezing out a little
> bit of performance.

Any figures?

> I ended up allocating the hash table data inline to the content
> vector, after some care looking at what the optimizer was doing, and
> that ended up being a win.  There are a couple of smaller wins there
> too.  Compilation is still a bit slower though.  Grrr.

Sounds like nice work.

Did you try fiddling with ‘block-growth-factor’, as suggested earlier?

Thanks,
Ludo’.