Re: LuaJIT2 performance for number crunching

[steve donovan <steve.j.donovan@...>]

On Tue, Feb 15, 2011 at 1:10 AM, Francesco Abbate
<francesco.bbt@gmail.com> wrote:
> Here the timing:
> rk4-unroll.lua with LuaJIT2: 0m0.240s
> C code with -O2 with GSL library: 0m0.638s

Very cool - looks like LHPC (Lua High Performance Computing) is
definitely a hot topic for 2011!  In fact, a great topic for the
workshop.

In the meantime we should start a wiki page and collect concrete tips.

I'm happy to look at the template issue, should be relatively
straightforward. This kind of notation is what I'm aiming for

fun = template [[ return function(y,n)
....
end]]

...

fun3 = fun(3) -- specialize the template, cache the result (sets T1,
T2 etc as template parameters)

-- use fun3

steve d.