So here is an example.
The FAIL function is very general and
will probably go in the library.
//
gen FAIL[ret,exn] : exn -> any = "(throw $1,*(?1*)0)";
//
Unfortunately, you have to give the type "ret" which is the
type of the C++ expression. Felix compiler actually knows this
type from
On 26/05/2013, at 12:43 PM, srean wrote:
> Now that you bring up parallelism, any plans of enabling OpenMP for Felix.
I have looked at it with intent. A flag to say if OpenMP is available is
generated.
However there's a problem. OpenMP primarily works by #pragma-ing subsequent
loops, which ar
On 26/05/2013, at 12:43 PM, srean wrote:
>
>
> In fact the one most relevant to YOU (srean) is the costing of
> parallelisation.
> You want to split a job up into parallel parts. Which parts do you split up?
> How many pieces? What shape are the pieces?
>
>
> Ah ! this is a difficult one and
On 26/05/2013, at 8:52 AM, srean wrote:
> Calling them bad compilers is, I think, unfair.
No its not. People that know little theory hack stuff together
without research and thought. I actually suggested on the clang mailing
list the slower optimisations turn off when they're having trouble.
"O
