Thanks!
On Sunday, October 11, 2015 at 7:45:35 PM UTC+2, Mauro wrote:
>
> > Sorry. I wanted to say AbstractInt not Int.
>
> `Integer` is the abstract datatype. Anyway the answer is: no, it does
> not help. To make fast code Julia must know the memory layout of a
> type, which can only be known for a concrete type.
>
> > Then will this abstract annotation help?
> >
> > On Sunday, October 11, 2015 at 6:07:35 PM UTC+2, Kristoffer Carlsson
> wrote:
> >>
> >> Int is a concrete type. On x64 it is Int64 and on x86 it is Int32.
> >>
> >> On Sunday, October 11, 2015 at 5:41:11 PM UTC+2, cheng wang wrote:
> >>>
> >>> In `performance tips`, there is an example:
> >>> function foo(a::Array{Any,1})
> >>> x = a[1]::Int32
> >>> b = x+1
> >>> ...
> >>> end
> >>> It say the annotation ::Int32 helps in this case.
> >>>
> >>> So I was wondering if it still helps in the following case with
> abstract
> >>> type annotation??
> >>> function foo(a::Array{Any,1})
> >>> x = a[1]::Int
> >>> b = x+1
> >>> ...
> >>> end
> >>> Here the annotation is ::Int.
> >>>
> >>> Thanks!
> >>>
> >>
>
