[boost] Re: Re: Re: Boost Mathematicians

From: Thorsten Ottosen (nesotto_at_[hidden])
Date: 2004-06-03 03:23:51

"Hurd, Matthew" <hurdm_at_[hidden]> wrote in message news:BA1220310B07EC4A8E321B451E07AF4784833F_at_msgbal501.ds.susq.com...

| There is an often seen misconception that C/C++ should be as fast as it
| gets as the programming model is close to the instruction model of the
| machine.  In practice FORTRAN is usually faster.
|
| At one level, in C++ this slowness this is usually because of
| temporaries and expression templates and modern optimizing compilers can
| often mitigate this for typical math calcs.  Other abstraction penalties
| apply.

yeah, we want move semantics.

| Another level is the fundamental differences in aliasing between the
| language definitions.
|
| FORTRAN as a language does not allow aliasing ambiguities which makes
| the optimizers job a lot easier and the code can be a lot faster.  C/C++
| assumes aliasing potential which makes optimizing very hard.

>From what I now, we cannot put C and C++ into the same basket.
See Type-Based Alias Analysis:

http://www.ddj.com/documents/s=880/ddj0010d/0010d.htm?temp=fXniOPb1aS

I my suggestion for Design by Contract I mentioned that aliasing (or lack of it)
could be expressed by a function precondition:

void foo( T* l, T* r, size_t sz )
precondition() {
std::less<T*>( l + sz, r ) || std::less<T*>( r + sz, l );
};

| You still need to wrap as you can't hope, nor should you try, to offer
| all algorithms from all languages.

probably not. But simply wrapping stuff is not always an option. For eaxample, in the
statistics lib I'm working on, I need to invert matrices. If I write wrappers, I will
probably only be ably to support float,double,complex leaving out higher precsion
datatypes.

br

Thorsten