From: Fernando Cacciola (fcacciola_at_[hidden])
Date: 2001-10-10 11:10:56
----- Original Message -----
From: Kevin Lynch <krlynch_at_[hidden]>
Sent: Wednesday, October 10, 2001 12:56 PM
Subject: Re: [boost] Re: Math Functions
> And we should see about fixing the optimization issues with numeric_cast
> soon, if at all possible; it could be a real performance killer in those
> cases where it prevents optimizers from removing all the code we've
> written :-) Looking at the generated assembly code, when the calls
> aren't optimized away the object code size doubles (at least under gcc).
I've uploaded (sometime ago), the files 'cvt_n.h' and 'numeric_bounds.h'
into the 'standard functions' branch of the Files Sections.
It contains my own numeric cast, which includes all the required
optimizations and can be used mixing integer and floating point types (it
actually works with any user defined number type). Unfortunately, that code
is pretty old, works only with Borland and requires a 'boostification'. I'm
supposed to modify numeric_cast based on that conversion code, but since it
will take some time, I couldn't do it yet.
Perhaps you want to look at those files and see if you can do the
modifications to numeric_cast<> yourself.
BTW: I've almost finished adding the Borland stuff to your latest update. I
had to apply a few changes, so I didn't post a message yet since I haven't
finished. I'll do that ASAP.
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk