[squeak-dev] Re: [ANN] Number comparison, hash, NaN, Point, and other partially ordered sets

bryce at kampjes.demon.co.uk bryce at kampjes.demon.co.uk
Thu Jan 8 20:48:40 UTC 2009


Nicolas Cellier writes:
 >  <bryce <at> kampjes.demon.co.uk> writes:
 > 
 > > 
 > > nicolas cellier writes:
 > >  > I don't even know which hardware supports or not these traps...
 > >  > I even wonder if such features were mandatory...
 > >  > That should not stop us: as suggested, Smalltalk is already sufficiently 
 > >  > slow to afford a sticky flag test at each Float primitive in case no 
 > >  > hardware trap is available.
 > > 
 > > Hopefully Smalltalk will not always be that slow.
 > > 
 > > All we'd need to do was remove the boxing and unboxing around
 > > floating point operations. Inside single statements, that is
 > > equivalent to the Boolean optimisations that Exupery already
 > > does. Across statements will require a proper optimiser.
 > > 
 > > Bryce
 > > 
 > > 
 > 
 > Great, but if FPU registers are using extra precision (extended double),
 > then the result of operations might depend on register/memory exchange.
 > 

The x86 SSE2 instruction set uses 64 bit floats. It also supports
16 floating point registers on 64 bit platforms. My plan is to use
that.

IMO the user needs control over the precision/portability/performance
tradeoffs but I haven't thought of any good ways of providing such 
control.

Bryce



More information about the Squeak-dev mailing list