🚀 go-pugleaf

> In article <DMGt0K.64B@cix.compulink.co.uk>, 
> mark_tbu@cix.compulink.co.uk ("Mark Morgan Lloyd ") writes:
> > The use of untagged variants is actually quite slow, since most of 
> > the work is done at run time.
> 
>  I am not sure what this means. Certainly in our compiler (Mod51) 
> untagged
> variants can be used to completely overlay variables in memory, with no
> RAM overhead for the TAG, nor code overhead for their access.
>  SO you can pack 4 BYTES onto a REAL, or 2 8 Bit BITSETs onto a WORD 
> etc...
> I can see no reason that any runtime process be involved.
>  As an aid to memory-overlay, and packing ( important when your target 
> may
> have just 64 BYTES of RAM ( including STACK! ), we also have extended 
> the
> BITSET concept to allow effectively packed arrays of BOOLEAN.
> 
>  
> > Mark Morgan Lloyd
> > mark_tbu@cix.compulink.co.uk
> > 
> > [Opinions above are the author's, not those of his employers or 
> > colleagues]
> - jim granville
> ===== Mandeno Granville FAX +64 9 6301 720, 128 Grange Rd Auckland 3 NZ 
> ======
> * Developers and suppliers of serious MicroController Embedded Control 
> Tools *
> * 89c2051, 89c1051 Emulator / ChipreProgrammers                         
   *
> * x51 C, Pascal & Modula-2 Compilers, Simulators, Emulators & FLASH 
> Pgmrs    *
> * Contact : Jim Granville . Email above.                                
>      *

I was thinking specifically of the general case, where an 
arbitrarily-complex data structure may be involved. If you are using 
untagged variants to extend the type-transfer mechanism without overhead 
then that's your business- but very much to the credit of your compilers 
:-)

Would you care to email me details of your product line for reference?

Mark Morgan Lloyd
mark_tbu@cix.compulink.co.uk

[Opinions above are the author's, not those of his employers or 
colleagues]