Re: Does Duff Device break C compilers

"Lambert Lum" <Lambert.Lum@eng.efi.com> writes:
>item called Duff's Device.
> case 0: do { *to = *from++;
> case 7: *to = *from++;
> case 6: *to = *from++;
>[...]


preston@tera.com (Preston Briggs) writes:
> Better is to rewrite it, like this
>
> /* no assumptions made about count */
> for (n = 0; n < count; n++)
> to[n] = from[n];


krste@ICSI.Berkeley.EDU (Krste Asanovic) writes:
>Better yet, "Krste's Kopy" :-)
>
> memcpy(from, to, count*(sizeof *from));


Errm, neither of these alternatives does what the original does.
Better to write it:


/* Jeremy's Junk Jumper */
volatile thing *to;
thing *from;
for(n = 0; n < count; n++)
*to = from[n];


The original was to jump on some hardware as quickly as possible; the
Device was the only way Duff could get the CPU/compiler to do it fast
enough. It was purely an expedient hack for that case alone, and was
never proposed as a general mechanism (though people tended to use it
as such for hack value, and it may well have helped them).


Tom Duff discusses the Device in
http://www.lysator.liu.se/c/duffs-device.html. This is the definitive
reference to all points raised here about Duff's Device.


J
[Oops. -John]




--

Return to the comp.compilers page.
Search the comp.compilers archives again.