Raymond Toy wrote:
> Yes, but you've declared INDEX-923 as a fixnum, and the compiler
> believes you. Thus, the compiler thinks that code isn't dead, so I
> think Doug's analysis is correct.
It better believe it. IT IS! Why should it think it's dead w/o a decl and
think it's not with it?
> Note that leaving out the declaration for INDEX-923 probably won't
> help either. The compiler isn't smart enough to derive the type of a
> variable if you setq the variable. A compiler deficiency.
That's the point. It's an important item for the TODO list.
Meanwhile, I'm happy I've been working in converting SETQs into initial
bindings for Series. Now it's confirmed it may make a big difference with
some compilers.
Note that I cannot get rid of all of them, like with this INCF. [BTW,
initializing the variable
with 0 and moving it just before the looping GO it's not as trivial as it may
look. It would
require quite a bit of work].
--
Fernando D. Mato Mira
Real-Time SW Eng & Networking
Advanced Systems Engineering Division
CSEM
Jaquet-Droz 1 email: matomira AT acm DOT org
CH-2007 Neuchatel tel: +41 (32) 720-5157
Switzerland FAX: +41 (32) 720-5720
http://www.csem.ch http://www.vrai.com ligwww.epfl.ch/matomira.html
|