Re: slot-definition

[Sam S. <sd...@gn...>]

> * Bruno Haible <oe...@py...t> [2004-05-11 22:26:46 +0200]:
>
> Sam wrote:
>> From the compilation POV, no, the compiled code may depend on many
>> factors, including global proclamations and the compilation status of
>> other parts of the system.  This is actually part of the raison d'etre
>> for WITH-COMPILATION-UNIT.
>
> In the small example, WITH-COMPILATION-UNIT was not used.
>
> The bug is that the compiler made assumptions about a function from
> file1 while compiling file2. It must not do this, according to CLHS
> 3.2.2.3:
>
>   "The definition of a function that is defined and declared inline in the
>    compilation environment must be the same at run time."
>
>   "A call within a file to a named function that is defined in the same
>    file refers to that function, unless that function has been declared
>    notinline. The consequences are unspecified if functions are redefined
>    individually at run time or multiply defined in the same file."
>
> Since the function from file1 was neither defined in the same
> compilation-unit as file2, nor declared inline, the compiler must not
> assume any details about it that could change if I re-defun the
> function after loading file2.fas.
>
> So you can save the seclass of a function for compiling other
> functions in the same compilation unit, or if the function was
> declared inline. But not in this case.

in that case we cannot fold (= 0 0) because this fas file might be
loaded by lisp.run with a redefined #'=.

-- 
Sam Steingold (http://www.podval.org/~sds) running w2k
<http://www.camera.org> <http://www.iris.org.il> <http://www.memri.org/>
<http://www.mideasttruth.com/> <http://www.honestreporting.com>
Lisp: it's here to save your butt.