Re: Windows 64 port

[Eli Zaretskii]

> Date: Thu, 01 Mar 2012 00:58:02 -0800
> From: Paul Eggert <address@hidden>
> CC: Eli Zaretskii <address@hidden>, address@hidden, 
>  address@hidden
> 
> >     Here is a more-detailed review of the latest patch you sent.
> > 
> >     > -      int i = 0, aligned = (intptr_t) ABLOCKS_BUSY (abase);
> >     > +      int i = 0, aligned = (ABLOCKS_BUSY (abase) != NULL);
> > 
> >     Not needed for Windows 64.  The change slows the code down a bit, and
> > 
> > 
> > Prove it.
> 
> I compiled it both ways (x86-64 GCC 4.6.2), and the "!= NULL" version
> has an extra "cmpl" instruction, so it is indeed a bit fatter.

Was that an optimized compile or an unoptimized one?  If the latter,
the results have no practical effect on the speed of the production
binary; if the former, I'm not sure you can reliably tell which
instruction belongs to which source line, especially since that code
is in a function that is inlined by GCC when you compile with
optimizations.

In any case, even if your claim _is_ true (and I very much doubt
that), it is beyond ridiculous to say that a single extra clock cycle
in a program such as Emacs should be seriously considered as an
important factor to weigh against code cleanness and maintainability.