Re: tags in the 3 lowest bits

[storm@cua.dk (Kim F. Storm)]

Stefan Monnier <monnier@IRO.UMontreal.CA> writes:

> > The only reason I can see for this is that there is a small
> > performance penalty on XFASTINT with LSB -- but I'd prefer that
> > penalty rather than having to support two different methods.
> 
> I doubt the performance penalty is an issue.  After all, manipulating MSB
> is generally slightly more costly than LSB, so the performance impact might
> even be positive.  But I expect it's a wash.

Once LSB is in place, I would like us to get rid of XFASTINT
all-together.  It's a potential danger lurking to hit us.


> 
> > Or are there systems which cannot use LSB?
> 
> The LSB code needs pointers values that are multiples of 8.
> On some systems, malloc does not guarantee it.  Probably we can use
> gmalloc.c for those systems, but we'll need to make sure that's an option
> and we'll need to figure out when that's nmecessary.  

We could wrap malloc and free on such systems to force 8 byte alignment like
this  (it's a little costly in memory, but no big deal):

char *malloc_wrap(size_t len)
{
        char *p = malloc(len + 16);
        char *p1 = p;
        p = (char *)((unsigned)(p + 15) & ~0x7);
        *(char **)(p - 8) = p1;
        return p;
}

char free_wrap(char *p)
{
        free(*(char **)(p - 8));
}


>                                                       Also, there might be
> systems that are word-addressed rather than byte-addressed.  I don't know
> if such systems are still in use and whether Emacs runs on them, but if
> yes, we'll probably need to find some other way to deal with them,
> maybe keeping the old behavior.
> 
> I also hope we can switch to LSB everywhere, but I don't think we can
> do that right now.
> 
> I suggest to start with something like along the lines of
> 
> #if defined GLIBC || defined GNU_MALLOC
> #define USE_LSB_TAG
> #endif

That's a good starting point, yes.
Or we could use the above trick like this:

#if !(defined GLIBC || defined GNU_MALLOC)
#define malloc malloc_wrap
#define free free_wrap
#endif

-- 
Kim F. Storm <storm@cua.dk> http://www.cua.dk