Re: gmp issues (long)

[Marius Vollmer]

Rob Browning <address@hidden> writes:

> The second (perhaps related) issue involves memory use.  It turns out
> that an mpz_t, which is the GMP integer type, takes up a minimum of 12
> bytes (when an int is 4 bytes), and will always be more than that
> because that 12 bytes doesn't include any space for the actual digits.

I'm not so much worried about the size of an mpz_t but about the fact
that it requires an additional malloc.  12 bytes is just right for our
double cells, tho.  So I would suggest making bignums a double cell
and putting the mpz_t structure into slots 1, 2, and 3.  There should
be a test somewhere whether mpz_t really fits into this space.  For
now we can just wait for this test to fail, I'd say.

This also removes the additional indirection.

-- 
GPG: D5D4E405 - 2F9B BCCC 8527 692A 04E3  331E FAF8 226A D5D4 E405