Re: Converting mpz_t to different integer representations

[Paul Thomas <paulthomas2 at wanadoo dot fr>]

Brooks,
> However, I still don't understand why you're trying to convert the 
> mpz_t constants into _host_ integer values and asking about what types 
> are available on the host, given that the host representation and the 
> target representation are likely to be different.  It looks like to me 
> that the native_encode_expr() function from gcc/fold-const.c is more 
> like what you want, since it takes a value and produces the 
> corresponding bytewise representation on the target.
You are absolutely right! I was so concentrated on doing the right thing 
with the lining up the initializer elements and putting them into the 
union that I was not thinking clearly about whose representations of 
numbers should be used.  I was aware of being ignorant - hence the call 
for help.

That said, which representation does mpz_export produce?  ie. is it not 
outputting in the target endian-ness and bitiness?
> Meanwhile, weren't you making some comments about going to not have 
> all that much time to be working on gfortran bug-fixing after the 
> first of the year?  :)   If you'd like (and aren't in too much of a 
> hurry with this), I can poke some at translating the 
> native_encode_expr stuff into a gfortran equivalent.
Please do - I have been raking through my bits box to make sure all the 
partially made patches see the light of day.  I would sleep a lot better 
(and a lot more :-) ) if you would take both patches on.
> - Brooks (with thanks to Andrew Pinski for pointing me at fold-const.c)
Indeed, thanks, Andrew.  native_encode_expr is very helpful for both 
PRs.  The version of the patch posted to the PR can be used with 
WORDS_BIG_ENDIAN to put the HOST_WIDE_INTs hi and lo into the right 
order.  Since the mpz_export function is used elswhere, if it is not 
right, we have all sorts of troubles that we do not know about.

Paul