This changes the code that deals with too large array sizes to
use int_fits_type_p instead of relying on the TREE_OVERFLOW setting
of the tree folder. The latter will break once we don't treat
sizetypes specially (and they keep being unsigned).
Bootstrapped and tested on x86_64-unknown-linux-gnu, ok for trunk?
Thanks,
Richard.
2011-05-02 Richard Guenther <rguent...@suse.de>
* c-decl.c (grokdeclarator): Instead of looking at
TREE_OVERFLOW check if the constant fits in the index type.
Index: trunk/gcc/c-decl.c
===================================================================
*** trunk.orig/gcc/c-decl.c 2011-05-02 14:50:36.000000000 +0200
--- trunk/gcc/c-decl.c 2011-05-02 15:12:15.000000000 +0200
*************** grokdeclarator (const struct c_declarato
*** 5368,5382 ****
convert (index_type,
size_one_node));
! /* If that overflowed, the array is too big. ???
! While a size of INT_MAX+1 technically shouldn't
! cause an overflow (because we subtract 1), the
! overflow is recorded during the conversion to
! index_type, before the subtraction. Handling
! this case seems like an unnecessary
! complication. */
! if (TREE_CODE (itype) == INTEGER_CST
! && TREE_OVERFLOW (itype))
{
if (name)
error_at (loc, "size of array %qE is too large",
--- 5368,5380 ----
convert (index_type,
size_one_node));
! /* The above overflows when size does not fit
! in index_type.
! ??? While a size of INT_MAX+1 technically shouldn't
! cause an overflow (because we subtract 1), handling
! this case seems like an unnecessary complication. */
! if (TREE_CODE (size) == INTEGER_CST
! && !int_fits_type_p (size, index_type))
{
if (name)
error_at (loc, "size of array %qE is too large",
