Port modiff changes to GCC 4.8.5

Problem reported by Spencer Baugh.
* src/lisp.h (elogb): Now simply a function on unsigned long long int.
This avoids problems porting to GCC 4.8.5 20150623 (Red Hat 4.8.5-44).
There may be a minor runtime cost on 32-bit platforms but it’s
not worth worrying about.
This commit is contained in:
Paul Eggert 2023-05-17 10:53:26 -07:00
parent 8a92d00470
commit 0bba1b8c3d

View file

@ -3906,18 +3906,11 @@ integer_to_uintmax (Lisp_Object num, uintmax_t *n)
}
/* Return floor (log2 (N)) as an int, where 0 < N <= ULLONG_MAX. */
#if (201112 <= __STDC_VERSION__ && INT_MAX <= UINT_MAX \
&& LONG_MAX <= ULONG_MAX && LLONG_MAX <= ULLONG_MAX)
# define elogb(n) \
_Generic (+(n), \
int: UINT_WIDTH - 1 - count_leading_zeros (n), \
unsigned int: UINT_WIDTH - 1 - count_leading_zeros (n), \
long: ULONG_WIDTH - 1 - count_leading_zeros_l (n), \
unsigned long: ULONG_WIDTH - 1 - count_leading_zeros_l (n), \
default: ULLONG_WIDTH - 1 - count_leading_zeros_ll (n))
#else
# define elogb(n) (ULLONG_WIDTH - 1 - count_leading_zeros_ll (n))
#endif
INLINE int
elogb (unsigned long long int n)
{
return ULLONG_WIDTH - 1 - count_leading_zeros_ll (n);
}
/* A modification count. These are wide enough, and incremented
rarely enough, so that they should never overflow a 60-bit counter