https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85747
Andrew Pinski <pinskia at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Last reconfirmed| |2021-12-27
Severity|normal |enhancement
Ever confirmed|0 |1
Status|UNCONFIRMED |NEW
--- Comment #9 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
size: 13-3, last_iteration: 13-3
Loop size: 13
Estimated size after unrolling: 40
Not unrolling loop 1: it is not innermost and code would grow.
There are a few others like this one.
Note LLVM is able even to handle:
template <class It>
constexpr void sort(It first, It last) {
for (;first != last; ++first) {
auto it = first;
++it;
for (; it != last; ++it) {
if (*it < *first) {
auto tmp = *it;
*it = *first;
*first = tmp;
}
}
}
}
static int generate() {
int a[] = {0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21};
a[5] = 55;
sort(a + 0, a + 21);
return a[0] + a[6] + a[1] + a[2] + a[3] + a[4];
}
I suspect the cost estimate it does for the loop is the removal of the load of
a[i] knowing that a is fully written to.
