https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115095
Richard Biener <rguenth at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Ever confirmed|0 |1
CC| |hubicka at gcc dot gnu.org,
| |rguenth at gcc dot gnu.org
Status|UNCONFIRMED |NEW
Last reconfirmed| |2024-05-15
--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
GCC isn't able to compute the number of iterations of a loop like the following
which would be required for us to peel it and do the required constant folding.
We're also not trying brute-force peeling for loops we cannot compute an
upper bound of its number of iterations (attempting to arrive at one when
evaluating the peeled stmts).
<bb 3> [local count: 955630224]:
# _26 = PHI <_9(8), 116(2)>
# h_27 = PHI <h_15(8), 0(2)>
# s_29 = PHI <s_14(8), "t.c"(2)>
_10 = (unsigned int) _26;
h_12 = _10 + h_27;
s_14 = s_29 + 1;
h_15 = _10 * h_12;
_9 = *s_14;
if (_9 != 0)
goto <bb 8>; [89.00%]
else
goto <bb 4>; [11.00%]
<bb 8> [local count: 850510900]:
goto <bb 3>; [100.00%]
