https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121143
Richard Biener changed:
What|Removed |Added
Keywords||false-positive,
|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121143
--- Comment #6 from Tomasz Kamiński ---
I was able to silence the warning by placing the bound on output range, instead
of input:
{
const auto __sz = ranges::distance(__rg);
if (__sz > (_Nm - size()))
__throw_bad_alloc();
// Bounded on
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121143
--- Comment #5 from Jakub Jelinek ---
My bet is that before cunroll we didn't know that __n_206 is 9 and during
cunroll we completely unroll the loop that computes it and when processing the
next loop we use ranger to ask about __n_206 value, ge
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121143
--- Comment #4 from Jakub Jelinek ---
That first loop corresponds to
struct __distance_fn final
{
template _Sent>
requires (!sized_sentinel_for<_Sent, _It>)
constexpr iter_difference_t<_It>
operator()[[nodiscard]](_It _
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121143
--- Comment #3 from Jakub Jelinek ---
Before cunroll that first loop is
[local count: 8680598903]:
# __n_274 = PHI <__n_206(43), 0(39)>
# ivtmp_255 = PHI
__n_206 = __n_274 + 1;
ivtmp_374 = ivtmp_255 - 1;
if (ivtmp_374 != 0)
got
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121143
Jakub Jelinek changed:
What|Removed |Added
CC||hubicka at gcc dot gnu.org,
