https://gcc.gnu.org/bugzilla/show_bug.cgi?id=120137
Li Pan <pan2.li at intel dot com> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |pan2.li at intel dot com
--- Comment #1 from Li Pan <pan2.li at intel dot com> ---
Reproduced from today's upstream, it comes from here:
inline poly_wide_int
rtx_vector_builder::step (rtx elt1, rtx elt2) const
{
return (wi::to_poly_wide (elt2, GET_MODE_INNER (m_mode))
- wi::to_poly_wide (elt1, GET_MODE_INNER (m_mode)));
}
the elt1 is -136 and while the m_mode is QImode, it exceeds the range of QImode
-128-127 and then triggers the assertion.
Maybe we need to add some check from stepped_sequence_p before goes to step().
