https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64829
--- Comment #7 from ktkachov at gcc dot gnu.org --- > > Index: gcc/tree-vect-patterns.c > =================================================================== > --- gcc/tree-vect-patterns.c (revision 220205) > +++ gcc/tree-vect-patterns.c (working copy) > @@ -1732,9 +1732,11 @@ vect_recog_widen_shift_pattern (vec<gimp > if (TREE_CODE (oprnd0) != SSA_NAME || TREE_CODE (oprnd1) != INTEGER_CST) > return NULL; > > - /* Check operand 0: it has to be defined by a type promotion. */ > - if (!type_conversion_p (oprnd0, last_stmt, false, &half_type0, &def_stmt0, > - &promotion) > + /* Check operand 0: it has to be defined by a type promotion and it > + should be only used by the shift. */ > + if (!has_single_use (oprnd0) > + || !type_conversion_p (oprnd0, last_stmt, false, &half_type0, > &def_stmt0, > + &promotion) > || !promotion) > return NULL; That fixes the ICE in the testcase for me. I'll give it a test on arm-none-eabi and an arm bootstrap
