[Bug target/83479] Register spilling in AVX code

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83479

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|WAITING                     |NEW
                 CC|                            |jakub at gcc dot gnu.org,
                   |                            |rguenth at gcc dot gnu.org

--- Comment #7 from Richard Biener <rguenth at gcc dot gnu.org> ---
Confirmed.  -fschedule-insns helps in both cases, so does -fno-tree-ter.

Just looking at the GIMPLE the AVX512 case is somewhat weird:

  __m512d v8;
  double _59;
  vector(2) double _303;
...
  _59 = BIT_FIELD_REF <v8_187, 64, 384>;
  _303 = {_59};
  _305 = __builtin_ia32_broadcastsd512 (_303, __Y_304(D), 255);
  _302 = __builtin_ia32_sqrtpd512_mask (_305, __Y_301(D), -1, 4);

but it seems this is how _mm512_set1_pd works:

extern __inline __m512d
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm512_set1_pd (double __A)
{
  return (__m512d) __builtin_ia32_broadcastsd512 (__extension__
                                                  (__v2df) { __A, },
                                                  (__v8df)
                                                  _mm512_undefined_pd (),
                                                  (__mmask8) -1);
}

given we now have VEC_DUPLICATE_EXPR it would be nice to open-code
those builtins somehow (or for GCC 9).