[Bug tree-optimization/122219] Missed store sinking when using memcpy with vector_size type in inlined functions

[rguenth at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122219

--- Comment #2 from Richard Biener <rguenth at gcc dot gnu.org> ---
LIM doesn't handle this nicely.  But I'd have expected that, for example,
FRE/PRE replace

  MEM <uint128_t> [(char * {ref-all})&ret] = _10;
  ret.1_14 = ret;

with

  MEM <uint128_t> [(char * {ref-all})&ret] = _10;
  ret.1_14 = VIEW_CONVERT<...> (_10);

possibly variably-sized fear gets in the way here?  This also looks like
possibly sth for aggregate copyprop.