[Bug middle-end/88963] gcc generates terrible code for vectors of 64+ length which are not natively supported

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

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88963

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |10.0
             Status|NEW                         |RESOLVED
         Resolution|---                         |FIXED

--- Comment #14 from Andrew Pinski <pinskia at gcc dot gnu.org> ---

        vmovdqa 32(%rsi), %ymm2
        vmovdqa (%rsi), %ymm1
        vpaddd  32(%rdx), %ymm2, %ymm0
        vpaddd  (%rdx), %ymm1, %ymm1
        vmovdqa %ymm0, 32(%rdi)
        vmovdqa %ymm1, (%rdi)

The stores look fine even in GCC 10.1.0 expand:
;; BIT_FIELD_REF <*a_7(D), 256, 256> = _14;

(insn 13 12 0 (set (mem/j:V8SI (plus:DI (reg/v/f:DI 88 [ a ])
                (const_int 32 [0x20])) [1 *a_7(D)+32 S32 A256])
        (reg:V8SI 87 [ _14 ])) "/app/example.cpp":6:8 -1
     (nil))

I think r10-3407 fixed it.  Note I notice the changelog is different from what
the patch did too.

Anyways fixed.