[Bug target/128200] [x86-64] Failure to vectorize a double-to-long converting store loop

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

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=128200

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Last reconfirmed|                            |2026-10-27
             Status|UNCONFIRMED                 |NEW
     Ever confirmed|0                           |1

--- Comment #3 from Richard Biener <rguenth at gcc dot gnu.org> ---
The issue is

t.c:8:23: note:   ==> examining statement: _6 = (long int) _4;
t.c:8:23: note:   vect_is_simple_use: operand _3->v, type of def: internal
t.c:8:23: missed:   conversion not supported by target.
t.c:8:23: note:   vect_is_simple_use: operand _3->v, type of def: internal
t.c:10:16: missed:   not vectorized: relevant stmt not supported: _6 = (long
int) _4;

it works with AVX512 which gets

.L4:
        vcvttpd2qq      (%rdi,%rax), %zmm0
        vmovdqu64       %zmm0, (%rsi,%rax)
        addq    $64, %rax
        cmpq    %rax, %rdx
        jne     .L4