[Bug tree-optimization/51499] vectorizer missing simple case

[dominiq at lps dot ens.fr]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51499

--- Comment #6 from Dominique d'Humieres <dominiq at lps dot ens.fr> 2011-12-11 14:14:01 UTC ---
> I think you are looking at the scalar epilogue. The number of iterations is
> unknown, so we need an epilogue loop for the case that number of iterations is
> not a multiple of 4.

While investigating pr51597, I have found that vectorized loops in programs as
simple as

subroutine spmmult(x,b,ad)
implicit none
integer, parameter :: nxyz=1008315
real(8),dimension(nxyz):: x,b,ad
b = ad*x
end subroutine spmmult               !=========================================

has always an additional non-vectorized loop, i.e. a vectorized one

L3:
        movsd   (%r9,%rax), %xmm1
        addq    $1, %rcx
        movapd  (%r10,%rax), %xmm0
        movhpd  8(%r9,%rax), %xmm1
        mulpd   %xmm1, %xmm0
        movlpd  %xmm0, (%r8,%rax)
        movhpd  %xmm0, 8(%r8,%rax)
        addq    $16, %rax
        cmpq    $504156, %rcx
        jbe     L3

and a non-vectorized one

L5:
        movsd   -8(%rdi,%rax,8), %xmm0
        mulsd   -8(%rdx,%rax,8), %xmm0
        movsd   %xmm0, -8(%rsi,%rax,8)
        addq    $1, %rax
        cmpq    %rcx, %rax
        jne     L5

even when the above loops are unrolled. How can the loop L5 be unrolled if it
is only there for a "scalar epilogue"?