Re: Popcount optimization using AVX512

On Fri, 5 Apr 2024 at 07:15, Nathan Bossart <nathandbossart(at)gmail(dot)com> wrote:
> Here is an updated patch set. IMHO this is in decent shape and is
> approaching committable.

I checked the code generation on various gcc and clang versions. It
looks mostly fine starting from versions where avx512 is supported,
gcc-7.1 and clang-5.

The main issue I saw was that clang was able to peel off the first
iteration of the loop and then eliminate the mask assignment and
replace masked load with a memory operand for vpopcnt. I was not able
to convince gcc to do that regardless of optimization options.
Generated code for the inner loop:

clang:
<L2>:
50: add rdx, 64
54: cmp rdx, rdi
57: jae <L1>
59: vpopcntq zmm1, zmmword ptr [rdx]
5f: vpaddq zmm0, zmm1, zmm0
65: jmp <L2>

gcc:
<L1>:
38: kmovq k1, rdx
3d: vmovdqu8 zmm0 {k1} {z}, zmmword ptr [rax]
43: add rax, 64
47: mov rdx, -1
4e: vpopcntq zmm0, zmm0
54: vpaddq zmm0, zmm0, zmm1
5a: vmovdqa64 zmm1, zmm0
60: cmp rax, rsi
63: jb <L1>

I'm not sure how much that matters in practice. Attached is a patch to
do this manually giving essentially the same result in gcc. As most
distro packages are built using gcc I think it would make sense to
have the extra code if it gives a noticeable benefit for large cases.

The visibility map patch has the same issue, otherwise looks good.

Regards,
Ants Aasma