Observations from a simple JMH benchmark

[Paul Sandoz]

Better! still some artifacts:

  0.05%   ↗    0x000000011ee23a72: mov    %eax,%r9d          ;*checkcast {reexecute=0 rethrow=0 return_oop=0}
          │                                                  ; - jdk.incubator.vector.Int128Vector$Int128Species::fromArray at 33 (line 673)
          │                                                  ; - jdk.incubator.vector.Int128Vector$Int128Species::fromArray at 3 (line 564)
          │                                                  ; - jmh.AddTest::add at 31 (line 53)
          │                                                  ; - jmh.generated.AddTest_add_jmhTest::add_avgt_jmhStub at 17 (line 186)
 25.26%   │ ↗  0x000000011ee23a75: vmovdqu 0x10(%rdx,%r9,4),%xmm0  ;*invokedynamic {reexecute=0 rethrow=0 return_oop=0}
          │ │                                                ; - jdk.incubator.vector.Int128Vector$Int128Species::fromArray at 25 (line 673)
          │ │                                                ; - jdk.incubator.vector.Int128Vector$Int128Species::fromArray at 3 (line 564)
          │ │                                                ; - jmh.AddTest::add at 19 (line 52)
          │ │                                                ; - jmh.generated.AddTest_add_jmhTest::add_avgt_jmhStub at 17 (line 186)
  2.67%   │ │  0x000000011ee23a7c: cmp    %r11d,%r9d
         ╭│ │  0x000000011ee23a7f: jae    0x000000011ee23b8c  ;*invokestatic store {reexecute=0 rethrow=0 return_oop=0}
         ││ │                                                ; - jdk.incubator.vector.Int128Vector::intoArray at 30 (line 250)
         ││ │                                                ; - jmh.AddTest::add at 45 (line 54)
         ││ │                                                ; - jmh.generated.AddTest_add_jmhTest::add_avgt_jmhStub at 17 (line 186)
  3.80%  ││ │  0x000000011ee23a85: vpaddd 0x10(%rsi,%r9,4),%xmm0,%xmm0
         ││ │                                                ;*invokestatic checkIndex {reexecute=0 rethrow=0 return_oop=0}
         ││ │                                                ; - java.util.Objects::checkIndex at 3 (line 372)
         ││ │                                                ; - jdk.incubator.vector.VectorIntrinsics::checkIndex at 45 (line 126)
         ││ │                                                ; - jdk.incubator.vector.Int128Vector$Int128Species::fromArray at 11 (line 672)
         ││ │                                                ; - jdk.incubator.vector.Int128Vector$Int128Species::fromArray at 3 (line 564)
         ││ │                                                ; - jmh.AddTest::add at 19 (line 52)
         ││ │                                                ; - jmh.generated.AddTest_add_jmhTest::add_avgt_jmhStub at 17 (line 186)
 33.52%  ││ │  0x000000011ee23a8c: cmp    %ecx,%r9d
         ││ │  0x000000011ee23a8f: jae    0x000000011ee23bb2  ;*invokestatic store {reexecute=0 rethrow=0 return_oop=0}
         ││ │                                                ; - jdk.incubator.vector.Int128Vector::intoArray at 30 (line 250)
         ││ │                                                ; - jmh.AddTest::add at 45 (line 54)
         ││ │                                                ; - jmh.generated.AddTest_add_jmhTest::add_avgt_jmhStub at 17 (line 186)
  0.10%  ││ │  0x000000011ee23a95: vmovdqu %xmm0,0x10(%rbx,%r9,4)  ;*invokestatic linkToStatic {reexecute=0 rethrow=0 return_oop=0}
         ││ │                                                ; - java.lang.invoke.DirectMethodHandle$Holder::invokeStatic at 10
         ││ │                                                ; - java.lang.invoke.LambdaForm$MH/787867107::linkToTargetMethod at 5
         ││ │                                                ; - jdk.incubator.vector.Int128Vector$Int128Species::fromArray at 25 (line 673)
         ││ │                                                ; - jdk.incubator.vector.Int128Vector$Int128Species::fromArray at 3 (line 564)
         ││ │                                                ; - jmh.AddTest::add at 19 (line 52)
         ││ │                                                ; - jmh.generated.AddTest_add_jmhTest::add_avgt_jmhStub at 17 (line 186)
 25.34%  ││ │  0x000000011ee23a9c: add    $0x4,%eax
  3.70%  ││ │  0x000000011ee23a9f: cmp    %edi,%eax
         │╰ │  0x000000011ee23aa1: jl     0x000000011ee23a72  ;*synchronization entry


I think you can drop the first index check for case 2. IIUC you are assuming constraints vlen > 0 and length >= 0, so for:

  Objects.checkIndex(ix, length - (vlen - 1));

the check will fail if length - (vlen - 1) < 0.
 
However, curiously, the single check makes no difference, the same code is generated.

Paul.

> On Feb 15, 2018, at 10:44 AM, Vladimir Ivanov <vladimir.x.ivanov at oracle.com> wrote:
> 
> Here's a quick prototype of enhanced array OOB checks:
> 
>  http://cr.openjdk.java.net/~vlivanov/panama/vector.oob/webrev.00
> 
> -Djdk.incubator.vector.VECTOR_ACCESS_OOB_CHECK:
> 
> =0: no checks
> 
> =1: use Objects.checkFromIndexSize (default)
> 
> =2: enhanced check
> +                // vlen > 0, length >=0
> +                //
> +                // 0 <= i, (i + vlen) <= length
> +                //   =>
> +                // 0 <= vlen <= length, 0 <= i < (length - vlen);
> +                Objects.checkIndex(vlen - 1, length);
> +                Objects.checkIndex(ix, length - (vlen - 1));
> 
> Best regards,
> Vladimir Ivanov
> 
> On 2/15/18 3:58 PM, Vladimir Ivanov wrote:
>>> I have been playing around with a simple benchmark and JMH (separately i can get asm hotspots working on the mac now via dtrace!)
>> ...
>>> Vladimir, i guess this is the kind of thing you were mentioning with regards to bounds checks?
>> Yes.
>>> Perhaps there are general optimization possibilities for such bounds checks. Only Preconditions.checkIndex is currently an intrinsic.
>> I proposed some tweaks earlier [1], but haven't done any experiments yet:
>>   0 <= i
>>   (i + vlen) <= length
>> ==>
>>   (a) 0 <= i <= (length - vlen)
>>   (b) (vlen <= length)
>> a - upper bound is loop invariant
>> b - loop invariant
>> Best regards,
>> Vladimir Ivanov
>> [1] http://mail.openjdk.java.net/pipermail/panama-dev/2017-December/000889.html