[lworld+vector] RFR: 8319972: [lworld+vector] Enable intrinsification of Unsafe.finishPrivateBuffer.

[Xiaohong Gong]

On Tue, 14 Nov 2023 15:00:37 GMT, Jatin Bhateja <jbhateja at openjdk.org> wrote:

> Re-enabling intrinsification of Unsafe.finishPrivateBuffer to fix performance degradation seen in Vector API JMH micro benchmarks. It's intrinsifation was disabled to fix incorrectness caused due to missing inductive InlineType Phi node in loops with Unsafe.put* operations. Since Unsafe.put* APIs returns a void hence JVM state is not altered and ciTypeFlow dataflow analysis does not encounter a inductive definition in the loop, due to this C2 parser misses creating an inductive phi node on
> encountering [loop header block](https://github.com/openjdk/valhalla/blob/lworld%2Bvector/src/hotspot/share/opto/parse1.cpp#L696). 
> 
> In order to maintain IR compliance with ciTypeFlow analysis C2 should prevent scalarizing value object b/w make and finish private buffer calls.
> 
> New behavior of unsafe inline expanders
> 1) makePrivate: Receive InlineTypeNode input and return initialized buffer in larval state.
> 2) finishPrivateBuffer: Receive value object buffer input and return rematerialize InlineTypeNode in non-larval state.
> 
> This may result into creation of unbalanced phi node at control flow merges where one of the phi input may InlineTypeNode and other is a buffer of compatible value type, but the IR still remains valid.
> 
> In addition compiler is now preventing elimination of allocation in larval state.
> 
> Validation Status:-
>   - All the VectorAPI JTREG tests are now passing at various AVX level with / wo -XX:+DoptimizeALot.
> 
> Kindly review and share your feedback.
> 
> Best Regards,
> Jatin

Do you think it's better to push the main changes (I mean the elimination scalarization of larval state buffer) to `lworld` branch, which may get more reviews from other people?

src/hotspot/share/opto/graphKit.cpp line 1968:

> 1966:         // and transition the allocation into larval state.
> 1967:         ret = ret->as_InlineType()->make_larval(this);
> 1968:       }

As my understanding, the expected behavior is compiler should guarantee the returned `InlineTypeNode` has a larval state buffer after calling `makePrivateBuffer()`, right? Why do we still need this here?

src/hotspot/share/opto/library_call.cpp line 2667:

> 2665:      return false;
> 2666:   }
> 2667:   const TypeInstPtr* ptr = buffer->bottom_type()->isa_instptr();

Please move this before line-2680.

src/hotspot/share/opto/macro.cpp line 964:

> 962:   if (alloc->_larval) {
> 963:     return false;
> 964:   }

Does it need to also add such checks in `InlineTypeNode::make_scalar_in_safepoints` ?

-------------

PR Comment: https://git.openjdk.org/valhalla/pull/952#issuecomment-1811888621
PR Review Comment: https://git.openjdk.org/valhalla/pull/952#discussion_r1393699005
PR Review Comment: https://git.openjdk.org/valhalla/pull/952#discussion_r1393706966
PR Review Comment: https://git.openjdk.org/valhalla/pull/952#discussion_r1393709519