RFR: 8341293: Split field loads through Nested Phis [v4]

[Emanuel Peter]

On Wed, 22 Jan 2025 01:32:22 GMT, Dhamoder Nalla <dhanalla at openjdk.org> wrote:

>> As an extension of the work done as part of https://github.com/openjdk/jdk/pull/12897, split the field loads (AddP -> Load*) with nested phi parent nodes to enable more scalar replacements, thereby reducing memory allocation.
>> 
>> 
>> Here are the sequence of Ideal graph transformations for Nested phi:
>> 
>>  
>> ![image](https://github.com/user-attachments/assets/c18e5ca0-c554-475c-814a-7cb288d96569)
>> 
>> ![image](https://github.com/user-attachments/assets/b279b5f2-9ec6-4d9b-a627-506451f1cf81)
>> 
>> ![image](https://github.com/user-attachments/assets/f506b918-2dd0-4dbe-a440-ff253afa3961)
>> 
>> JMH results:
>> with disabled RAM
>> 
>> Benchmark Mode Cnt Score Error Units
>> NestedPhiAndRematerialize.NopRAM.testBailOut_runner avgt 15 13.969 ± 0.248 ms/op
>> NestedPhiAndRematerialize.NopRAM.testFieldEscapeWithMerge_runner avgt 15 80.300 ± 4.306 ms/op
>> NestedPhiAndRematerialize.NopRAM.testMerge_TryCatchFinally_runner avgt 15 72.182 ± 1.781 ms/op
>> NestedPhiAndRematerialize.NopRAM.testMultiParentPhi_runner avgt 15 2.983 ± 0.001 ms/op
>> NestedPhiAndRematerialize.NopRAM.testNestedPhiPolymorphic_runner avgt 15 18.342 ± 0.731 ms/op
>> NestedPhiAndRematerialize.NopRAM.testNestedPhiProcessOrder_runner avgt 15 14.315 ± 0.443 ms/op
>> NestedPhiAndRematerialize.NopRAM.testNestedPhiWithLambda_runner avgt 15 18.511 ± 1.212 ms/op
>> NestedPhiAndRematerialize.NopRAM.testNestedPhiWithTrap_runner avgt 15 66.277 ± 1.478 ms/op
>> NestedPhiAndRematerialize.NopRAM.testNestedPhi_FieldLoad_runner avgt 15 17.968 ± 0.306 ms/op
>> NestedPhiAndRematerialize.NopRAM.testNestedPhi_TryCatch_runner avgt 15 14.186 ± 0.247 ms/op
>> NestedPhiAndRematerialize.NopRAM.testRematerialize_MultiObj_runner avgt 15 88.435 ± 4.869 ms/op
>> NestedPhiAndRematerialize.NopRAM.testRematerialize_SingleObj_runner avgt 15 29560.130 ± 48.797 ms/op
>> NestedPhiAndRematerialize.NopRAM.testRematerialize_TryCatch_runner avgt 15 49.150 ± 2.307 ms/op
>> NestedPhiAndRematerialize.NopRAM.testThreeLevelNestedPhi_runner avgt 15 18.236 ± 0.308 ms/op
>> 
>> with enabled RAM
>> Benchmark Mode Cnt Score Error Units
>> NestedPhiAndRematerialize.YesRAM.testBailOut_runner avgt 15 3.257 ± 0.423 ms/op
>> NestedPhiAndRematerialize.YesRAM.testFieldEscapeWithMerge_runner avgt 15 79.916 ± 3.477 ms/op
>> NestedPhiAndRematerialize.YesRAM.testMerge_TryCatchFinally_runner avgt 15 72.053 ± 1.916 ms/op
>> NestedPhiAndRematerialize.YesRAM.testMultiParentPhi_runner avgt 15 2.984 ± 0.001 ms/op
>> NestedPhiAndRematerialize.YesRAM.testNestedPhiPolymorphic_runner avgt ...
>
> Dhamoder Nalla has updated the pull request incrementally with one additional commit since the last revision:
> 
>   add IR rule applyIf

test/micro/org/openjdk/bench/vm/compiler/AllocationMergesNestedPhi.java line 2:

> 1: /*
> 2:  * Copyright (c) 2024, Oracle and/or its affiliates. All rights reserved.

Drive-by comment: update the copyright years to 2025 ;)

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

PR Review Comment: https://git.openjdk.org/jdk/pull/21270#discussion_r1924799652