RFR: 8327963: C2: fix construction of memory graph around Initialize node to prevent incorrect execution if allocation is removed [v7]

[Roberto Castañeda Lozano]

On Wed, 21 May 2025 09:21:26 GMT, Roland Westrelin <roland at openjdk.org> wrote:

>> An `Initialize` node for an `Allocate` node is created with a memory
>> `Proj` of adr type raw memory. In order for stores to be captured, the
>> memory state out of the allocation is a `MergeMem` with slices for the
>> various object fields/array element set to the raw memory `Proj` of
>> the `Initialize` node. If `Phi`s need to be created during later
>> transformations from this memory state, The `Phi` for a particular
>> slice gets its adr type from the type of the `Proj` which is raw
>> memory. If during macro expansion, the `Allocate` is found to have no
>> use and so can be removed, the `Proj` out of the `Initialize` is
>> replaced by the memory state on input to the `Allocate`. A `Phi` for
>> some slice for a field of an object will end up with the raw memory
>> state on input to the `Allocate` node. As a result, memory state at
>> the `Phi` is incorrect and incorrect execution can happen.
>> 
>> The fix I propose is, rather than have a single `Proj` for the memory
>> state out of the `Initialize` with adr type raw memory, to use one
>> `Proj` per slice added to the memory state after the `Initalize`. Each
>> of the `Proj` should return the right adr type for its slice. For that
>> I propose having a new type of `Proj`: `NarrowMemProj` that captures
>> the right adr type.
>> 
>> Logic for the construction of the `Allocate`/`Initialize` subgraph is
>> tweaked so the right adr type captured in is own `NarrowMemProj` is
>> added to the memory sugraph. Code that removes an allocation or moves
>> it also has to be changed so it correctly takes the multiple memory
>> projections out of the `Initialize` node into account.
>> 
>> One tricky issue is that when EA split types for a scalar replaceable
>> `Allocate` node:
>> 
>> 1- the adr type captured in the `NarrowMemProj` becomes out of sync
>>   with the type of the slices for the allocation
>>   
>> 2- before EA, the memory state for one particular field out of the
>>   `Initialize` node can be used for a `Store` to the just allocated
>>   object or some other. So we can have a chain of `Store`s, some to
>>   the newly allocated object, some to some other objects, all of them
>>   using the state of `NarrowMemProj` out of the `Initialize`. After
>>   split unique types, the `NarrowMemProj` is for the slice of a
>>   particular allocation. So `Store`s to some other objects shouldn't
>>   use that memory state but the memory state before the `Allocate`.
>>   
>> For that, I added logic to update the adr type of `NarrowMemProj`
>> during split uni...
>
> Roland Westrelin has updated the pull request with a new target base due to a merge or a rebase. The incremental webrev excludes the unrelated changes brought in by the merge/rebase. The pull request contains 33 additional commits since the last revision:
> 
>  - new test tweak
>  - new test
>  - Merge branch 'master' into JDK-8327963
>  - Merge branch 'master' into JDK-8327963
>  - typo
>  - more
>  - more
>  - more
>  - more
>  - more
>  - ... and 23 more: https://git.openjdk.org/jdk/compare/4f595bfb...a6c6c044

test/hotspot/jtreg/compiler/macronodes/TestEarlyEliminationOfAllocationWithoutUse.java line 54:

> 52:     private static void test1(A otherA, boolean[] flags) {
> 53:         if (flags == null) {
> 54:         }

Consider removing these two lines, which do not seem essential to reproduce the issue. Removing them also gets rid of the deoptimization, which simplifies the failure analysis.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/24570#discussion_r2099904747