RFR: 8326962: C2 SuperWord: cache VPointer

[Johan Sjölen]

On Tue, 2 Apr 2024 09:04:45 GMT, Emanuel Peter <epeter at openjdk.org> wrote:

> This is a subtask of [JDK-8315361](https://bugs.openjdk.org/browse/JDK-8315361).
> 
> Parsing `VPointer` currently happens all over SuperWord. And often in quadratic loops, where we compare all-with-all loads/stores.
> 
> I propose to cache the `VPointer`s, then we can do a constant-time cache lookup rather than parsing the pointer subgraph every time.
> 
> There are now only a few cases where we cannot use the cached `VPointer`:
> - `SuperWord::unrolling_analysis`: we have no `VLoopAnalyzer`, and so no submodules like `VLoopPointers`. We don't need to cache, since we only iterate over the loop body once, and create only a single `VPointer` per memop.
> - `SuperWord::output`: when we have a `Load`, and try to bypass `StoreVector` nodes. The `StoreVector` nodes are new, and so we have no cached `VPointer` for them. This could be fixed somehow, but I don't want to deal with it now. I intend to refactor `SuperWord::output` soon, and can look into options at that point (either I bypass before we insert the vector nodes, or I remember what scalar memop the vector was created from, and then get the cached pointer this way).
> 
> This changeset is also a preparation step for [JDK-8325155](https://bugs.openjdk.org/browse/JDK-8325155). I will have a list of pointers, and sort them such that creating adjacent refs is much more efficient.
> 
> **Benchmarking SuperWord Compile Time**
> 
> I use the same benchmark from https://github.com/openjdk/jdk/pull/18532.
> 
> On master:
> 
>     C2 Compile Time:       56.816 s
>          IdealLoop:            56.604 s
>            AutoVectorize:      56.192 s
> 
> 
> With this patch:
> 
>     C2 Compile Time:       49.719 s
>          IdealLoop:            49.509 s
>            AutoVectorize:      49.106 s
> 
> 
> This saves us about `7 sec`, which is significant. I will have to see what it effect it has once we also apply https://github.com/openjdk/jdk/pull/18532, but I think the combined effect will be very significant.

Hi Emanuel,

I've some general questions regarding naming and Arena usage, I hope you don't mind some runtime team input.

src/hotspot/share/opto/superword.hpp line 499:

> 497: 
> 498:   // VLoopDependencyGraph Accessors
> 499:   const VPointer& get_pointer(const MemNode* mem) const {

This can't just be called `vpointer` or `vpointer_of`?

src/hotspot/share/opto/vectorization.hpp line 458:

> 456: class VLoopPointers : public StackObj {
> 457: private:
> 458:   Arena*                   _arena;

Will the pointer ever change? Could potentially change this to a reference.

src/hotspot/share/opto/vectorization.hpp line 483:

> 481: 
> 482:   void compute_and_cache();
> 483:   const VPointer& get(const MemNode* mem) const;

Questioning naming of this also :-).

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

PR Review: https://git.openjdk.org/jdk/pull/18577#pullrequestreview-1973893348
PR Review Comment: https://git.openjdk.org/jdk/pull/18577#discussion_r1547931080
PR Review Comment: https://git.openjdk.org/jdk/pull/18577#discussion_r1547937229
PR Review Comment: https://git.openjdk.org/jdk/pull/18577#discussion_r1547934364