RFR: 8317121: vector_masked_load instruction is moved too early after JDK-8286941

[Vladimir Kozlov]

On Mon, 2 Oct 2023 19:17:40 GMT, Vladimir Kozlov <kvn at openjdk.org> wrote:

> [JDK-8286941](https://bugs.openjdk.org/browse/JDK-8286941) changed how masked vector operations are created and processed. It calls `LoadNode::Ideal()` which tries to find best place in code for it.
> Unfortunately some places do not expect to have vector access to modified memory.
> In `LoadNode::find_previous_arraycopy()` the call to `ArrayCopyNode::modifies()` assumes that it access only one element and did not take into account that a vector may access wider memory.
> The same issue exists with Unsafe mismatching (bigger size) access.
> 
> The fix is to take into account size of memory access.
> 
> New regression test is added which show the issue.
> 
> Tested tier1-5. xcomp, stress. Testing tier6-7 in progress.

White space in test will be fixed in next update.

GHA has 2 failures.

On macos-x64 `gc/TestAllocHumongousFragment.java#aggressive` test failed when run with `-XX:+UseShenandoahGC`. It looks like [JDK-8298781](https://bugs.openjdk.org/browse/JDK-8298781)

On windows-x64 `gc/TestJNICriticalStressTest.java#id0` failed with `-XX:+UseG1GC -Xms1g -Xmx1g`:

java.lang.OutOfMemoryError: Java heap space
	at TestJNICriticalStressTest$JNICriticalWorker.doStep(TestJNICriticalStressTest.java:122)
	at TestJNICriticalStressTest$JNICriticalWorker.run(TestJNICriticalStressTest.java:140)

I can't find similar issue in JBS or our internal testing. But I doubt it is related to this changes.

The `gc/TestJNICriticalStressTest.java#id0` failure could be due to [JDK-8308507](https://bugs.openjdk.org/browse/JDK-8308507)
The last GC messages were:

[10.094s][info][gc] GC(45) Pause Young (Normal) (G1 Evacuation Pause) 492M->13M(1024M) 43.488ms
Warning: [10.353s][warning][gc,alloc] Thread-5: Retried waiting for GCLocker too often allocating 2050 words
[10.397s][info   ][gc      ] GC(46) Pause Young (Normal) (GCLocker Initiated GC) 500M->13M(1024M) 43.932ms

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

PR Comment: https://git.openjdk.org/jdk/pull/16015#issuecomment-1743983127
PR Comment: https://git.openjdk.org/jdk/pull/16015#issuecomment-1743986334