RFR: 8287788: reuse intermediate segments allocated during FFM stub invocations [v4]

[Maurizio Cimadamore]

On Mon, 20 Jan 2025 16:34:15 GMT, Matthias Ernst <duke at openjdk.org> wrote:

>> Certain signatures for foreign function calls (e.g. HVA return by value) require allocation of an intermediate buffer to adapt the FFM's to the native stub's calling convention. In the current implementation, this buffer is malloced and freed on every FFM invocation, a non-negligible overhead.
>> 
>> Sample stack trace:
>> 
>>    java.lang.Thread.State: RUNNABLE
>> 	at jdk.internal.misc.Unsafe.allocateMemory0(java.base at 25-ea/Native Method)
>> ...
>> 	at jdk.internal.foreign.abi.SharedUtils.newBoundedArena(java.base at 25-ea/SharedUtils.java:386)
>> 	at jdk.internal.foreign.abi.DowncallStub/0x000001f001084c00.invoke(java.base at 25-ea/Unknown Source)
>> ...
>> 	at java.lang.invoke.Invokers$Holder.invokeExact_MT(java.base at 25-ea/Invokers$Holder)
>> 
>> 
>> To alleviate this, this PR remembers and reuses up to two small intermediate buffers per carrier-thread in subsequent calls.
>> 
>> Performance (MBA M3):
>> 
>> 
>> master at 764d70b7df18e288582e616c62b0d7078f1ff3aa
>> Benchmark                         Mode  Cnt    Score    Error  Units
>> PointsAlloc.circle_by_ptr         avgt   30    9.197 ?  0.037  ns/op
>> PointsAlloc.circle_by_value       avgt   30   42.195 ?  0.088  ns/op   <= #######
>> PointsAlloc.jni_ByteBuffer_alloc  avgt   30  226.127 ? 35.378  ns/op
>> PointsAlloc.jni_long_alloc        avgt   30   25.297 ?  2.457  ns/op
>> PointsAlloc.panama_alloc          avgt   30   27.053 ?  1.915  ns/op
>> 
>> After:
>> Benchmark                         Mode  Cnt    Score    Error  Units
>> PointsAlloc.circle_by_ptr         avgt   30    9.156 ?  0.021  ns/op
>> PointsAlloc.circle_by_value       avgt   30   11.995 ?  0.051  ns/op   <= #######
>> PointsAlloc.jni_ByteBuffer_alloc  avgt   30  211.161 ? 23.284  ns/op
>> PointsAlloc.jni_long_alloc        avgt   30   24.885 ?  2.461  ns/op
>> PointsAlloc.panama_alloc          avgt   30   26.905 ?  1.935  ns/op
>> 
>> 
>> `-prof gc` also shows that the new call path is fully scalar-replaced vs 160 byte/call before.
>
> Matthias Ernst has updated the pull request incrementally with one additional commit since the last revision:
> 
>   reduce visibility

src/java.base/share/classes/jdk/internal/foreign/abi/CallBufferCache.java line 112:

> 110: 
> 111:     @SuppressWarnings("restricted")
> 112:     public static MemorySegment acquireOrAllocate(long requestedSize) {

An even higher-level alternative here would be for this method to return a custom arena, whose `close` implementation does the `release`. This way, no cleanup action is needed. The arena could use a slicing allocator on a pre-allocated, per-thread segment. If the segment is not big enough for the allocation the linker wants to do, we can just return a plain confined arena.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/23142#discussion_r1922709682