Even with the fix for JDK-8364579, the tests from JDK-8364579 (both the ones attached to the bug as well as the one pushed as regression test) still trigger issues on AArch64.

# A fatal error has been detected by the Java Runtime Environment:
#
# Internal Error (/home/opc/valhalla/open/src/hotspot/share/oops/compressedKlass.inline.hpp:80), pid=958046, tid=958048
# assert(is_encodable(addr)) failed: Address 0x00000000c38726ce is not encodable (Klass range: [0x000000009f000000 - 0x00000000df000000), (1073741824 bytes), klass alignment: 8)
#
# JRE version: Java(TM) SE Runtime Environment (25.0) (fastdebug build 25-lworld5ea-LTS-2025-08-22-1512040.opc...)
# Java VM: Java HotSpot(TM) 64-Bit Server VM (fastdebug 25-lworld5ea-LTS-2025-08-22-1512040.opc..., mixed mode, tiered, compressed oops, compressed class ptrs, g1 gc, linux-aarch64)
# Problematic frame:
# V [libjvm.so+0x36542c] CompressedKlassPointers::decode_not_null(unsigned int)+0x16c
#

Stack: [0x0000ffffc6108000,0x0000ffffc6306000], sp=0x0000ffffc6303b50, free space=2030k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native code)
V [libjvm.so+0x36542c] CompressedKlassPointers::decode_not_null(unsigned int)+0x16c (compressedKlass.inline.hpp:80)
V [libjvm.so+0xc74654] G1FullGCMarker::mark_object(oop)+0x174

#
# A fatal error has been detected by the Java Runtime Environment:
#
# SIGSEGV (0xb) at pc=0x0000ffff6b4a94c4, pid=958122, tid=958144
#
# JRE version: Java(TM) SE Runtime Environment (25.0) (fastdebug build 25-lworld5ea-LTS-2025-08-22-1512040.opc...)
# Java VM: Java HotSpot(TM) 64-Bit Server VM (fastdebug 25-lworld5ea-LTS-2025-08-22-1512040.opc..., mixed mode, tiered, compressed oops, compressed class ptrs, g1 gc, linux-aarch64)
# Problematic frame:
# J 41 c1 TestVirtualThreads2$LargeValueWithOops.verify(Ljava/lang/String;Ljava/lang/Object;)V (63 bytes) @ 0x0000ffff6b4a94c4 [0x0000ffff6b4a8c40+0x0000000000000884]
#

The issues reproduce with:
-XX:-InlineTypeReturnedAsFields

But not with
-XX:-InlineTypePassFieldsAsArgs

-XX:+VerifyOops also triggers which suggests that there is (at least) an invalid oop.

Also reproduces with -XX:+UseParallelGC but then we always seem to hit this:

Exception thrown: java.lang.NullPointerException: Cannot invoke "Object.toString()" because "obj" is null
java.lang.NullPointerException: Cannot invoke "Object.toString()" because "obj" is null
        at java.base/java.lang.String.valueOf(String.java:4525)
        at TestVirtualThreads$LargeValueWithOops.verify(TestVirtualThreads.java:235)
        at TestVirtualThreads$LargeValueWithOops.verify(TestVirtualThreads.java:245)
        at TestVirtualThreads.testLargeValueWithOopsHelper(TestVirtualThreads.java:432)
        at TestVirtualThreads.lambda$main$0(TestVirtualThreads.java:592)
        at java.base/java.lang.Thread.run(Thread.java:1447)

We should enable test/hotspot/jtreg/compiler/valhalla/inlinetypes/TestOopsInReturnConvention.java on AArch64 once this issue is fixed.