-
Bug
-
Resolution: Fixed
-
P2
-
21
-
b26
-
aarch64
-
Verified
jdk/incubator/vector/Float64VectorTests.java fails with:
# A fatal error has been detected by the Java Runtime Environment:
#
# Internal Error (assembler_aarch64.hpp:2641), pid=586848, tid=586863
# guarantee(T != T2S) failed: incorrect arrangement
#
# JRE version: Java(TM) SE Runtime Environment (21.0+25) (build 21-ea+25-LTS-2170)
# Java VM: Java HotSpot(TM) 64-Bit Server VM (21-ea+25-LTS-2170, mixed mode, sharing, compressed oops, compressed class ptrs, g1 gc, linux-aarch64)
# Problematic frame:
# V [libjvm.so+0x36ef04] Assembler::addv(FloatRegister, Assembler::SIMD_Arrangement, FloatRegister)+0xe4
Current CompileTask:
C2: 9077 2118 % b Float64VectorTests::maskCompressFloat64VectorTestsSmokeTest @ 22 (135 bytes)
Stack: [0x0000ffff92dfd000,0x0000ffff92ffd000], sp=0x0000ffff92ff8c40, free space=2031k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native code)
V [libjvm.so+0x36ef04] Assembler::addv(FloatRegister, Assembler::SIMD_Arrangement, FloatRegister)+0xe4 (assembler_aarch64.hpp:2641)
V [libjvm.so+0x35a2a4] vstoremask_truecount_neonNode::emit(CodeBuffer&, PhaseRegAlloc*) const+0x324 (aarch64_vector.ad:5531)
V [libjvm.so+0xb89918] PhaseOutput::scratch_emit_size(Node const*)+0x1a8 (output.cpp:3245)
V [libjvm.so+0xb862d8] PhaseOutput::shorten_branches(unsigned int*)+0x1d8 (output.cpp:539)
V [libjvm.so+0xb8cec0] PhaseOutput::Output()+0x480 (output.cpp:340)
V [libjvm.so+0x598cc8] Compile::Code_Gen()+0x458 (compile.cpp:2971)
V [libjvm.so+0x59c134] Compile::Compile(ciEnv*, ciMethod*, int, Options, DirectiveSet*)+0x10b4 (compile.cpp:875)
V [libjvm.so+0x4d2cac] C2Compiler::compile_method(ciEnv*, ciMethod*, int, bool, DirectiveSet*)+0xf8 (c2compiler.cpp:118)
V [libjvm.so+0x5a1614] CompileBroker::invoke_compiler_on_method(CompileTask*)+0x9b4 (compileBroker.cpp:2265)
V [libjvm.so+0x5a3ed8] CompileBroker::compiler_thread_loop()+0x384 (compileBroker.cpp:1944)
V [libjvm.so+0x80b354] JavaThread::thread_main_inner() [clone .part.0]+0xa4 (javaThread.cpp:719)
V [libjvm.so+0xced608] Thread::call_run()+0xa8 (thread.cpp:217)
V [libjvm.so+0xb767dc] thread_native_entry(Thread*)+0xdc (os_linux.cpp:775)
# A fatal error has been detected by the Java Runtime Environment:
#
# Internal Error (assembler_aarch64.hpp:2641), pid=586848, tid=586863
# guarantee(T != T2S) failed: incorrect arrangement
#
# JRE version: Java(TM) SE Runtime Environment (21.0+25) (build 21-ea+25-LTS-2170)
# Java VM: Java HotSpot(TM) 64-Bit Server VM (21-ea+25-LTS-2170, mixed mode, sharing, compressed oops, compressed class ptrs, g1 gc, linux-aarch64)
# Problematic frame:
# V [libjvm.so+0x36ef04] Assembler::addv(FloatRegister, Assembler::SIMD_Arrangement, FloatRegister)+0xe4
Current CompileTask:
C2: 9077 2118 % b Float64VectorTests::maskCompressFloat64VectorTestsSmokeTest @ 22 (135 bytes)
Stack: [0x0000ffff92dfd000,0x0000ffff92ffd000], sp=0x0000ffff92ff8c40, free space=2031k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native code)
V [libjvm.so+0x36ef04] Assembler::addv(FloatRegister, Assembler::SIMD_Arrangement, FloatRegister)+0xe4 (assembler_aarch64.hpp:2641)
V [libjvm.so+0x35a2a4] vstoremask_truecount_neonNode::emit(CodeBuffer&, PhaseRegAlloc*) const+0x324 (aarch64_vector.ad:5531)
V [libjvm.so+0xb89918] PhaseOutput::scratch_emit_size(Node const*)+0x1a8 (output.cpp:3245)
V [libjvm.so+0xb862d8] PhaseOutput::shorten_branches(unsigned int*)+0x1d8 (output.cpp:539)
V [libjvm.so+0xb8cec0] PhaseOutput::Output()+0x480 (output.cpp:340)
V [libjvm.so+0x598cc8] Compile::Code_Gen()+0x458 (compile.cpp:2971)
V [libjvm.so+0x59c134] Compile::Compile(ciEnv*, ciMethod*, int, Options, DirectiveSet*)+0x10b4 (compile.cpp:875)
V [libjvm.so+0x4d2cac] C2Compiler::compile_method(ciEnv*, ciMethod*, int, bool, DirectiveSet*)+0xf8 (c2compiler.cpp:118)
V [libjvm.so+0x5a1614] CompileBroker::invoke_compiler_on_method(CompileTask*)+0x9b4 (compileBroker.cpp:2265)
V [libjvm.so+0x5a3ed8] CompileBroker::compiler_thread_loop()+0x384 (compileBroker.cpp:1944)
V [libjvm.so+0x80b354] JavaThread::thread_main_inner() [clone .part.0]+0xa4 (javaThread.cpp:719)
V [libjvm.so+0xced608] Thread::call_run()+0xa8 (thread.cpp:217)
V [libjvm.so+0xb767dc] thread_native_entry(Thread*)+0xdc (os_linux.cpp:775)
- relates to
-
JDK-8307795 AArch64: Optimize VectorMask.truecount() on Neon
- Resolved