Bug ID: JDK-8257192 Integrate AArch64 JIT port into 8u

JDK-8257192 : Integrate AArch64 JIT port into 8u

Type: Enhancement
Component: hotspot
Sub-Component: compiler
Affected Version: 8

Priority: P3
Status: Resolved
Resolution: Fixed
OS: linux
CPU: aarch64

Submitted: 2020-11-27
Updated: 2022-05-12
Resolved: 2021-02-01

Versions (Unresolved/Resolved/Fixed)

The Version table provides details related to the release that this issue/RFE will be addressed.

Unresolved : Release in which this issue/RFE will be addressed.
Resolved: Release in which this issue/RFE has been resolved.
Fixed : Release in which this issue/RFE has been fixed. The release containing this fix may be available for download as an Early Access Release or a General Availability Release.

To download the current JDK release, click here.

Other
openjdk8u292 b01Fixed

Related Reports

Relates :	JDK-8155653 - TestVectorUnalignedOffset.java not pushed with 8155612
Relates :	JDK-8211064 - [AArch64] Interpreter and c1 don't correctly handle jboolean results in native calls
Relates :	JDK-8215951 - AArch64: jtreg test vmTestbase/nsk/jvmti/PopFrame/popframe005 segfaults
Relates :	JDK-8224671 - AArch64: mauve System.arraycopy test failure
Relates :	JDK-8246482 - Build failures with +JFR -PCH
Relates :	JDK-8130309 - Need to bailout cleanly if creation of stubs fails when codecache is out of space
Relates :	JDK-8161190 - AArch64: Fix overflow in immediate cmp instruction
Relates :	JDK-8171537 - aarch64: compiler/c1/Test6849574.java generates guarantee failure in C1
Relates :	JDK-8203481 - Incorrect constraint for unextended_sp in frame:safe_for_sender
Relates :	JDK-8209414 - AArch64: method handle invocation does not respect JVMTI interp_only mode
Relates :	JDK-8211233 - MemBarNode::trailing_membar() and MemBarNode::leading_membar() need to handle dying subgraphs better
Relates :	JDK-8213419 - C2 may hang in MulLNode::Ideal()/MulINode::Ideal() with gcc 8.2.1
Relates :	JDK-8214857 - "bad trailing membar" assert failure at memnode.cpp:3220
Relates :	JDK-8248219 - aarch64: missing memory barrier in fast_storefield and fast_accessfield
Relates :	JDK-8132875 - AArch64: Fix error introduced into AArch64 CodeCache by commit for 8130309
Relates :	JDK-8135018 - AARCH64: Missing memory barriers for CMS collector
Relates :	JDK-8148783 - aarch64: SEGV running SpecJBB2013
Relates :	JDK-8148948 - aarch64: generate_copy_longs calls align() incorrectly
Relates :	JDK-8150045 - AArch64: arraycopy causes segfaults in SATB during garbage collection
Relates :	JDK-8153172 - aarch64: hotspot crashes after the 8.1 LSE patch is merged
Relates :	JDK-8153797 - aarch64: Add Arrays.fill stub code
Relates :	JDK-8154413 - AArch64: Better byte behaviour
Relates :	JDK-8154537 - AArch64: some integer rotate instructions are never emitted
Relates :	JDK-8154739 - AArch64: TemplateTable::fast_xaccess loads in wrong mode
Relates :	JDK-8155015 - Aarch64: bad assert in spill generation code
Relates :	JDK-8155100 - AArch64: Relax alignment requirement for byte_map_base
Relates :	JDK-8155612 - Aarch64: vector nodes need to support misaligned offset
Relates :	JDK-8155617 - aarch64: ClearArray does not use DC ZVA
Relates :	JDK-8155627 - Enable SA on AArch64
Relates :	JDK-8156731 - aarch64: java/util/Arrays/Correct.java fails due to _generic_arraycopy stub routine
Relates :	JDK-8157841 - aarch64: prefetch ignores cache line size
Relates :	JDK-8157906 - aarch64: some more integer rotate instructions are never emitted
Relates :	JDK-8158913 - aarch64: SEGV running Spark terasort
Relates :	JDK-8161072 - AArch64: jtreg compiler/uncommontrap/TestDeoptOOM failure
Relates :	JDK-8164113 - AArch64: follow-up the fix for 8161598
Relates :	JDK-8165673 - AArch64: Fix JNI floating point argument handling
Relates :	JDK-8167200 - AArch64: Broken stack pointer adjustment in interpreter
Relates :	JDK-8167421 - AArch64: in one core system, fatal error: Illegal threadstate encountered
Relates :	JDK-8167595 - AArch64: SEGV in stub code cipherBlockChaining_decryptAESCrypt
Relates :	JDK-8168888 - AArch64 port of JDK-8160591
Relates :	JDK-8170100 - AArch64: Crash in C1-compiled code accessing References
Relates :	JDK-8170188 - aarch64: jtreg test compiler/types/TestMeetIncompatibleInterfaceArrays.java causes JVM crash
Relates :	JDK-8170873 - PPC64/aarch64: Poor StrictMath performance due to non-optimized compilation
Relates :	JDK-8182581 - aarch64: fix for crash caused by earlyret of compiled method
Relates :	JDK-8186325 - AArch64: jtreg test hotspot/test/gc/g1/TestJNIWeakG1/TestJNIWeakG1.java SEGV
Relates :	JDK-8203699 - java/lang/invoke/SpecialInterfaceCall fails with SIGILL on aarch64
Relates :	JDK-8209413 - AArch64: NPE in clhsdb jstack command
Relates :	JDK-8209415 - Fix JVMTI test failure HS202
Relates :	JDK-8213134 - AArch64: vector shift failed with MaxVectorSize=8
Relates :	JDK-8215961 - jdk/jfr/event/os/TestCPUInformation.java fails on AArch64
Relates :	JDK-8216989 - CardTableBarrierSetAssembler::gen_write_ref_array_post_barrier() does not check for zero length on AARCH64
Relates :	JDK-8217368 - AArch64: C2 recursive stack locking optimisation not triggered
Relates :	JDK-8218185 - aarch64: missing LoadStore barrier in TemplateTable::putfield_or_static
Relates :	JDK-8219635 - aarch64: missing LoadStore barrier in TemplateTable::fast_storefield
Relates :	JDK-8221220 - AArch64: Add StoreStore membar explicitly for Volatile Writes in TemplateTable
Relates :	JDK-8221658 - aarch64: add necessary predicate for ubfx patterns
Relates :	JDK-8224828 - aarch64: rflags is not correct after safepoint poll
Relates :	JDK-8224880 - AArch64: java/javac error with AllocatePrefetchDistance
Relates :	JDK-8228406 - Superfluous change in chaitin.hpp
Relates :	JDK-8228718 - Revert incorrect backport of JDK-8129757 to 8-aarch64
Relates :	JDK-8233839 - aarch64: missing memory barrier in NewObjectArrayStub and NewTypeArrayStub
Relates :	JDK-8237512 - AArch64: aarch64TestHook leaks a BufferBlob
Relates :	JDK-8247979 - aarch64: missing side effect of killing flags for clearArray_reg_reg
Relates :	JDK-8195685 - AArch64 cannot build with JDK-8174962
Relates :	JDK-8195859 - AArch64: vtableStubs gtest fails after 8174962
Relates :	JDK-8196136 - AArch64: Correct register use in patch for JDK-8195685
Relates :	JDK-8196221 - AArch64: Mistake in committed patch for JDK-8195859
Relates :	JDK-8206163 - AArch64: incorrect code generation for StoreCM
Relates :	JDK-8207838 - AArch64: Float registers incorrectly restored in JNI call
Relates :	JDK-7009641 - Don't use CodeCache for allocations if it is already full
Relates :	JDK-8172881 - AArch64: assertion failure: the int pressure is incorrect
Relates :	JDK-8173472 - AArch64: C1 comparisons with null only use 32-bit instructions
Relates :	JDK-8176100 - [REDO][REDO] G1 Needs pre barrier on dereference of weak JNI handles
Relates :	JDK-8177661 - AArch64: Incorrect C2 patterns cause system register corruption
Relates :	JDK-8179954 - AArch64: C1 and C2 volatile accesses are not sequentially consistent
Relates :	JDK-8207345 - AArch64: Trampoline generation code reads from uninitialized memory
Relates :	JDK-8187224 - aarch64: some inconsistency between aarch64_ad.m4 and aarch64.ad
Relates :	JDK-8193133 - Assertion failure because 0xDEADDEAD can be in-heap
Relates :	JDK-8224851 - AArch64: fix warnings and errors with Clang and GCC 8.3
Relates :	JDK-8073108 - Use x86 and SPARC CPU instructions for GHASH acceleration
Relates :	JDK-8216350 - AArch64: monitor unlock fast path not called
Relates :	JDK-8131779 - AARCH64: add Montgomery multiply intrinsic
Relates :	JDK-8145320 - Create unsafe_arraycopy and generic_arraycopy for AArch64
Relates :	JDK-8148328 - aarch64: redundant lsr instructions in stub code
Relates :	JDK-8149080 - AArch64: Recognize disjoint array copy in stub code
Relates :	JDK-8149365 - aarch64: memory copy does not prefetch on backwards copy
Relates :	JDK-8149907 - aarch64: use load/store pair instructions in call_stub
Relates :	JDK-8150038 - aarch64: Make use of CBZ and CBNZ when comparing narrow pointer with zero
Relates :	JDK-8150082 - aarch64: optimise small array copy
Relates :	JDK-8150229 - aarch64: pipeline class for several instructions is not set correctly
Relates :	JDK-8150313 - aarch64: optimise array copy using SIMD instructions
Relates :	JDK-8150394 - aarch64: add support for 8.1 LSE CAS instructions
Relates :	JDK-8150652 - Remove unused code in AArch64 back end
Relates :	JDK-8151340 - aarch64: prefetch the destination word for write prior to ldxr/stxr loops
Relates :	JDK-8151502 - aarch64: optimize pd_disjoint_words and pd_conjoint_words
Relates :	JDK-8151775 - aarch64: add support for 8.1 LSE atomic operations
Relates :	JDK-8152537 - aarch64: Make use of CBZ and CBNZ when comparing unsigned values with zero
Relates :	JDK-8152840 - aarch64: improve _unsafe_arraycopy stub routine
Relates :	JDK-8153713 - aarch64: improve short array clearing using store pair
Relates :	JDK-8159052 - AArch64: Optimise unaligned copies in pd_disjoint_words and pd_conjoint_words
Relates :	JDK-8159063 - aarch64: optimise unaligned array copy long
Relates :	JDK-8160748 - Inconsistent types for ideal_reg
Relates :	JDK-8183925 - Decouple crash protection from watcher thread
Relates :	JDK-8189170 - Add option to disable stack overflow checking in primordial thread for use with JNI_CreateJavaJVM
Relates :	JDK-8194686 - Placeholder for AArch64 January 2018 CPU
Relates :	JDK-8205421 - AARCH64: StubCodeMark should be placed after alignment
Relates :	JDK-8209420 - Track membars for volatile accesses so they can be properly optimized
Relates :	JDK-8209835 - Aarch64: elide barriers on all volatile operations
Relates :	JDK-8219011 - Implement MacroAssembler::warn method on AArch64
Relates :	JDK-8228400 - Remove built-in AArch64 simulator
Relates :	JDK-8228593 - Revert explicit JDK 7 support additions
Relates :	JDK-8228725 - AArch64: Purge method call format support
Relates :	JDK-8228747 - Revert "unused" attribute from test_arraycopy_func
Relates :	JDK-8228767 - Revert ResourceMark additions
Relates :	JDK-8228770 - Revert development hsdis changes
Relates :	JDK-8229123 - Revert build fixes for aarch64/zero
Relates :	JDK-8229124 - Revert disassembler.cpp changes
Relates :	JDK-8229145 - Revert TemplateTable::bytecode() visibility change
Relates :	JDK-8199712 - Flight Recorder
Relates :	JDK-8228716 - Revert InstanceKlass::print_on debug additions
Relates :	JDK-8221725 - AArch64 build failures after JDK-8221408 (Windows 32bit build build errors/warnings in hotspot)
Relates :	JDK-8264171 - Missing aarch64 parts of JDK-8236179 (C1 register allocation failure with T_ADDRESS)
Relates :	JDK-8186090 - java.nio.Bits.unaligned() doesn't handle aarch64

Description

The aarch64 port was originally developed at Red Hat against OpenJDK 7 & 8, then later integrated into OpenJDK 9.

The 8u port has continued to be maintained in the aarch64-port/jdk8u and aarch64-port/shenandoah-jdk8u trees since and has been used for Red Hat's AArch64 binaries.

The port has been stable for a long time and finally integrating it into 8u upstream should be low risk. There are few changes to shared HotSpot code (mostly header inclusion) and the 8u version of the port has had over half a decade of production use.

Comments

URL: https://hg.openjdk.java.net/jdk8u/jdk8u/hotspot/rev/f57189b7648d User: andrew Date: 2021-02-02 05:24:08 +0000
02-02-2021
URL: https://hg.openjdk.java.net/jdk8u/jdk8u/jdk/rev/56a21314ba4b User: andrew Date: 2021-02-02 05:23:57 +0000
02-02-2021
URL: https://hg.openjdk.java.net/jdk8u/jdk8u/rev/6812bfa1824c User: andrew Date: 2021-02-02 05:23:05 +0000
02-02-2021
URL: https://hg.openjdk.java.net/jdk8u/jdk8u-dev/hotspot/rev/f57189b7648d User: andrew Date: 2021-02-01 20:00:27 +0000
01-02-2021
URL: https://hg.openjdk.java.net/jdk8u/jdk8u-dev/jdk/rev/56a21314ba4b User: andrew Date: 2021-02-01 20:00:16 +0000
01-02-2021
URL: https://hg.openjdk.java.net/jdk8u/jdk8u-dev/rev/6812bfa1824c User: andrew Date: 2021-02-01 19:59:32 +0000
01-02-2021
Fix Request [8u]: Merge has been reviewed by Aleksey Shipilev: https://mail.openjdk.java.net/pipermail/jdk8u-dev/2021-January/013364.html Informal approval from Andrew Haley: https://mail.openjdk.java.net/pipermail/jdk8u-dev/2020-November/013144.html
29-01-2021
Please don't change this to '8-pool', that does not work for OpenJDK 8u.
30-11-2020