Mercurial > hg > icedtea8-forest > hotspot
view make/linux/makefiles/gcc.make @ 10905:f57189b7648d
8257192: Integrate AArch64 JIT port into 8u
7009641: Don't fail VM when CodeCache is full
8073108: [AArch64] Use x86 and SPARC CPU instructions for GHASH acceleration
8130309: Need to bailout cleanly if creation of stubs fails when codecache is out of space (AArch64 changes)
8131779: AARCH64: add Montgomery multiply intrinsic
8132875: AArch64: Fix error introduced into AArch64 CodeCache by commit for 8130309
8135018: AARCH64: Missing memory barriers for CMS collector
8145320: Create unsafe_arraycopy and generic_arraycopy for AArch64
8148328: aarch64: redundant lsr instructions in stub code.
8148783: aarch64: SEGV running SpecJBB2013
8148948: aarch64: generate_copy_longs calls align() incorrectly
8149080: AArch64: Recognise disjoint array copy in stub code
8149365: aarch64: memory copy does not prefetch on backwards copy
8149907: aarch64: use load/store pair instructions in call_stub
8150038: aarch64: make use of CBZ and CBNZ when comparing narrow pointer with zero
8150045: arraycopy causes segfaults in SATB during garbage collection
8150082: aarch64: optimise small array copy
8150229: aarch64: pipeline class for several instructions is not set correctly
8150313: aarch64: optimise array copy using SIMD instructions
8150394: aarch64: add support for 8.1 LSE CAS instructions
8150652: Remove unused code in AArch64 back end
8151340: aarch64: prefetch the destination word for write prior to ldxr/stxr loops.
8151502: optimize pd_disjoint_words and pd_conjoint_words
8151775: aarch64: add support for 8.1 LSE atomic operations
8152537: aarch64: Make use of CBZ and CBNZ when comparing unsigned values with zero.
8152840: aarch64: improve _unsafe_arraycopy stub routine
8153172: aarch64: hotspot crashes after the 8.1 LSE patch is merged
8153713: aarch64: improve short array clearing using store pair
8153797: aarch64: Add Arrays.fill stub code
8154413: AArch64: Better byte behaviour
8154537: AArch64: some integer rotate instructions are never emitted
8154739: AArch64: TemplateTable::fast_xaccess loads in wrong mode
8155015: Aarch64: bad assert in spill generation code
8155100: AArch64: Relax alignment requirement for byte_map_base
8155612: Aarch64: vector nodes need to support misaligned offset
8155617: aarch64: ClearArray does not use DC ZVA
8155627: Enable SA on AArch64
8155653: TestVectorUnalignedOffset.java not pushed with 8155612
8156731: aarch64: java/util/Arrays/Correct.java fails due to _generic_arraycopy stub routine
8157841: aarch64: prefetch ignores cache line size
8157906: aarch64: some more integer rotate instructions are never emitted
8158913: aarch64: SEGV running Spark terasort
8159052: aarch64: optimise unaligned copies in pd_disjoint_words and pd_conjoint_words
8159063: aarch64: optimise unaligned array copy long
8160748: [AArch64] Inconsistent types for ideal_reg
8161072: AArch64: jtreg compiler/uncommontrap/TestDeoptOOM failure
8161190: AArch64: Fix overflow in immediate cmp instruction
8164113: AArch64: follow-up the fix for 8161598
8165673: AArch64: Fix JNI floating point argument handling
8167200: AArch64: Broken stack pointer adjustment in interpreter
8167421: AArch64: in one core system, fatal error: Illegal threadstate encountered
8167595: AArch64: SEGV in stub code cipherBlockChaining_decryptAESCrypt
8168699: Validate special case invocations [AArch64 support]
8168888: Port 8160591: Improve internal array handling to AArch64.
8170100: AArch64: Crash in C1-compiled code accessing References
8170188: jtreg test compiler/types/TestMeetIncompatibleInterfaceArrays.java causes JVM crash
8170873: PPC64/aarch64: Poor StrictMath performance due to non-optimized compilation
8171537: aarch64: compiler/c1/Test6849574.java generates guarantee failure in C1
8172881: AArch64: assertion failure: the int pressure is incorrect
8173472: AArch64: C1 comparisons with null only use 32-bit instructions
8176100: [AArch64] [REDO][REDO] G1 Needs pre barrier on dereference of weak JNI handles
8177661: Correct ad rule output register types from iRegX to iRegXNoSp
8179954: AArch64: C1 and C2 volatile accesses are not sequentially consistent
8182581: aarch64: fix for crash caused by earlyret of compiled method
8183925: [AArch64] Decouple crash protection from watcher thread
8186325: AArch64: jtreg test hotspot/test/gc/g1/TestJNIWeakG1/TestJNIWeakG1.java SEGV
8187224: aarch64: some inconsistency between aarch64_ad.m4 and aarch64.ad
8189170: [AArch64] Add option to disable stack overflow checking in primordial thread for use with JNI_CreateJavaJVM
8193133: Assertion failure because 0xDEADDEAD can be in-heap
8195685: AArch64 port of 8174962: Better interface invocations
8195859: AArch64: vtableStubs gtest fails after 8174962
8196136: AArch64: Correct register use in patch for JDK-8194686
8196221: AArch64: Mistake in committed patch for JDK-8195859
8199712: [AArch64] Flight Recorder
8203481: Incorrect constraint for unextended_sp in frame:safe_for_sender
8203699: java/lang/invoke/SpecialInterfaceCall fails with SIGILL on aarch64
8205421: AARCH64: StubCodeMark should be placed after alignment
8206163: AArch64: incorrect code generation for StoreCM
8207345: Trampoline generation code reads from uninitialized memory
8207838: AArch64: Float registers incorrectly restored in JNI call
8209413: AArch64: NPE in clhsdb jstack command
8209414: [AArch64] method handle invocation does not respect JVMTI interp_only mode
8209415: Fix JVMTI test failure HS202
8209420: Track membars for volatile accesses so they can be properly optimized
8209835: Aarch64: elide barriers on all volatile operations
8210425: [AArch64] sharedRuntimeTrig/sharedRuntimeTrans compiled without optimization
8211064: [AArch64] Interpreter and c1 don't correctly handle jboolean results in native calls
8211233: MemBarNode::trailing_membar() and MemBarNode::leading_membar() need to handle dying subgraphs better
8213134: AArch64: vector shift failed with MaxVectorSize=8
8213419: [AArch64] C2 may hang in MulLNode::Ideal()/MulINode::Ideal() with gcc 8.2.1
8214857: "bad trailing membar" assert failure at memnode.cpp:3220
8215951: AArch64: jtreg test vmTestbase/nsk/jvmti/PopFrame/popframe005 segfaults
8215961: jdk/jfr/event/os/TestCPUInformation.java fails on AArch64
8216350: AArch64: monitor unlock fast path not called
8216989: CardTableBarrierSetAssembler::gen_write_ref_array_post_barrier() does not check for zero length on AARCH64
8217368: AArch64: C2 recursive stack locking optimisation not triggered
8218185: aarch64: missing LoadStore barrier in TemplateTable::putfield_or_static
8219011: Implement MacroAssembler::warn method on AArch64
8219635: aarch64: missing LoadStore barrier in TemplateTable::fast_storefield
8221220: AArch64: Add StoreStore membar explicitly for Volatile Writes in TemplateTable
8221658: aarch64: add necessary predicate for ubfx patterns
8224671: AArch64: mauve System.arraycopy test failure
8224828: aarch64: rflags is not correct after safepoint poll
8224851: AArch64: fix warnings and errors with Clang and GCC 8.3
8224880: AArch64: java/javac error with AllocatePrefetchDistance
8228400: Remove built-in AArch64 simulator
8228406: Superfluous change in chaitin.hpp
8228593: Revert explicit JDK 7 support additions
8228716: Revert InstanceKlass::print_on debug additions
8228718: Revert incorrect backport of JDK-8129757 to 8-aarch64
8228725: AArch64: Purge method call format support
8228747: Revert "unused" attribute from test_arraycopy_func
8228767: Revert ResourceMark additions
8228770: Revert development hsdis changes
8229123: Revert build fixes for aarch64/zero
8229124: Revert disassembler.cpp changes
8229145: Revert TemplateTable::bytecode() visibility change
8233839: aarch64: missing memory barrier in NewObjectArrayStub and NewTypeArrayStub
8237512: AArch64: aarch64TestHook leaks a BufferBlob
8246482: Build failures with +JFR -PCH
8247979: aarch64: missing side effect of killing flags for clearArray_reg_reg
8248219: aarch64: missing memory barrier in fast_storefield and fast_accessfield
Reviewed-by: shade, aph
author | andrew |
---|---|
date | Mon, 01 Feb 2021 03:48:36 +0000 |
parents | 007d682bf5d0 |
children | f79e943d15a7 |
line wrap: on
line source
# # Copyright (c) 1999, 2018, Oracle and/or its affiliates. All rights reserved. # DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. # # This code is free software; you can redistribute it and/or modify it # under the terms of the GNU General Public License version 2 only, as # published by the Free Software Foundation. # # This code is distributed in the hope that it will be useful, but WITHOUT # ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or # FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License # version 2 for more details (a copy is included in the LICENSE file that # accompanied this code). # # You should have received a copy of the GNU General Public License version # 2 along with this work; if not, write to the Free Software Foundation, # Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. # # Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA # or visit www.oracle.com if you need additional information or have any # questions. # # #------------------------------------------------------------------------ # CC, CXX & AS # If a SPEC is not set already, then use these defaults. ifeq ($(SPEC),) # When cross-compiling the ALT_COMPILER_PATH points # to the cross-compilation toolset ifdef CROSS_COMPILE_ARCH CXX = $(ALT_COMPILER_PATH)/g++ CC = $(ALT_COMPILER_PATH)/gcc HOSTCXX = g++ HOSTCC = gcc STRIP = $(ALT_COMPILER_PATH)/strip else ifeq ($(USE_CLANG), true) CXX = clang++ CC = clang else CXX = g++ CC = gcc endif HOSTCXX = $(CXX) HOSTCC = $(CC) STRIP = strip endif AS = $(CC) -c endif ifeq ($(USE_CLANG), true) CC_VER_MAJOR := $(shell $(CC) -v 2>&1 | grep version | sed "s/.*version \([0-9]*\.[0-9]*\).*/\1/" | cut -d'.' -f1) CC_VER_MINOR := $(shell $(CC) -v 2>&1 | grep version | sed "s/.*version \([0-9]*\.[0-9]*\).*/\1/" | cut -d'.' -f2) else # -dumpversion in gcc-2.91 shows "egcs-2.91.66". In later version, it only # prints the numbers (e.g. "2.95", "3.2.1") CC_VER_MAJOR := $(shell $(CC) -dumpversion | sed 's/egcs-//' | cut -d'.' -f1) CC_VER_MINOR := $(shell $(CC) -dumpversion | sed 's/egcs-//' | cut -d'.' -f2) endif ifeq ($(USE_CLANG), true) # Clang has precompiled headers support by default, but the user can switch # it off by using 'USE_PRECOMPILED_HEADER=0'. ifdef LP64 ifeq ($(USE_PRECOMPILED_HEADER),) USE_PRECOMPILED_HEADER=1 endif else # We don't support precompiled headers on 32-bit builds because there some files are # compiled with -fPIC while others are compiled without (see 'NONPIC_OBJ_FILES' rules.make) # Clang produces an error if the PCH file was compiled with other options than the actual compilation unit. USE_PRECOMPILED_HEADER=0 endif ifeq ($(USE_PRECOMPILED_HEADER),1) ifndef LP64 $(error " Precompiled Headers only supported on 64-bit platforms!") endif PRECOMPILED_HEADER_DIR=. PRECOMPILED_HEADER_SRC=$(GAMMADIR)/src/share/vm/precompiled/precompiled.hpp PRECOMPILED_HEADER=$(PRECOMPILED_HEADER_DIR)/precompiled.hpp.pch PCH_FLAG = -include precompiled.hpp PCH_FLAG/DEFAULT = $(PCH_FLAG) PCH_FLAG/NO_PCH = -DNO_PCH PCH_FLAG/BY_FILE = $(PCH_FLAG/$@)$(PCH_FLAG/DEFAULT$(PCH_FLAG/$@)) VM_PCH_FLAG/LIBJVM = $(PCH_FLAG/BY_FILE) VM_PCH_FLAG/AOUT = VM_PCH_FLAG = $(VM_PCH_FLAG/$(LINK_INTO)) # We only use precompiled headers for the JVM build CFLAGS += $(VM_PCH_FLAG) # There are some files which don't like precompiled headers # The following files are build with 'OPT_CFLAGS/NOOPT' (-O0) in the opt build. # But Clang doesn't support a precompiled header which was compiled with -O3 # to be used in a compilation unit which uses '-O0'. We could also prepare an # extra '-O0' PCH file for the opt build and use it here, but it's probably # not worth the effoert as long as only two files need this special handling. PCH_FLAG/loopTransform.o = $(PCH_FLAG/NO_PCH) PCH_FLAG/sharedRuntimeTrig.o = $(PCH_FLAG/NO_PCH) PCH_FLAG/sharedRuntimeTrans.o = $(PCH_FLAG/NO_PCH) endif else # ($(USE_CLANG), true) # check for precompiled headers support ifneq "$(shell expr \( $(CC_VER_MAJOR) \> 3 \) \| \( \( $(CC_VER_MAJOR) = 3 \) \& \( $(CC_VER_MINOR) \>= 4 \) \))" "0" # Allow the user to turn off precompiled headers from the command line. ifneq ($(USE_PRECOMPILED_HEADER),0) PRECOMPILED_HEADER_DIR=. PRECOMPILED_HEADER_SRC=$(GAMMADIR)/src/share/vm/precompiled/precompiled.hpp PRECOMPILED_HEADER=$(PRECOMPILED_HEADER_DIR)/precompiled.hpp.gch endif endif endif # -DDONT_USE_PRECOMPILED_HEADER will exclude all includes in precompiled.hpp. ifeq ($(USE_PRECOMPILED_HEADER),0) CFLAGS += -DDONT_USE_PRECOMPILED_HEADER endif #------------------------------------------------------------------------ # Compiler flags # position-independent code PICFLAG = -fPIC VM_PICFLAG/LIBJVM = $(PICFLAG) VM_PICFLAG/AOUT = VM_PICFLAG = $(VM_PICFLAG/$(LINK_INTO)) ifeq ($(JVM_VARIANT_ZERO), true) CFLAGS += $(LIBFFI_CFLAGS) endif ifeq ($(JVM_VARIANT_ZEROSHARK), true) CFLAGS += $(LIBFFI_CFLAGS) CFLAGS += $(LLVM_CFLAGS) endif CFLAGS += $(VM_PICFLAG) CFLAGS += -fno-rtti CFLAGS += -fno-exceptions CFLAGS += -D_REENTRANT ifeq ($(USE_CLANG),) CFLAGS += -fcheck-new # version 4 and above support fvisibility=hidden (matches jni_x86.h file) # except 4.1.2 gives pointless warnings that can't be disabled (afaik) ifneq "$(shell expr \( $(CC_VER_MAJOR) \> 4 \) \| \( \( $(CC_VER_MAJOR) = 4 \) \& \( $(CC_VER_MINOR) \>= 3 \) \))" "0" CFLAGS += -fvisibility=hidden endif else CFLAGS += -fvisibility=hidden endif ifeq ($(USE_CLANG), true) # Before Clang 3.1, we had to pass the stack alignment specification directly to llvm with the help of '-mllvm' # Starting with version 3.1, Clang understands the '-mstack-alignment' (and rejects '-mllvm -stack-alignment') ifneq "$(shell expr \( $(CC_VER_MAJOR) \> 3 \) \| \( \( $(CC_VER_MAJOR) = 3 \) \& \( $(CC_VER_MINOR) \>= 1 \) \))" "0" STACK_ALIGNMENT_OPT = -mno-omit-leaf-frame-pointer -mstack-alignment=16 else STACK_ALIGNMENT_OPT = -mno-omit-leaf-frame-pointer -mllvm -stack-alignment=16 endif endif ARCHFLAG = $(ARCHFLAG/$(BUILDARCH)) ARCHFLAG/i486 = -m32 -march=i586 ARCHFLAG/amd64 = -m64 $(STACK_ALIGNMENT_OPT) ARCHFLAG/aarch64 = ARCHFLAG/ia64 = ARCHFLAG/sparc = -m32 -mcpu=v9 ARCHFLAG/sparcv9 = -m64 -mcpu=v9 ARCHFLAG/zero = $(ZERO_ARCHFLAG) ARCHFLAG/ppc64 = -m64 CFLAGS += $(ARCHFLAG) AOUT_FLAGS += $(ARCHFLAG) LFLAGS += $(ARCHFLAG) ASFLAGS += $(ARCHFLAG) ifeq ($(DEBUG_BINARIES), true) ASFLAGS += $(ASFLAGS_DEBUG_SYMBOLS) endif # Use C++ Interpreter ifdef CC_INTERP CFLAGS += -DCC_INTERP endif # Keep temporary files (.ii, .s) ifdef NEED_ASM CFLAGS += -save-temps else CFLAGS += -pipe endif # Compiler warnings are treated as errors WARNINGS_ARE_ERRORS = -Werror ifeq ($(USE_CLANG), true) # However we need to clean the code up before we can unrestrictedly enable this option with Clang WARNINGS_ARE_ERRORS += -Wno-logical-op-parentheses -Wno-parentheses-equality -Wno-parentheses WARNINGS_ARE_ERRORS += -Wno-switch -Wno-tautological-constant-out-of-range-compare -Wno-tautological-compare WARNINGS_ARE_ERRORS += -Wno-delete-non-virtual-dtor -Wno-deprecated -Wno-format -Wno-dynamic-class-memaccess WARNINGS_ARE_ERRORS += -Wno-return-type -Wno-empty-body endif WARNING_FLAGS = -Wpointer-arith -Wsign-compare -Wundef -Wunused-function -Wunused-value -Wformat=2 -Wreturn-type ifeq ($(USE_CLANG),) # Since GCC 4.3, -Wconversion has changed its meanings to warn these implicit # conversions which might affect the values. Only enable it in earlier versions. ifeq "$(shell expr \( $(CC_VER_MAJOR) \> 4 \) \| \( \( $(CC_VER_MAJOR) = 4 \) \& \( $(CC_VER_MINOR) \>= 3 \) \))" "0" WARNING_FLAGS += -Wconversion endif endif CFLAGS_WARN/DEFAULT = $(WARNINGS_ARE_ERRORS) $(WARNING_FLAGS) # Special cases CFLAGS_WARN/BYFILE = $(CFLAGS_WARN/$@)$(CFLAGS_WARN/DEFAULT$(CFLAGS_WARN/$@)) # The flags to use for an Optimized g++ build OPT_CFLAGS/SIZE=-Os OPT_CFLAGS/SPEED=-O3 # Hotspot uses very unstrict aliasing turn this optimization off # This option is added to CFLAGS rather than OPT_CFLAGS # so that OPT_CFLAGS overrides get this option too. CFLAGS += -fno-strict-aliasing OPT_CFLAGS_DEFAULT ?= SPEED ifdef OPT_CFLAGS ifneq ("$(origin OPT_CFLAGS)", "command line") $(error " Use OPT_EXTRAS instead of OPT_CFLAGS to add extra flags to OPT_CFLAGS.") endif endif OPT_CFLAGS = $(OPT_CFLAGS/$(OPT_CFLAGS_DEFAULT)) $(OPT_EXTRAS) # The gcc compiler segv's on ia64 when compiling bytecodeInterpreter.cpp # if we use expensive-optimizations ifeq ($(BUILDARCH), ia64) OPT_CFLAGS += -fno-expensive-optimizations endif OPT_CFLAGS/NOOPT=-O0 # Work around some compiler bugs. ifeq ($(USE_CLANG), true) ifeq ($(shell expr $(CC_VER_MAJOR) = 4 \& $(CC_VER_MINOR) = 2), 1) OPT_CFLAGS/loopTransform.o += $(OPT_CFLAGS/NOOPT) endif else # 6835796. Problem in GCC 4.3.0 with mulnode.o optimized compilation. ifeq ($(shell expr $(CC_VER_MAJOR) = 4 \& $(CC_VER_MINOR) = 3), 1) OPT_CFLAGS/mulnode.o += $(OPT_CFLAGS/NOOPT) endif endif # Flags for generating make dependency flags. DEPFLAGS = -MMD -MP -MF $(DEP_DIR)/$(@:%=%.d) ifeq ($(USE_CLANG),) ifneq ("${CC_VER_MAJOR}", "2") DEPFLAGS += -fpch-deps endif endif #------------------------------------------------------------------------ # Linker flags # statically link libstdc++.so, work with gcc but ignored by g++ STATIC_STDCXX = -Wl,-Bstatic -lstdc++ -Wl,-Bdynamic # While the VM needs the above line, adlc needs a separate setting: ADLC_STATIC_STDCXX = -static-libstdc++ ifeq ($(USE_CLANG),) # statically link libgcc and/or libgcc_s, libgcc does not exist before gcc-3.x. ifneq ("${CC_VER_MAJOR}", "2") STATIC_LIBGCC += -static-libgcc endif ifeq ($(BUILDARCH), ia64) LFLAGS += -Wl,-relax endif endif # Enable linker optimization LFLAGS += -Xlinker -O1 ifeq ($(USE_CLANG),) # If this is a --hash-style=gnu system, use --hash-style=both # The gnu .hash section won't work on some Linux systems like SuSE 10. _HAS_HASH_STYLE_GNU:=$(shell $(CC) -dumpspecs | grep -- '--hash-style=gnu') ifneq ($(_HAS_HASH_STYLE_GNU),) LDFLAGS_HASH_STYLE = -Wl,--hash-style=both endif else # Don't know how to find out the 'hash style' of a system as '-dumpspecs' # doesn't work for Clang. So for now we'll alwys use --hash-style=both LDFLAGS_HASH_STYLE = -Wl,--hash-style=both endif LFLAGS += $(LDFLAGS_HASH_STYLE) LDFLAGS_NO_EXEC_STACK="-Wl,-z,noexecstack" # Use $(MAPFLAG:FILENAME=real_file_name) to specify a map file. MAPFLAG = -Xlinker --version-script=FILENAME # Use $(SONAMEFLAG:SONAME=soname) to specify the intrinsic name of a shared obj SONAMEFLAG = -Xlinker -soname=SONAME # Build shared library SHARED_FLAG = -shared # Keep symbols even they are not used AOUT_FLAGS += -Xlinker -export-dynamic #------------------------------------------------------------------------ # Debug flags ifeq ($(USE_CLANG), true) # Restrict the debug information created by Clang to avoid # too big object files and speed the build up a little bit # (see http://llvm.org/bugs/show_bug.cgi?id=7554) CFLAGS += -flimit-debug-info endif # DEBUG_BINARIES uses full -g debug information for all configs ifeq ($(DEBUG_BINARIES), true) CFLAGS += -g else DEBUG_CFLAGS += $(DEBUG_CFLAGS/$(BUILDARCH)) ifeq ($(DEBUG_CFLAGS/$(BUILDARCH)),) DEBUG_CFLAGS += -g endif ifeq ($(ENABLE_FULL_DEBUG_SYMBOLS),1) FASTDEBUG_CFLAGS += $(FASTDEBUG_CFLAGS/$(BUILDARCH)) ifeq ($(FASTDEBUG_CFLAGS/$(BUILDARCH)),) FASTDEBUG_CFLAGS/$(BUILDARCH) = -g endif OPT_CFLAGS += $(OPT_CFLAGS/$(BUILDARCH)) ifeq ($(OPT_CFLAGS/$(BUILDARCH)),) OPT_CFLAGS/$(BUILDARCH) = -g endif endif endif # If we are building HEADLESS, pass on to VM # so it can set the java.awt.headless property ifdef HEADLESS CFLAGS += -DHEADLESS endif # We are building Embedded for a small device # favor code space over speed ifdef MINIMIZE_RAM_USAGE CFLAGS += -DMINIMIZE_RAM_USAGE endif # Stack walking in the JVM relies on frame pointer (%rbp) to walk thread stack. # Explicitly specify -fno-omit-frame-pointer because it is off by default # starting with gcc 4.6. ifndef USE_SUNCC CFLAGS += -fno-omit-frame-pointer endif -include $(HS_ALT_MAKE)/linux/makefiles/gcc.make