java-21-openjdk/SOURCES/java-21-openjdk-portable.specfile

2300 lines
92 KiB
Plaintext
Raw Normal View History

# debug_package %%{nil} is portable-jdks specific
%define debug_package %{nil}
# RPM conditionals so as to be able to dynamically produce
# slowdebug/release builds. See:
# http://rpm.org/user_doc/conditional_builds.html
#
# Examples:
#
# Produce release, fastdebug *and* slowdebug builds on x86_64 (default):
# $ rpmbuild -ba java-21-openjdk.spec
#
# Produce only release builds (no debug builds) on x86_64:
# $ rpmbuild -ba java-21-openjdk.spec --without slowdebug --without fastdebug
#
# Only produce a release build on x86_64:
# $ fedpkg mockbuild --without slowdebug --without fastdebug
# Enable fastdebug builds by default on relevant arches.
%bcond_without fastdebug
# Enable slowdebug builds by default on relevant arches.
%bcond_without slowdebug
# Enable release builds by default on relevant arches.
%bcond_without release
# Enable static library builds by default.
%bcond_without staticlibs
# Build a fresh libjvm.so for use in a copy of the bootstrap JDK
%bcond_without fresh_libjvm
# Build with system libraries
%bcond_with system_libs
# This is RHEL 7 specific as it doesn't seem to have the
# __brp_strip_static_archive macro.
%define __os_install_post %{nil}
# Workaround for stripping of debug symbols from static libraries
%if %{with staticlibs}
%define __brp_strip_static_archive %{nil}
%global include_staticlibs 1
%else
%global include_staticlibs 0
%endif
%if %{with system_libs}
%global system_libs 1
%global link_type system
%global freetype_lib %{nil}
%else
%global system_libs 0
%global link_type bundled
%global freetype_lib |libfreetype[.]so.*
%endif
# The -g flag says to use strip -g instead of full strip on DSOs or EXEs.
# This fixes detailed NMT and other tools which need minimal debug info.
# See: https://bugzilla.redhat.com/show_bug.cgi?id=1520879
%global _find_debuginfo_opts -g
# note: parametrized macros are order-sensitive (unlike not-parametrized) even with normal macros
# also necessary when passing it as parameter to other macros. If not macro, then it is considered a switch
# see the difference between global and define:
# See https://github.com/rpm-software-management/rpm/issues/127 to comments at "pmatilai commented on Aug 18, 2017"
# (initiated in https://bugzilla.redhat.com/show_bug.cgi?id=1482192)
%global debug_suffix_unquoted -slowdebug
%global fastdebug_suffix_unquoted -fastdebug
%global main_suffix_unquoted -main
%global staticlibs_suffix_unquoted -staticlibs
# quoted one for shell operations
%global debug_suffix "%{debug_suffix_unquoted}"
%global fastdebug_suffix "%{fastdebug_suffix_unquoted}"
%global normal_suffix ""
%global main_suffix "%{main_suffix_unquoted}"
%global staticlibs_suffix "%{staticlibs_suffix_unquoted}"
%global debug_warning This package is unoptimised with full debugging. Install only as needed and remove ASAP.
%global fastdebug_warning This package is optimised with full debugging. Install only as needed and remove ASAP.
%global debug_on unoptimised with full debugging on
%global fastdebug_on optimised with full debugging on
%global for_fastdebug for packages with debugging on and optimisation
%global for_debug for packages with debugging on and no optimisation
%if %{with release}
%global include_normal_build 1
%else
%global include_normal_build 0
%endif
%if %{include_normal_build}
%global normal_build %{normal_suffix}
%else
%global normal_build %{nil}
%endif
# We have hardcoded list of files, which is appearing in alternatives, and in files
# in alternatives those are slaves and master, very often triplicated by man pages
# in files all masters and slaves are ghosted
# the ghosts are here to allow installation via query like `dnf install /usr/bin/java`
# you can list those files, with appropriate sections: cat *.spec | grep -e --install -e --slave -e post_
# TODO - fix those hardcoded lists via single list
# Those files must *NOT* be ghosted for *slowdebug* packages
# FIXME - if you are moving jshell or jlink or similar, always modify all three sections
# you can check via headless and devels:
# rpm -ql --noghost java-11-openjdk-headless-11.0.1.13-8.fc29.x86_64.rpm | grep bin
# == rpm -ql java-11-openjdk-headless-slowdebug-11.0.1.13-8.fc29.x86_64.rpm | grep bin
# != rpm -ql java-11-openjdk-headless-11.0.1.13-8.fc29.x86_64.rpm | grep bin
# similarly for other %%{_jvmdir}/{jre,java} and %%{_javadocdir}/{java,java-zip}
%define is_release_build() %( if [ "%{?1}" == "%{debug_suffix_unquoted}" -o "%{?1}" == "%{fastdebug_suffix_unquoted}" ]; then echo "0" ; else echo "1"; fi )
# while JDK is a techpreview(is_system_jdk=0), some provides are turned off. Once jdk stops to be an techpreview, move it to 1
# as sytem JDK, we mean any JDK which can run whole system java stack without issues (like bytecode issues, module issues, dependencies...)
%global is_system_jdk 0
%global aarch64 aarch64 arm64 armv8
# we need to distinguish between big and little endian PPC64
%global ppc64le ppc64le
%global ppc64be ppc64 ppc64p7
# Set of architectures which support multiple ABIs
%global multilib_arches %{power64} sparc64 x86_64
# Set of architectures for which we build slowdebug builds
%global debug_arches %{ix86} x86_64 sparcv9 sparc64 %{aarch64} %{power64} s390x
# Set of architectures for which we build fastdebug builds
%global fastdebug_arches x86_64 ppc64le aarch64
# Set of architectures with a Just-In-Time (JIT) compiler
%global jit_arches %{arm} %{aarch64} %{ix86} %{power64} s390x sparcv9 sparc64 x86_64
# Set of architectures which use the Zero assembler port (!jit_arches)
%global zero_arches ppc s390
# Set of architectures which run a full bootstrap cycle
%global bootstrap_arches %{jit_arches}
# Set of architectures which support SystemTap tapsets
%global systemtap_arches %{jit_arches}
# Set of architectures with a Ahead-Of-Time (AOT) compiler
%global aot_arches x86_64 %{aarch64}
# Set of architectures which support the serviceability agent
%global sa_arches %{ix86} x86_64 sparcv9 sparc64 %{aarch64} %{power64} %{arm}
# Set of architectures which support class data sharing
# As of JDK-8005165 in OpenJDK 10, class sharing is not arch-specific
# However, it does segfault on the Zero assembler port, so currently JIT only
%global share_arches %{jit_arches}
# Set of architectures for which we build the Shenandoah garbage collector
%global shenandoah_arches x86_64 %{aarch64}
# Set of architectures for which we build the Z garbage collector
%global zgc_arches x86_64
# Set of architectures for which alt-java has SSB mitigation
%global ssbd_arches x86_64
# Set of architectures for which java has short vector math library (libsvml.so)
%global svml_arches x86_64
# Set of architectures where we verify backtraces with gdb
# s390x fails on RHEL 7 so we exclude it there
%if (0%{?rhel} > 0 && 0%{?rhel} < 8)
%global gdb_arches %{arm} %{aarch64} %{ix86} %{power64} sparcv9 sparc64 x86_64 %{zero_arches}
%else
%global gdb_arches %{jit_arches} %{zero_arches}
%endif
# By default, we build a slowdebug build during main build on JIT architectures
%if %{with slowdebug}
%ifarch %{debug_arches}
%global include_debug_build 1
%else
%global include_debug_build 0
%endif
%else
%global include_debug_build 0
%endif
# On certain architectures, we compile the Shenandoah GC
%ifarch %{shenandoah_arches}
%global use_shenandoah_hotspot 1
%else
%global use_shenandoah_hotspot 0
%endif
# By default, we build a fastdebug build during main build only on fastdebug architectures
%if %{with fastdebug}
%ifarch %{fastdebug_arches}
%global include_fastdebug_build 1
%else
%global include_fastdebug_build 0
%endif
%else
%global include_fastdebug_build 0
%endif
%if %{include_debug_build}
%global slowdebug_build %{debug_suffix}
%else
%global slowdebug_build %{nil}
%endif
%if %{include_fastdebug_build}
%global fastdebug_build %{fastdebug_suffix}
%else
%global fastdebug_build %{nil}
%endif
# If you disable all builds, then the build fails
# Build and test slowdebug first as it provides the best diagnostics
%global build_loop %{slowdebug_build} %{fastdebug_build} %{normal_build}
%if %{include_staticlibs}
%global staticlibs_loop %{staticlibs_suffix}
%else
%global staticlibs_loop %{nil}
%endif
%ifarch %{bootstrap_arches}
%global bootstrap_build true
%else
%global bootstrap_build false
%endif
%if %{include_staticlibs}
# Extra target for producing the static-libraries. Separate from
# other targets since this target is configured to use in-tree
# AWT dependencies: lcms, libjpeg, libpng, libharfbuzz, giflib
# and possibly others
%global static_libs_target static-libs-image
%else
%global static_libs_target %{nil}
%endif
# The static libraries are produced under the same configuration as the main
# build for portables, as we expect in-tree libraries to be used throughout.
# If system libraries are enabled, the static libraries will also use them
# which may cause issues.
%global bootstrap_targets images %{static_libs_target} legacy-jre-image
%global release_targets images docs-zip %{static_libs_target} legacy-jre-image
# No docs nor bootcycle for debug builds
%global debug_targets images %{static_libs_target} legacy-jre-image
# Target to use to just build HotSpot
%global hotspot_target hotspot
# DTS toolset to use to provide gcc & binutils
%global dtsversion 10
# Disable LTO as this causes build failures at the moment.
# See RHBZ#1861401
%define _lto_cflags %{nil}
# Filter out flags from the optflags macro that cause problems with the OpenJDK build
# We filter out -O flags so that the optimization of HotSpot is not lowered from O3 to O2
# We filter out -Wall which will otherwise cause HotSpot to produce hundreds of thousands of warnings (100+mb logs)
# We replace it with -Wformat (required by -Werror=format-security) and -Wno-cpp to avoid FORTIFY_SOURCE warnings
# We filter out -fexceptions as the HotSpot build explicitly does -fno-exceptions and it's otherwise the default for C++
%global ourflags %(echo %optflags | sed -e 's|-Wall|-Wformat -Wno-cpp|' | sed -r -e 's|-O[0-9]*||')
%global ourcppflags %(echo %ourflags | sed -e 's|-fexceptions||')
%global ourldflags %{__global_ldflags}
# In some cases, the arch used by the JDK does
# not match _arch.
# Also, in some cases, the machine name used by SystemTap
# does not match that given by _target_cpu
%ifarch x86_64
%global archinstall amd64
%global stapinstall x86_64
%endif
%ifarch ppc
%global archinstall ppc
%global stapinstall powerpc
%endif
%ifarch %{ppc64be}
%global archinstall ppc64
%global stapinstall powerpc
%endif
%ifarch %{ppc64le}
%global archinstall ppc64le
%global stapinstall powerpc
%endif
%ifarch %{ix86}
%global archinstall i686
%global stapinstall i386
%endif
%ifarch ia64
%global archinstall ia64
%global stapinstall ia64
%endif
%ifarch s390
%global archinstall s390
%global stapinstall s390
%endif
%ifarch s390x
%global archinstall s390x
%global stapinstall s390
%endif
%ifarch %{arm}
%global archinstall arm
%global stapinstall arm
%endif
%ifarch %{aarch64}
%global archinstall aarch64
%global stapinstall arm64
%endif
# 32 bit sparc, optimized for v9
%ifarch sparcv9
%global archinstall sparc
%global stapinstall %{_target_cpu}
%endif
# 64 bit sparc
%ifarch sparc64
%global archinstall sparcv9
%global stapinstall %{_target_cpu}
%endif
# Need to support noarch for srpm build
%ifarch noarch
%global archinstall %{nil}
%global stapinstall %{nil}
%endif
%ifarch %{systemtap_arches}
%global with_systemtap 1
%else
%global with_systemtap 0
%endif
# New Version-String scheme-style defines
%global featurever 21
%global interimver 0
%global updatever 3
%global patchver 0
# buildjdkver is usually same as %%{featurever},
# but in time of bootstrap of next jdk, it is featurever-1,
# and this it is better to change it here, on single place
%global buildjdkver %{featurever}
# We don't add any LTS designator for STS packages (Fedora and EPEL).
# We need to explicitly exclude EPEL as it would have the %%{rhel} macro defined.
%if 0%{?rhel} && !0%{?epel}
%global lts_designator "LTS"
%global lts_designator_zip -%{lts_designator}
%else
%global lts_designator ""
%global lts_designator_zip ""
%endif
# JDK to use for bootstrapping
%global bootjdk /usr/lib/jvm/java-%{buildjdkver}-openjdk
# Define whether to use the bootstrap JDK directly or with a fresh libjvm.so
# This will only work where the bootstrap JDK is the same major version
# as the JDK being built
%if %{with fresh_libjvm} && %{buildjdkver} == %{featurever}
%global build_hotspot_first 1
%else
%global build_hotspot_first 0
%endif
# Define vendor information used by OpenJDK
%global oj_vendor Red Hat, Inc.
%global oj_vendor_url https://www.redhat.com/
# Define what url should JVM offer in case of a crash report
# order may be important, epel may have rhel declared
%if 0%{?epel}
%global oj_vendor_bug_url https://bugzilla.redhat.com/enter_bug.cgi?product=Fedora%20EPEL&component=%{name}&version=epel%{epel}
%else
%if 0%{?fedora}
# Does not work for rawhide, keeps the version field empty
%global oj_vendor_bug_url https://bugzilla.redhat.com/enter_bug.cgi?product=Fedora&component=%{name}&version=%{fedora}
%else
%if 0%{?rhel}
%global oj_vendor_bug_url https://access.redhat.com/support/cases/
%else
%global oj_vendor_bug_url https://bugzilla.redhat.com/enter_bug.cgi
%endif
%endif
%endif
%global oj_vendor_version (Red_Hat-%{version}-%{rpmrelease})
# Define IcedTea version used for SystemTap tapsets and desktop file
%global icedteaver 6.0.0pre00-c848b93a8598
# Define current Git revision for the FIPS support patches
%global fipsver 0a42e29b391
# Define JDK versions
%global newjavaver %{featurever}.%{interimver}.%{updatever}.%{patchver}
%global javaver %{featurever}
# Strip up to 6 trailing zeros in newjavaver, as the JDK does, to get the correct version used in filenames
%global filever %(svn=%{newjavaver}; for i in 1 2 3 4 5 6 ; do svn=${svn%%.0} ; done; echo ${svn})
# The tag used to create the OpenJDK tarball
%global vcstag jdk-%{filever}+%{buildver}%{?tagsuffix:-%{tagsuffix}}
# Standard JPackage naming and versioning defines
%global origin openjdk
%global origin_nice OpenJDK
%global top_level_dir_name %{vcstag}
%global top_level_dir_name_backup %{top_level_dir_name}-backup
%global buildver 9
%global rpmrelease 1
#%%global tagsuffix %%{nil}
# Priority must be 8 digits in total; up to openjdk 1.8, we were using 18..... so when we moved to 11, we had to add another digit
%if %is_system_jdk
# Using 10 digits may overflow the int used for priority, so we combine the patch and build versions
# It is very unlikely we will ever have a patch version > 4 or a build version > 20, so we combine as (patch * 20) + build.
# This means 11.0.9.0+11 would have had a priority of 11000911 as before
# A 11.0.9.1+1 would have had a priority of 11000921 (20 * 1 + 1), thus ensuring it is bigger than 11.0.9.0+11
%global combiver $( expr 20 '*' %{patchver} + %{buildver} )
%global priority %( printf '%02d%02d%02d%02d' %{featurever} %{interimver} %{updatever} %{combiver} )
%else
# for techpreview, using 1, so slowdebugs can have 0
%global priority %( printf '%08d' 1 )
%endif
# Define milestone (EA for pre-releases, GA for releases)
# Release will be (where N is usually a number starting at 1):
# - 0.N%%{?extraver}%%{?dist} for EA releases,
# - N%%{?extraver}{?dist} for GA releases
%global is_ga 1
%if %{is_ga}
%global build_type GA
%global ea_designator ""
%global ea_designator_zip %{nil}
%global extraver %{nil}
%global eaprefix %{nil}
%else
%global build_type EA
%global ea_designator ea
%global ea_designator_zip -%{ea_designator}
%global extraver .%{ea_designator}
%global eaprefix 0.
%endif
# parametrized macros are order-sensitive
%global compatiblename java-%{featurever}-%{origin}
%global fullversion %{compatiblename}-%{version}-%{release}
# images directories from upstream build
%global jdkimage jdk
%global static_libs_image static-libs
# output dir stub
%define buildoutputdir() %{expand:build/jdk%{featurever}.build%{?1}}
%define installoutputdir() %{expand:install/jdk%{featurever}.install%{?1}}
%global altjavaoutputdir install/altjava.install
%define packageoutputdir() %{expand:packages/jdk%{featurever}.packages%{?1}}
# we can copy the javadoc to not arched dir, or make it not noarch
%define uniquejavadocdir() %{expand:%{fullversion}.%{_arch}%{?1}}
# main id and dir of this jdk
%define uniquesuffix() %{expand:%{fullversion}.%{_arch}%{?1}}
# portable only declarations
%global jreimage jre
%define jreportablenameimpl() %(echo %{uniquesuffix ""} | sed "s;el%{rhel}\\(_[0-9]\\)*;portable%{1}.jre;g")
%define jdkportablenameimpl() %(echo %{uniquesuffix ""} | sed "s;el%{rhel}\\(_[0-9]\\)*;portable%{1}.jdk;g")
%define staticlibsportablenameimpl() %(echo %{uniquesuffix ""} | sed "s;el%{rhel}\\(_[0-9]\\)*;portable%{1}.static-libs;g")
%define jreportablearchive() %{expand:%{jreportablenameimpl -- %%{1}}.tar.xz}
%define jdkportablearchive() %{expand:%{jdkportablenameimpl -- %%{1}}.tar.xz}
%define staticlibsportablearchive() %{expand:%{staticlibsportablenameimpl -- %%{1}}.tar.xz}
%define jreportablename() %{expand:%{jreportablenameimpl -- %%{1}}}
%define jdkportablename() %{expand:%{jdkportablenameimpl -- %%{1}}}
# Intentionally use jdkportablenameimpl here since we want to have static-libs files overlayed on
# top of the JDK archive
%define staticlibsportablename() %{expand:%{jdkportablenameimpl -- %%{1}}}
%define docportablename() %(echo %{uniquesuffix ""} | sed "s;el%{rhel}\\(_[0-9]\\)*;portable.docs;g")
%define docportablearchive() %{docportablename}.tar.xz
%define miscportablename() %(echo %{uniquesuffix ""} | sed "s;el%{rhel}\\(_[0-9]\\)*;portable.misc;g")
%define miscportablearchive() %{miscportablename}.tar.xz
#################################################################
# fix for https://bugzilla.redhat.com/show_bug.cgi?id=1111349
# https://bugzilla.redhat.com/show_bug.cgi?id=1590796#c14
# https://bugzilla.redhat.com/show_bug.cgi?id=1655938
%global _privatelibs libsplashscreen[.]so.*|libawt_xawt[.]so.*|libjli[.]so.*|libattach[.]so.*|libawt[.]so.*|libextnet[.]so.*|libawt_headless[.]so.*|libdt_socket[.]so.*|libfontmanager[.]so.*|libinstrument[.]so.*|libj2gss[.]so.*|libj2pcsc[.]so.*|libj2pkcs11[.]so.*|libjaas[.]so.*|libjavajpeg[.]so.*|libjdwp[.]so.*|libjimage[.]so.*|libjsound[.]so.*|liblcms[.]so.*|libmanagement[.]so.*|libmanagement_agent[.]so.*|libmanagement_ext[.]so.*|libmlib_image[.]so.*|libnet[.]so.*|libnio[.]so.*|libprefs[.]so.*|librmi[.]so.*|libsaproc[.]so.*|libsctp[.]so.*|libsystemconf[.]so.*|libzip[.]so.*%{freetype_lib}
%global _publiclibs libjawt[.]so.*|libjava[.]so.*|libjvm[.]so.*|libverify[.]so.*|libjsig[.]so.*
%if %is_system_jdk
%global __provides_exclude ^(%{_privatelibs})$
%global __requires_exclude ^(%{_privatelibs})$
# Never generate lib-style provides/requires for slowdebug packages
%global __provides_exclude_from ^.*/%{uniquesuffix -- %{debug_suffix_unquoted}}/.*$
%global __requires_exclude_from ^.*/%{uniquesuffix -- %{debug_suffix_unquoted}}/.*$
%global __provides_exclude_from ^.*/%{uniquesuffix -- %{fastdebug_suffix_unquoted}}/.*$
%global __requires_exclude_from ^.*/%{uniquesuffix -- %{fastdebug_suffix_unquoted}}/.*$
%else
# Don't generate provides/requires for JDK provided shared libraries at all.
%global __provides_exclude ^(%{_privatelibs}|%{_publiclibs})$
%global __requires_exclude ^(%{_privatelibs}|%{_publiclibs})$
%endif
# VM variant being built
%ifarch %{zero_arches}
%global vm_variant zero
%else
%global vm_variant server
%endif
%global etcjavasubdir %{_sysconfdir}/java/java-%{javaver}-%{origin}
%define etcjavadir() %{expand:%{etcjavasubdir}/%{uniquesuffix -- %{?1}}}
# Standard JPackage directories and symbolic links.
%define sdkdir() %{expand:%{uniquesuffix -- %{?1}}}
%define jrelnk() %{expand:jre-%{javaver}-%{origin}-%{version}-%{release}.%{_arch}%{?1}}
%define sdkbindir() %{expand:%{_jvmdir}/%{sdkdir -- %{?1}}/bin}
%define jrebindir() %{expand:%{_jvmdir}/%{sdkdir -- %{?1}}/bin}
%global alt_java_name alt-java
%global rpm_state_dir %{_localstatedir}/lib/rpm-state/
# For flatpack builds hard-code /usr/sbin/alternatives,
# otherwise use %%{_sbindir} relative path.
%if 0%{?flatpak}
%global alternatives_requires /usr/sbin/alternatives
%else
%global alternatives_requires %{_sbindir}/alternatives
%endif
# Portables have no repo (requires/provides), but these are awesome for orientation in spec
# Also scriptlets are happily missing and files are handled old fashion
# not-duplicated requires/provides/obsoletes for normal/debug packages
%define java_rpo() %{expand:
}
%define java_devel_rpo() %{expand:
}
%define java_static_libs_rpo() %{expand:
}
%define java_unstripped_rpo() %{expand:
}
%define java_docs_rpo() %{expand:
}
%define java_misc_rpo() %{expand:
}
# Prevent brp-java-repack-jars from being run
%global __jar_repack 0
# Define the architectures on which we build
ExclusiveArch: %{aarch64} %{ppc64le} s390x x86_64
Name: java-%{javaver}-%{origin}-portable%{?pkgos:-%{pkgos}}
Version: %{newjavaver}.%{buildver}
Release: %{?eaprefix}%{rpmrelease}%{?extraver}%{?dist}
# java-1.5.0-ibm from jpackage.org set Epoch to 1 for unknown reasons
# and this change was brought into RHEL-4. java-1.5.0-ibm packages
# also included the epoch in their virtual provides. This created a
# situation where in-the-wild java-1.5.0-ibm packages provided "java =
# 1:1.5.0". In RPM terms, "1.6.0 < 1:1.5.0" since 1.6.0 is
# interpreted as 0:1.6.0. So the "java >= 1.6.0" requirement would be
# satisfied by the 1:1.5.0 packages. Thus we need to set the epoch in
# JDK package >= 1.6.0 to 1, and packages referring to JDK virtual
# provides >= 1.6.0 must specify the epoch, "java >= 1:1.6.0".
Epoch: 1
# portables have grown out of its component, moving back to java-x-vendor
# this expression, when declared as global, filled component with java-x-vendor portable
%define component %(echo %{name} | sed "s;-portable%{?pkgos:-%{pkgos}};;g")
Summary: %{origin_nice} %{featurever} Runtime Environment portable edition
# Groups are only used up to RHEL 8 and on Fedora versions prior to F30
%if (0%{?rhel} > 0 && 0%{?rhel} <= 8) || (0%{?fedora} >= 0 && 0%{?fedora} < 30)
Group: Development/Languages
%endif
# HotSpot code is licensed under GPLv2
# JDK library code is licensed under GPLv2 with the Classpath exception
# The Apache license is used in code taken from Apache projects (primarily xalan & xerces)
# DOM levels 2 & 3 and the XML digital signature schemas are licensed under the W3C Software License
# The JSR166 concurrency code is in the public domain
# The BSD and MIT licenses are used for a number of third-party libraries (see ADDITIONAL_LICENSE_INFO)
# The OpenJDK source tree includes:
# - JPEG library (IJG), zlib & libpng (zlib), giflib (MIT), harfbuzz (ISC),
# - freetype (FTL), jline (BSD) and LCMS (MIT)
# - jquery (MIT), jdk.crypto.cryptoki PKCS 11 wrapper (RSA)
# - public_suffix_list.dat from publicsuffix.org (MPLv2.0)
# The test code includes copies of NSS under the Mozilla Public License v2.0
# The PCSClite headers are under a BSD with advertising license
# The elliptic curve cryptography (ECC) source code is licensed under the LGPLv2.1 or any later version
License: ASL 1.1 and ASL 2.0 and BSD and BSD with advertising and GPL+ and GPLv2 and GPLv2 with exceptions and IJG and LGPLv2+ and MIT and MPLv2.0 and Public Domain and W3C and zlib and ISC and FTL and RSA
URL: http://openjdk.java.net/
# The source tarball, generated using generate_source_tarball.sh
Source0: https://openjdk-sources.osci.io/openjdk%{featurever}/open%{vcstag}%{ea_designator_zip}.tar.xz
# Use 'icedtea_sync.sh' to update the following
# They are based on code contained in the IcedTea project (6.x).
# Systemtap tapsets. Zipped up to keep it small.
Source8: tapsets-icedtea-%%{icedteaver}.tar.xz
# Desktop files. Adapted from IcedTea
# Disabled in portables
#Source9: jconsole.desktop.in
# Release notes
Source10: NEWS
# Source code for alt-java
Source11: alt-java.c
# Removed libraries that we link instead
Source12: remove-intree-libraries.sh
# Ensure we aren't using the limited crypto policy
Source13: TestCryptoLevel.java
# Ensure ECDSA is working
Source14: TestECDSA.java
# Verify system crypto (policy) can be disabled via a property
Source15: TestSecurityProperties.java
# Ensure vendor settings are correct
Source16: CheckVendor.java
# Ensure translations are available for new timezones
Source18: TestTranslations.java
############################################
#
# RPM/distribution specific patches
#
############################################
# Crypto policy and FIPS support patches
# Patch is generated from the fips-21u tree at https://github.com/rh-openjdk/jdk/tree/fips-21u
# as follows: git diff %%{vcstag} src make test > fips-21u-$(git show -s --format=%h HEAD).patch
# Diff is limited to src and make subdirectories to exclude .github changes
# Fixes currently included:
# PR3183, RH1340845: Follow system wide crypto policy
# PR3695: Allow use of system crypto policy to be disabled by the user
# RH1655466: Support RHEL FIPS mode using SunPKCS11 provider
# RH1818909: No ciphersuites availale for SSLSocket in FIPS mode
# RH1860986: Disable TLSv1.3 with the NSS-FIPS provider until PKCS#11 v3.0 support is available
# RH1915071: Always initialise JavaSecuritySystemConfiguratorAccess
# RH1929465: Improve system FIPS detection
# RH1995150: Disable non-FIPS crypto in SUN and SunEC security providers
# RH1996182: Login to the NSS software token in FIPS mode
# RH1991003: Allow plain key import unless com.redhat.fips.plainKeySupport is set to false
# RH2021263: Resolve outstanding FIPS issues
# RH2052819: Fix FIPS reliance on crypto policies
# RH2052829: Detect NSS at Runtime for FIPS detection
# RH2052070: Enable AlgorithmParameters and AlgorithmParameterGenerator services in FIPS mode
# RH2023467: Enable FIPS keys export
# RH2094027: SunEC runtime permission for FIPS
# RH2036462: sun.security.pkcs11.wrapper.PKCS11.getInstance breakage
# RH2090378: Revert to disabling system security properties and FIPS mode support together
# RH2104724: Avoid import/export of DH private keys
# RH2092507: P11Key.getEncoded does not work for DH keys in FIPS mode
# Build the systemconf library on all platforms
# RH2048582: Support PKCS#12 keystores [now part of JDK-8301553 upstream]
# RH2020290: Support TLS 1.3 in FIPS mode
# Add nss.fips.cfg support to OpenJDK tree
# RH2117972: Extend the support for NSS DBs (PKCS11) in FIPS mode
# Remove forgotten dead code from RH2020290 and RH2104724
# OJ1357: Fix issue on FIPS with a SecurityManager in place
# RH2134669: Add missing attributes when registering services in FIPS mode.
# test/jdk/sun/security/pkcs11/fips/VerifyMissingAttributes.java: fixed jtreg main class
# RH1940064: Enable XML Signature provider in FIPS mode
# RH2173781: Avoid calling C_GetInfo() too early, before cryptoki is initialized [now part of JDK-8301553 upstream]
Patch1001: fips-%{featurever}u-%{fipsver}.patch
#############################################
#
# OpenJDK patches in need of upstreaming
#
#############################################
# Currently empty
#############################################
#
# OpenJDK patches which missed last update
#
#############################################
#############################################
#
# Portable build specific patches
#
#############################################
# Currently empty
BuildRequires: autoconf
BuildRequires: automake
BuildRequires: alsa-lib-devel
BuildRequires: binutils
BuildRequires: cups-devel
BuildRequires: desktop-file-utils
# elfutils only are OK for build without AOT
BuildRequires: elfutils-devel
BuildRequires: file
BuildRequires: fontconfig-devel
BuildRequires: gcc-c++
BuildRequires: gdb
BuildRequires: libxslt
BuildRequires: libX11-devel
BuildRequires: libXi-devel
BuildRequires: libXinerama-devel
BuildRequires: libXrandr-devel
BuildRequires: libXrender-devel
BuildRequires: libXt-devel
BuildRequires: libXtst-devel
# Requirement for setting up nss.fips.cfg
BuildRequires: nss-devel
# Requirement for system security property test
# N/A for portable. RHEL7 doesn't provide them
#BuildRequires: crypto-policies
BuildRequires: pkgconfig
BuildRequires: xorg-x11-proto-devel
BuildRequires: zip
# to pack portable tarballs
BuildRequires: tar
BuildRequires: unzip
BuildRequires: javapackages-tools
BuildRequires: java-%{buildjdkver}-%{origin}%{?pkgos:-%{pkgos}}-devel
# Zero-assembler build requirement
%ifarch %{zero_arches}
BuildRequires: libffi-devel
%endif
# Full documentation build requirements
# pandoc is not available on RHEL 7
%if 0%{?rhel} >= 8
BuildRequires: graphviz
BuildRequires: pandoc
%endif
# 2024a required as of JDK-8325150
# Use 2023d until 2024a is in the buildroot
BuildRequires: tzdata-java >= 2023d
# cacerts build requirement in portable mode
BuildRequires: ca-certificates
# Earlier versions have a bug in tree vectorization on PPC
BuildRequires: gcc >= 4.8.3-8
%if %{with_systemtap}
BuildRequires: systemtap-sdt-devel
%endif
BuildRequires: make
%if %{system_libs}
BuildRequires: freetype-devel
BuildRequires: giflib-devel
BuildRequires: harfbuzz-devel
BuildRequires: lcms2-devel
BuildRequires: libjpeg-devel
BuildRequires: libpng-devel
%else
# Version in src/java.desktop/share/legal/freetype.md
Provides: bundled(freetype) = 2.13.2
# Version in src/java.desktop/share/native/libsplashscreen/giflib/gif_lib.h
Provides: bundled(giflib) = 5.2.1
# Version in src/java.desktop/share/native/libharfbuzz/hb-version.h
Provides: bundled(harfbuzz) = 8.2.2
# Version in src/java.desktop/share/native/liblcms/lcms2.h
Provides: bundled(lcms2) = 2.15.0
# Version in src/java.desktop/share/native/libjavajpeg/jpeglib.h
Provides: bundled(libjpeg) = 6b
# Version in src/java.desktop/share/native/libsplashscreen/libpng/png.h
Provides: bundled(libpng) = 1.6.40
# We link statically against libstdc++ to increase portability
BuildRequires: libstdc++-static
%endif
# this is always built, also during debug-only build
# when it is built in debug-only this package is just placeholder
%{java_rpo %{nil}}
%description
The %{origin_nice} %{featurever} runtime environment - portable edition.
%if %{include_debug_build}
%package slowdebug
Summary: %{origin_nice} %{featurever} Runtime Environment portable edition %{debug_on}
%if (0%{?rhel} > 0 && 0%{?rhel} <= 8) || (0%{?fedora} >= 0 && 0%{?fedora} < 30)
Group: Development/Languages
%endif
%{java_rpo -- %{debug_suffix_unquoted}}
%description slowdebug
The %{origin_nice} %{featurever} runtime environment - portable edition.
%{debug_warning}
%endif
%if %{include_fastdebug_build}
%package fastdebug
Summary: %{origin_nice} %{featurever} Runtime Environment portable edition %{fastdebug_on}
%if (0%{?rhel} > 0 && 0%{?rhel} <= 8) || (0%{?fedora} >= 0 && 0%{?fedora} < 30)
Group: Development/Languages
%endif
%{java_rpo -- %{fastdebug_suffix_unquoted}}
%description fastdebug
The %{origin_nice} %{featurever} runtime environment - portable edition.
%{fastdebug_warning}
%endif
%if %{include_normal_build}
%package devel
Summary: %{origin_nice} %{featurever} Development Environment portable edition
%if (0%{?rhel} > 0 && 0%{?rhel} <= 8) || (0%{?fedora} >= 0 && 0%{?fedora} < 30)
Group: Development/Languages
%endif
%{java_devel_rpo %{nil}}
%description devel
The %{origin_nice} %{featurever} development tools - portable edition.
%endif
%if %{include_debug_build}
%package devel-slowdebug
Summary: %{origin_nice} %{featurever} Runtime and Development Environment portable edition %{debug_on}
%if (0%{?rhel} > 0 && 0%{?rhel} <= 8) || (0%{?fedora} >= 0 && 0%{?fedora} < 30)
Group: Development/Languages
%endif
%{java_devel_rpo -- %{debug_suffix_unquoted}}
%description devel-slowdebug
The %{origin_nice} %{featurever} development tools - portable edition.
%{debug_warning}
%endif
%if %{include_fastdebug_build}
%package devel-fastdebug
Summary: %{origin_nice} %{featurever} Runtime and Development Environment portable edition %{fastdebug_on}
%if (0%{?rhel} > 0 && 0%{?rhel} <= 8) || (0%{?fedora} >= 0 && 0%{?fedora} < 30)
Group: Development/Tools
%endif
%{java_devel_rpo -- %{fastdebug_suffix_unquoted}}
%description devel-fastdebug
The %{origin_nice} %{featurever} runtime environment and development tools - portable edition
%{fastdebug_warning}
%endif
%if %{include_staticlibs}
%if %{include_normal_build}
%package static-libs
Summary: %{origin_nice} %{featurever} libraries for static linking - portable edition
%{java_static_libs_rpo %{nil}}
%description static-libs
The %{origin_nice} %{featurever} libraries for static linking - portable edition.
%endif
%if %{include_debug_build}
%package static-libs-slowdebug
Summary: %{origin_nice} %{featurever} libraries for static linking - portable edition %{debug_on}
%{java_static_libs_rpo -- %{debug_suffix_unquoted}}
%description static-libs-slowdebug
The %{origin_nice} %{featurever} libraries for static linking - portable edition
%{debug_warning}
%endif
%if %{include_fastdebug_build}
%package static-libs-fastdebug
Summary: %{origin_nice} %{featurever} libraries for static linking - portable edition %{fastdebug_on}
%{java_static_libs_rpo -- %{fastdebug_suffix_unquoted}}
%description static-libs-fastdebug
The %{origin_nice} %{featurever} libraries for static linking - portable edition
%{fastdebug_warning}
%endif
# staticlibs
%endif
%if %{include_normal_build}
%package unstripped
Summary: The %{origin_nice} %{featurever} runtime environment.
%{java_unstripped_rpo %{nil}}
%description unstripped
The %{origin_nice} %{featurever} runtime environment.
%endif
%package docs
Summary: %{origin_nice} %{featurever} API documentation
%{java_docs_rpo %{nil}}
%description docs
The %{origin_nice} %{featurever} API documentation.
%package misc
Summary: %{origin_nice} %{featurever} miscellany
%{java_misc_rpo %{nil}}
%description misc
The %{origin_nice} %{featurever} miscellany.
%prep
echo "Preparing %{oj_vendor_version}"
# Using the echo macro breaks rpmdev-bumpspec, as it parses the first line of stdout :-(
%if 0%{?_build_cpu:1}
echo "CPU: %{_target_cpu}, arch install directory: %{archinstall}, SystemTap install directory: %{_build_cpu}"
%else
%{error:Unrecognised architecture %{_build_cpu}}
%endif
if [ %{include_normal_build} -eq 0 -o %{include_normal_build} -eq 1 ] ; then
echo "include_normal_build is %{include_normal_build}"
else
echo "include_normal_build is %{include_normal_build}, that is invalid. Use 1 for yes or 0 for no"
exit 11
fi
if [ %{include_debug_build} -eq 0 -o %{include_debug_build} -eq 1 ] ; then
echo "include_debug_build is %{include_debug_build}"
else
echo "include_debug_build is %{include_debug_build}, that is invalid. Use 1 for yes or 0 for no"
exit 12
fi
if [ %{include_fastdebug_build} -eq 0 -o %{include_fastdebug_build} -eq 1 ] ; then
echo "include_fastdebug_build is %{include_fastdebug_build}"
else
echo "include_fastdebug_build is %{include_fastdebug_build}, that is invalid. Use 1 for yes or 0 for no"
exit 13
fi
if [ %{include_debug_build} -eq 0 -a %{include_normal_build} -eq 0 -a %{include_fastdebug_build} -eq 0 ] ; then
echo "You have disabled all builds (normal,fastdebug,slowdebug). That is a no go."
exit 14
fi
%if %{with fresh_libjvm} && ! %{build_hotspot_first}
echo "WARNING: The build of a fresh libjvm has been disabled due to a JDK version mismatch"
echo "Build JDK version is %{buildjdkver}, feature JDK version is %{featurever}"
%endif
export XZ_OPT="-T0"
%setup -q -c -n %{uniquesuffix ""} -T -a 0
# https://bugzilla.redhat.com/show_bug.cgi?id=1189084
prioritylength=`expr length %{priority}`
if [ $prioritylength -ne 8 ] ; then
echo "priority must be 8 digits in total, violated"
exit 14
fi
# OpenJDK patches
%if %{system_libs}
# Remove libraries that are linked by both static and dynamic builds
sh %{SOURCE12} %{top_level_dir_name}
%endif
# Patch the JDK
# This syntax is deprecated:
# %patchN [...]
# and should be replaced with:
# %patch -PN [...]
# For example:
# %patch1001 -p1
# becomes:
# %patch -P1001 -p1
# The replacement format suggested by recent (circa Fedora 38) RPM
# deprecation messages:
# %patch N [...]
# is not backward-compatible with prior (circa RHEL-8) versions of
# rpmbuild.
pushd %{top_level_dir_name}
# Add crypto policy and FIPS support
%patch -P1001 -p1
popd # openjdk
# The OpenJDK version file includes the current
# upstream version information. For some reason,
# configure does not automatically use the
# default pre-version supplied there (despite
# what the file claims), so we pass it manually
# to configure
VERSION_FILE=$(pwd)/%{top_level_dir_name}/make/conf/version-numbers.conf
if [ -f ${VERSION_FILE} ] ; then
UPSTREAM_EA_DESIGNATOR=$(grep '^DEFAULT_PROMOTED_VERSION_PRE' ${VERSION_FILE} | cut -d '=' -f 2)
else
echo "Could not find OpenJDK version file.";
exit 16
fi
if [ "x${UPSTREAM_EA_DESIGNATOR}" != "x%{ea_designator}" ] ; then
echo "WARNING: Designator mismatch";
echo "Spec file is configured for a %{build_type} build with designator '%{ea_designator}'"
echo "Upstream version-pre setting is '${UPSTREAM_EA_DESIGNATOR}'";
exit 17
fi
# Extract systemtap tapsets
%if %{with_systemtap}
tar --strip-components=1 -x -I xz -f %{SOURCE8}
%if %{include_debug_build}
cp -r tapset tapset%{debug_suffix}
%endif
%if %{include_fastdebug_build}
cp -r tapset tapset%{fastdebug_suffix}
%endif
for suffix in %{build_loop} ; do
for file in "tapset"$suffix/*.in; do
sed -i -e "s:@JAVA_SPEC_VER@:%{javaver}:g" $file
sed -i -e "s:@INSTALL_ARCH_DIR@:%{archinstall}:g" $file
done
done
# systemtap tapsets ends
%endif
# Prepare desktop files
# Portables do not have desktop integration
%build
# How many CPU's do we have?
export NUM_PROC=%(/usr/bin/getconf _NPROCESSORS_ONLN 2> /dev/null || :)
export NUM_PROC=${NUM_PROC:-1}
%if 0%{?_smp_ncpus_max}
# Honor %%_smp_ncpus_max
[ ${NUM_PROC} -gt %{?_smp_ncpus_max} ] && export NUM_PROC=%{?_smp_ncpus_max}
%endif
export XZ_OPT="-T0"
%ifarch s390x sparc64 alpha %{power64} %{aarch64}
export ARCH_DATA_MODEL=64
%endif
%ifarch alpha
export CFLAGS="$CFLAGS -mieee"
%endif
# We use ourcppflags because the OpenJDK build seems to
# pass EXTRA_CFLAGS to the HotSpot C++ compiler...
# Explicitly set the C++ standard as the default has changed on GCC >= 6
EXTRA_CFLAGS="%ourcppflags"
EXTRA_CPP_FLAGS="%ourcppflags"
%ifarch %{power64} ppc
# fix rpmlint warnings
EXTRA_CFLAGS="$EXTRA_CFLAGS -fno-strict-aliasing"
%endif
%ifarch %{ix86}
# Align stack boundary on x86_32
EXTRA_CFLAGS="$(echo ${EXTRA_CFLAGS} | sed -e 's|-mstackrealign|-mincoming-stack-boundary=2 -mpreferred-stack-boundary=4|')"
EXTRA_CPP_FLAGS="$(echo ${EXTRA_CPP_FLAGS} | sed -e 's|-mstackrealign|-mincoming-stack-boundary=2 -mpreferred-stack-boundary=4|')"
%endif
export EXTRA_CFLAGS EXTRA_CPP_FLAGS
# Set modification times (mtimes) of files within JAR files generated
# by the OpenJDK build to a timestamp that is constant across RPM
# rebuilds. OpenJDK provides the --with-source-date configure option
# for this purpose. Potential arguments in the RPM build context are:
#
# A) --with-source-date="${SOURCE_DATE_EPOCH}"
# B) --with-source-date=version
# C) --with-source-date="${OPENJDK_UPSTREAM_TAG_EPOCH}"
#
# Consider Option A. Fedora 38 (rpm-4.18.2) and RHEL-8 (rpm-4.14.3)
# have different support for SOURCE_DATE_EPOCH. To keep
# SOURCE_DATE_EPOCH constant across RPM rebuilds, one could set the
# source_date_epoch_from_changelog macro to 1 on both Fedora 38 and
# RHEL-8. However, on RHEL-8, this results in the RPM build times
# being set to the timestamp of the most recent changelog. This is
# bad for tracing when RPMs were actually built. Fedora 38 supports a
# better behaviour via the introduction of the
# use_source_date_epoch_as_buildtime macro, set to 0 by default.
# There is no way to make this work on RHEL-8 as well though, so
# option A is suboptimal.
#
# Option B uses the value of the DEFAULT_VERSION_DATE field from
# make/conf/version-numbers.conf. DEFAULT_VERSION_DATE represents the
# aspirational eventual JDK general availability (GA) release date.
# When the RPM build occurs prior to GA, generated JAR files will have
# payload mtimes in the future relative to the RPM build time.
# Whereas for tarballs some tools will issue warnings about future
# mtimes, per OPENJDK-2583 apparently this is no problem for Java and
# JAR files.
#
# Option C uses the modification timestamp of files in the source
# tarball. The reproducibility logic in generate_source_tarball.sh
# sets them all to the commit time of the release-tagged OpenJDK
# commit, as archived in the tarball. This timestamp is deterministic
# across RPM rebuilds and is reliably in the past. Any file's mtime
# will do, so use version-numbers.conf's.
#
# Use option B for JAR files, based on the discussion in OPENJDK-2583.
#
# For portable tarballs, use option C (OPENJDK_UPSTREAM_TAG_EPOCH) for
# the modification times of all files in the portable tarballs. Doing
# so eliminates one source of variability across RPM rebuilds.
VERSION_FILE="$(pwd)"/"%{top_level_dir_name}"/make/conf/version-numbers.conf
OPENJDK_UPSTREAM_TAG_EPOCH="$(stat --format=%Y "${VERSION_FILE}")"
echo "Building %{SOURCE11}"
mkdir -p %{altjavaoutputdir}
gcc ${EXTRA_CFLAGS} -o %{altjavaoutputdir}/%{alt_java_name} %{SOURCE11}
echo "Generating %{alt_java_name} man page"
altjavamanpage=%{altjavaoutputdir}/%{alt_java_name}.1
echo "Hardened java binary recommended for launching untrusted code from the Web e.g. javaws" > ${altjavamanpage}
cat %{top_level_dir_name}/src/java.base/share/man/java.1 >> ${altjavamanpage}
echo "Building %{newjavaver}-%{buildver}, pre=%{ea_designator}, opt=%{lts_designator}"
function buildjdk() {
local outputdir=${1}
local buildjdk=${2}
local maketargets="${3}"
local debuglevel=${4}
local link_opt=${5}
local debug_symbols=${6}
local top_dir_abs_src_path=$(pwd)/%{top_level_dir_name}
local top_dir_abs_build_path=$(pwd)/${outputdir}
# This must be set using the global, so that the
# static libraries still use a dynamic stdc++lib
if [ "x%{link_type}" = "xbundled" ] ; then
libc_link_opt="static";
else
libc_link_opt="dynamic";
fi
echo "Using output directory: ${outputdir}";
echo "Checking build JDK ${buildjdk} is operational..."
${buildjdk}/bin/java -version
echo "Using make targets: ${maketargets}"
echo "Using debuglevel: ${debuglevel}"
echo "Using link_opt: ${link_opt}"
echo "Using debug_symbols: ${debug_symbols}"
echo "Building %{newjavaver}-%{buildver}, pre=%{ea_designator}, opt=%{lts_designator}"
mkdir -p ${outputdir}
pushd ${outputdir}
# Note: zlib and freetype use %{link_type}
# rather than ${link_opt} as the system versions
# are always used in a system_libs build, even
# for the static library build
2023-11-16 08:11:49 +00:00
bash -c "bash ${top_dir_abs_src_path}/configure \
%ifarch %{zero_arches}
--with-jvm-variants=zero \
%endif
--with-cacerts-file=`readlink -f %{_sysconfdir}/pki/java/cacerts` \
--with-version-build=%{buildver} \
--with-version-pre=\"%{ea_designator}\" \
--with-version-opt=\"%{lts_designator}\" \
--with-vendor-version-string=\"%{oj_vendor_version}\" \
--with-vendor-name=\"%{oj_vendor}\" \
--with-vendor-url=\"%{oj_vendor_url}\" \
--with-vendor-bug-url=\"%{oj_vendor_bug_url}\" \
--with-vendor-vm-bug-url=\"%{oj_vendor_bug_url}\" \
--with-boot-jdk=${buildjdk} \
--with-debug-level=${debuglevel} \
--with-native-debug-symbols="${debug_symbols}" \
--disable-sysconf-nss \
--enable-unlimited-crypto \
--with-zlib=%{link_type} \
--with-freetype=%{link_type} \
--with-libjpeg=${link_opt} \
--with-giflib=${link_opt} \
--with-libpng=${link_opt} \
--with-lcms=${link_opt} \
--with-harfbuzz=${link_opt} \
--with-stdc++lib=${libc_link_opt} \
--with-extra-cxxflags=\"$EXTRA_CPP_FLAGS\" \
--with-extra-cflags=\"$EXTRA_CFLAGS\" \
--with-extra-ldflags=\"%{ourldflags}\" \
--with-num-cores=\"$NUM_PROC\" \
--with-source-date=\"version\" \
--disable-javac-server \
%ifarch %{zgc_arches}
--with-jvm-features=zgc \
%endif
--disable-warnings-as-errors"
cat spec.gmk
2023-11-16 08:11:49 +00:00
bash -c "make LOG=trace $maketargets || \
( pwd; find ${top_dir_abs_src_path} ${top_dir_abs_build_path} -name \"hs_err_pid*.log\" | xargs cat && false )"
popd
}
function stripjdk() {
local outputdir=${1}
local jdkimagepath=${outputdir}/images/%{jdkimage}
local jreimagepath=${outputdir}/images/%{jreimage}
local jmodimagepath=${outputdir}/images/jmods
local supportdir=${outputdir}/support
local modulebuildpath=${outputdir}/jdk/modules
if [ "x$suffix" = "x" ] ; then
# Keep the unstripped version for consumption by RHEL RPMs
cp -a ${jdkimagepath}{,.unstripped}
# Strip the files
for file in $(find ${jdkimagepath} ${jreimagepath} ${supportdir} ${modulebuildpath} -type f) ; do
if file ${file} | cut -d ':' -f 2 | grep -q 'ELF'; then
noextfile=${file/.so/};
2023-11-16 08:11:49 +00:00
bash -c "objcopy --only-keep-debug ${file} ${noextfile}.debuginfo";
bash -c "objcopy --add-gnu-debuglink=${noextfile}.debuginfo ${file}";
bash -c "strip -g ${file}";
fi
done
# Rebuild jmod files against the stripped binaries
if [ ! -d ${supportdir} ] ; then
echo "Support directory missing.";
exit 15
fi
for cmd in $(find ${supportdir} -name '*.jmod_exec.cmdline') ; do
pre=${cmd/_exec/_pre};
post=${cmd/_exec/_post};
jmod=$(echo ${cmd}|sed 's#.*_create_##'|sed 's#_exec.cmdline##')
echo "Rebuilding ${jmod} against stripped binaries...";
if [ -e ${pre} ] ; then
echo "Executing ${pre}...";
cat ${pre} | sh -s ;
fi
echo "Executing ${cmd}...";
cat ${cmd} | sh -s ;
if [ -e ${post} ] ; then
echo "Executing ${post}...";
cat ${post} | sh -s ;
fi
done
rm -rf ${jdkimagepath}/jmods
cp -a ${jmodimagepath} ${jdkimagepath}
fi
}
function installjdk() {
local outputdir=${1}
local installdir=${2}
local jdkimagepath=${installdir}/images/%{jdkimage}
local jreimagepath=${installdir}/images/%{jreimage}
local unstripped=${jdkimagepath}.unstripped
echo "Installing build from ${outputdir} to ${installdir}..."
mkdir -p ${installdir}
echo "Installing images..."
mv ${outputdir}/images ${installdir}
if [ -d ${outputdir}/bundles ] ; then
echo "Installing bundles...";
mv ${outputdir}/bundles ${installdir} ;
fi
%if !%{with artifacts}
echo "Removing output directory...";
rm -rf ${outputdir}
%endif
# legacy-jre-image target does not install any man pages for the JRE
# We copy the jdk man directory and then remove pages for binaries that
# don't exist in the JRE
cp -a ${jdkimagepath}/man ${jreimagepath}
for manpage in $(find ${jreimagepath}/man -name '*.1'); do
filename=$(basename ${manpage});
binary=${filename/.1/};
if [ ! -f ${jreimagepath}/bin/${binary} ] ; then
echo "Removing ${manpage} from JRE for which no binary ${binary} exists";
rm -f ${manpage};
fi;
done
for imagepath in ${jdkimagepath} ${jreimagepath} ${unstripped}; do
if [ -d ${imagepath} ] ; then
# the build (erroneously) removes read permissions from some jars
# this is a regression in OpenJDK 7 (our compiler):
# http://icedtea.classpath.org/bugzilla/show_bug.cgi?id=1437
find ${imagepath} -iname '*.jar' -exec chmod ugo+r {} \;
# Build screws up permissions on binaries
# https://bugs.openjdk.java.net/browse/JDK-8173610
find ${imagepath} -iname '*.so' -exec chmod +x {} \;
find ${imagepath}/bin/ -exec chmod +x {} \;
# Install local files which are distributed with the JDK
install -m 644 %{SOURCE10} ${imagepath}
# Print release information
cat ${imagepath}/release
fi
done
}
function genchecksum() {
local checkedfile=${1}
checkdir=$(dirname ${1})
checkfile=$(basename ${1})
echo "Generating checksum for ${checkfile} in ${checkdir}..."
pushd ${checkdir}
sha256sum ${checkfile} > ${checkfile}.sha256sum
sha256sum --check ${checkfile}.sha256sum
popd
}
# Create a reproducible tarball in an appropriate way for
# the version of tar in use
function createtar() {
local directory=${1}
local archive=${2}
local filter=${3}
local transform=${4}
if [ "x${filter}" != "x" ] ; then
local filteroption="-name ${filter}";
fi
if [ "x${transform}" != "x" ] ; then
local transoption="--transform ${transform}";
fi
local common_tar_opts="--owner=0 --group=0 --numeric-owner \
${transoption} --create --xz"
# Capture tar version, removing the decimal point (so 1.28 => 128)
tarver=$(tar --version|head -n1|sed -re 's|tar \(GNU tar\) ([0-9]).([0-9]*)|\1\2|')
echo "Detected tar ${tarver}"
if [ ${tarver} -gt 128 ] ; then
local tar_time="$(date --utc --iso-8601=seconds --date=@"${OPENJDK_UPSTREAM_TAG_EPOCH}")"
local tar_opts="--mtime=${tar_time} --sort=name ${common_tar_opts}"
if test "x${filteroption}" = "x"; then
tar ${tar_opts} --file ${archive} ${directory}
else
tar ${tar_opts} --file ${archive} $(find ${directory} ${filteroption})
fi
else
# See https://reproducible-builds.org/docs/archives/
# RHEL-7 has tar 1.26 which does not support --sort=name (added
# in 1.28), so use find-piped-through-sort instead. Omit
# --pax-option since it made the docs package not reproducible
# due to PaxHeaders timestamp differences.
local tar_opts="--mtime=@${OPENJDK_UPSTREAM_TAG_EPOCH} \
--no-recursion --null --files-from - \
${common_tar_opts}"
find ${directory} ${filteroption} -print0 | \
LC_ALL=C sort -z | \
tar ${tar_opts} --file ${archive}
fi
}
function packagejdk() {
local imagesdir=$(pwd)/${1}/images
local docdir=$(pwd)/${1}/images/docs
local bundledir=$(pwd)/${1}/bundles
local packagesdir=$(pwd)/${2}
local srcdir=$(pwd)/%{top_level_dir_name}
local tapsetdir=$(pwd)/tapset
local altjavadir=$(pwd)/${3}
echo "Packaging build from ${imagesdir} to ${packagesdir}..."
mkdir -p ${packagesdir}
pushd ${imagesdir}
if [ "x$suffix" = "x" ] ; then
nameSuffix=""
else
nameSuffix=`echo "$suffix"| sed s/-/./`
fi
jdkname=%{jdkportablename -- "$nameSuffix"}
jdkarchive=${packagesdir}/%{jdkportablearchive -- "$nameSuffix"}
jrename=%{jreportablename -- "$nameSuffix"}
jrearchive=${packagesdir}/%{jreportablearchive -- "$nameSuffix"}
staticname=%{staticlibsportablename -- "$nameSuffix"}
staticarchive=${packagesdir}/%{staticlibsportablearchive -- "$nameSuffix"}
debugarchive=${packagesdir}/%{jdkportablearchive -- "${nameSuffix}.debuginfo"}
unstrippedarchive=${packagesdir}/%{jdkportablearchive -- "${nameSuffix}.unstripped"}
# We only use docs for the release build
docname=%{docportablename}
docarchive=${packagesdir}/%{docportablearchive}
built_doc_archive=jdk-%{filever}%{ea_designator_zip}+%{buildver}%{lts_designator_zip}-docs.zip
# These are from the source tree so no debug variants
miscname=%{miscportablename}
miscarchive=${packagesdir}/%{miscportablearchive}
if [ "x$suffix" = "x" ] ; then
# Keep the unstripped version for consumption by RHEL RPMs
mv %{jdkimage}.unstripped ${jdkname}
createtar ${jdkname} ${unstrippedarchive}
genchecksum ${unstrippedarchive}
mv ${jdkname} %{jdkimage}.unstripped
fi
# Rename directories for packaging
mv %{jdkimage} ${jdkname}
mv %{jreimage} ${jrename}
# Release images have external debug symbols
if [ "x$suffix" = "x" ] ; then
createtar ${jdkname} ${debugarchive} \*.debuginfo
genchecksum ${debugarchive}
mkdir ${docname}
mv ${docdir} ${docname}
mv ${bundledir}/${built_doc_archive} ${docname}
createtar ${docname} ${docarchive}
genchecksum ${docarchive}
mkdir ${miscname}
for s in 16 24 32 48 ; do
cp -av ${srcdir}/src/java.desktop/unix/classes/sun/awt/X11/java-icon${s}.png ${miscname}
done
%if %{with_systemtap}
cp -a ${tapsetdir}* ${miscname}
%endif
cp -av ${altjavadir}/%{alt_java_name}{,.1} ${miscname}
createtar ${miscname} ${miscarchive}
genchecksum ${miscarchive}
fi
createtar ${jdkname} ${jdkarchive}
genchecksum ${jdkarchive}
createtar ${jrename} ${jrearchive}
genchecksum ${jrearchive}
%if %{include_staticlibs}
# Static libraries (needed for building graal vm with native image)
# Tar as overlay. Transform to the JDK name, since we just want to "add"
# static libraries to that folder
createtar "%{static_libs_image}/lib" ${staticarchive} "" \
"s|^%{static_libs_image}/lib/*|${staticname}/lib/static/linux-%{archinstall}/glibc/|"
genchecksum ${staticarchive}
%endif
# Revert directory renaming so testing will run
# TODO: testing should run on the packaged JDK
mv ${jdkname} %{jdkimage}
mv ${jrename} %{jreimage}
popd #images
}
%if %{build_hotspot_first}
# Build a fresh libjvm.so first and use it to bootstrap
cp -LR --preserve=mode,timestamps %{bootjdk} newboot
systemjdk=$(pwd)/newboot
buildjdk build/newboot ${systemjdk} %{hotspot_target} "release" "bundled" "internal"
mv build/newboot/jdk/lib/%{vm_variant}/libjvm.so newboot/lib/%{vm_variant}
%else
systemjdk=%{bootjdk}
%endif
for suffix in %{build_loop} ; do
if [ "x$suffix" = "x" ] ; then
debugbuild=release
else
# change --something to something
debugbuild=`echo $suffix | sed "s/-//g"`
fi
# We build with internal debug symbols and do
# our own stripping for one version of the
# release build
debug_symbols=internal
builddir=%{buildoutputdir -- ${suffix}}
bootbuilddir=boot${builddir}
installdir=%{installoutputdir -- ${suffix}}
bootinstalldir=boot${installdir}
packagesdir=%{packageoutputdir -- ${suffix}}
link_opt="%{link_type}"
%if %{system_libs}
# Copy the source tree so we can remove all in-tree libraries
cp -a %{top_level_dir_name} %{top_level_dir_name_backup}
# Remove all libraries that are linked
sh %{SOURCE12} %{top_level_dir_name} full
%endif
# Debug builds don't need same targets as release for
# build speed-up. We also avoid bootstrapping these
# slower builds.
if echo $debugbuild | grep -q "debug" ; then
maketargets="%{debug_targets}"
run_bootstrap=false
else
maketargets="%{release_targets}"
run_bootstrap=%{bootstrap_build}
fi
if ${run_bootstrap} ; then
buildjdk ${bootbuilddir} ${systemjdk} "%{bootstrap_targets}" ${debugbuild} ${link_opt} ${debug_symbols}
installjdk ${bootbuilddir} ${bootinstalldir}
buildjdk ${builddir} $(pwd)/${bootinstalldir}/images/%{jdkimage} "${maketargets}" ${debugbuild} ${link_opt} ${debug_symbols}
stripjdk ${builddir}
installjdk ${builddir} ${installdir}
%{!?with_artifacts:rm -rf ${bootinstalldir}}
else
buildjdk ${builddir} ${systemjdk} "${maketargets}" ${debugbuild} ${link_opt} ${debug_symbols}
stripjdk ${builddir}
installjdk ${builddir} ${installdir}
fi
packagejdk ${installdir} ${packagesdir} %{altjavaoutputdir}
%if %{system_libs}
# Restore original source tree we modified by removing full in-tree sources
rm -rf %{top_level_dir_name}
mv %{top_level_dir_name_backup} %{top_level_dir_name}
%endif
# build cycles
done # end of release / debug cycle loop
%check
# We test debug first as it will give better diagnostics on a crash
for suffix in %{build_loop} ; do
# portable builds have static_libs embedded, thus top_dir_abs_main_build_path is same as top_dir_abs_staticlibs_build_path
top_dir_abs_main_build_path=$(pwd)/%{installoutputdir -- ${suffix}}
%if %{include_staticlibs}
top_dir_abs_staticlibs_build_path=${top_dir_abs_main_build_path}
%endif
export JAVA_HOME=${top_dir_abs_main_build_path}/images/%{jdkimage}
# Pre-test setup
# System security properties are disabled by default on portable.
# Turn on system security properties
#sed -i -e "s:^security.useSystemPropertiesFile=.*:security.useSystemPropertiesFile=true:" \
#${JAVA_HOME}/conf/security/java.security
# Check Shenandoah is enabled
%if %{use_shenandoah_hotspot}
$JAVA_HOME//bin/java -XX:+UnlockExperimentalVMOptions -XX:+UseShenandoahGC -version
%endif
# Check unlimited policy has been used
$JAVA_HOME/bin/javac -d . %{SOURCE13}
$JAVA_HOME/bin/java --add-opens java.base/javax.crypto=ALL-UNNAMED TestCryptoLevel
# Check ECC is working
$JAVA_HOME/bin/javac -d . %{SOURCE14}
$JAVA_HOME/bin/java $(echo $(basename %{SOURCE14})|sed "s|\.java||")
# Check system crypto (policy) is active and can be disabled
# Test takes a single argument - true or false - to state whether system
# security properties are enabled or not.
$JAVA_HOME/bin/javac -d . %{SOURCE15}
export PROG=$(echo $(basename %{SOURCE15})|sed "s|\.java||")
export SEC_DEBUG="-Djava.security.debug=properties"
# Specific to portable:System security properties to be off by default
$JAVA_HOME/bin/java ${SEC_DEBUG} ${PROG} false
$JAVA_HOME/bin/java ${SEC_DEBUG} -Djava.security.disableSystemPropertiesFile=true ${PROG} false
# Check correct vendor values have been set
$JAVA_HOME/bin/javac -d . %{SOURCE16}
$JAVA_HOME/bin/java $(echo $(basename %{SOURCE16})|sed "s|\.java||") "%{oj_vendor}" "%{oj_vendor_url}" "%{oj_vendor_bug_url}" "%{oj_vendor_version}"
# Check java launcher has no SSB mitigation
if ! nm $JAVA_HOME/bin/java | grep set_speculation ; then true ; else false; fi
# Check alt-java launcher has SSB mitigation on supported architectures
# set_speculation function exists in both cases, so check for prctl call
%ifarch %{ssbd_arches}
nm %{altjavaoutputdir}/%{alt_java_name} | grep prctl
%else
if ! nm %{altjavaoutputdir}/%{alt_java_name} | grep prctl ; then true ; else false; fi
%endif
%if ! 0%{?flatpak}
# Check translations are available for new timezones (during flatpak builds, the
# tzdb.dat used by this test is not where the test expects it, so this is
# disabled for flatpak builds)
# Disable test until we are on the latest JDK
$JAVA_HOME/bin/javac -d . %{SOURCE18}
$JAVA_HOME/bin/java $(echo $(basename %{SOURCE18})|sed "s|\.java||") JRE
$JAVA_HOME/bin/java -Djava.locale.providers=CLDR $(echo $(basename %{SOURCE18})|sed "s|\.java||") CLDR
%endif
%if %{include_staticlibs}
# Check debug symbols in static libraries (smoke test)
export STATIC_LIBS_HOME=${top_dir_abs_staticlibs_build_path}/images/%{static_libs_image}
ls -l $STATIC_LIBS_HOME
ls -l $STATIC_LIBS_HOME/lib
readelf --debug-dump $STATIC_LIBS_HOME/lib/libnet.a | grep Inet4AddressImpl.c
readelf --debug-dump $STATIC_LIBS_HOME/lib/libnet.a | grep Inet6AddressImpl.c
%endif
# Release builds strip the debug symbols into external .debuginfo files
if [ "x$suffix" = "x" ] ; then
so_suffix="debuginfo"
else
so_suffix="so"
fi
# Check debug symbols are present and can identify code
find "$JAVA_HOME" -iname "*.$so_suffix" -print0 | while read -d $'\0' lib
do
if [ -f "$lib" ] ; then
echo "Testing $lib for debug symbols"
# All these tests rely on RPM failing the build if the exit code of any set
# of piped commands is non-zero.
# Test for .debug_* sections in the shared object. This is the main test
# Stripped objects will not contain these
eu-readelf -S "$lib" | grep "] .debug_"
test $(eu-readelf -S "$lib" | grep -E "\]\ .debug_(info|abbrev)" | wc --lines) == 2
# Test FILE symbols. These will most likely be removed by anything that
# manipulates symbol tables because it's generally useless. So a nice test
# that nothing has messed with symbols
old_IFS="$IFS"
IFS=$'\n'
for line in $(eu-readelf -s "$lib" | grep "00000000 0 FILE LOCAL DEFAULT")
do
# We expect to see .cpp and .S files, except for architectures like aarch64 and
# s390 where we expect .o and .oS files
echo "$line" | grep -E "ABS ((.*/)?[-_a-zA-Z0-9]+\.(c|cc|cpp|cxx|o|S|oS))?$"
done
IFS="$old_IFS"
# If this is the JVM, look for javaCalls.(cpp|o) in FILEs, for extra sanity checking
if [ "`basename $lib`" = "libjvm.so" ]; then
eu-readelf -s "$lib" | \
grep -E "00000000 0 FILE LOCAL DEFAULT ABS javaCalls.(cpp|o)$"
fi
# Test that there are no .gnu_debuglink sections pointing to another
# debuginfo file. There shouldn't be any debuginfo files, so the link makes
# no sense either
eu-readelf -S "$lib" | grep 'gnu'
if eu-readelf -S "$lib" | grep '] .gnu_debuglink' | grep PROGBITS; then
echo "bad .gnu_debuglink section."
eu-readelf -x .gnu_debuglink "$lib"
false
fi
fi
done
# Make sure gdb can do a backtrace based on line numbers on libjvm.so
# javaCalls.cpp:58 should map to:
# http://hg.openjdk.java.net/jdk8u/jdk8u/hotspot/file/ff3b27e6bcc2/src/share/vm/runtime/javaCalls.cpp#l58
# Using line number 1 might cause build problems. See:
# https://bugzilla.redhat.com/show_bug.cgi?id=1539664
# https://bugzilla.redhat.com/show_bug.cgi?id=1538767
gdb -q "$JAVA_HOME/bin/java" <<EOF | tee gdb.out
handle SIGSEGV pass nostop noprint
handle SIGILL pass nostop noprint
set breakpoint pending on
break javaCalls.cpp:58
commands 1
backtrace
quit
end
run -version
EOF
%ifarch %{gdb_arches}
grep 'JavaCallWrapper::JavaCallWrapper' gdb.out
%endif
# Check src.zip has all sources. See RHBZ#1130490
unzip -l $JAVA_HOME/lib/src.zip | grep 'sun.misc.Unsafe'
# Check class files include useful debugging information
$JAVA_HOME/bin/javap -l java.lang.Object | grep "Compiled from"
$JAVA_HOME/bin/javap -l java.lang.Object | grep LineNumberTable
$JAVA_HOME/bin/javap -l java.lang.Object | grep LocalVariableTable
# Check generated class files include useful debugging information
$JAVA_HOME/bin/javap -l java.nio.ByteBuffer | grep "Compiled from"
$JAVA_HOME/bin/javap -l java.nio.ByteBuffer | grep LineNumberTable
$JAVA_HOME/bin/javap -l java.nio.ByteBuffer | grep LocalVariableTable
# build cycles check
done
%install
for suffix in %{build_loop} ; do
packagesdir=%{packageoutputdir -- ${suffix}}
if [ "x$suffix" == "x" ] ; then
nameSuffix=""
else
nameSuffix=`echo "$suffix"| sed s/-/./`
fi
# These definitions should match those in installjdk
jdkarchive=${packagesdir}/%{jdkportablearchive -- "$nameSuffix"}
jrearchive=${packagesdir}/%{jreportablearchive -- "$nameSuffix"}
staticarchive=${packagesdir}/%{staticlibsportablearchive -- "$nameSuffix"}
debugarchive=${packagesdir}/%{jdkportablearchive -- "${nameSuffix}.debuginfo"}
unstrippedarchive=${packagesdir}/%{jdkportablearchive -- "${nameSuffix}.unstripped"}
mkdir -p $RPM_BUILD_ROOT%{_jvmdir}
mv ${jdkarchive} $RPM_BUILD_ROOT%{_jvmdir}/
mv ${jdkarchive}.sha256sum $RPM_BUILD_ROOT%{_jvmdir}/
mv ${jrearchive} $RPM_BUILD_ROOT%{_jvmdir}/
mv ${jrearchive}.sha256sum $RPM_BUILD_ROOT%{_jvmdir}/
%if %{include_staticlibs}
mv ${staticarchive} $RPM_BUILD_ROOT%{_jvmdir}/
mv ${staticarchive}.sha256sum $RPM_BUILD_ROOT%{_jvmdir}/
%endif
if [ "x$suffix" = "x" ] ; then
mv ${debugarchive} $RPM_BUILD_ROOT%{_jvmdir}/
mv ${debugarchive}.sha256sum $RPM_BUILD_ROOT%{_jvmdir}/
mv ${unstrippedarchive} $RPM_BUILD_ROOT%{_jvmdir}/
mv ${unstrippedarchive}.sha256sum $RPM_BUILD_ROOT%{_jvmdir}/
fi
done
# These definitions should match those in installjdk
# Install outside the loop as there are no debug variants
docarchive=${packagesdir}/%{docportablearchive}
miscarchive=${packagesdir}/%{miscportablearchive}
mv ${docarchive} $RPM_BUILD_ROOT%{_jvmdir}/
mv ${docarchive}.sha256sum $RPM_BUILD_ROOT%{_jvmdir}/
mv ${miscarchive} $RPM_BUILD_ROOT%{_jvmdir}/
mv ${miscarchive}.sha256sum $RPM_BUILD_ROOT%{_jvmdir}/
# To show sha in the build log
for file in `ls $RPM_BUILD_ROOT%{_jvmdir}/*.sha256sum` ; do
ls -l $file ;
cat $file ;
done
%if %{include_normal_build}
%files
# main package builds always
%{_jvmdir}/%{jreportablearchive -- %%{nil}}
%{_jvmdir}/%{jreportablearchive -- %%{nil}}.sha256sum
%else
%files
# placeholder
%endif
%files devel
%{_jvmdir}/%{jdkportablearchive -- %%{nil}}
%{_jvmdir}/%{jdkportablearchive -- .debuginfo}
%{_jvmdir}/%{jdkportablearchive -- %%{nil}}.sha256sum
%{_jvmdir}/%{jdkportablearchive -- .debuginfo}.sha256sum
%if %{include_staticlibs}
%files static-libs
%{_jvmdir}/%{staticlibsportablearchive -- %%{nil}}
%{_jvmdir}/%{staticlibsportablearchive -- %%{nil}}.sha256sum
%endif
%files unstripped
%{_jvmdir}/%{jdkportablearchive -- .unstripped}
%{_jvmdir}/%{jdkportablearchive -- .unstripped}.sha256sum
%if %{include_debug_build}
%files slowdebug
%{_jvmdir}/%{jreportablearchive -- .slowdebug}
%{_jvmdir}/%{jreportablearchive -- .slowdebug}.sha256sum
%files devel-slowdebug
%{_jvmdir}/%{jdkportablearchive -- .slowdebug}
%{_jvmdir}/%{jdkportablearchive -- .slowdebug}.sha256sum
%if %{include_staticlibs}
%files static-libs-slowdebug
%{_jvmdir}/%{staticlibsportablearchive -- .slowdebug}
%{_jvmdir}/%{staticlibsportablearchive -- .slowdebug}.sha256sum
%endif
%endif
%if %{include_fastdebug_build}
%files fastdebug
%{_jvmdir}/%{jreportablearchive -- .fastdebug}
%{_jvmdir}/%{jreportablearchive -- .fastdebug}.sha256sum
%files devel-fastdebug
%{_jvmdir}/%{jdkportablearchive -- .fastdebug}
%{_jvmdir}/%{jdkportablearchive -- .fastdebug}.sha256sum
%if %{include_staticlibs}
%files static-libs-fastdebug
%{_jvmdir}/%{staticlibsportablearchive -- .fastdebug}
%{_jvmdir}/%{staticlibsportablearchive -- .fastdebug}.sha256sum
%endif
%endif
%files docs
%{_jvmdir}/%{docportablearchive}
%{_jvmdir}/%{docportablearchive}.sha256sum
%files misc
%{_jvmdir}/%{miscportablearchive}
%{_jvmdir}/%{miscportablearchive}.sha256sum
%changelog
* Sat Apr 13 2024 Andrew Hughes <gnu.andrew@redhat.com> - 1:21.0.3.0.9-1
- Update to jdk-21.0.3+9 (GA)
- Update release notes to 21.0.3+9
- Switch to GA mode.
- Change --with-source-date value to 'version' to match Temurin builds
- ** This tarball is embargoed until 2024-04-16 @ 1pm PT. **
- Resolves: OPENJDK-2585
* Thu Apr 04 2024 Andrew Hughes <gnu.andrew@redhat.com> - 1:21.0.3.0.7-0.1.ea
- Update to jdk-21.0.3+7 (EA)
- Update release notes to 21.0.3+7
- Require tzdata 2024a due to upstream inclusion of JDK-8322725
- Only require tzdata 2023d for now as 2024a is unavailable in buildroot
- Drop JDK-8009550 which is now available upstream
- Re-generate FIPS patch against 21.0.3+7 following backport of JDK-8325254
* Thu Mar 21 2024 Andrew Hughes <gnu.andrew@redhat.com> - 1:21.0.3.0.1-0.2.ea
- generate_source_tarball.sh: Update examples in header for clarity
- generate_source_tarball.sh: Cleanup message issued when checkout already exists
- generate_source_tarball.sh: Create directory in TMPDIR when using WITH_TEMP
- generate_source_tarball.sh: Only add --depth=1 on non-local repositories
- icedtea_sync.sh: Reinstate from rhel-8.9.0 branch
- Move maintenance scripts to a scripts subdirectory
- discover_trees.sh: Set compile-command and indentation instructions for Emacs
- discover_trees.sh: shellcheck: Do not use -o (SC2166)
- discover_trees.sh: shellcheck: Remove x-prefixes since we use Bash (SC2268)
- discover_trees.sh: shellcheck: Double-quote variable references (SC2086)
- generate_source_tarball.sh: Add authorship
- icedtea_sync.sh: Set compile-command and indentation instructions for Emacs
- icedtea_sync.sh: shellcheck: Double-quote variable references (SC2086)
- icedtea_sync.sh: shellcheck: Remove x-prefixes since we use Bash (SC2268)
- openjdk_news.sh: Set compile-command and indentation instructions for Emacs
- openjdk_news.sh: shellcheck: Double-quote variable references (SC2086)
- openjdk_news.sh: shellcheck: Remove x-prefixes since we use Bash (SC2268)
- openjdk_news.sh: shellcheck: Remove deprecated egrep usage (SC2196)
- generate_source_tarball.sh: Output values of new options WITH_TEMP and OPENJDK_LATEST
- generate_source_tarball.sh: Double-quote DEPTH reference (SC2086)
- Vary reproducible tar creation by version of tar detected
- Set OPENJDK_UPSTREAM_TAG_EPOCH & VERSION_FILE at start of build section as in 17u
- generate_source_tarball.sh: Avoid empty DEPTH reference while still appeasing shellcheck
* Wed Mar 20 2024 Thomas Fitzsimmons <fitzsim@redhat.com> - 1:21.0.3.0.1-0.2.ea
- generate_source_tarball.sh: Add WITH_TEMP environment variable
- generate_source_tarball.sh: Multithread xz on all available cores
- generate_source_tarball.sh: Add OPENJDK_LATEST environment variable
- generate_source_tarball.sh: Update comment about tarball naming
- generate_source_tarball.sh: Reformat comment header
- generate_source_tarball.sh: Reformat and update help output
- generate_source_tarball.sh: Do a shallow clone, for speed
- generate_source_tarball.sh: Append -ea designator when required
- generate_source_tarball.sh: Eliminate some removal prompting
- generate_source_tarball.sh: Make tarball reproducible
- generate_source_tarball.sh: Prefix temporary directory with temp-
- generate_source_tarball.sh: Remove temporary directory exit conditions
- generate_source_tarball.sh: Fix -ea logic to add dash
- generate_source_tarball.sh: Set compile-command in Emacs
- generate_source_tarball.sh: Remove REPO_NAME from FILE_NAME_ROOT
- generate_source_tarball.sh: Move PROJECT_NAME and REPO_NAME checks
- generate_source_tarball.sh: shellcheck: Remove x-prefixes since we use Bash (SC2268)
- generate_source_tarball.sh: shellcheck: Double-quote variable references (SC2086)
- generate_source_tarball.sh: shellcheck: Do not use -a (SC2166)
- generate_source_tarball.sh: shellcheck: Do not use $ on arithmetic variables (SC2004)
- Use backward-compatible patch syntax
- generate_source_tarball.sh: Ignore -ga tags with OPENJDK_LATEST
- generate_source_tarball.sh: Fix whitespace
- generate_source_tarball.sh: Remove trailing period in echo
- generate_source_tarball.sh: Use long-style argument to grep
- generate_source_tarball.sh: Add license
- generate_source_tarball.sh: Add indentation instructions for Emacs
- Remove -T0 argument from systemtap tar invocation
* Mon Mar 11 2024 Andrew Hughes <gnu.andrew@redhat.com> - 1:21.0.3.0.1-0.2.ea
- Introduce tar_opts to avoid repetition of lengthy tar creation options
- Add module build path to stripped directories to catch jpackageapplauncher files
- Move alt-java man page to the misc tarball so it is not in the JDK image
- Resolves: OPENJDK-2820
- Resolves: OPENJDK-2821
* Thu Feb 08 2024 Thomas Fitzsimmons <fitzsim@redhat.com> - 1:21.0.3.0.1-0.2.ea
- Invoke xz in multi-threaded mode
- Remove ppc64le with-jobs=1 workaround
- Make portable tarball modification times reproducible
- Use RHEL-7 tar-1.26-compatible invocations for reproducible tarballs
* Fri Feb 02 2024 Andrew Hughes <gnu.andrew@redhat.com> - 1:21.0.3.0.1-0.1.ea
- Update to jdk-21.0.3+1 (EA)
- Update release notes to 21.0.3+1
- Switch to EA mode
- Require tzdata 2023d due to upstream inclusion of JDK-8322725
- Bump FreeType version to 2.13.2 following JDK-8316028
* Sat Jan 27 2024 Andrew Hughes <gnu.andrew@redhat.com> - 1:21.0.2.0.13-2
- Sync with upstream release notes
* Tue Jan 09 2024 Andrew Hughes <gnu.andrew@redhat.com> - 1:21.0.2.0.13-1
- Update to jdk-21.0.2+13 (GA)
- Update release notes to 21.0.2+13
- Drop no longer needed local patch to fix versioning
- ** This tarball is embargoed until 2024-01-16 @ 1pm PT. **
* Mon Jan 08 2024 Andrew Hughes <gnu.andrew@redhat.com> - 1:21.0.2.0.12-1
- Update to jdk-21.0.2+12 (GA)
- Update release notes to 21.0.2+12
- ** This tarball is embargoed until 2024-01-16 @ 1pm PT. **
* Sat Jan 06 2024 Andrew Hughes <gnu.andrew@redhat.com> - 1:21.0.2.0.11-1
- Update to jdk-21.0.2+11 (GA)
- Update release notes to 21.0.2+11
- Bump libpng version to 1.6.40 following JDK-8316030
- Bump HarfBuzz version to 8.2.2 following JDK-8313643
- Drop local JDK-8311630 patch which is now upstream
- Locally patch versioning to be 21.0.2 released on 2014-01-16
- ** This tarball is embargoed until 2024-01-16 @ 1pm PT. **
* Mon Nov 06 2023 Andrew Hughes <gnu.andrew@redhat.com> - 1:21.0.1.0.12-2
- Include JDK-8311630 patch to implement Foreign Function & Memory preview API on s390x
* Sun Oct 29 2023 Andrew Hughes <gnu.andrew@redhat.com> - 1:21.0.1.0.12-1
- Update to jdk-21.0.1.0+12 (GA)
- Update release notes to 21.0.1.0+12
- Update openjdk_news script to specify subdirectory last
- Add missing discover_trees script required by openjdk_news
- Synchronise bundled versions with 21u sources (FreeType, LCMS, HarfBuzz, libpng)
- Sync generate_tarball.sh with 11u & 17u version
- Update bug URL for RHEL to point to the Red Hat customer portal
- Fix upstream release URL for OpenJDK source
- Update buildjdkver to match the featurever
- Re-enable SystemTap support and perform only substitutions possible without final NVR available
- Fix typo which stops the EA designator being included in the build
- Include tapsets in the miscellaneous tarball
- Drop unused globals for tapset installation
- Rebuild jmods using the stripped binaries in release builds
- Make sure the unstripped JDK is customised by the installjdk function
* Sat Oct 28 2023 Andrew Hughes <gnu.andrew@redhat.com> - 1:21.0.0.0.35-1
- Update to jdk-21.0.0+35
- Update release notes to 21.0.0+35
- Update documentation (README.md)
- Update system crypto policy & FIPS patch from new fips-21u tree
- Update generate_tarball.sh to sync with upstream vanilla script inc. no more ECC removal
- Drop fakefeaturever now it is no longer needed
- Hardcode buildjdkver while the build JDK is not yet 21
- Change top_level_dir_name to use the VCS tag, matching new upstream release style tarball
- Use upstream release URL for OpenJDK source
- Re-enable tzdata tests now we are on the latest JDK and things are back in sync
- Fix trailing '.' in tarball name
- Use rpmrelease in vendor version to avoid inclusion of dist tag
- Replace alt-java patch with a binary separate from the JDK
- Drop stale patches that are of little use any more:
- * nss.cfg has been disabled since early PKCS11 work and long superseded by FIPS work
- * No accessibility subpackage to warrant RH1648242 patch any more
- * No use of system libjpeg turbo to warrant RH649512 patch any more
- Replace RH1684077 pcsc-lite-libs patch with better JDK-8009550 fix being upstreamed
* Sat Oct 28 2023 Petra Alice Mikova <pmikova@redhat.com> - 1:21.0.0.0.35-1
- Replace smoke test files used in the staticlibs test, as fdlibm was removed by JDK-8303798
- Related: rhbz#2192749
* Fri Oct 27 2023 Andrew Hughes <gnu.andrew@redhat.com> - 1:20.0.2.0.9-1.1
- Update to jdk-20.0.2+9
- Update release notes to 20.0.2+9
- Update system crypto policy & FIPS patch from new fips-20u tree
- Update generate_tarball.sh ICEDTEA_VERSION
- Update CLDR reference data following update to 42 (Rocky Mountain-Normalzeit => Rocky-Mountain-Normalzeit)
- Related: rhbz#2192749
* Fri Oct 27 2023 Jiri Vanek <jvanek@redhat.com> - 1:20.0.0.0.36-1
- Dropped JDK-8295447, JDK-8296239 & JDK-8299439 patches now upstream
- Adapted rh1750419-redhat_alt_java.patch
- Related: rhbz#2192749
* Fri Oct 27 2023 Andrew Hughes <gnu.andrew@redhat.com> - 1:19.0.1.0.10-1
- Update to jdk-19.0.2 release
- Update release notes to 19.0.2
- Rebase FIPS patches from fips-19u branch
- Remove references to sample directory removed by JDK-8284999
- Add local patch JDK-8295447 (javac NPE) which was accepted into 19u upstream but not in the GA tag
- Add local patches for JDK-8296239 & JDK-8299439 (Croatia Euro update) which are present in 8u, 11u & 17u releases
- Related: rhbz#2192749
* Tue Oct 24 2023 Andrew Hughes <gnu.andrew@redhat.com> - 1:18.0.2.0.9-1
- Update to jdk-18.0.2 release
- Update release notes to actually reflect OpenJDK 18
- Support JVM variant zero following JDK-8273494 no longer installing Zero's libjvm.so in the server directory
- Rebase FIPS patches from fips-18u branch
- Rebase RH1648249 nss.cfg patch so it applies after the FIPS patch
- Automatically turn off building a fresh HotSpot first, if the bootstrap JDK is not the same major version as that being built
- Drop tzdata patches added for 17.0.7 which will eventually appear in the upstream tarball when we reach OpenJDK 21
- Switch bootjdkver to java-21-openjdk
- Disable tzdata tests until we are on the latest JDK and things are back in sync
- Drop bootstrap JDKs and use the java-21-openjdk-rhel7 build
- Related: rhbz#2192749
* Tue Oct 24 2023 Petra Alice Mikova <pmikova@redhat.com> - 1:18.0.0.0.37-1
- Update to ea version of jdk18
- Adjust rh1684077-openjdk_should_depend_on_pcsc-lite-libs_instead_of_pcsc-lite-devel.patch
* Tue Aug 22 2023 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.7.0.7-2
- Define architectures we build on to avoid those without DTS 10 (e.g. s390)
* Tue Aug 22 2023 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.7.0.7-2
- Switch to DTS 10
- Related: rhbz#2192749
* Mon May 15 2023 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.7.0.7-2
- Create java-21-openjdk-portable package based on java-17-openjdk-portable
- Related: rhbz#2192749
* Thu Apr 13 2023 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.7.0.7-1
- Update to jdk-17.0.7.0+7
- Update release notes to 17.0.7.0+7
- Require tzdata 2023c due to local inclusion of JDK-8274864 & JDK-8305113
- Reintroduce generate_source_tarball.sh from RHEL 9
- Update generate_tarball.sh to add support for passing a boot JDK to the configure run
- Add POSIX-friendly error codes to generate_tarball.sh and fix whitespace
- Remove .jcheck and GitHub support when generating tarballs, as done in upstream release tarballs
- Update FIPS support against 17.0.7+6 and bring in latest changes:
- * RH2134669: Add missing attributes when registering services in FIPS mode.
- * test/jdk/sun/security/pkcs11/fips/VerifyMissingAttributes.java: fixed jtreg main class
- * RH1940064: Enable XML Signature provider in FIPS mode
- * RH2173781: Avoid calling C_GetInfo() too early, before cryptoki is initialized
- ** This tarball is embargoed until 2023-04-18 @ 1pm PT. **
- Resolves: rhbz#2185182
- Resolves: rhbz#2134669
- Resolves: rhbz#1940064
- Resolves: rhbz#2173781
* Tue Feb 21 2023 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.6.0.10-6
- Add docs, icons and samples to the portable output
- Make sure generated checksums work and don't include full path
- The docs directory is a subdirectory of images, so remove confusing separate copying
* Wed Feb 15 2023 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.6.0.10-5
- Build with internal debuginfo as in RHEL and then create a stripped variant ourselves for the portable release build
- Restore compiler flags to those used in RHEL
- Drop unused static library patch
- Drop syslookup workaround which was fixed by JDK-8276572 over a year ago
* Tue Feb 14 2023 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.6.0.10-4
- Separate JDK packaging into a separate function
- Use variables to make it clearer what is going on
- Use a package output directory as we do for building and installing
- Workaround missing manpage directory in the JRE image
* Sun Feb 12 2023 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.6.0.10-3
- Adapt the portable build to use the same system library handling as RHEL builds
* Sat Jan 14 2023 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.6.0.10-3
- Add missing release note for JDK-8295687
- Resolves: rhbz#2160111
* Fri Jan 13 2023 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.6.0.10-2
- Update FIPS support to bring in latest changes
- * Add nss.fips.cfg support to OpenJDK tree
- * RH2117972: Extend the support for NSS DBs (PKCS11) in FIPS mode
- * Remove forgotten dead code from RH2020290 and RH2104724
- * OJ1357: Fix issue on FIPS with a SecurityManager in place
- Drop local nss.fips.cfg.in handling now this is handled in the patched OpenJDK build
- Resolves: rhbz#2118493
* Fri Jan 13 2023 Stephan Bergmann <sbergman@redhat.com> - 1:17.0.6.0.10-2
- Fix flatpak builds by disabling TestTranslations test due to missing tzdb.dat
- Related: rhbz#2160111
* Wed Jan 11 2023 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.6.0.10-1
- Update to jdk-17.0.6.0+10
- Update release notes to 17.0.6.0+10
- Re-enable EA upstream status check now it is being actively maintained.
- Drop JDK-8294357 (tzdata2022d) & JDK-8295173 (tzdata2022e) local patches which are now upstream
- Drop JDK-8275535 local patch now this has been accepted and backported upstream
- Drop local copy of JDK-8293834 now this is upstream
- Require tzdata 2022g due to inclusion of JDK-8296108, JDK-8296715 & JDK-8297804
- Update TestTranslations.java to test the new America/Ciudad_Juarez zone
- ** This tarball is embargoed until 2023-01-17 @ 1pm PT. **
- Resolves: rhbz#2160111
* Sat Oct 15 2022 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.5.0.8-2
- Update in-tree tzdata to 2022e with JDK-8294357 & JDK-8295173
- Update CLDR data with Europe/Kyiv (JDK-8293834)
- Drop JDK-8292223 patch which we found to be unnecessary
- Update TestTranslations.java to use public API based on TimeZoneNamesTest upstream
- Related: rhbz#2160111
* Thu Oct 13 2022 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.5.0.8-1
- Update to jdk-17.0.5+8 (GA)
- Update release notes to 17.0.5+8 (GA)
- Switch to GA mode for final release.
- * This tarball is embargoed until 2022-10-18 @ 1pm PT. *
- Resolves: rhbz#2133695
* Fri Sep 02 2022 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.4.1.1-2
- Update FIPS support to bring in latest changes
- * RH2023467: Enable FIPS keys export
- * RH2104724: Avoid import/export of DH private keys
- * RH2092507: P11Key.getEncoded does not work for DH keys in FIPS mode
- * Build the systemconf library on all platforms
- * RH2048582: Support PKCS#12 keystores
- * RH2020290: Support TLS 1.3 in FIPS mode
- Resolves: rhbz#2123579
- Resolves: rhbz#2123580
- Resolves: rhbz#2123581
- Resolves: rhbz#2123583
- Resolves: rhbz#2123584
* Sun Aug 21 2022 Jayashree Huttanagoudar <jhuttana@redhat.com> - 1:17.0.4.1.1-1
- Added a missing change to portable NEWS file from upstream.
* Sun Aug 21 2022 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.4.1.1-1
- Update to jdk-17.0.4.1+1
- Update release notes to 17.0.4.1+1
- Add patch to provide translations for Europe/Kyiv added in tzdata2022b
- Add test to ensure timezones can be translated
- Resolves: rhbz#2119532
* Mon Jul 18 2022 Jayashree Huttanagoudar <jhuttana@redhat.com> - 1:17.0.4.0.8-1
- Commented out: fipsver f8142a23d0a which was from rhel-9-main
- Picked 17.0.4+8 GA tag from rhel-9.0.0
- For Jul 2022 CPU fipsver is 765f970aef1 on rhel-9.0.0
* Mon Jul 18 2022 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.4.0.8-1
- Update to jdk-17.0.4.0+8 (GA)
- Update release notes to 17.0.4.0+8
- Need to include the '.S' suffix in debuginfo checks after JDK-8284661
- Switch to GA mode for release
- ** This tarball is embargoed until 2022-07-19 @ 1pm PT. **
* Thu Jul 14 2022 Jayashree Huttanagoudar <jhuttana@redhat.com> - 1:17.0.4.0.1-0.2.ea
- Fix issue where CheckVendor.java test erroneously passes when it should fail.
- Add proper quoting so '&' is not treated as a special character by the shell.
- Related: rhbz#2084779
* Tue Jul 12 2022 Jayashree Huttanagoudar <jhuttana@redhat.com> - 1:17.0.4.0.1-0.1.ea
- Tweaked line to print release information for portable
* Tue Jul 12 2022 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.4.0.1-0.1.ea
- Update to jdk-17.0.4.0+1
- Update release notes to 17.0.4.0+1
- Switch to EA mode for 17.0.4 pre-release builds.
- Print release file during build, which should now include a correct SOURCE value from .src-rev
- Update tarball script with IcedTea GitHub URL and .src-rev generation
- Include script to generate bug list for release notes
- Update tzdata requirement to 2022a to match JDK-8283350
- Move EA designator check to prep so failures can be caught earlier
- Make EA designator check non-fatal while upstream is not maintaining it
- Related: rhbz#2084218
* Thu Jun 30 2022 Jayashree Huttanagoudar <jhuttana@redhat.com> - 1:17.0.3.0.7-8
- Comment line for portable: System security properties to be off by default
* Thu Jun 30 2022 Francisco Ferrari Bihurriet <fferrari@redhat.com> - 1:17.0.3.0.7-8
- RH2007331: SecretKey generate/import operations don't add the CKA_SIGN attribute in FIPS mode
- Resolves: rhbz#2102433
* Wed Jun 29 2022 Jayashree Huttanagoudar <jhuttana@redhat.com> - 1:17.0.3.0.7-7
- System security properties are disabled by default on portable.
- Commented out lines which are not applicable for portable.
* Wed Jun 29 2022 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.3.0.7-7
- Update FIPS support to bring in latest changes
- * RH2036462: sun.security.pkcs11.wrapper.PKCS11.getInstance breakage
- * RH2090378: Revert to disabling system security properties and FIPS mode support together
- Rebase RH1648249 nss.cfg patch so it applies after the FIPS patch
- Enable system security properties in the RPM (now disabled by default in the FIPS repo)
- Improve security properties test to check both enabled and disabled behaviour
- Run security properties test with property debugging on
- Resolves: rhbz#2099844
- Resolves: rhbz#2100677
* Tue Jun 28 2022 Jayashree Huttanagoudar <jhuttana@redhat.com> - 1:17.0.3.0.7-6
- Removed upstreamed patch2001: aqaCheckSecurityAndProviderFileSocketPermissions.patch
* Sun Jun 12 2022 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.3.0.7-6
- Rebase FIPS patches from fips-17u branch and simplify by using a single patch from that repository
- Rebase RH1648249 nss.cfg patch so it applies after the FIPS patch
- RH2023467: Enable FIPS keys export
- RH2094027: SunEC runtime permission for FIPS
- Resolves: rhbz#2029657
- Resolves: rhbz#2096117
* Wed May 25 2022 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.3.0.7-5
- Exclude s390x from the gdb test on RHEL 7 where we see failures with the portable build
* Tue May 24 2022 Jiri Vanek <jvanek@redhat.com> - 1:17.0.3.0.7-4
- to pass aqa, fixing genuie failure in :
- java/lang/SecurityManager/CheckAccessClassInPackagePermissions.java#CheckAccessClassInPackagePermissions
- javax/xml/crypto/dsig/FileSocketPermissions.java#FileSocketPermissions
- added and applied patch2001: aqaCheckSecurityAndProviderFileSocketPermissions.patch
- this, properly named, patch must go to all our jdk17 builds, and to the fips repo
* Thu May 19 2022 Jiri Vanek <jvanek@redhat.com> - 1:17.0.3.0.7-3
- to pass aqa:
- removed copy system tzdb in favour of in-tree
- removed Patch2: rh1648644-java_access_bridge_privileged_security.patch
- This is not intended to release untill we decide proper steps
* Thu May 19 2022 Jayashree Huttanagoudar <jhuttana@redhat.com> - 1:17.0.3.0.7-2
- Include BOOT_JDK for s390x for portable
- BOOT_JDK downlaoded form hydra as
java-17-temurin-17.0.3.7-0.private.ojdk17~upstream.hotspot.release.sdk.el7.s390x.tarxz
and renamed
- Added cosmetic changes to bypass a failure for s390x
* Wed Apr 20 2022 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.3.0.7-1
- April 2022 security update to jdk 17.0.3+7
- Remove JDK-8284548 and JDK-8284920 they are upstreamed now
- Resolves: rhbz#2073579
* Sat Apr 16 2022 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.3.0.6-3
- Add JDK-8284920 fix for XPath regression
- Related: rhbz#2073575
* Fri Apr 15 2022 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.3.0.6-2
- Remove the patch jdk8283911-default_promoted_version_pre.patch which missed in previous commit
- JDK-8275082 should be listed as also resolving JDK-8278008 & CVE-2022-21476
- Related: rhbz#2073575
* Mon Apr 11 2022 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.3.0.6-1
- April 2022 security update to jdk 17.0.3+6
- Update to jdk-17.0.3.0+6 pre-release tarball (17usec.17.0.3+5-220408)
- Add JDK-8284548 regression fix missing from pre-release tarball but in jdk-17.0.3+6/jdk-17.0.3-ga
- Update release notes to 17.0.3.0+6
- Add missing README.md and generate_source_tarball.sh
- Introduce tests/tests.yml, based on the one in java-11-openjdk
- JDK-8283911 patch no longer needed now we're GA...
- Switch to GA mode for release
- ** This tarball is embargoed until 2022-04-19 @ 1pm PT. **
- Resolves: rhbz#2073575
* Wed Apr 06 2022 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.3.0.5-0.1.ea
- Update to jdk-17.0.3.0+5
- Update release notes to 17.0.3.0+5
- Resolves: rhbz#2050460
* Tue Mar 29 2022 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.3.0.1-0.1.ea
- Update to jdk-17.0.3.0+1
- Update release notes to 17.0.3.0+1
- Switch to EA mode for 17.0.3 pre-release builds.
- Add JDK-8283911 to fix bad DEFAULT_PROMOTED_VERSION_PRE value
- Related: rhbz#2050456
* Mon Feb 28 2022 Jayashree Huttanagoudar <jhuttana@redhat.com> - 1:17.0.2.0.8-10
- Update icedtea_sync.sh with suitable message for portable
* Mon Feb 28 2022 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.2.0.8-10
- Restructure the build so a minimal initial build is then used for the final build (with docs)
- This reduces pressure on the system JDK and ensures the JDK being built can do a full build
- Turn off bootstrapping for slow debug builds, which are particularly slow on ppc64le.
- Handle Fedora in distro conditionals that currently only pertain to RHEL.
- Run OpenJDK normalizer script on the spec file to fix further rogue whitespace
- Sync gdb test with java-1.8.0-openjdk and improve architecture restrictions.
- Introduce stapinstall variable to set SystemTap arch directory correctly (e.g. arm64 on aarch64)
- Need to support noarch for creating source RPMs for non-scratch builds.
- Replace -mstackrealign with -mincoming-stack-boundary=2 -mpreferred-stack-boundary=4 on x86_32 for stack alignment
- Support a HotSpot-only build so a freshly built libjvm.so can then be used in the bootstrap JDK.
- Explicitly list JIT architectures rather than relying on those with slowdebug builds
- Disable the serviceability agent on Zero architectures even when the architecture itself is supported
- Resolves: rhbz#2022822
* Mon Feb 28 2022 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.2.0.8-9
- Enable AlgorithmParameters and AlgorithmParameterGenerator services in FIPS mode
- Correction to previous changelog entry
- Resolves: rhbz#2052070
* Sun Feb 27 2022 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.2.0.8-8
- Detect NSS at runtime for FIPS detection
- Resolves: rhbz#2051605
* Wed Feb 23 2022 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.2.0.8-7
- Add JDK-8275535 patch to fix LDAP authentication issue.
- Resolves: rhbz#2053521
* Tue Feb 08 2022 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.2.0.8-6
- Minor cosmetic improvements to make spec more comparable between variants
- Related: rhbz#2022822
* Thu Feb 03 2022 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.2.0.8-5
- Update tapsets from IcedTea 6.x repository with fix for JDK-8015774 changes (_heap->_heaps) and @JAVA_SPEC_VER@
- Related: rhbz#2022822
* Thu Feb 03 2022 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.2.0.8-4
- Extend LTS check to exclude EPEL.
- Related: rhbz#2022822
* Tue Jan 18 2022 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.2.0.8-3
- Separate crypto policy initialisation from FIPS initialisation, now they are no longer interdependent
* Mon Jan 17 2022 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.2.0.8-2
- Fix FIPS issues in native code and with initialisation of java.security.Security
- Related: rhbz#2039366
* Wed Jan 12 2022 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.2.0.8-1
- January 2022 security update to jdk 17.0.2+8
- Rebase RH1995150 & RH1996182 patches following JDK-8275863 addition to module-info.java
- Resolves: rhbz#2039366
- Minor change to the OUTPUT_FILE value to separate the name from the version with '-'
* Mon Nov 29 2021 Severin Gehwolf <sgehwolf@redhat.com> - 1:17.0.1.0.12-3
- Use 'sql:' prefix in nss.fips.cfg as F35+ no longer ship the legacy
secmod.db file as part of nss
- Resolves: rhbz#2023537
* Tue Oct 26 2021 Andrew Hughes <gnu.andrew@redhat.com> - 1:17.0.1.0.12-2
- Drop JDK-8272332/RH2004078 patch which is upstream in 17.0.1
- October CPU update to jdk 17.0.1+12
- Allow plain key import to be disabled with -Dcom.redhat.fips.plainKeySupport=false
- Add patch to allow plain key import.
* Mon Oct 25 2021 Jiri Vanek <jvanek@redhat.com> - 1:17.0.0.0.35-5
- cacerts symlink is resolved before passed to configure
- https://issues.redhat.com/browse/OPENJDK-487
- Disable FIPS mode detection using NSS in favour of using /proc/sys/crypto/fips_enabled for now, so we don't link against NSS
-- effectively disabled Patch1008: rh1929465-improve_system_FIPS_detection.patch by settng --enable-sysconf-nss to --disable-sysconf-nss
-- the enable-sysconf-nss was bringing in hard depndence on nss. Without nss, even in non fips, jvm had not even started
* Thu Sep 30 2021 Jiri Vanek <jvanek@redhat.com> - 1:17.0.0.0.35-4
- initial import, based on jdk11 portbale, merged with jdk17 rpms and java-latest-openjdk for epel7