| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
| |
This was breaking on hosts with dash as /bin/sh.
Signed-off-by: Christopher Larson <chris_larson@mentor.com>
Signed-off-by: Sujith H <sujith.h@gmail.com>
Signed-off-by: Maxin B. John <maxin.john@intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Simplifies duplicated implementation across multiple recipes that all
inherited from java.bbclass.
Previously implementation was not flexible in supporting other make
job-limiting flags (like -l for load) that are typically passed in
through PARALLEL_MAKE. (OpenJDK doesn't know about these other flags
that might have gotten tagged on after the value for -j.)
Signed-off-by: Kyle Russell <bkylerussell@gmail.com>
Signed-off-by: Maxin B. John <maxin.john@intel.com>
|
|
|
|
|
|
|
|
|
| |
rhino's do_compile task passes
${STAGING_DATADIR_NATIVE}/classpath/glibj.zip to javac. This glibj.zip
archive comes from the classpath-native package.
Signed-off-by: Kyle Russell <bkylerussell@gmail.com>
Signed-off-by: Maxin B. John <maxin.john@intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
OpenJDK 7 is updated from from 91b01 (IcedTea 2.6.2) to 99b00 (IcedTea
2.6.5). This update introduces following security updates:
* Mar 2016 (IcedTea 2.6.5, OpenJDK 7 u99) [1]
* Jan 2016 (IcedTea 2.6.4, OpenJDK 7 u95) [2]
* Oct 2015 (IcedTea 2.6.3, OpenJDK 7 u91) [3]
Due to the fact some "byte behavior improvements" were not ported to
the ARM32-JIT in IcedTea 2.6.6 yet [6], [7] we need to stick IcedTea
2.6.5 for now.
[1]http://blog.fuseyism.com/index.php/2016/03/25/security-icedtea-2-6-5-for-openjdk-7-released/
[2]http://blog.fuseyism.com/index.php/2016/01/21/security-icedtea-2-6-4-for-openjdk-7-released/
[3]http://blog.fuseyism.com/index.php/2015/11/17/security-icedtea-2-6-3-for-openjdk-7-released/
[6]http://icedtea.classpath.org/bugzilla/show_bug.cgi?id=2942
[7]http://icedtea.classpath.org/hg/release/icedtea7-forest-2.6/hotspot/rev/1c5c4564204e
Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
Signed-off-by: Maxin B. John <maxin.john@intel.com>
|
|
|
|
|
|
|
|
|
| |
EXTRA_LDFLAGS were not used when building certain shared objects. This
means that the Yocto specific linker flags were lost, which caused build
failures due to a bad hash style QA error.
Signed-off-by: Erkka Kääriä <erkka.kaaria@intel.com>
Signed-off-by: Maxin B. John <maxin.john@intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
readdir_r has been deprecated since glibc 2.24. As openjdk uses this
function, a warning is generated during build that gets turned into
an error when -Werror is used.
While the readdir function is suggested as an replacement, it has
slightly different API. The warning is ignored instead to prevent any
unexpected breakage from patching the offending functions.
Signed-off-by: Erkka Kääriä <erkka.kaaria@intel.com>
Signed-off-by: Maxin B. John <maxin.john@intel.com>
|
|
|
|
|
|
|
|
|
|
| |
Openjdk is upgraded from u72 to u102. Two patches are dropped, as these
were backports and the fixes are present in the current version. One
patch was refreshed. One patch was backported from the upcoming u112 to
fix a zero mode crash.
Signed-off-by: Erkka Kääriä <erkka.kaaria@intel.com>
Signed-off-by: Maxin B. John <maxin.john@intel.com>
|
|
|
|
|
|
|
|
|
| |
Pass "--enable-downloading=no" to IcedTea's make. This ensures
IcedTea uses the versions and checksums we defined in the openjdk-7-jre
recipe for its "sub-modules".
Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
Signed-off-by: Maxin B. John <maxin.john@intel.com>
|
|
|
|
|
|
|
|
| |
Javasqlite is a Java wrapper including a basic JDBC driver
for SQLite 2/3.
Signed-off-by: Feng Mu <Feng.Mu@windriver.com>
Signed-off-by: Maxin B. John <maxin.john@intel.com>
|
|
|
|
|
|
|
|
|
| |
During my recipe parsing, GCCVERSION in the cross case seems to
be 'l' this leads to a ValueError during recipe parsing. Therefore
check type before trying to parse the version flag.
Signed-off-by: Sven Ebenfeld <sven.ebenfeld@vaillant-group.com>
Signed-off-by: Maxin B. John <maxin.john@intel.com>
|
|
|
|
|
|
|
|
|
|
| |
bitbake rev 67a7b8b02 "build: don't use $B as the default cwd for
functions" (included in current bitbake master) breaks the assumption
that do_unpackpost runs inside the build directory. Now that has to be
set explicitly, which is also okay for older bitbake versions.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When GCC is at version 4 or 5, parsing the recipe fails with:
ERROR: ExpansionError during parsing
.../ostro-os/meta-java/recipes-core/openjdk/openjdk-8_72b05.bb
...
bb.data_smart.ExpansionError: Failure expanding variable CFLAGS,
expression was -O2 -pipe -g -feliminate-unused-debug-types
-fdebug-prefix-map=.../ostro-os/build/tmp-glibc/work/corei7-64-ostro-
linux/openjdk-8/72b05-r0=/usr/src/debug/openjdk-8/72b05-r0
-fdebug-prefix-map=.../ostro-os/build/tmp-glibc/sysroots/x86_64-linux=
-fdebug-prefix-map=.../ostro-os/build/tmp-glibc/sysroots/intel-corei7-64=
-fstack-protector-strong -D_FORTIFY_SOURCE=2
${ <at> version_specific_cflags(d)} -Wno-error=deprecated-declarations which
triggered exception TypeError: can only join an iterable
That's because FLAGS_GCC<version> may be unset, thus leading to
d.getVar() returning None and ''.join(extraflags) failing.
The join() is also redundant: extraflags already is a string. It
happened to work because Python treats a string as sequence of
single-character strings, and thus ''.join() re-created the original
string.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Some supported hosts still use GCC 4.X. These don't support the flags
needed to make GCC 6 work, so check the GCC version and add appropriate
compiler flags.
This implementation will append flags for any gcc version, but it's only
used for GCC 6 right now.
Signed-off-by: Daniel McGregor <daniel.mcgregor@vecima.com>
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
|
|
|
|
|
|
|
|
|
| |
Previously if PARALLEL_MAKE changed, so did the task signature
for do_configure. Use vardepsvalue so that the signature remains
constant.
Signed-off-by: Daniel McGregor <daniel.mcgregor@vecima.com>
Signed-off-by: Maxin B. John <maxin.john@intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
GCC 6 sets the default C++ standard to C++14 and introduces dead store
elimination by default. OpenJDK 8 is not ready for either of these
changes, so set the C++ standard back to gnu++98 and disable dead
store elimination.
Switched to using --with-extra-cflags, cxxflags, and ldflags. The added
patch fixes building when using those flags, and are needed to get
CFLAGS into the JDK build in the native case.
Signed-off-by: Daniel McGregor <daniel.mcgregor@vecima.com>
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
|
|
|
|
|
|
|
|
|
|
|
| |
GCC 6 sets the default C++ standard to C++14 and introduces dead store
elimination by default. icedtea7 is not ready for either of these
changes, so just set the C++ standard back to gnu++98 and disable dead
store elimination.
Signed-off-by: Daniel McGregor <daniel.mcgregor@vecima.com>
Signed-off-by: Dan McGregor <dan.mcgregor@usask.ca>
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
|
|
|
|
|
|
|
|
|
|
| |
The OpenJDK builds did not include the update version and build number
leading to problems when applications check for the minimum needed version.
This commit fixes it by adding the update version and build number to the
configure parameters.
Signed-off-by: Bernhard Dick <bernhard@bdick.de>
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
|
|
|
|
|
|
|
| |
Replace all instances of base_contains with bb.utils.contains to avoid
warnings.
Signed-off-by: Maxin B. John <maxin.john@intel.com>
|
|
|
|
|
|
|
| |
Two patches were only applied to openjdk-8. Apply these to jre as well
Signed-off-by: Erkka Kääriä <erkka.kaaria@intel.com>
Signed-off-by: Maxin B. John <maxin.john@intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The security flag '-D_FORTIFY_SOURCE' requires at least -O to work,
otherwise a warning is given. If CFLAGS additionally contains -Werror,
this warning turns into an error. As Openjdk build system intentionally
deoptimizes certains files due to potential bad codegen during
optimization,
build will fail when both '-D_FORTIFY_SOURCE' and '-Werror' are used.
As turning the optimizations back on will likely break things, the
warning is silenced instead.
Signed-off-by: Erkka Kääriä <erkka.kaaria@intel.com>
Signed-off-by: Maxin B. John <maxin.john@intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Various paths to files are hardcoded in either helper scripts or the
resulting binaries (like path to Java class libraries). While this works
well when using the files on the build machine, things start breaking
down when the sstate gets shared between multiple workers in CI, as
the paths between workers can differ.
Instead of using these hardcoded paths, form the paths at runtime and pass them as parameters when required.
Signed-off-by: Erkka Kääriä <erkka.kaaria@intel.com>
Signed-off-by: Maxin B. John <maxin.john@intel.com>
|
|
|
|
|
|
|
|
|
|
| |
libfreetype.so.6 is copied from STAGING_LIBDIR and
binaries/libraries in sysroot are already stripped
since the following commit in oe-core:
"30f3774 staging: Strip files in sysroot"
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
Signed-off-by: Maxin B. John <maxin.john@intel.com>
|
|
|
|
|
|
|
|
|
| |
Prevent a non-deterministic run time dependency between openjdk-8
and openjre-8. Without PRIVATE_LIBS, if openjre is built before
openjdk, openjdk might RDEPEND on openjre.
Signed-off-by: Daniel McGregor <daniel.mcgregor@vecima.com>
Signed-off-by: Maxin B. John <maxin.john@intel.com>
|
|
|
|
|
|
|
|
|
| |
Repacking the runtime jar files takes a long time, and is only done
to save some space. Make it optional to speed up builds that don't
need it.
Signed-off-by: Daniel McGregor <daniel.mcgregor@vecima.com>
Signed-off-by: Maxin B. John <maxin.john@intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Backport patch of jaxb from
https://java.net/projects/jaxb/sources/v2/revision/1864898
to fix NullPointerException.
See https://java.net/jira/browse/JAXB-860 as well.
Signed-off-by: Kai Kang <kai.kang@windriver.com>
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Adlc is a native tool that openjdk builds and uses during its build process.
Bitbake however passes target machine specific CFLAGS and CXXFLAGS because we
are crosscompiling openjdk for the target architecture. This can cause issues,
if these flags contains values, that work for the Yocto provided crosscompiler
but not for the system compiler.
As an example, compilation will fail on Ubuntu 14.04 if
-fstack-protector-strong is specified in the distro security flags. Ubuntu
14.04 ships with GCC 4.8, whereas this flag is only supported by GCC 4.9+.
Signed-off-by: Erkka Kääriä <erkka.kaaria@intel.com>
Signed-off-by: Maxin B. John <maxin.john@intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Builds currently fail, if CXX also contains assembler arguments. These
arguments are in form "-Wa,args,are,here". When this string is then used in
openjdk make process, it is inserted before the make evaluates its argument
lists. As these lists are comma separated, the commas in assembler argument list
are interpreted as list separators, which breaks the build.
This is fixed by adding the extra '$', which means the string is inserted after
make evaluates the list.
Signed-off-by: Erkka Kääriä <erkka.kaaria@intel.com>
Signed-off-by: Maxin B. John <maxin.john@intel.com>
|
|
|
|
|
|
|
|
| |
There is no 'Classpath' license, set it as
the same with classpath-native.
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
Signed-off-by: Maxin B. John <maxin.john@intel.com>
|
|
|
|
|
|
|
|
|
| |
The license statement is in the middle of source file
Main.java, it's not correct to referrence the COREBASE
MIT license in LIC_FILES_CHKSUM.
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
Signed-off-by: Maxin B. John <maxin.john@intel.com>
|
|
|
|
|
|
|
|
|
|
| |
license.amd(AMD): applies to the TJC compiler source and test files
license.itcl(CDS): applies to the Itcl source and test files
license.ucb(UCB): applies to portions of Jacl and Tcl Blend sources
license.terms(SUN): applies to the other cource files
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
Signed-off-by: Maxin B. John <maxin.john@intel.com>
|
|
|
|
|
|
|
|
| |
The license is: IBM Public License Version 1.0
and it is named IPL-1.0 instead of IBM in common-licenses.
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
Signed-off-by: Maxin B. John <maxin.john@intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
These configure options are invalid:
--with-classpath-libdir
--with-classpath-classes
--with-classpath-includedir
They had been replaced by:
--with-java-runtime-library-libdir
--with-java-runtime-library-classes
--with-jni_h and --with-jni_md_h
Remove them to avoid QA Issue:
WARNING: QA Issue: cacao-native: configure was passed unrecognised
options: --with-classpath-libdir --with-classpath-includedir
--with-classpath-classes [unknown-configure-option]
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
Signed-off-by: Maxin B. John <maxin.john@intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
After updated to version 0.99, the following options
became invalid:
* --with-ecj: use JAVAC env variable instead
* --with-fastjar: change to --with-jar
* --with-vm: uae JAVA env variable instead
* --disable-qt4-peer: invalid even in 0.93, just drop it.
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
|
|
|
|
|
|
|
|
|
| |
* ${PN}-jdk doesn't exist at all so the alternative
for javac is invalid.
* remove the useless commented lines as well.
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The error in do_configure is like:
openjdk.tar.bz2: FAILED
/usr/bin/sha256sum: WARNING: 1 of 1 computed checksum did NOT match
That's because the URIs and checksums do not match the ones defined
in Makefile.am, then the configure will fail the test and download
again from ICEDTEA_DROP_URL.
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
|
|
|
|
|
|
|
|
| |
Backport from upstream from jdk9
Hotspot build system looking for sdt.h in the wrong place
Signed-off-by: Amy Fong <amy.fong@windriver.com>
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Fails with an error message /usr/bin/unzip not found.
Additionally, icedtea7 enables you to specify an unzip to use by setting
the variable UNZIP, this results in an error because unzip uses UNZIP
to specify options, hence you'll end up with unzip trying to unzip unzip.
Note: Same issue applies to openjdk
Signed-off-by: Amy Fong <amy.fong@windriver.com>
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
openjdk depends on attr, this can result in a configure
failure if the dependency is missing.
checking for sys/epoll.h... (cached) yes
checking attr/xattr.h usability... no
checking attr/xattr.h presence... no
checking for attr/xattr.h... no
configure: error: Could not find required system headers;
install the appropriate files from glibc-headers,
libc6-dev and/or libattr-devel, libattr1-dev or use
--disable-compile-against-syscalls if necessary.
Configure failed. The contents of all config.log files follows to aid debugging
Signed-off-by: Amy Fong <amy.fong@windriver.com>
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
|
|
|
|
|
| |
Signed-off-by: Amy Fong <amy.fong@windriver.com>
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
|
|
|
|
|
|
|
|
|
|
| |
Rather than creating /tmp/unpack.log and insecure permissions,
if unpack cannot create teh specified log file, it defaults to writing
to /dev/null, failing that, stderr. (These are the default options if
it cannot write to /tmp/unpack.log)
Signed-off-by: Amy Fong <amy.fong@windriver.com>
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
On hosts without /usr/include/X11/extensions, icedtea7-native compile
results in a build failure while trying to create an object file:
+/data/home/.../git/csu/../sysdeps/x86_64/start.S:118: undefined reference to `main'
[2015-03-31 15:22:53.115930161+00:00] | collect2: error: ld returned 1 exit status
The compile line shows: (-c gets eaten)
... -I -c -o
The makefile tries to do the following which results in a faulty build
line if /usr/include/X11/extensions doesn't exist:
CPPFLAGS += -I$(firstword $(wildcard $(OPENWIN_HOME)/include/X11/extensions) \
$(wildcard /usr/include/X11/extensions))
This looks like a missed line from the following patch which removes
the need for X11/extensions:
icedtea-ecj-disable-compilation.patch
Signed-off-by: Amy Fong <amy.fong@windriver.com>
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
|
|
|
|
|
|
|
|
| |
Some of the postinst rules results in a build error because "ERR" is in
the first line.
Signed-off-by: Amy Fong <amy.fong@windriver.com>
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
|
|
|
|
|
|
|
| |
Add alternative link keytool
Signed-off-by: Amy Fong <amy.fong@windriver.com>
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
|
|
|
|
|
|
|
|
|
| |
The jdk package was missing tools like javac and tools.jar. Adding
jvm to the packaging pulls in the tool set.
Signed-off-by: Paul Barrette <paul.barrette@windriver.com>
Signed-off-by: Amy Fong <amy.fong@windriver.com>
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
icedtea7-native, openjdk-7, and jamvm-native requires pkg-config to
configure without running into an error similar to
7554: syntax error near unexpected token `NSS,'
7554: `PKG_CHECK_MODULES(NSS, nss, NSS_FOUND=yes, NSS_FOUND=no)'
classpath-native and jamvm-native both needs to depend on
java-initial-native
Signed-off-by: Amy Fong <amy.fong@windriver.com>
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
|
|
|
|
|
|
|
|
| |
In the default /etc/profile, search for and define JAVA_HOME
in the environment variables.
Signed-off-by: Amy Fong <amy.fong@windriver.com>
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
jamvm.inc's SRC_URI isn't used anymore, all jamvm bb recipes have been
defining their own.
In a rare instance, a build failed because it was using the SRC_URI
defined from jamvm.inc instead (jamvm_git.bb), this cleanup should
prevent this oddity from occuring.
Signed-off-by: Amy Fong <amy.fong@windriver.com>
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
${B} is set before autotools and thus is overrided
by autotools. Some of the patches within icedtea are
expected to be patched against ${S}/build (during
do_configure() phase). As autotools to change B
= "${WORKDIR}/build", so this causes some patches failed
to apply and hence breaks do_configure() with the following
errors:
| patching file openjdk/jdk/make/tools/src/build/tools/generatecurrencydata/GenerateCurrencyData.java
| can't find file to patch at input line 13
| Perhaps you used the wrong -p or --strip option?
| The text leading up to this was:
| --------------------------
| |---
| | build/openjdk/corba/make/common/shared/Defs-utils.gmk | 6 +++---
| | build/openjdk/jdk/make/common/Demo.gmk | 2 +-
| | build/openjdk/jdk/make/common/Release.gmk | 2 +-
| | build/openjdk/jdk/make/common/internal/ImportComponents.gmk | 4 ++--
| | build/openjdk/jdk/make/common/shared/Defs-utils.gmk | 6 +++---
| | build/openjdk/jdk/make/common/shared/Sanity.gmk | 2 +-
| | build/openjdk/jdk/make/java/redist/Makefile | 4 ++--
| | 7 files changed, 13 insertions(+), 13 deletions(-)
| |
| |--- a/build/openjdk/corba/make/common/shared/Defs-utils.gmk
| |+++ b/build/openjdk/corba/make/common/shared/Defs-utils.gmk
| --------------------------
| File to patch:
| Skip this patch? [y]
| File to patch:
| Skip this patch? [y]
| Skipping patch.
Signed-off-by: Vu Tran <vu.tran@windriver.com>
Signed-off-by: Amy Fong <amy.fong@windriver.com>
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Fix a dependency issue when building many parallel threads for
inetlib. The build failure looks like:
1. ERROR in source/gnu/inet/util/SaslCallbackHandler.java (at line 43)
import javax.security.auth.callback.Callback;
^^^^^^^^^^^^^^
The issue is due to the jar linking before all the objects are
complete.
Signed-off-by: Paul Barrette <paul.barrette@windriver.com>
Signed-off-by: Amy Fong <amy.fong@windriver.com>
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
We had:
virtual/jar-native provided by fastjar-native
virtual/java-native provided by cacao-native and jamvm-native
virtual/javac-native provided by ecj-bootstrap-native
Then openjdk-8-native added and also provides:
virtual/java-native and virtual/javac-native
(it also installs ${STAGING_BINDIR_NATIVE}/jar but
doesn't add virtual/jar-native in the PROVIDES)
which introduces circular dependencies,
java-bootstrap-components.bbclass is to avoid the
circular dependencies but it's not a correct way
since it make two conflicted packages build into
the same sysroot:
DEPENDS_prepend_class-native = " ecj-bootstrap-native "
DEPENDS_prepend_class-target = " virtual/javac-native "
virtual/javac-native is provided by openjdk-8-native,
and it assumed that ecj-bootstrap-native, fastjar-native and
cacao-native/jamvm-native are always built before openjdk-8-native,
then in the do_install of penjdk-8-native, it removes them from
staging dirs first, then install (create symlinks for) these files
in staging dirs directly, which results with empty sstate and
builds from sstate cache or builds not in this assumption
will fail.
The fixes here includes:
- Revert "meta-java: rely on well known bootstrap-path" in
previous commit to remove the improper
java-bootstrap-components.bbclass
- Do not let openjdk-8-native provides jar, java and javac,
just like what we do in icedtea7-native, which can provides
but not to avoid circular dependencies and conflicts.
- make cacao-native back to provide virtual/java-native,
and ecj-bootstrap-native back to provide virtual/javac-native
- Add back provider settings in README.
- Remove the update-alternatives in openjdk-8-native,
it doesn't work for native package.
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
|