aboutsummaryrefslogtreecommitdiffstats
AgeCommit message (Collapse)AuthorFilesLines
2020-11-09openjdk-8: fix changed upstream checksumsHEADmaster-nextmastergatesgarthdunfellRichard Leitner1-8/+8
The upstream archives and therefore their checksums for 8u272 have changed. This is caused by a new "jdk8u275-b00" tag which was added on the same revision as our "jdk8u272-ga" tag. Therefore the .hg_archival.txt file changed. This patch adds the new upstream checksums. Reported-by: Chencheng Zhang <chencheng@wittra.se> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-11-03layer.conf: Mark as compatible with gatesgarthGeoff Parker1-1/+1
Signed-off-by: Geoff Parker <geoffrey.parker@arthrex.com> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-11-03icedtea-7: add fix for deprecated sys/sysctl.h includeRichard Leitner2-0/+24
As glibc deprecated sys/sysctl.h we use linux/sysctl.h for now. Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-10-22openjdk-8: update to latest ga version 272Richard Leitner42-124/+167
For aarch32 we need to fall back to b09 as there is no ga tag available yet. Patch related changes: * 0011, 0012 & 0013 were changed to valid git patches Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-10-22openjdk-8-native: Don't warn for stripped libfreetype.so.6Jonatan Pålsson1-0/+3
Mimic the approach used in openjdk-8-cross.inc and add INSANE_SKIP for "already-skipped" to avoid the following warning: WARNING: openjdk-8-native-252-r0 do_populate_sysroot: File '..snip../libfreetype.so.6' from openjdk-8-native was already stripped, this will prevent future debugging! Signed-off-by: Jonatan Pålsson <jonatan.p@gmail.com> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-10-22openjdk-8-common: fix soundless evaluationRichard Leitner1-4/+1
If only alsa and not pulseaudio was in PACKAGECONFIG soundless was configured. This is incorrect for obvious reasons, therefore set soundless only if alsa and pulseaudio isn't set. Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-10-22openjre-8: add ALTERNATIVE for "keytool"Richard Leitner1-2/+6
As we have ALTERNATIVEs for java and javac we should also provide one for the keytool. Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-10-22openjdk-8: add ALTERNATIVE for "keytool"Richard Leitner1-1/+4
As we have ALTERNATIVEs for java and javac we should also provide one for the keytool. Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-08-24openjdk-8: update to latest ga version 265Richard Leitner48-451/+268
Patch related changes: * The hotspot patch 1004 was mainlined in changeset 3a3803a0c789 [1] and is therfore dropped. * The jdk patch 2010 was mainlined in changeset c4418d567028 [2] and is therefore dropped. * Rename hotspot/aarch64 patches to start with number 1401. * Rename hotspot/aarch32 patches to start with number 1201. * Merge aarch32-hotspot-fix-shark-build-pt2.patch and openjdk8-fix-shark-build-pt2.patch to hotspot patch 1013 as they were the same. [1] https://hg.openjdk.java.net/jdk8u/jdk8u/hotspot/rev/3a3803a0c789 [2] https://hg.openjdk.java.net/jdk8u/jdk8u/jdk/rev/c4418d567028 Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-08-24cacao: update to latest git versionRichard Leitner3-25/+41
As the cacao-staging mercurial repository was removed update the recipe to use the git repository instead. Furthermore update to the latest revision. Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-08-20openjdk-8: add a glibc 2.32 compatibility patchAlexander Kanavin2-0/+40
Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-08-20openjdk-8: refresh patchesRichard Leitner15-57/+76
Refresh all openjdk8 patches as a preparation for the upcoming update. Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-07-02icedtea7-native: set ALT_CUPS_HEADERS_PATH to stagingJeremy Puhlman1-0/+1
ALT_CUPS_HEADERS_PATH eventually gets added to some links as -I/usr/include If you are using the buildtools-extended-tarball, the limits.h from the system gets included first then later the limits.h from the buildtools tarball gets added again, but is missing glibc definitions, if the system glibc is much older(centos7). This leads to: In file included from /usr/include/limits.h:124, from /usr/include/sys/param.h:26, from ../../../src/solaris/javavm/export/jvm_md.h:35, from ../../../src/share/javavm/export/jvm.h:32, from ../../../src/solaris/native/sun/awt/awt.h:33, from ../../../src/solaris/native/sun/awt/awt_p.h:75, from ../../../src/solaris/native/sun/awt/awt_MToolkit.c:26: /build/buildtools/sysroots/x86_64-montavistasdk-linux/usr/include/limits.h:145:17: error: missing binary operator before token "(" 145 | #if __GLIBC_USE (IEC_60559_BFP_EXT_C2X) Signed-off-by: Jeremy Puhlman <jpuhlman@mvista.com> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-07-02icedtea7-native: set unique download file namesVyacheslav Yurkov2-41/+46
Add package names to download file names. This allows to avoid name collisions in case PREMIRRORS are used and all source archives reside in the same directory. Fixed mixed indentation. Signed-off-by: Vyacheslav Yurkov <Vyacheslav.Yurkov@bruker.com> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-07-02jlex: set unique file name for downloadVyacheslav Yurkov1-2/+5
Signed-off-by: Vyacheslav Yurkov <Vyacheslav.Yurkov@bruker.com> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-06-10openjdk-8-native: Build cups-native when using cups packageconfigFabio Berton1-1/+1
Signed-off-by: Fabio Berton <fabio.berton@ossystems.com.br> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-06-10openjdk-8-native: Add libxtst-native to x11 packageconfigFabio Berton1-1/+1
libxtst-native it's needed to build openjdk-8-native. Signed-off-by: Fabio Berton <fabio.berton@ossystems.com.br> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-06-10WORKDIR and ARCHIVER_WORKDIR supportRobert Berger1-2/+17
do_unpack_extract_submodules was called via postfuncs[do_unpack]. This breaks the build when the archiver.bbclass is used in a way where do_unpack_and_patch is called by it. In this mode the archiver class modifies WORKDIR, S, B,... and those modifications are lost via postfuncs. In order to pick up WORKDIR, S, B,... (potentially modified by archiver.bbclass) do_patch_prepend is used instead of pre/postfuncs and calls do_unpack_extract_submodules now. Without this patch the build will break when you add this to local.conf: INHERIT += "archiver" ARCHIVER_MODE[src] = "original" ARCHIVER_MODE[diff] = "1" ARCHIVER_MODE[dumpdata] = "1" ARCHIVER_MODE[recipe] = "1" COPYLEFT_LICENSE_INCLUDE = "GPL* LGPL*" Signed-off-by: Robert Berger <robert.berger@ReliableEmbeddedSystems.com> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-05-09icedtea7-native: host gcc-9, extra CFLAGS neededRobert Berger1-0/+5
Signed-off-by: Robert Berger <robert.berger@ReliableEmbeddedSystems.com> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-05-06openjdk-8: remove -WerrorRichard Leitner4-13/+227
Add patches to remove Werror from the build process. Therefore we also drop the "-Wno-*" exceptions. Signed-off-by: Richard Leitner <richard.leitner@skidata.com> Tested-by: Jacob Kroon <jacob.kroon@gmail.com>
2020-05-06openjdk-8: require GCC >= v6Richard Leitner2-15/+9
As nobody should use older versions of GCC anymore this patch drops all version dependent workarounds and requiers host gcc >=6 for building openjdk-8. Signed-off-by: Richard Leitner <richard.leitner@skidata.com> Tested-by: Jacob Kroon <jacob.kroon@gmail.com>
2020-05-01openjdk-8: Fix building with GCC 10Jacob Kroon1-0/+4
Signed-off-by: Jacob Kroon <jacob.kroon@gmail.com> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-05-01icedtea7-native: Fix building with GCC 10Jacob Kroon1-0/+3
Signed-off-by: Jacob Kroon <jacob.kroon@gmail.com> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-05-01openjdk-build-helper: Fix decoding GCC major versionJacob Kroon1-1/+1
GCC in Fedora 32 is up to version 10, so return the whole word instead of the first letter. Signed-off-by: Jacob Kroon <jacob.kroon@gmail.com> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-05-01openjdk-8: update to latest ga version 252Richard Leitner6-50/+26
This patch also drops the md5 checksums as sha256 checksums are sufficient. Signed-off-by: Richard Leitner <richard.leitner@skidata.com> Tested-by: Jacob Kroon <jacob.kroon@gmail.com>
2020-04-28docs: UPDATING: pipe curl calls to "head"Richard Leitner1-2/+2
Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-04-09layer.conf: Mark as compatible with dunfellJacob Kroon1-1/+1
Signed-off-by: Jacob Kroon <jacob.kroon@gmail.com> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-02-10openjdk-8: Don't set TOPDIR bitbake variableJacob Kroon1-1/+4
It is used to set the default sstate cache location. Just export it to the environment. Signed-off-by: Jacob Kroon <jacob.kroon@gmail.com> [richard.leitner@skidata.com: adapted commit message] Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-01-28openjdk-8: update to latest ga version 242Richard Leitner70-4018/+1390
As OpenJDK-8 is now tagging "ga" versions in addition to the "build" version the recipes are adapted to use those "ga" versions. All existing patches got re-applied and renamed. For better handling Hotspot patches now start at patch number 1001 and jdk patches at 2001. Furthermore architecture dependent patches are prefixed with the architecture they apply on. Following patches/hunks were completely dropped: - 0002-hotspot-fix-compilation-with-security-flags-enabled.patch got backported to hotspot jdk8u https://hg.openjdk.java.net/jdk8u/jdk8u/hotspot/rev/c40a28e54185 - 0011-hotspot-backport-patch-to-fix-misuses-of-strncpy-str.patch ostream.cpp:112 got fixed in hotspot jdk8u https://hg.openjdk.java.net/jdk8u/jdk8u/hotspot/rev/f3108e56b502 - 0014-hotspot-zero-fix-undefined-behaviour-gcc-v8-fix.patch got backported to hotspot jdk8u https://hg.openjdk.java.net/jdk8u/jdk8u/hotspot/rev/ca4663e58916 - 0018-hotspot-Fix-debug-build-after-8062808-Turn-on-the-Wr.patch fixed in hotspot jdk8u https://hg.openjdk.java.net/jdk8u/jdk8u/hotspot/rev/32bc598624bd - 0020-Enable-HotSpot-builds-on-5.x-Linux-kernels.patch fixed in hotspot jdk8u https://hg.openjdk.java.net/jdk8u/jdk8u/hotspot/rev/5af8ec63c21c - openjdk8-add-missing-linker-flags.patch fixed in hotspot jdk8u https://hg.openjdk.java.net/jdk8u/jdk8u/hotspot/rev/f175513c2c3a - openjdk8-fix-shark-stdc++11.patch fixed in hotspot jdk8u - openjdk8-fix-libpng-neon-build.patch fixed in jdk https://hg.openjdk.java.net/jdk8u/jdk8u/jdk/rev/5d57817931e1 - aarch64-hotspot-fix-undefined-behaviour-gcc-v8-fix.patch got backported to hotspot jdk8u Following patches were newly added: - 0011-autoconf-fix-CC-with-arguments-detection.patch needed because of jdk8u commit "8038340: Cleanup and fix sysroot and devkit handling on Linux and Solaris" - 0012-autoconf-NativeCompilation-remove-sysroot.patch needed because of jdk8u commit "8038340: Cleanup and fix sysroot and devkit handling on Linux and Solaris" Additionally add UPDATING.md which describes the openjdk8 update process for this layer and update8checksums.sh, a script that updates the openjdk8 source archive checksums in the corresponding .inc files. Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-01-27icedtea7-native: inherit python nativeKraag Gorim1-0/+1
icedtea7-native relys on python to enable building it. Therefor it should inherit python instead of assuming python being part of HOSTTOOLS. Signed-off-by: Kraag Gorim <kraaggorim@gmail.com> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-01-27java-library: inherit python nativeKraag Gorim1-0/+1
Several components make use of python in their build process. To not rely on python being part of the HOSTTOOLS python should be inherited. Signed-off-by: Kraag Gorim <kraaggorim@gmail.com> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-01-21openjdk-7: add patch to fix xattr include pathRichard Leitner3-0/+25
As attr dropped the <attr/xattr.h> header use <sys/xattr.h> instead. http://git.savannah.nongnu.org/cgit/attr.git/commit/include?id=7921157890d07858d092f4003ca4c6bae9fd2c38 Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-01-21icedtea7-native: add patch to fix xattr include pathRichard Leitner3-0/+25
As attr dropped the <attr/xattr.h> header use <sys/xattr.h> instead. http://git.savannah.nongnu.org/cgit/attr.git/commit/include?id=7921157890d07858d092f4003ca4c6bae9fd2c38 Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-01-21layer.conf: add LAYERDEPENDS for openembedded-layer aka meta-oeRichard Leitner1-0/+1
Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-01-21README: mention dependency on meta-poky for oeqa image testingRichard Leitner1-0/+9
As the mejate distributions are based on poky we need meta-poky for oeqa image testing. Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-01-21java.bbclass: remove bashism in comparisonRichard Leitner1-1/+1
Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2020-01-21java.bbclass: oe_makeclasspath does nothing without -sDavis, Michael1-5/+5
oe_makeclasspath is supposed to output for the target platform by default and the staging platform with -s. However it checks if the file exists even if -s is not defined which is always false. Additionaly remove comment around an error that never occurs. Signed-off-by: Michael Davis <michael.davis@essvote.com> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2019-12-27icedtea7-native: set default x11 PACKAGECONFIG from DISTRO_FEATURESKyle Russell1-1/+1
If x11 is specified in DISTRO_FEATURES, then make it a default PACKAGECONFIG option when building icedtea7-native; otherwise, the libx11-native dependency won't be satisfied. The recipe reads as though icedtea7-native can build completely without X11, but if the libx11-native headers are not installed on the host system, compile errors in multiple units because of a missing X11/Xlib.h. In file included from ../../../src/share/native/sun/awt/../java2d/pipe/Region.h:34:0, from ../../../src/share/native/sun/awt/../java2d/loops/GraphicsPrimitiveMgr.c:31: ../../../src/solaris/native/sun/awt/utility/rect.h:31:22: fatal error: X11/Xlib.h: No such file or directory In this example, it appears that GraphicsPrimitiveMgr.c is supposed to be compiled even in headless mode, so the dependency still seems to be needed. Signed-off-by: Kyle Russell <bkylerussell@gmail.com> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2019-12-27icedtea7-native: use openjdk build helper functionsRichard Leitner1-12/+2
Replace target arch selection switch with functions from the openjdk build helper. Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2019-12-27jamvm: Disable inliningMarek Belisko1-1/+1
This will fix crashes seen when build openjdk-8: ERROR: jaxp1.3-native-1.4.01-r0 do_compile: Function failed: do_compile (log file is located at /home/jenkins/my_build/tmp/work/x86_64-linux/jaxp1.3-native/1.4.01-r0/temp/log.do_compile.28894) ERROR: Logfile of failure stored in: /home/jenkins/my_build/tmp/work/x86_64-linux/jaxp1.3-native/1.4.01-r0/temp/log.do_compile.28894 Log data follows: | DEBUG: Executing shell function do_compile | Illegal instruction (core dumped) | WARNING: exit code 132 from a shell command. | ERROR: Function failed: do_compile (log file is located at /home/jenkins/my_build/tmp/work/x86_64-linux/jaxp1.3-native/1.4.01-r0/temp/log.do_compile.28894) ERROR: Task (virtual:native:/home/jenkins/meta-java/recipes-core/xml-commons/jaxp1.3_1.4.01.bb:do_compile) failed with exit code '1' Signed-off-by: Marek Belisko <marek.belisko@open-nandra.com> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2019-12-27openjdk-8: fix a random crash on JNI_CreateJavaVMWenlin Kang2-0/+37
When call JNI_CreateJavaVM API, sometimes it will crash(we have reproduced it on quemuppc), this patch fixes it. Signed-off-by: Wenlin Kang <wenlin.kang@windriver.com> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2019-12-27remove True option to getVar callsAndré Draszik5-15/+15
getVar() has been defaulting to expanding by default for a long time (2016), thus remove the True option from getVar() calls with a regex search and replace. Search & replace made using the following command: sed -e 's|\(d\.getVar \?\)( \?\([^,()]*\), \?True)|\1(\2)|g' \ -i $(git grep -E 'getVar ?\( ?([^,()]*), ?True\)' \ | cut -d':' -f1 \ | sort -u) Signed-off-by: André Draszik <andre.draszik@jci.com> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2019-12-27class/java-native: removeMohamad Noor Alim Hussin3-11/+2
Java-native class used NATIVE_INSTALL_WORKS which was removed from oe-core around 2010. Also updated recipes that inherit java-native to native class. Signed-off-by: Mohamad Noor Alim Hussin <mohamad.noor.alim.hussin@intel.com> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2019-12-27cacao: Use python3 in java wrapper scriptJacob Kroon1-2/+2
OE-Core has dropped python2 so switch to using python3. Signed-off-by: Jacob Kroon <jacob.kroon@gmail.com> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2019-12-27icedtea7-native: add patch to fix friend declaration of prefix_relocInfoRichard Leitner2-0/+22
This fixes the following error: build/openjdk-boot/hotspot/src/share/vm/code/relocInfo.hpp:374:27: error: friend declaration of ‘relocInfo prefix_relocInfo(int)’ specifies default arguments and isn’t a definition [-fpermissive] Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2019-12-27recipes: rename distro_features_check to features_checkAlex Kiernan4-4/+4
Avoid warning due to the class rename in OE-Core. Signed-off-by: Alex Kiernan <alex.kiernan@gmail.com> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2019-11-11classpath: add patch to fix BigDecimal.stripTrailingZeros()'s handling of 0.Chris Laplante2-0/+52
Previously, 'new BigDecimal("0").stripTrailingZeros()' would blow up: Caused by: java.lang.StringIndexOutOfBoundsException: String index out of range: -1 at java.lang.String.charAt at java.math.BigDecimal.stripTrailingZeros Fixes https://sourceforge.net/p/saxon/mailman/message/27204592/ Signed-off-by: Chris Laplante <chris.laplante@agilent.com> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2019-11-11classpath: add patch to fix BigDecimal.compareTo()Chris Laplante2-5/+74
Prior to this patch compareTo couldn't handle operands with negative scales. It passes the following unit test from JDK8: http://hg.openjdk.java.net/jdk8/jdk8/jdk/file/tip/test/java/math/BigDecimal/CompareToTests.java Signed-off-by: Chris Laplante <chris.laplante@agilent.com> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2019-11-05layer.conf: remove compatibility with thud and warriorRichard Leitner1-1/+1
As we want to start some cleanup work on meta-java stay only compatible with zeus. Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
2019-11-05layer.conf: Mark as compatible with zeusAlistair Francis1-1/+1
Signed-off-by: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Richard Leitner <richard.leitner@skidata.com>