Re: [oe] [meta-java][PATCH] jsch: switch to inclusive language
On 11/25/22 13:35, Ulrich Ölmann wrote:
Signed-off-by: Ulrich Ölmann
---
recipes-core/jcraft/jsch_0.1.40.bb | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/recipes-core/jcraft/jsch_0.1.40.bb
b/recipes-core/jcraft/jsch_0.1.40.bb
index 07d5b9e3108e..259712661f4d 100644
--- a/recipes-core/jcraft/jsch_0.1.40.bb
+++ b/recipes-core/jcraft/jsch_0.1.40.bb
@@ -25,8 +25,8 @@ do_compile() {
SRC_URI[md5sum] = "b59cec19a487e95aed68378976b4b566"
SRC_URI[sha256sum] =
"ca9d2ae08fd7a8983fb00d04f0f0c216a985218a5eb364ff9bee73870f28e097"
-# Whitelist the CVE because it only affects Windows platforms
-CVE_CHECK_WHITELIST += "CVE-2016-5725"
+# Ignore the CVE because it only affects Windows platforms
+CVE_CHECK_IGNORE += "CVE-2016-5725"
BBCLASSEXTEND = "native"
Also your commit:
3161106f0879a6be0a016df3a8bb8fa800bd6044
openjdk-8: refresh patches
seems to have *introduced* fuzz warnings, rather than fix any on master
branch ...
Jacob
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#99731):
https://lists.openembedded.org/g/openembedded-devel/message/99731
Mute This Topic: https://lists.openembedded.org/mt/95253071/21656
Group Owner: openembedded-devel+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
Re: [oe] [OE-core] [meta-java] icedtea7-native build fails on kirkstone and in master
On 6/9/22 11:28, Bartosz Golaszewski wrote: Hi! I'm trying to build java support with the following configuration: Build Configuration: BB_VERSION = "2.0.0" BUILD_SYS= "x86_64-linux" NATIVELSBSTRING = "universal" TARGET_SYS = "x86_64-poky-linux" MACHINE = "qemux86-64" DISTRO = "poky" DISTRO_VERSION = "4.0.1" TUNE_FEATURES= "m64 core2" TARGET_FPU = "" meta meta-poky= "kirkstone:45d7615dfef8093a2467d52ca1130e62db6a1187" meta-oe = "kirkstone:fcc7d7eae82be4c180f2e8fa3db90a8ab3be07b7" meta-java= "kirkstone:1a8059f6b257ebe6fcae6416e499784d976afd24" --- PREFERRED_PROVIDER_virtual/java-initial-native = "cacao-initial-native" PREFERRED_PROVIDER_virtual/java-native = "jamvm-native" PREFERRED_PROVIDER_virtual/javac-native = "ecj-bootstrap-native" --- The build fails for icedtea7-native but it's hard to tell why exactly from the logs. Here is the full output: https://pastebin.com/3dRt6J0Y It seems to be failing here: | Checking for mapfile use in: /home/brgl/workspace/target-build-yocto/build/tmp/work/x86_64-linux/icedtea7-native/2.1.3-r1.0/icedtea-2.1.3/build/openjdk.build-boot/lib/amd64/libmanagement.so | Library loads for: /home/brgl/workspace/target-build-yocto/build/tmp/work/x86_64-linux/icedtea7-native/2.1.3-r1.0/icedtea-2.1.3/build/openjdk.build-boot/lib/amd64/libmanagement.so | make[5]: Leaving directory '/home/brgl/workspace/target-build-yocto/build/tmp/work/x86_64-linux/icedtea7-native/2.1.3-r1.0/icedtea-2.1.3/build/openjdk-boot/jdk/make/java/management' | make[5]: *** [../../common/Library.gmk:225: /home/brgl/workspace/target-build-yocto/build/tmp/work/x86_64-linux/icedtea7-native/2.1.3-r1.0/icedtea-2.1.3/build/openjdk.build-boot/lib/amd64/libmanagement.so] Error 1 | make[4]: Leaving directory '/home/brgl/workspace/target-build-yocto/build/tmp/work/x86_64-linux/icedtea7-native/2.1.3-r1.0/icedtea-2.1.3/build/openjdk-boot/jdk/make/java' | make[4]: *** [Makefile:63: all] Error 1 | make[3]: Leaving directory '/home/brgl/workspace/target-build-yocto/build/tmp/work/x86_64-linux/icedtea7-native/2.1.3-r1.0/icedtea-2.1.3/build/openjdk-boot/jdk/make' | make[3]: *** [Makefile:247: all] Error 1 | make[2]: Leaving directory '/home/brgl/workspace/target-build-yocto/build/tmp/work/x86_64-linux/icedtea7-native/2.1.3-r1.0/icedtea-2.1.3/build/openjdk-boot' | make[2]: *** [make/jdk-rules.gmk:79: jdk-build] Error 2 | make[1]: Leaving directory '/home/brgl/workspace/target-build-yocto/build/tmp/work/x86_64-linux/icedtea7-native/2.1.3-r1.0/icedtea-2.1.3/build/openjdk-boot' | make[1]: *** [Makefile:244: build_product_image] Error 2 | make: *** [Makefile:2252: stamps/icedtea-boot.stamp] Error 2 But not sure why. Thanks in advance for any help. Bart I gave up building icedtea7-native on my Fedora 36 system due to first: https://sourceware.org/bugzilla/show_bug.cgi?id=28868 That one is fixed now, but then I ran into: https://mail.openjdk.java.net/pipermail/distro-pkg-dev/2022-May/042638.html so for now I'm using podman for building in a Fedora 35 container image. Jacob -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#97464): https://lists.openembedded.org/g/openembedded-devel/message/97464 Mute This Topic: https://lists.openembedded.org/mt/91642102/21656 Group Owner: openembedded-devel+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
[oe] [meta-java] Merge master-next into master ?
Hi, Can we get master-next merged into master ? Thanks in advance, Jacob -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#95115): https://lists.openembedded.org/g/openembedded-devel/message/95115 Mute This Topic: https://lists.openembedded.org/mt/88716734/21656 Group Owner: openembedded-devel+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
Re: [oe] [meta-oe][PATCH] Revert "ccid: Fix floating runtime dependency on flex"
On 12/6/21 10:08, Jacob Kroon wrote:
> With the switch to recipe-specific sysroots this patch is no longer
> needed. Tested building with and without the patch and the resulting
> binaries are identical.
>
> This reverts commit 960314be5d854212ee65604a68a92a3616b2a544.
>
> Signed-off-by: Jacob Kroon
> ---
> .../ccid/ccid/no-dep-on-libfl.patch| 18 --
> meta-oe/recipes-support/ccid/ccid_1.4.33.bb| 1 -
> 2 files changed, 19 deletions(-)
> delete mode 100644 meta-oe/recipes-support/ccid/ccid/no-dep-on-libfl.patch
>
> diff --git a/meta-oe/recipes-support/ccid/ccid/no-dep-on-libfl.patch
> b/meta-oe/recipes-support/ccid/ccid/no-dep-on-libfl.patch
> deleted file mode 100644
> index 477cc31ad..0
> --- a/meta-oe/recipes-support/ccid/ccid/no-dep-on-libfl.patch
> +++ /dev/null
> @@ -1,18 +0,0 @@
> -No need to link with libfl.
> -
> -Upstream-Status: Pending
> -Signed-off-by: Jacob Kroon
> -
> -Index: ccid-1.4.8/src/Makefile.am
> -===
> ccid-1.4.8.orig/src/Makefile.am
> -+++ ccid-1.4.8/src/Makefile.am
> -@@ -53,7 +53,7 @@ PROVIDED_BY_PCSC = debug.c
> - endif
> -
> - libccid_la_SOURCES = $(COMMON) $(USB) $(TOKEN_PARSER) $(PROVIDED_BY_PCSC)
> $(T1)
> --libccid_la_LIBADD = $(LEXLIB) $(LIBUSB_LIBS) $(PTHREAD_LIBS)
> -+libccid_la_LIBADD = $(LIBUSB_LIBS) $(PTHREAD_LIBS)
> - libccid_la_CFLAGS = $(PCSC_CFLAGS) $(LIBUSB_CFLAGS) $(PTHREAD_CFLAGS) \
> - $(SYMBOL_VISIBILITY) -D$(CCID_VERSION) -DSIMCLIST_NO_DUMPRESTORE
> - libccid_la_LDFLAGS = -avoid-version
> diff --git a/meta-oe/recipes-support/ccid/ccid_1.4.33.bb
> b/meta-oe/recipes-support/ccid/ccid_1.4.33.bb
> index 4b1c42173..e3cb7cef1 100644
> --- a/meta-oe/recipes-support/ccid/ccid_1.4.33.bb
> +++ b/meta-oe/recipes-support/ccid/ccid_1.4.33.bb
> @@ -7,7 +7,6 @@ DEPENDS = "virtual/libusb0 pcsc-lite"
> RDEPENDS:${PN} = "pcsc-lite"
>
> SRC_URI = "https://ccid.apdu.fr/files/ccid-${PV}.tar.bz2 \
> -file://no-dep-on-libfl.patch \
> file://0001-Add-build-rule-for-README.patch \
> "
>
>
ping
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#94312):
https://lists.openembedded.org/g/openembedded-devel/message/94312
Mute This Topic: https://lists.openembedded.org/mt/87536557/21656
Group Owner: openembedded-devel+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
Re: [oe] [meta-oe][PATCH] freerdp: Use DEPENDS:append
On 12/7/21 12:39, Marek Vasut wrote: > The build system might have put something into DEPENDS already, > adhere to the recommended best practice and use DEPENDS:append > to avoid overriding the DEPENDS set by the build system. > Which document states that this is considered best practice ? Jacob -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#94228): https://lists.openembedded.org/g/openembedded-devel/message/94228 Mute This Topic: https://lists.openembedded.org/mt/87562882/21656 Group Owner: openembedded-devel+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
[oe] [meta-oe][PATCH] Revert "ccid: Fix floating runtime dependency on flex"
With the switch to recipe-specific sysroots this patch is no longer
needed. Tested building with and without the patch and the resulting
binaries are identical.
This reverts commit 960314be5d854212ee65604a68a92a3616b2a544.
Signed-off-by: Jacob Kroon
---
.../ccid/ccid/no-dep-on-libfl.patch| 18 --
meta-oe/recipes-support/ccid/ccid_1.4.33.bb| 1 -
2 files changed, 19 deletions(-)
delete mode 100644 meta-oe/recipes-support/ccid/ccid/no-dep-on-libfl.patch
diff --git a/meta-oe/recipes-support/ccid/ccid/no-dep-on-libfl.patch
b/meta-oe/recipes-support/ccid/ccid/no-dep-on-libfl.patch
deleted file mode 100644
index 477cc31ad..0
--- a/meta-oe/recipes-support/ccid/ccid/no-dep-on-libfl.patch
+++ /dev/null
@@ -1,18 +0,0 @@
-No need to link with libfl.
-
-Upstream-Status: Pending
-Signed-off-by: Jacob Kroon
-
-Index: ccid-1.4.8/src/Makefile.am
-===
ccid-1.4.8.orig/src/Makefile.am
-+++ ccid-1.4.8/src/Makefile.am
-@@ -53,7 +53,7 @@ PROVIDED_BY_PCSC = debug.c
- endif
-
- libccid_la_SOURCES = $(COMMON) $(USB) $(TOKEN_PARSER) $(PROVIDED_BY_PCSC)
$(T1)
--libccid_la_LIBADD = $(LEXLIB) $(LIBUSB_LIBS) $(PTHREAD_LIBS)
-+libccid_la_LIBADD = $(LIBUSB_LIBS) $(PTHREAD_LIBS)
- libccid_la_CFLAGS = $(PCSC_CFLAGS) $(LIBUSB_CFLAGS) $(PTHREAD_CFLAGS) \
- $(SYMBOL_VISIBILITY) -D$(CCID_VERSION) -DSIMCLIST_NO_DUMPRESTORE
- libccid_la_LDFLAGS = -avoid-version
diff --git a/meta-oe/recipes-support/ccid/ccid_1.4.33.bb
b/meta-oe/recipes-support/ccid/ccid_1.4.33.bb
index 4b1c42173..e3cb7cef1 100644
--- a/meta-oe/recipes-support/ccid/ccid_1.4.33.bb
+++ b/meta-oe/recipes-support/ccid/ccid_1.4.33.bb
@@ -7,7 +7,6 @@ DEPENDS = "virtual/libusb0 pcsc-lite"
RDEPENDS:${PN} = "pcsc-lite"
SRC_URI = "https://ccid.apdu.fr/files/ccid-${PV}.tar.bz2 \
-file://no-dep-on-libfl.patch \
file://0001-Add-build-rule-for-README.patch \
"
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#94214):
https://lists.openembedded.org/g/openembedded-devel/message/94214
Mute This Topic: https://lists.openembedded.org/mt/87536557/21656
Group Owner: openembedded-devel+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
Re: [oe] [meta-java] Request to pull honister branch and override syntax commits from master-next to master branch
Hi Richard, On 11/17/21 16:57, Richard Leitner wrote: > Hi Geoff, > > On Sun, Nov 14, 2021 at 08:55:41PM -0800, Geoff Parker via > lists.openembedded.org wrote: >> I made this request a month ago, still no change to the master branch >> so that builds can be made using meta-java with honister or master >> openembedded layers. > > Thanks for the reminder. > It seems my CI script missed the final push. > I think your CI script needs some more love :-D Or in case of errors, perhaps publish them here. Jacob -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#94206): https://lists.openembedded.org/g/openembedded-devel/message/94206 Mute This Topic: https://lists.openembedded.org/mt/87063255/21656 Group Owner: openembedded-devel+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
[oe] [meta-java][PATCH] jdepend: Set protocol=https and branch=master
Fixes:
WARNING: jdepend-native-2.9.1-r0 do_fetch: URL:
git://github.com/clarkware/jdepend uses git protocol which is no longer
supported by github. Please change to ;protocol=https in the url.
WARNING: jdepend-native-2.9.1-r0 do_fetch: URL:
git://github.com/clarkware/jdepend does not set any branch parameter. The
future default branch used by tools and repositories is uncertain and we will
therefore soon require this is set in all git urls.
Signed-off-by: Jacob Kroon
---
recipes-core/jdepend/jdepend_2.9.1.bb | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/recipes-core/jdepend/jdepend_2.9.1.bb
b/recipes-core/jdepend/jdepend_2.9.1.bb
index 7c6cc8c..deae6a8 100644
--- a/recipes-core/jdepend/jdepend_2.9.1.bb
+++ b/recipes-core/jdepend/jdepend_2.9.1.bb
@@ -6,7 +6,7 @@ LIC_FILES_CHKSUM =
"file://LICENSE;md5=f5777d32a7709d558c2877d4a6616230"
HOMEPAGE = "https://github.com/clarkware/jdepend;
-SRC_URI = "git://github.com/clarkware/jdepend"
+SRC_URI = "git://github.com/clarkware/jdepend;protocol=https;branch=master"
SRCREV = "57980590313a5dbde236a3eb2c8958e9e53e6a10"
S = "${WORKDIR}/git"
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#94019):
https://lists.openembedded.org/g/openembedded-devel/message/94019
Mute This Topic: https://lists.openembedded.org/mt/87139361/21656
Group Owner: openembedded-devel+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[oe] [meta-java][PATCH] cacao: Set git branch
Signed-off-by: Jacob Kroon
---
recipes-core/cacao/cacao_git.bb | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/recipes-core/cacao/cacao_git.bb b/recipes-core/cacao/cacao_git.bb
index 29aa24f..fef7f34 100644
--- a/recipes-core/cacao/cacao_git.bb
+++ b/recipes-core/cacao/cacao_git.bb
@@ -11,7 +11,7 @@ DEPENDS = "zlib libtool classpath virtual/javac-native bdwgc"
RPROVIDES:${PN} = "java2-runtime"
SRCREV = "6c4694f9bd175386a8c451531e9a5ad97aa23b6f"
-SRC_URI = "git://bitbucket.org/cacaovm/cacao.git;protocol=https \
+SRC_URI = "git://bitbucket.org/cacaovm/cacao.git;protocol=https;branch=master \
file://0001-java.in-Do-not-use-hardcode-paths.patch \
file://0002-cacao-use-system-s-boehm-garbage-collector.patch \
"
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#94017):
https://lists.openembedded.org/g/openembedded-devel/message/94017
Mute This Topic: https://lists.openembedded.org/mt/87139168/21656
Group Owner: openembedded-devel+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
Re: [oe] [meta-java][PATCH 1/2] jamvm: Set git branch
Hi Richard, On 11/17/21 17:20, Richard Leitner - SKIDATA wrote: Hi Jacob, thanks for the patches! I've just added them to master-next. Another question as you might know advice: How may I apply those messages from groups.io using "git am" without having to adapt the "From:" for every single message? Is there any "trick" you know about? Hmm sorry no, I don't use the groups.io web-interface. I figured maintainers usually use their local mbox:es when applying patches from email ? /Jacob -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#94004): https://lists.openembedded.org/g/openembedded-devel/message/94004 Mute This Topic: https://lists.openembedded.org/mt/87114731/21656 Group Owner: openembedded-devel+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
[oe] [meta-java][PATCH 2/2] jikes: Avoid append override together with += operator
Signed-off-by: Jacob Kroon
---
recipes-core/jikes/jikes_1.22.bb | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/recipes-core/jikes/jikes_1.22.bb b/recipes-core/jikes/jikes_1.22.bb
index 5e6e470..3e83ff8 100644
--- a/recipes-core/jikes/jikes_1.22.bb
+++ b/recipes-core/jikes/jikes_1.22.bb
@@ -20,7 +20,7 @@ RDEPENDS:${PN}:class-native = ""
EXTRA_OECONF = "--disable-fp-emulation --enable-source15"
# configure script incorrectly defines these when cross compiling for ARM
-CXXFLAGS:append:arm += "-UHAVE_64BIT_TYPES -DWORDS_BIGENDIAN=1"
+CXXFLAGS:append:arm = " -UHAVE_64BIT_TYPES -DWORDS_BIGENDIAN=1"
do_install() {
oe_runmake 'DESTDIR=${D}' install
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#93987):
https://lists.openembedded.org/g/openembedded-devel/message/93987
Mute This Topic: https://lists.openembedded.org/mt/87114732/21656
Group Owner: openembedded-devel+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[oe] [meta-java][PATCH 1/2] jamvm: Set git branch
Signed-off-by: Jacob Kroon
---
recipes-core/jamvm/jamvm_git.bb | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/recipes-core/jamvm/jamvm_git.bb b/recipes-core/jamvm/jamvm_git.bb
index 3c9d47b..34861ac 100644
--- a/recipes-core/jamvm/jamvm_git.bb
+++ b/recipes-core/jamvm/jamvm_git.bb
@@ -6,7 +6,7 @@ require jamvm.inc
SRCREV = "6cef41d859fbc9ce7868a97cb2cb5dd2b10b9103"
PV = "2.0.0-devel+git${SRCPV}"
-SRC_URI = "git://git.code.sf.net/p/jamvm/code;protocol=git \
+SRC_URI = "git://git.code.sf.net/p/jamvm/code;protocol=git;branch=master \
file://jamvm-jni_h-noinst.patch \
file://libffi.patch \
file://jamvm-minmax-heap.patch \
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#93986):
https://lists.openembedded.org/g/openembedded-devel/message/93986
Mute This Topic: https://lists.openembedded.org/mt/87114731/21656
Group Owner: openembedded-devel+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[oe] [meta-java][PATCHv2] junit/xom: Fix task dependency typo
Signed-off-by: Jacob Kroon
---
recipes-core/junit/junit_3.8.2.bb | 4 ++--
recipes-core/xml-commons/xom_1.2.10.bb | 2 +-
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/recipes-core/junit/junit_3.8.2.bb
b/recipes-core/junit/junit_3.8.2.bb
index feaddc3..57dd459 100644
--- a/recipes-core/junit/junit_3.8.2.bb
+++ b/recipes-core/junit/junit_3.8.2.bb
@@ -16,14 +16,14 @@ inherit java-library
do_unpackpost[dirs] = "${B}"
do_unpackpost() {
mkdir -p src
- # Prevent deletion by do_deletebinraries.
+ # Prevent deletion by do_deletebinaries.
if [ -e src.jar ]; then
mv src.jar src.zip
fi
unzip -o src.zip -d src
}
-addtask unpackpost before do_deletebinraries after do_unpack
+addtask unpackpost before do_deletebinaries after do_unpack
do_compile() {
mkdir -p build
diff --git a/recipes-core/xml-commons/xom_1.2.10.bb
b/recipes-core/xml-commons/xom_1.2.10.bb
index 2520405..42ad957 100644
--- a/recipes-core/xml-commons/xom_1.2.10.bb
+++ b/recipes-core/xml-commons/xom_1.2.10.bb
@@ -36,7 +36,7 @@ do_move_jaxen() {
}
-addtask move_jaxen after do_deletebinraries before do_patch
+addtask move_jaxen after do_deletebinaries before do_patch
do_compile() {
mkdir -p build
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#93027):
https://lists.openembedded.org/g/openembedded-devel/message/93027
Mute This Topic: https://lists.openembedded.org/mt/85751867/21656
Group Owner: openembedded-devel+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[oe] [meta-java][PATCH 2/2] junit: Fix task dependency typo
Signed-off-by: Jacob Kroon
---
recipes-core/junit/junit_3.8.2.bb | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/recipes-core/junit/junit_3.8.2.bb
b/recipes-core/junit/junit_3.8.2.bb
index feaddc3..36ca453 100644
--- a/recipes-core/junit/junit_3.8.2.bb
+++ b/recipes-core/junit/junit_3.8.2.bb
@@ -23,7 +23,7 @@ do_unpackpost() {
unzip -o src.zip -d src
}
-addtask unpackpost before do_deletebinraries after do_unpack
+addtask unpackpost before do_deletebinaries after do_unpack
do_compile() {
mkdir -p build
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#93026):
https://lists.openembedded.org/g/openembedded-devel/message/93026
Mute This Topic: https://lists.openembedded.org/mt/85751788/21656
Group Owner: openembedded-devel+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[oe] [meta-java][PATCH 1/2] openjdk7: Use bb.parse.vars_from_file()
bb.parse.BBHandler.vars_from_file() was removed in:
https://git.openembedded.org/bitbake/commit/?id=aaa5292ef96ea27f505bc5c5a4b1eb4f497ed061
Signed-off-by: Jacob Kroon
---
recipes-core/openjdk/openjdk-7-common.inc | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/recipes-core/openjdk/openjdk-7-common.inc
b/recipes-core/openjdk/openjdk-7-common.inc
index 61b62b9..0435233 100644
--- a/recipes-core/openjdk/openjdk-7-common.inc
+++ b/recipes-core/openjdk/openjdk-7-common.inc
@@ -4,7 +4,7 @@ LICENSE = "GPL-2.0-with-classpath-exception"
PRIORITY = "optional"
SECTION = "libs"
-ICEDTEA_VERSION = "${@bb.parse.BBHandler.vars_from_file(d.getVar('FILE',
False),d)[1].split("-")[1] or '1.0'}"
+ICEDTEA_VERSION = "${@bb.parse.vars_from_file(d.getVar('FILE',
False),d)[1].split("-")[1] or '1.0'}"
S = "${WORKDIR}/${ICEDTEA}"
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#93025):
https://lists.openembedded.org/g/openembedded-devel/message/93025
Mute This Topic: https://lists.openembedded.org/mt/85751785/21656
Group Owner: openembedded-devel+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
Re: [oe] Fedora 34, shadow-native/icedtea7-native, umask problems ?
On 6/2/21 6:01 PM, Jacob Kroon via lists.openembedded.org wrote:
On 6/2/21 8:32 AM, Jacob Kroon via lists.openembedded.org wrote:
Hi,
I'm using Fedora 34 and OE-Core/Bitbake/layers from git master as of
today, with reproducible builds. Sometimes when rebuilding my image I
see sudden changes in file permissions (jumping back and forth) in the
buildhistory output for two native recipes:
shadow-native: (OE-Core)
-drwxr-xr-x - - 40 ./var/spool/mail
+drwxrwxr-x - - 40 ./var/spool/mail
I can reproduce it reliably with these commands:
After I do:
# bitbake -c cleansstate shadow-native && bitbake shadow-native
I get:
drwxrwxr-x - - 40 ./var/spool/mail
After I do:
# bitbake -c clean shadow-native && bitbake shadow-native
I get:
drwxr-xr-x - - 40 ./var/spool/mail
shadow.inc does:
do_install_append() {
...
install -m 0775 -d ${D}${localstatedir}/spool/mail
So it looks to me like those permissions are either not recorded in the
sstate cache, or they are not restored when extracting the cache.
I see this issue in master and hardknot builds, but not in dunfell builds.
/Jacob
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#91835):
https://lists.openembedded.org/g/openembedded-devel/message/91835
Mute This Topic: https://lists.openembedded.org/mt/83252995/21656
Group Owner: openembedded-devel+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
Re: [oe] Fedora 34, shadow-native/icedtea7-native, umask problems ?
On 6/2/21 8:32 AM, Jacob Kroon via lists.openembedded.org wrote:
Hi,
I'm using Fedora 34 and OE-Core/Bitbake/layers from git master as of
today, with reproducible builds. Sometimes when rebuilding my image I
see sudden changes in file permissions (jumping back and forth) in the
buildhistory output for two native recipes:
shadow-native: (OE-Core)
-drwxr-xr-x - - 40 ./var/spool/mail
+drwxrwxr-x - - 40 ./var/spool/mail
I can reproduce it reliably with these commands:
After I do:
# bitbake -c cleansstate shadow-native && bitbake shadow-native
I get:
drwxrwxr-x - - 40 ./var/spool/mail
After I do:
# bitbake -c clean shadow-native && bitbake shadow-native
I get:
drwxr-xr-x - - 40 ./var/spool/mail
shadow.inc does:
do_install_append() {
...
install -m 0775 -d ${D}${localstatedir}/spool/mail
So it looks to me like those permissions are either not recorded in the
sstate cache, or they are not restored when extracting the cache.
/Jacob
icedtea7-native: (meta-java)
-drwxrwxr-x - - 120
./usr/lib/jvm/icedtea7-native
-drwxrwxr-x - - 800
./usr/lib/jvm/icedtea7-native/bin
--rwxrwxr-x - - 15016
./usr/lib/jvm/icedtea7-native/bin/apt
--rwxrwxr-x - - 15016
./usr/lib/jvm/icedtea7-native/bin/extcheck
--rwxrwxr-x - - 15016
./usr/lib/jvm/icedtea7-native/bin/idlj
...
+drwxr-xr-x - - 120
./usr/lib/jvm/icedtea7-native
+drwxr-xr-x - - 800
./usr/lib/jvm/icedtea7-native/bin
+-rwxr-xr-x - - 15016
./usr/lib/jvm/icedtea7-native/bin/apt
+-rwxr-xr-x - - 15016
./usr/lib/jvm/icedtea7-native/bin/extcheck
+-rwxr-xr-x - - 15016
./usr/lib/jvm/icedtea7-native/bin/idlj
...
Before I start debugging them, does anyone have an obvious fix or hint
for any of these ?
Regards,
Jacob
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#91693):
https://lists.openembedded.org/g/openembedded-devel/message/91693
Mute This Topic: https://lists.openembedded.org/mt/83252995/21656
Group Owner: openembedded-devel+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[oe] Fedora 34, shadow-native/icedtea7-native, umask problems ?
Hi, I'm using Fedora 34 and OE-Core/Bitbake/layers from git master as of today, with reproducible builds. Sometimes when rebuilding my image I see sudden changes in file permissions (jumping back and forth) in the buildhistory output for two native recipes: shadow-native: (OE-Core) -drwxr-xr-x - - 40 ./var/spool/mail +drwxrwxr-x - - 40 ./var/spool/mail icedtea7-native: (meta-java) -drwxrwxr-x - - 120 ./usr/lib/jvm/icedtea7-native -drwxrwxr-x - - 800 ./usr/lib/jvm/icedtea7-native/bin --rwxrwxr-x - - 15016 ./usr/lib/jvm/icedtea7-native/bin/apt --rwxrwxr-x - - 15016 ./usr/lib/jvm/icedtea7-native/bin/extcheck --rwxrwxr-x - - 15016 ./usr/lib/jvm/icedtea7-native/bin/idlj ... +drwxr-xr-x - - 120 ./usr/lib/jvm/icedtea7-native +drwxr-xr-x - - 800 ./usr/lib/jvm/icedtea7-native/bin +-rwxr-xr-x - - 15016 ./usr/lib/jvm/icedtea7-native/bin/apt +-rwxr-xr-x - - 15016 ./usr/lib/jvm/icedtea7-native/bin/extcheck +-rwxr-xr-x - - 15016 ./usr/lib/jvm/icedtea7-native/bin/idlj ... Before I start debugging them, does anyone have an obvious fix or hint for any of these ? Regards, Jacob -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#91678): https://lists.openembedded.org/g/openembedded-devel/message/91678 Mute This Topic: https://lists.openembedded.org/mt/83252995/21656 Group Owner: openembedded-devel+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
Re: [oe] [meta-java] Changes from master-next merged to master
On 8/25/20 7:58 AM, Richard Leitner wrote: On Mon, Aug 24, 2020 at 09:36:42PM +0200, Alexander Kanavin wrote: I think the site and the downloads are now properly back. Thanks for the info and sorry for my misunderstanding regarding the java.net site. So you are able to build 8u265 with the current master? I can build master now. Must have been temporary server problems as you stated. Sorry for the noise. Regards Jacob -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#86557): https://lists.openembedded.org/g/openembedded-devel/message/86557 Mute This Topic: https://lists.openembedded.org/mt/76380470/21656 Group Owner: openembedded-devel+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
Re: [oe] [meta-java] Changes from master-next merged to master
On 8/24/20 8:22 AM, Richard Leitner wrote: Hi, this is a note to let you know that I've just merged the following changes from master-next to master: Alexander Kanavin (1): openjdk-8: add a glibc 2.32 compatibility patch Fabio Berton (2): openjdk-8-native: Add libxtst-native to x11 packageconfig openjdk-8-native: Build cups-native when using cups packageconfig Jeremy Puhlman (1): icedtea7-native: set ALT_CUPS_HEADERS_PATH to staging Richard Leitner (2): openjdk-8: refresh patches cacao: update to latest git version openjdk-8: update to latest ga version 265 Robert Berger (2): icedtea7-native: host gcc-9, extra CFLAGS needed WORKDIR and ARCHIVER_WORKDIR support Vyacheslav Yurkov (2): jlex: set unique file name for download icedtea7-native: set unique download file names If you want to see any of those changes backported to stable branches please just reply to this mail or send a patch. The patch "icedtea7-native: set ALT_CUPS_HEADERS_PATH to staging" by Jeremy is already scheduled for dunfell as mentioned in the patch. After updating master I'm getting: WARNING: openjdk-8-native-265-r0 do_fetch: Failed to fetch URL https://hg.openjdk.java.net/jdk8u/jdk8u/archive/jdk8u265-ga.tar.bz2;downloadfilename=openjdk8-265-jdk8u-jdk8u265-ga.tar.bz2;name=openjdk, attempting MIRRORS if available ERROR: openjdk-8-native-265-r0 do_fetch: Fetcher failure: Fetch command export PSEUDO_DISABLED=1; unset _PYTHON_SYSCONFIGDATA_NAME; export DBUS_SESSION_BUS_ADDRESS="unix:path=/run/user/1000/bus"; export SSH_AUTH_SOCK="/run/user/1000/keyring/ssh"; export PATH="/home/jkroon/Projects/foo-linux/openembedded-core/scripts/native-intercept:/home/jkroon/Projects/foo-linux/openembedded-core/scripts:/tmp/ramdisk/foo-linux-master-glibc/work/x86_64-linux/openjdk-8-native/265-r0/recipe-sysroot-native/usr/bin/x86_64-linux:/tmp/ramdisk/foo-linux-master-glibc/work/x86_64-linux/openjdk-8-native/265-r0/recipe-sysroot-native/usr/bin:/tmp/ramdisk/foo-linux-master-glibc/work/x86_64-linux/openjdk-8-native/265-r0/recipe-sysroot-native/usr/sbin:/tmp/ramdisk/foo-linux-master-glibc/work/x86_64-linux/openjdk-8-native/265-r0/recipe-sysroot-native/usr/bin:/tmp/ramdisk/foo-linux-master-glibc/work/x86_64-linux/openjdk-8-native/265-r0/recipe-sysroot-native/sbin:/tmp/ramdisk/foo-linux-master-glibc/work/x86_64-linux/openjdk-8-native/265-r0/recipe-sysroot-native/bin:/home/jkroon/Projects/foo-linux/bitbake/bin:/tmp/ramdisk/foo-linux-master-glibc/hosttools"; export HOME="/home/jkroon"; /usr/bin/env wget -t 2 -T 30 --passive-ftp --no-check-certificate -O /home/jkroon/Downloads/yocto/openjdk8-265-jdk8u-jdk8u265-ga.tar.bz2 -P /home/jkroon/Downloads/yocto 'https://hg.openjdk.java.net/jdk8u/jdk8u/archive/jdk8u265-ga.tar.bz2' --progress=dot -v failed with exit code 4, output: --2020-08-24 07:30:12-- https://hg.openjdk.java.net/jdk8u/jdk8u/archive/jdk8u265-ga.tar.bz2 Resolving hg.openjdk.java.net (hg.openjdk.java.net)... 137.254.56.63 Connecting to hg.openjdk.java.net (hg.openjdk.java.net)|137.254.56.63|:443... connected. HTTP request sent, awaiting response... Read error (Success.) in headers. Retrying. --2020-08-24 07:30:44-- (try: 2) https://hg.openjdk.java.net/jdk8u/jdk8u/archive/jdk8u265-ga.tar.bz2 Connecting to hg.openjdk.java.net (hg.openjdk.java.net)|137.254.56.63|:443... connected. HTTP request sent, awaiting response... Read error (Success.) in headers. Giving up. ERROR: openjdk-8-native-265-r0 do_fetch: Fetcher failure for URL: 'https://hg.openjdk.java.net/jdk8u/jdk8u/archive/jdk8u265-ga.tar.bz2;downloadfilename=openjdk8-265-jdk8u-jdk8u265-ga.tar.bz2;name=openjdk'. Unable to fetch URL from any source. ERROR: Logfile of failure stored in: /tmp/ramdisk/foo-linux-master-glibc/work/x86_64-linux/openjdk-8-native/265-r0/temp/log.do_fetch.756296 ERROR: Task (/home/jkroon/Projects/foo-linux/build/../meta-java/recipes-core/openjdk/openjdk-8-native_265.bb:do_fetch) failed with exit code '1' WARNING: openjre-8-265-r0 do_fetch: Failed to fetch URL https://hg.openjdk.java.net/jdk8u/jdk8u/archive/jdk8u265-ga.tar.bz2;downloadfilename=openjdk8-265-jdk8u-jdk8u265-ga.tar.bz2;name=openjdk, attempting MIRRORS if available ERROR: openjre-8-265-r0 do_fetch: Fetcher failure: Fetch command export PSEUDO_DISABLED=1; unset _PYTHON_SYSCONFIGDATA_NAME; export DBUS_SESSION_BUS_ADDRESS="unix:path=/run/user/1000/bus"; export SSH_AUTH_SOCK="/run/user/1000/keyring/ssh"; export
[oe] [PATCH] abseil-cpp: Use SkipRecipe exception
SkipPackage exception is deprecated.
Signed-off-by: Jacob Kroon
---
meta-oe/recipes-devtools/abseil-cpp/abseil-cpp_git.bb | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/meta-oe/recipes-devtools/abseil-cpp/abseil-cpp_git.bb
b/meta-oe/recipes-devtools/abseil-cpp/abseil-cpp_git.bb
index e874e4a5e..8efa31e95 100644
--- a/meta-oe/recipes-devtools/abseil-cpp/abseil-cpp_git.bb
+++ b/meta-oe/recipes-devtools/abseil-cpp/abseil-cpp_git.bb
@@ -37,22 +37,22 @@ python () {
if arch == "aarch64":
tunes = d.getVar("TUNE_FEATURES")
if not tunes:
-raise bb.parse.SkipPackage("%s-%s Needs support for crypto on
armv8" % (pkgn, pkgv))
+raise bb.parse.SkipRecipe("%s-%s Needs support for crypto on
armv8" % (pkgn, pkgv))
return
pkgn = d.getVar("PN")
pkgv = d.getVar("PV")
if "crypto" not in tunes:
-raise bb.parse.SkipPackage("%s-%s Needs support for crypto on
armv8" % (pkgn, pkgv))
+raise bb.parse.SkipRecipe("%s-%s Needs support for crypto on
armv8" % (pkgn, pkgv))
if arch == "x86_64":
tunes = d.getVar("TUNE_FEATURES")
if not tunes:
- raise bb.parse.SkipPackage("%s-%s Needs support for corei7 on
x86_64" % (pkgn, pkgv))
+ raise bb.parse.SkipRecipe("%s-%s Needs support for corei7 on
x86_64" % (pkgn, pkgv))
return
pkgn = d.getVar("PN")
pkgv = d.getVar("PV")
if "corei7" not in tunes:
-raise bb.parse.SkipPackage("%s-%s Needs support for corei7 on
x86_64" % (pkgn, pkgv))
+raise bb.parse.SkipRecipe("%s-%s Needs support for corei7 on
x86_64" % (pkgn, pkgv))
}
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#86152):
https://lists.openembedded.org/g/openembedded-devel/message/86152
Mute This Topic: https://lists.openembedded.org/mt/75983364/21656
Group Owner: openembedded-devel+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[oe] [PATCH] meta-oe: Don't inherit 'features_check' in recipes that don't utilize it
Signed-off-by: Jacob Kroon
---
meta-oe/recipes-connectivity/transmission/transmission_git.bb | 2 +-
meta-oe/recipes-extended/fluentbit/fluentbit_1.3.5.bb | 2 +-
meta-oe/recipes-support/upower/upower_0.99.11.bb | 2 +-
3 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/meta-oe/recipes-connectivity/transmission/transmission_git.bb
b/meta-oe/recipes-connectivity/transmission/transmission_git.bb
index ee5578f8b..412bd9d3f 100644
--- a/meta-oe/recipes-connectivity/transmission/transmission_git.bb
+++ b/meta-oe/recipes-connectivity/transmission/transmission_git.bb
@@ -19,7 +19,7 @@ PV = "2.94"
S = "${WORKDIR}/git"
-inherit autotools gettext update-rc.d systemd mime-xdg features_check
+inherit autotools gettext update-rc.d systemd mime-xdg
PACKAGECONFIG = "${@bb.utils.contains('DISTRO_FEATURES', 'x11', 'gtk', '', d)}
\
${@bb.utils.contains('DISTRO_FEATURES','systemd','systemd','',d)}"
diff --git a/meta-oe/recipes-extended/fluentbit/fluentbit_1.3.5.bb
b/meta-oe/recipes-extended/fluentbit/fluentbit_1.3.5.bb
index 06895a03a..21d13ac93 100644
--- a/meta-oe/recipes-extended/fluentbit/fluentbit_1.3.5.bb
+++ b/meta-oe/recipes-extended/fluentbit/fluentbit_1.3.5.bb
@@ -52,7 +52,7 @@ EXTRA_OECMAKE_append_riscv64 = " -DFLB_DEPS='atomic'"
# DEPENDS += "openssl "
# EXTRA_OECMAKE += "-DFLB_OUT_KAFKA=On "
-inherit cmake systemd features_check
+inherit cmake systemd
SYSTEMD_SERVICE_${PN} = "td-agent-bit.service"
TARGET_CC_ARCH_append = " ${SELECTED_OPTIMIZATION}"
diff --git a/meta-oe/recipes-support/upower/upower_0.99.11.bb
b/meta-oe/recipes-support/upower/upower_0.99.11.bb
index 640841792..d6c89a65d 100644
--- a/meta-oe/recipes-support/upower/upower_0.99.11.bb
+++ b/meta-oe/recipes-support/upower/upower_0.99.11.bb
@@ -10,7 +10,7 @@ SRC_URI = " \
SRC_URI[md5sum] = "abe6acb617f11f2e8dbd9846fcf86e24"
SRC_URI[sha256sum] =
"64b5ffbfccd5bdb15d925777979a4dbee1a957f9eaeb158dc76175267eddbdef"
-inherit autotools pkgconfig gettext gobject-introspection systemd
features_check
+inherit autotools pkgconfig gettext gobject-introspection systemd
PACKAGECONFIG ??= "${@bb.utils.filter('DISTRO_FEATURES', 'systemd', d)}"
PACKAGECONFIG[idevice] = "--with-idevice,--without-idevice,libimobiledevice
libplist"
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#84966):
https://lists.openembedded.org/g/openembedded-devel/message/84966
Mute This Topic: https://lists.openembedded.org/mt/74804636/21656
Group Owner: openembedded-devel+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[oe] [meta-java] g0hl1n/jdk8-fix
Hi Richard, Just a FYI, I successfully built branch g0hl1n/jdk8-fix, 00fa8f5e330679d00275010c7fda976caae44a00, on my Fedora 32 system with gcc v10. I did notice a comment typo in commit, "openjdk-8: require GCC >= v6", 0207ab72c58662cf7ac60229a5e14879b73f461d: diff --git a/recipes-core/openjdk/openjdk-8-common.inc b/recipes-core/openjdk/openjdk-8-common.inc index 084ead9..c78bb2a 100644 --- a/recipes-core/openjdk/openjdk-8-common.inc +++ b/recipes-core/openjdk/openjdk-8-common.inc @@ -171,7 +171,7 @@ EXTRA_OECONF_append = "\ # Since v6, GCC sets the default C++ standard to C++14 and introduces # dead store elimination by default. -# Since v10, GCC defaults to -fcommon. +# Since v10, GCC defaults to -fno-common. # OpenJDK 8 is not ready for either of these changes. GLOBAL_FLAGS = "-fno-lifetime-dse -fno-delete-null-pointer-checks -fcommon" /Jacob -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#84220): https://lists.openembedded.org/g/openembedded-devel/message/84220 Mute This Topic: https://lists.openembedded.org/mt/74023103/21656 Group Owner: openembedded-devel+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/leave/8024792/1994799631/xyzzy [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
Re: [oe] [PATCH][meta-java 3/3] openjdk-8: Fix building with GCC 10
On 4/30/20 11:10 PM, Richard Leitner wrote: Hi Jacob, thanks for the patch! On Thu, Apr 30, 2020 at 03:15:56PM +0200, Jacob Kroon wrote: Signed-off-by: Jacob Kroon --- recipes-core/openjdk/openjdk-8-common.inc | 4 1 file changed, 4 insertions(+) diff --git a/recipes-core/openjdk/openjdk-8-common.inc b/recipes-core/openjdk/openjdk-8-common.inc index d8b30b8..5228338 100644 --- a/recipes-core/openjdk/openjdk-8-common.inc +++ b/recipes-core/openjdk/openjdk-8-common.inc @@ -176,6 +176,10 @@ FLAGS_GCC6 = "-fno-lifetime-dse -fno-delete-null-pointer-checks" FLAGS_GCC7 = "-fno-lifetime-dse -fno-delete-null-pointer-checks" FLAGS_GCC8 = "-fno-lifetime-dse -fno-delete-null-pointer-checks" FLAGS_GCC9 = "-fno-lifetime-dse -fno-delete-null-pointer-checks" +FLAGS_GCC10 = "-fno-lifetime-dse -fno-delete-null-pointer-checks" + +# GCC 10 defaults to -fno-common, but OpenJDK 8 is not ready for this. +FLAGS_GCC10_append = " -fcommon" Does setting "-fcommon" also work for GCC version > 5 and < 10? In icedtea7-native you've appended it to CFLAGS it w/o GCC version. Background of the question is that I'm currently planning to drop support for GCC versions < 6 in the recipes. If done so, we currently have no need for version specific flags. Just want to check if that would be an exception. Thanks. regards;rl A quick check shows that support for -fcommon/-fno-common exists since at least gcc 4.7.2: https://gcc.gnu.org/onlinedocs/gcc-4.7.2/gcc/Code-Gen-Options.html So yes, it should work for 5 < GCC version < 10. /Jacob -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#84169): https://lists.openembedded.org/g/openembedded-devel/message/84169 Mute This Topic: https://lists.openembedded.org/mt/73374911/21656 Group Owner: openembedded-devel+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/leave/8024792/1994799631/xyzzy [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
[oe] [PATCH][meta-java 2/3] icedtea7-native: Fix building with GCC 10
Signed-off-by: Jacob Kroon --- recipes-core/icedtea/icedtea7-native.inc | 3 +++ 1 file changed, 3 insertions(+) diff --git a/recipes-core/icedtea/icedtea7-native.inc b/recipes-core/icedtea/icedtea7-native.inc index 8d0dc71..b0b9419 100644 --- a/recipes-core/icedtea/icedtea7-native.inc +++ b/recipes-core/icedtea/icedtea7-native.inc @@ -28,6 +28,9 @@ CXX_append = " -std=gnu++98" # WORKAROUND: ignore errors from new compilers CFLAGS_append = " -Wno-error=stringop-overflow -Wno-error=return-type" +# Explicitly pass -fcommon since GCC 10 defaults to -fno-common for C sources +CFLAGS_append = " -fcommon" + inherit native java autotools pkgconfig inherit openjdk-build-helper inherit python3native -- 2.26.2 -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#84160): https://lists.openembedded.org/g/openembedded-devel/message/84160 Mute This Topic: https://lists.openembedded.org/mt/73374910/21656 Group Owner: openembedded-devel+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/leave/8024792/1994799631/xyzzy [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
[oe] [PATCH][meta-java 3/3] openjdk-8: Fix building with GCC 10
Signed-off-by: Jacob Kroon
---
recipes-core/openjdk/openjdk-8-common.inc | 4
1 file changed, 4 insertions(+)
diff --git a/recipes-core/openjdk/openjdk-8-common.inc
b/recipes-core/openjdk/openjdk-8-common.inc
index d8b30b8..5228338 100644
--- a/recipes-core/openjdk/openjdk-8-common.inc
+++ b/recipes-core/openjdk/openjdk-8-common.inc
@@ -176,6 +176,10 @@ FLAGS_GCC6 = "-fno-lifetime-dse
-fno-delete-null-pointer-checks"
FLAGS_GCC7 = "-fno-lifetime-dse -fno-delete-null-pointer-checks"
FLAGS_GCC8 = "-fno-lifetime-dse -fno-delete-null-pointer-checks"
FLAGS_GCC9 = "-fno-lifetime-dse -fno-delete-null-pointer-checks"
+FLAGS_GCC10 = "-fno-lifetime-dse -fno-delete-null-pointer-checks"
+
+# GCC 10 defaults to -fno-common, but OpenJDK 8 is not ready for this.
+FLAGS_GCC10_append = " -fcommon"
# flags for -native, and for bits that need a host-tool during -cross
BUILD_CFLAGS_append = " ${@openjdk_build_helper_get_build_cflags(d)}"
--
2.26.2
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#84161):
https://lists.openembedded.org/g/openembedded-devel/message/84161
Mute This Topic: https://lists.openembedded.org/mt/73374911/21656
Group Owner: openembedded-devel+ow...@lists.openembedded.org
Unsubscribe:
https://lists.openembedded.org/g/openembedded-devel/leave/8024792/1994799631/xyzzy
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[oe] [PATCH][meta-java 1/3] openjdk-build-helper: Fix decoding GCC major version
GCC in Fedora 32 is up to version 10, so return the whole word
instead of the first letter.
Signed-off-by: Jacob Kroon
---
classes/openjdk-build-helper.bbclass | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/classes/openjdk-build-helper.bbclass
b/classes/openjdk-build-helper.bbclass
index 8d83e02..4e334c2 100644
--- a/classes/openjdk-build-helper.bbclass
+++ b/classes/openjdk-build-helper.bbclass
@@ -31,7 +31,7 @@ def openjdk_build_helper_get_build_cflags(d):
from subprocess import Popen, PIPE
cmd = d.expand('%s -dumpversion' % build_cc).split()
cc = Popen(cmd, stdout=PIPE, stderr=PIPE)
-return cc.communicate()[0].decode('utf-8')[0]
+return cc.communicate()[0].decode('utf-8').split()[0]
build_cc = d.getVar('BUILD_CC')
version = get_build_cc_version(build_cc)
--
2.26.2
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#84159):
https://lists.openembedded.org/g/openembedded-devel/message/84159
Mute This Topic: https://lists.openembedded.org/mt/73374906/21656
Group Owner: openembedded-devel+ow...@lists.openembedded.org
Unsubscribe:
https://lists.openembedded.org/g/openembedded-devel/leave/8024792/1994799631/xyzzy
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
Re: [oe] [meta-java][PATCH v2] openjdk-8: update to latest ga version 252
Hi Richard, On 4/28/20 8:14 AM, Richard Leitner wrote: This patch also drops the md5 checksums as sha256 checksums are sufficient. Signed-off-by: Richard Leitner --- CHANGES v2: + added missing openjre recipe update --- Builds and runs fine on my setup. Thanks, Jacob -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#84125): https://lists.openembedded.org/g/openembedded-devel/message/84125 Mute This Topic: https://lists.openembedded.org/mt/73322007/21656 Group Owner: openembedded-devel+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
Re: [oe] [meta-java][PATCH] openjdk-8: update to latest ga version 252
On 4/27/20 8:27 AM, Richard Leitner wrote: fyi: This patch has been added to master-next. Any test results are welcome! I think this breaks building openjre: WARNING: openjdk-8-native-252-r0 do_fetch: Checksum mismatch for local file /home/jkroon/Downloads/yocto/openjdk8-252-langtools-jdk8u252-ga.tar.bz2 Cleaning and trying again. WARNING: openjdk-8-native-252-r0 do_fetch: Renaming /home/jkroon/Downloads/yocto/openjdk8-252-langtools-jdk8u252-ga.tar.bz2 to /home/jkroon/Downloads/yocto/openjdk8-252-langtools-jdk8u252-ga.tar.bz2_bad-checksum_2cde0f920e96038c8fe27af3e802bcd90b894a2feb37fed0604cc53787807dfe WARNING: openjre-8-242-r0 do_fetch: Checksum mismatch for local file /home/jkroon/Downloads/yocto/openjdk8-242-jdk8u-jdk8u242-ga.tar.bz2 Cleaning and trying again. WARNING: openjre-8-242-r0 do_fetch: Renaming /home/jkroon/Downloads/yocto/openjdk8-242-jdk8u-jdk8u242-ga.tar.bz2 to /home/jkroon/Downloads/yocto/openjdk8-242-jdk8u-jdk8u242-ga.tar.bz2_bad-checksum_633f1d20d4d8d36fea1680140b5514ab153096a19c306b812fbbd19314c8dece WARNING: openjre-8-242-r0 do_fetch: Checksum failure encountered with download of https://hg.openjdk.java.net/jdk8u/jdk8u/archive/jdk8u242-ga.tar.bz2;downloadfilename=openjdk8-242-jdk8u-jdk8u242-ga.tar.bz2;name=openjdk - will attempt other sources if available ERROR: openjre-8-242-r0 do_fetch: Fetcher failure for URL: 'https://hg.openjdk.java.net/jdk8u/jdk8u/archive/jdk8u242-ga.tar.bz2;downloadfilename=openjdk8-242-jdk8u-jdk8u242-ga.tar.bz2;name=openjdk'. Checksum mismatch! File: '/home/jkroon/Downloads/yocto/openjdk8-242-jdk8u-jdk8u242-ga.tar.bz2' has sha256 checksum 633f1d20d4d8d36fea1680140b5514ab153096a19c306b812fbbd19314c8dece when a83f8daa2afece63bf7ae95cfbca91ef765424c40b9f5b45e4a3c2d4da0f7a95 was expected If this change is expected (e.g. you have upgraded to a new version without updating the checksums) then you can use these lines within the recipe: SRC_URI[openjdk.sha256sum] = "633f1d20d4d8d36fea1680140b5514ab153096a19c306b812fbbd19314c8dece" Otherwise you should retry the download and/or check with upstream to determine if the file has become corrupted or otherwise unexpectedly modified. ERROR: openjre-8-242-r0 do_fetch: Fetcher failure for URL: 'https://hg.openjdk.java.net/jdk8u/jdk8u/archive/jdk8u242-ga.tar.bz2;downloadfilename=openjdk8-242-jdk8u-jdk8u242-ga.tar.bz2;name=openjdk'. Unable to fetch URL from any source. ERROR: Logfile of failure stored in: /tmp/ramdisk/tmp-master-glibc/work/i686-oe-linux/openjre-8/242-r0/temp/log.do_fetch.15248 ERROR: Task (/home/jkroon/Projects/tmp-linux/build/../meta-java/recipes-core/openjdk/openjre-8_242.bb:do_fetch) failed with exit code '1' Waiting for 1 running tasks to finish: 0: openjdk-8-native-252-r0 do_fetch (pid 15246) 0% | /Jacob -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#84113): https://lists.openembedded.org/g/openembedded-devel/message/84113 Mute This Topic: https://lists.openembedded.org/mt/73145297/21656 Group Owner: openembedded-devel+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
[oe] [meta-java][PATCH] layer.conf: Mark as compatible with dunfell
Signed-off-by: Jacob Kroon
---
conf/layer.conf | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/conf/layer.conf b/conf/layer.conf
index 0ddc5f9..294c319 100644
--- a/conf/layer.conf
+++ b/conf/layer.conf
@@ -7,7 +7,7 @@ BBFILES += "${LAYERDIR}/recipes*/*/*.bb
${LAYERDIR}/recipes*/*/*.bbappend"
BBFILE_COLLECTIONS += "meta-java"
BBFILE_PATTERN_meta-java := "^${LAYERDIR}/"
BBFILE_PRIORITY_meta-java = "10"
-LAYERSERIES_COMPAT_meta-java = "zeus"
+LAYERSERIES_COMPAT_meta-java = "dunfell"
LAYERDEPENDS_meta-java += "openembedded-layer"
LICENSE_PATH += "${LAYERDIR}/licenses"
--
2.25.1
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#83575):
https://lists.openembedded.org/g/openembedded-devel/message/83575
Mute This Topic: https://lists.openembedded.org/mt/72672432/21656
Group Owner: openembedded-devel+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
Re: [oe] Request to Revert 'recipes: rename distro_features_check to features_check' comit on meta-java zeus branch
On 3/31/20 8:17 AM, Alex Kiernan wrote: On Tue, Mar 31, 2020 at 5:46 AM Geoff Parker wrote: Hi Alex, We moved one of our products to the zeus Yocto 3.0 meta layer branches. It recently stopped building due to this commit on the meta-java zeus branch: commit d2e962a795e8713e5443659669842298453d938f Author: Alex Kiernan Date: Sat Dec 7 18:07:38 2019 + recipes: rename distro_features_check to features_check Avoid warning due to the class rename in OE-Core. Signed-off-by: Alex Kiernan Signed-off-by: Richard Leitner Although this change is appropriate for the Yocto 3.1 branches, it breaks our build on Yocto 3.0 zeus because the zeus branch of the openembedded-core layer does not have the features_check class. The following recipes fail to parse: openjdk-7_99b00-2.6.5.bb classpath_0.99.bb cacao_2d6fc614daf9.bb With errors like: ERROR: ParseError at /home/geoff/yocto-ccu/poky/build/../meta-java/recipes-core/openjdk/openjdk-7_99b00-2.6.5.bb:7: Could not inherit file classes/features_check.bbclass Can this commit be reverted? @OpenEmbedded Devel List this shouldn't have been pulled into zeus as it relies on changes in master on oe-core I've sent: https://lists.openembedded.org/g/openembedded-devel/message/83557 Anyone know how I can make the new mailing list archive use a sane font when displaying patches ? /Jacob -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#83574): https://lists.openembedded.org/g/openembedded-devel/message/83574 Mute This Topic: https://lists.openembedded.org/mt/72671623/21656 Group Owner: openembedded-devel+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
[oe] [meta-java][zeus][PATCH] Revert "recipes: rename distro_features_check to features_check"
The corresponding changes in oe-core has not been backported to zeus.
This reverts commit d2e962a795e8713e5443659669842298453d938f.
Signed-off-by: Jacob Kroon
---
recipes-core/cacao/cacao_2d6f6c14daf9.bb | 2 +-
recipes-core/classpath/classpath_0.99.bb | 2 +-
recipes-core/jamvm/jamvm.inc | 2 +-
recipes-core/openjdk/openjdk-7_99b00-2.6.5.bb | 2 +-
4 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/recipes-core/cacao/cacao_2d6f6c14daf9.bb
b/recipes-core/cacao/cacao_2d6f6c14daf9.bb
index 0e50b68..6783894 100644
--- a/recipes-core/cacao/cacao_2d6f6c14daf9.bb
+++ b/recipes-core/cacao/cacao_2d6f6c14daf9.bb
@@ -19,7 +19,7 @@ S = "${WORKDIR}/cacaovm-cacao-staging-${PV}"
SRC_URI[md5sum] = "5157d0bb9eb1332f7a6b6186eaac01aa"
SRC_URI[sha256sum] =
"34b51d660d69c2a83225fd75eab5e1fac002fb5974cb4a74b7478baf923a76cd"
-inherit java autotools-brokensep update-alternatives pkgconfig features_check
+inherit java autotools-brokensep update-alternatives pkgconfig
distro_features_check
REQUIRED_DISTRO_FEATURES = "x11"
REQUIRED_DISTRO_FEATURES_class-native := ""
diff --git a/recipes-core/classpath/classpath_0.99.bb
b/recipes-core/classpath/classpath_0.99.bb
index 5b9d904..714503a 100644
--- a/recipes-core/classpath/classpath_0.99.bb
+++ b/recipes-core/classpath/classpath_0.99.bb
@@ -1,6 +1,6 @@
require classpath.inc
-inherit features_check
+inherit distro_features_check
REQUIRED_DISTRO_FEATURES = "x11"
diff --git a/recipes-core/jamvm/jamvm.inc b/recipes-core/jamvm/jamvm.inc
index b03a5b9..d47b2f1 100644
--- a/recipes-core/jamvm/jamvm.inc
+++ b/recipes-core/jamvm/jamvm.inc
@@ -15,7 +15,7 @@ RPROVIDES_${PN}_class-native = ""
BBCLASSEXTEND = "native"
-inherit java autotools update-alternatives pkgconfig relative_symlinks
features_check
+inherit java autotools update-alternatives pkgconfig relative_symlinks
distro_features_check
REQUIRED_DISTRO_FEATURES_class-target = "x11"
diff --git a/recipes-core/openjdk/openjdk-7_99b00-2.6.5.bb
b/recipes-core/openjdk/openjdk-7_99b00-2.6.5.bb
index d774cc2..cb670cb 100644
--- a/recipes-core/openjdk/openjdk-7_99b00-2.6.5.bb
+++ b/recipes-core/openjdk/openjdk-7_99b00-2.6.5.bb
@@ -4,7 +4,7 @@ LIC_FILES_CHKSUM =
"file://COPYING;md5=59530bdf33659b29e73d4adb9f9f6552"
FILESPATH =. "${FILE_DIRNAME}/patches-openjdk-7:"
-inherit features_check
+inherit distro_features_check
REQUIRED_DISTRO_FEATURES = "x11"
--
2.25.1
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#83557):
https://lists.openembedded.org/g/openembedded-devel/message/83557
Mute This Topic: https://lists.openembedded.org/mt/72646949/21656
Group Owner: openembedded-devel+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[oe] [meta-java][master-next][PATCH] openjdk-8: Don't set TOPDIR bitbake variable
It is used to set the default sstate cache location.
Just export it to the environment.
Richard, feel free to squash this or someting similar into the
openjdk-8 update commit.
Signed-off-by: Jacob Kroon
---
recipes-core/openjdk/openjdk-8-common.inc | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/recipes-core/openjdk/openjdk-8-common.inc
b/recipes-core/openjdk/openjdk-8-common.inc
index 0ef4750..d8b30b8 100644
--- a/recipes-core/openjdk/openjdk-8-common.inc
+++ b/recipes-core/openjdk/openjdk-8-common.inc
@@ -12,7 +12,6 @@ FILESPATH =. "${FILE_DIRNAME}/patches-openjdk-8:"
# for weird openjdk-common.inc
S = "${WORKDIR}/${OPENJDK_HG_U}-${OPENJDK_CHANGESET}"
-export TOPDIR="${S}"
SRC_URI = "\
${OPENJDK_URI} \
@@ -27,6 +26,10 @@ SRC_URI = "\
${PATCHES_URI} \
"
+do_configure_prepend () {
+export TOPDIR=${S}
+}
+
do_unpack_extract_submodules () {
cd "${S}"
# tar --transform
--
2.24.1
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-java][PATCH 0/4] openjdk-8: update to jdk8u242ga
On 1/28/20 1:01 PM, Jacob Kroon wrote: On 1/28/20 6:43 AM, Jacob Kroon wrote: On 1/27/20 5:05 PM, Jacob Kroon wrote: On 1/27/20 3:57 PM, Richard Leitner wrote: On Mon, Jan 27, 2020 at 03:10:04PM +0100, Jacob Kroon wrote: On 1/27/20 12:24 PM, Richard Leitner wrote: This series updates Openjdk 8 to the latest "ga" release 242. Successful testing has been done on a fedora-31 build host for qemuarm64, qemux86-64, qemuarm32 and armv7a. I don't know if it is related to my setup, but "bitbake openjre-8" insists on rebuilding itself every time I run it, and I don't see any sstate cache being generated for it. I just tried to reproduce it, but neither for armv7a nor for aarch64 openjre-8 is rebuiled on the latest poky/oe-core master when nothing was changed in meta-java. So in your case it also rebuilds if you call "bitbake openjre-8" directly one-after-another without any change to the layers? Yes, precisely. But.. I think you can ignore this. I suspect I have confused bitbake's hashequiv database by pruning old cache-files using the sstate-cache-management.sh script. Joshua, do you know if that script is supported when using hashequiv ? Gonna try and remove cache/ aswell, and see if that helps. I just retried on a clean build, I wiped everything, and openjre-8 still insists on rebuilding on the second run. I don't know whats going on here.. Looks like it is caused by me using rm_work.bbclass, it seems to wipe something that causes the rebuilds. Actually it looks like openjre-8 and openjdk-8-native cannot rebuild from sstate cache, even with rm_work disabled. -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-java][PATCH v2 1/4] openjdk-8: update to latest ga version 242
On 1/29/20 10:17 PM, Jacob Kroon wrote:
On 1/28/20 6:22 AM, Richard Leitner wrote:
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"
Signed-off-by: Richard Leitner
---
docs/UPDATING.md | 18 +
recipes-core/openjdk/openjdk-8-common.inc | 13 +-
recipes-core/openjdk/openjdk-8-cross.inc | 27 +-
recipes-core/openjdk/openjdk-8-native.inc | 5 +-
.../openjdk/openjdk-8-native_172b11.bb | 2 -
recipes-core/openjdk/openjdk-8-native_242.bb | 2 +
.../openjdk-8-release-172b11-aarch32.inc | 58 -
.../openjdk-8-release-172b11-aarch64.inc | 35 -
.../openjdk/openjdk-8-release-172b11.inc | 39 -
.../openjdk/openjdk-8-release-aarch32.inc | 60 +
.../openjdk/openjdk-8-release-aarch64.inc | 37 +
...ommon.inc => openjdk-8-release-common.inc} | 96 +-
recipes-core/openjdk/openjdk-8-release.inc | 39 +
.../{openjdk-8_172b11.bb => openjdk-8_242.bb} | 4 +-
.../{openjre-8_172b11.bb => openjre-8_242.bb} | 4 +-
...low-using-a-system-installed-libjpeg.patch | 77 +
...llow-using-a-system-installed-libpng.patch | 80 +
...pilation-with-security-flags-enabled.patch | 41 -
...ild-on-as-needed-toolchains-generic.patch} | 12 +-
... => 0004-don-t-expect-fqpn-for-make.patch} | 21 +-
...llow-using-a-system-installed-libpng.patch | 148 -
...005-autoconf-filter-aclocal-copy-too.patch | 25 +
...> 0006-autoconf-handle-extra-output.patch} | 19 +-
...-assembler-flag-handling-in-makefile.patch | 36 +
...0008-autoconf-fix-shark-build-common.patch | 37 +
...-in-favour-of-openembedded-package-.patch} | 98 +-
...ove-shell-variables-from-autoheader.patch} | 25 +-
...conf-fix-CC-with-arguments-detection.patch | 43 +
...onf-NativeCompilation-remove-sysroot.patch | 24 +
...4-fix-undefined-behaviour-gcc-v8-fix.patch | 41 -
...o-fix-undefined-behaviour-gcc-v8-fix.patch | 42 -
...g-build-after-8062808-Turn-on-the-Wr.patch | 47 -
...-HotSpot-builds-on-5.x-Linux-kernels.patch | 43 -
...tspot-fix-crash-on-JNI_CreateJavaVM.patch} | 13 +-
...otspot-use-correct-include-for-poll.patch} | 24 +-
...tspot-don-t-rely-on-old-SysV-SIGCLD.patch} | 12 +-
...patch-to-fix-misuses-of-strncpy-str.patch} | 63 +-
...rn-type-issues-introduced-by-806165.patch} | 21 +-
...ro-build-requires-disabled-warnings.patch} | 36 +-
...eturn-statement-in-__sync_val_compa.patch} | 18 +-
...ot-Turn-on-the-Wreturn-type-warning.patch} | 70 +-
...otspot-handle-format-error-for-GCC-7.patch | 34 +
...atch => 1010-hotspot-fix-adlc-flags.patch} | 33 +-
...1011-hotspot-fix-shark-build-common.patch} | 322 +-
...1012-hotspot-restrict-to-staging-dir.patch | 26 +
...parison-between-pointer-and-integer.patch} | 34 +-
...ow-using-a-system-installed-libjpeg.patch} | 84 +-
...llow-using-a-system-installed-libpng.p
Re: [oe] [meta-java][PATCH 0/4] openjdk-8: update to jdk8u242ga
On 1/28/20 6:43 AM, Jacob Kroon wrote: On 1/27/20 5:05 PM, Jacob Kroon wrote: On 1/27/20 3:57 PM, Richard Leitner wrote: On Mon, Jan 27, 2020 at 03:10:04PM +0100, Jacob Kroon wrote: On 1/27/20 12:24 PM, Richard Leitner wrote: This series updates Openjdk 8 to the latest "ga" release 242. Successful testing has been done on a fedora-31 build host for qemuarm64, qemux86-64, qemuarm32 and armv7a. I don't know if it is related to my setup, but "bitbake openjre-8" insists on rebuilding itself every time I run it, and I don't see any sstate cache being generated for it. I just tried to reproduce it, but neither for armv7a nor for aarch64 openjre-8 is rebuiled on the latest poky/oe-core master when nothing was changed in meta-java. So in your case it also rebuilds if you call "bitbake openjre-8" directly one-after-another without any change to the layers? Yes, precisely. But.. I think you can ignore this. I suspect I have confused bitbake's hashequiv database by pruning old cache-files using the sstate-cache-management.sh script. Joshua, do you know if that script is supported when using hashequiv ? Gonna try and remove cache/ aswell, and see if that helps. I just retried on a clean build, I wiped everything, and openjre-8 still insists on rebuilding on the second run. I don't know whats going on here.. Looks like it is caused by me using rm_work.bbclass, it seems to wipe something that causes the rebuilds. -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-java][PATCH v2 0/4] openjdk-8: update to jdk8u242ga
Hi Richard,
On 1/28/20 6:22 AM, Richard Leitner wrote:
This series updates Openjdk 8 to the latest "ga" release 242.
Successful testing has been done on a fedora-31 build host for
qemuarm64, qemux86-64, qemuarm32 and armv7a.
CHANGES v2:
- remove openjre debuginfo fixup completely as it targeted JDK_HOME
instead of JRE_HOME and jre doen't create debuginfo files.
Richard Leitner (4):
openjdk-8: update to latest ga version 242
docs: add update8checksums.sh and mention it in UPDATING.md
openjre-8: remove debuginfo fixup from do_install
openjdk-8: fix jdk linking for non aarch architectures
docs/UPDATING.md | 16 +
docs/update8checksums.sh | 64 +
recipes-core/openjdk/openjdk-8-common.inc | 13 +-
recipes-core/openjdk/openjdk-8-cross.inc | 27 +-
recipes-core/openjdk/openjdk-8-native.inc |5 +-
.../openjdk/openjdk-8-native_172b11.bb|2 -
recipes-core/openjdk/openjdk-8-native_242.bb |2 +
.../openjdk-8-release-172b11-aarch32.inc | 58 -
.../openjdk-8-release-172b11-aarch64.inc | 35 -
.../openjdk/openjdk-8-release-172b11.inc | 39 -
.../openjdk/openjdk-8-release-aarch32.inc | 60 +
.../openjdk/openjdk-8-release-aarch64.inc | 37 +
...ommon.inc => openjdk-8-release-common.inc} | 102 +-
recipes-core/openjdk/openjdk-8-release.inc| 39 +
.../{openjdk-8_172b11.bb => openjdk-8_242.bb} |4 +-
.../{openjre-8_172b11.bb => openjre-8_242.bb} |3 +-
...low-using-a-system-installed-libjpeg.patch | 77 +
...llow-using-a-system-installed-libpng.patch | 80 +
...pilation-with-security-flags-enabled.patch | 41 -
...ild-on-as-needed-toolchains-generic.patch} | 12 +-
... => 0004-don-t-expect-fqpn-for-make.patch} | 21 +-
...llow-using-a-system-installed-libpng.patch | 148 -
...005-autoconf-filter-aclocal-copy-too.patch | 25 +
...> 0006-autoconf-handle-extra-output.patch} | 19 +-
...-assembler-flag-handling-in-makefile.patch | 36 +
...0008-autoconf-fix-shark-build-common.patch | 37 +
...-in-favour-of-openembedded-package-.patch} | 98 +-
...ove-shell-variables-from-autoheader.patch} | 25 +-
...conf-fix-CC-with-arguments-detection.patch | 43 +
...onf-NativeCompilation-remove-sysroot.patch | 24 +
...4-fix-undefined-behaviour-gcc-v8-fix.patch | 41 -
...o-fix-undefined-behaviour-gcc-v8-fix.patch | 42 -
...g-build-after-8062808-Turn-on-the-Wr.patch | 47 -
...-HotSpot-builds-on-5.x-Linux-kernels.patch | 43 -
...tspot-fix-crash-on-JNI_CreateJavaVM.patch} | 13 +-
...otspot-use-correct-include-for-poll.patch} | 24 +-
...tspot-don-t-rely-on-old-SysV-SIGCLD.patch} | 12 +-
...patch-to-fix-misuses-of-strncpy-str.patch} | 63 +-
...rn-type-issues-introduced-by-806165.patch} | 21 +-
...ro-build-requires-disabled-warnings.patch} | 36 +-
...eturn-statement-in-__sync_val_compa.patch} | 18 +-
...ot-Turn-on-the-Wreturn-type-warning.patch} | 70 +-
...otspot-handle-format-error-for-GCC-7.patch | 34 +
...atch => 1010-hotspot-fix-adlc-flags.patch} | 33 +-
...1011-hotspot-fix-shark-build-common.patch} | 322 +-
...1012-hotspot-restrict-to-staging-dir.patch | 26 +
...parison-between-pointer-and-integer.patch} | 34 +-
...ow-using-a-system-installed-libjpeg.patch} | 84 +-
...llow-using-a-system-installed-libpng.patch | 83 +
...04-jdk-use-correct-include-for-poll.patch} | 47 +-
...-jdk-use-correct-include-for-signal.patch} | 23 +-
...dk-disable-backtrace-musl-build-fix.patch} | 13 +-
.../2007-jdk-no-genx11-in-headless.patch | 32 +
...ps.patch => 2008-jdk-no-unused-deps.patch} | 66 +-
...f-ld-to-link-genSocketOptionRegistry.patch | 13 +
...2-hotspot-fix-missing-return-values.patch} |0
...aarch32-hotspot-fix-shark-build-pt2.patch} |0
...arch32-hotspot-remaining-musl-fixes.patch} |0
...ix-further-functions-with-a-missing.patch} |0
...arch64-hotspot-remaining-musl-fixes.patch} |0
.../filter-aclocal-copy-too.patch | 11 -
.../hotspot-handle-gcc7-format-overflow.patch | 25 -
.../openjdk8-add-missing-linker-flags.patch | 57 -
...enjdk8-find-compiler-fix-env-respect.patch | 140 -
...-assembler-flag-handling-in-makefile.patch | 20 -
.../openjdk8-fix-libpng-neon-build.patch | 29 -
.../openjdk8-fix-shark-build-pt2.patch| 35 +-
.../openjdk8-fix-shark-stdc++11.patch | 2631 -
.../openjdk8-no-genx11-in-headless.patch | 17 -
.../openjdk8-restrict-to-staging-dir.patch| 11 -
70 files changed, 1390 insertions(+), 4018 deletions(-)
create mode 100644 docs/UPDATING.md
create mode 100755 docs/update8checksums.sh
delete mode 100644 recipes-core/openjdk/openjdk-8-native_172b11.bb
create mode 100644 recipes-core/openjdk/openjdk-8-native_242.bb
delete mode 100644 recipes-core/openjdk/openjdk-8-release-172b11-aarch32.inc
Re: [oe] [meta-java][PATCH 0/4] openjdk-8: update to jdk8u242ga
On 1/27/20 5:05 PM, Jacob Kroon wrote: On 1/27/20 3:57 PM, Richard Leitner wrote: On Mon, Jan 27, 2020 at 03:10:04PM +0100, Jacob Kroon wrote: On 1/27/20 12:24 PM, Richard Leitner wrote: This series updates Openjdk 8 to the latest "ga" release 242. Successful testing has been done on a fedora-31 build host for qemuarm64, qemux86-64, qemuarm32 and armv7a. I don't know if it is related to my setup, but "bitbake openjre-8" insists on rebuilding itself every time I run it, and I don't see any sstate cache being generated for it. I just tried to reproduce it, but neither for armv7a nor for aarch64 openjre-8 is rebuiled on the latest poky/oe-core master when nothing was changed in meta-java. So in your case it also rebuilds if you call "bitbake openjre-8" directly one-after-another without any change to the layers? Yes, precisely. But.. I think you can ignore this. I suspect I have confused bitbake's hashequiv database by pruning old cache-files using the sstate-cache-management.sh script. Joshua, do you know if that script is supported when using hashequiv ? Gonna try and remove cache/ aswell, and see if that helps. I just retried on a clean build, I wiped everything, and openjre-8 still insists on rebuilding on the second run. I don't know whats going on here.. -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-java][PATCH 0/4] openjdk-8: update to jdk8u242ga
On 1/27/20 3:57 PM, Richard Leitner wrote: On Mon, Jan 27, 2020 at 03:10:04PM +0100, Jacob Kroon wrote: On 1/27/20 12:24 PM, Richard Leitner wrote: This series updates Openjdk 8 to the latest "ga" release 242. Successful testing has been done on a fedora-31 build host for qemuarm64, qemux86-64, qemuarm32 and armv7a. I don't know if it is related to my setup, but "bitbake openjre-8" insists on rebuilding itself every time I run it, and I don't see any sstate cache being generated for it. I just tried to reproduce it, but neither for armv7a nor for aarch64 openjre-8 is rebuiled on the latest poky/oe-core master when nothing was changed in meta-java. So in your case it also rebuilds if you call "bitbake openjre-8" directly one-after-another without any change to the layers? Yes, precisely. But.. I think you can ignore this. I suspect I have confused bitbake's hashequiv database by pruning old cache-files using the sstate-cache-management.sh script. Joshua, do you know if that script is supported when using hashequiv ? Gonna try and remove cache/ aswell, and see if that helps. Jacob -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-java][PATCH 0/4] openjdk-8: update to jdk8u242ga
On 1/27/20 12:24 PM, Richard Leitner wrote: This series updates Openjdk 8 to the latest "ga" release 242. Successful testing has been done on a fedora-31 build host for qemuarm64, qemux86-64, qemuarm32 and armv7a. I don't know if it is related to my setup, but "bitbake openjre-8" insists on rebuilding itself every time I run it, and I don't see any sstate cache being generated for it. /Jacob -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-java][PATCH 3/4] openjre-8: fix debuginfo fixup in do_install
On 1/27/20 12:37 PM, André Draszik wrote:
Hi,
On Mon, 2020-01-27 at 12:24 +0100, Richard Leitner wrote:
Based on a patch sent by Jacob Kroon.
Signed-off-by: Richard Leitner
Cc: Jacob Kroon
---
recipes-core/openjdk/openjre-8_242.bb | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/recipes-core/openjdk/openjre-8_242.bb
b/recipes-core/openjdk/openjre-8_242.bb
index 4ed2e6b..2c8a6a5 100644
--- a/recipes-core/openjdk/openjre-8_242.bb
+++ b/recipes-core/openjdk/openjre-8_242.bb
@@ -11,7 +11,7 @@ do_install() {
cp -rp ${B}/images/j2re-image/* ${D}${JRE_HOME}
chown -R root:root ${D}${JRE_HOME}
install -m644 ${WORKDIR}/jvm.cfg ${D}${JRE_HOME}/lib/${JDK_ARCH}/
-find ${D}${JDK_HOME} -name "*.debuginfo" -exec rm {} \;
+[ -d ${D}${JDK_HOME} ] && find ${D}${JDK_HOME} -name "*.debuginfo" -exec
rm {} \;
Does this line make sense? The JRE build should not produce any files in the
JDK_HOME
directory, I think?
So the only 'fix' this is doing is to silence 'find' to not print a warning
about a
directory it can't find?
It wasn't a warning - the build fails.
Shouldn't the line be dropped completely instead?
Should we worry if the JRE build (sometimes?) places files into JDK_HOME?
Right, I'm all for removing the line completely if this is the case.
Cheers,
Jacob
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-java][g0hl1n/wip-jdk8u242ga][PATCH] Fix building openjre-8 242
On 1/26/20 12:26 PM, Jacob Kroon wrote: On 1/26/20 12:21 PM, Richard Leitner wrote: On Sun, Jan 26, 2020 at 12:10:35PM +0100, Jacob Kroon wrote: Hi Richard, I need something like this in order to build on my Fedora 31 host. I'm using master branches of bitbake, oe-core, meta-openembedded. Ok, then please just ignore the mail I sent some minutes ago ;-) Good thing is the crash I reported earlier seems to be fixed. Great! I'll do some tests on my side and give you feedback afterwards. Is it OK for you if i rename the "fix-build.patch" to something more meaningful? Yes, feel free to rename/rework anything here to your liking; this was just hacked together to get the build going. Cheers, Jacob Btw, as I mentioned my build host is an up2date Fedora 31, x86-64. Any idea why these errors aren't showing up on your test builds ? Cheers, Jacob -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-java][g0hl1n/wip-jdk8u242ga][PATCH] Fix building openjre-8 242
On 1/26/20 12:21 PM, Richard Leitner wrote: On Sun, Jan 26, 2020 at 12:10:35PM +0100, Jacob Kroon wrote: Hi Richard, I need something like this in order to build on my Fedora 31 host. I'm using master branches of bitbake, oe-core, meta-openembedded. Ok, then please just ignore the mail I sent some minutes ago ;-) Good thing is the crash I reported earlier seems to be fixed. Great! I'll do some tests on my side and give you feedback afterwards. Is it OK for you if i rename the "fix-build.patch" to something more meaningful? Yes, feel free to rename/rework anything here to your liking; this was just hacked together to get the build going. Cheers, Jacob -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-java] openjre-8: java from master/master-next crashes on startup
On 1/26/20 12:11 PM, Richard Leitner wrote: On Fri, Jan 24, 2020 at 01:28:52PM +0100, Jacob Kroon wrote: On 1/24/20 12:44 PM, Richard Leitner wrote: On Fri, Jan 24, 2020 at 12:37:58PM +0100, Jacob Kroon wrote: On 1/24/20 11:37 AM, Richard Leitner wrote: ... I've just pushed a temporary branch named "g0hl1n/wip-jdk8u242ga" to the meta-java repository [1]. It contains (as you might have guessed) the update to openjdk 8u242ga. Can you please give it a try if it works for you? I tried but I get a linker error during openjre-8_242.bb:do_compile: Making adlc ... /tmp/ramdisk/build/hosttools/ld: cannot find -lstdc++ collect2: error: ld returned 1 exit status This is using MACHINE=qemux86-64. Gonna try and fix and get the build going. Do you have "libstdc++ static" installed on your buildhost as stated in the README? ...the error let me suggest you don't have it ;-) Oops, yeah I missed that. That got me further, but I got a new linking error: As this happens in the "native" part of the openjre build: What distribution and gcc version are you using on your host? For me (Fedora 31) the openjdk build and runtime tests for aarch32, aarch64 and x86_64 work. [cut] Odd.. I'm on plain Fedora 31, with Fedora's regular gcc 9.2.1. I sent a patch to the ml that fixes the build for me, forgot to CC you. Maybe I have some extra headers installed that triggers something extra to build.. Cheers, Jacob -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-java][g0hl1n/wip-jdk8u242ga][PATCH] Fix building openjre-8 242
Hi Richard,
I need something like this in order to build on my Fedora 31 host.
I'm using master branches of bitbake, oe-core, meta-openembedded.
Good thing is the crash I reported earlier seems to be fixed.
Cheers,
Jacob
Signed-off-by: Jacob Kroon
---
recipes-core/openjdk/openjdk-8-release-common.inc | 1 +
recipes-core/openjdk/openjre-8_242.bb | 2 +-
.../openjdk/patches-openjdk-8/fix-build.patch | 13 +
3 files changed, 15 insertions(+), 1 deletion(-)
create mode 100644 recipes-core/openjdk/patches-openjdk-8/fix-build.patch
diff --git a/recipes-core/openjdk/openjdk-8-release-common.inc
b/recipes-core/openjdk/openjdk-8-release-common.inc
index a8a653b..bd1b9c0 100644
--- a/recipes-core/openjdk/openjdk-8-release-common.inc
+++ b/recipes-core/openjdk/openjdk-8-release-common.inc
@@ -9,6 +9,7 @@ PATCHES_URI = "\
file://0008-autoconf-fix-shark-build-common.patch \
file://0009-prevent-debuginfo-in-favour-of-openembedded-package-.patch \
file://0010-autoconf-remove-shell-variables-from-autoheader.patch \
+file://fix-build.patch \
${HOTSPOT_UB_PATCH} \
file://2001-jdk-comparison-between-pointer-and-integer.patch \
file://2002-jdk-Allow-using-a-system-installed-libjpeg.patch \
diff --git a/recipes-core/openjdk/openjre-8_242.bb
b/recipes-core/openjdk/openjre-8_242.bb
index 4ed2e6b..2c8a6a5 100644
--- a/recipes-core/openjdk/openjre-8_242.bb
+++ b/recipes-core/openjdk/openjre-8_242.bb
@@ -11,7 +11,7 @@ do_install() {
cp -rp ${B}/images/j2re-image/* ${D}${JRE_HOME}
chown -R root:root ${D}${JRE_HOME}
install -m644 ${WORKDIR}/jvm.cfg ${D}${JRE_HOME}/lib/${JDK_ARCH}/
-find ${D}${JDK_HOME} -name "*.debuginfo" -exec rm {} \;
+[ -d ${D}${JDK_HOME} ] && find ${D}${JDK_HOME} -name "*.debuginfo" -exec
rm {} \;
}
FILES_${PN}_append = "\
diff --git a/recipes-core/openjdk/patches-openjdk-8/fix-build.patch
b/recipes-core/openjdk/patches-openjdk-8/fix-build.patch
new file mode 100644
index 000..e35d0eb
--- /dev/null
+++ b/recipes-core/openjdk/patches-openjdk-8/fix-build.patch
@@ -0,0 +1,13 @@
+Index: jdk8u-jdk8u242-ga/jdk/make/gensrc/GensrcMisc.gmk
+===
+--- jdk8u-jdk8u242-ga.orig/jdk/make/gensrc/GensrcMisc.gmk
jdk8u-jdk8u242-ga/jdk/make/gensrc/GensrcMisc.gmk
+@@ -76,7 +76,7 @@ $(eval $(call SetupNativeCompilation,BUI
+ INCLUDE_FILES := $(GENSRC_SOR_SRC_FILE), \
+ LANG := C, \
+ CC := $(BUILD_CC), \
+-LDEXE := $(BUILD_LD), \
++LDEXE := $(BUILD_CC), \
+ OBJECT_DIR := $(GENSRC_SOR_BIN), \
+ OUTPUT_DIR := $(GENSRC_SOR_BIN), \
+ PROGRAM := genSocketOptionRegistry))
--
2.24.1
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-java] openjre-8: java from master/master-next crashes on startup
On 1/24/20 12:44 PM, Richard Leitner wrote: On Fri, Jan 24, 2020 at 12:37:58PM +0100, Jacob Kroon wrote: On 1/24/20 11:37 AM, Richard Leitner wrote: ... I've just pushed a temporary branch named "g0hl1n/wip-jdk8u242ga" to the meta-java repository [1]. It contains (as you might have guessed) the update to openjdk 8u242ga. Can you please give it a try if it works for you? I tried but I get a linker error during openjre-8_242.bb:do_compile: Making adlc ... /tmp/ramdisk/build/hosttools/ld: cannot find -lstdc++ collect2: error: ld returned 1 exit status This is using MACHINE=qemux86-64. Gonna try and fix and get the build going. Do you have "libstdc++ static" installed on your buildhost as stated in the README? ...the error let me suggest you don't have it ;-) Oops, yeah I missed that. That got me further, but I got a new linking error: | Linking executable genSocketOptionRegistry | gensrc/GensrcMisc.gmk:75: Building /tmp/ramdisk/build/work/core2-64-oe-linux/openjre-8/242-r0/build/jdk/btnative/genSocketOptionRegistry/genSocketOptionRegistry (from /tmp/ramdisk/build/work/core2-64-oe-linux/openjre-8/242-r0/build/jdk/btnative/genSocketOptionRegistry/genSocketOptionRegi stry.o) (/tmp/ramdisk/build/work/core2-64-oe-linux/openjre-8/242-r0/build/jdk/btnative/genSocketOptionRegistry/genSocketOptionRegistry.o newer) | + /tmp/ramdisk/build/hosttools/gcc /tmp/ramdisk/build/work/core2-64-oe-linux/openjre-8/242-r0/build/jdk/btnative/genUnixConstants/genUnixConstants.o -o /tmp/ramdisk/build/work/core2-64-oe-linux/openjre-8/242-r0/build/jdk/btnative/genUnixConstants/genUnixConstants | + /tmp/ramdisk/build/hosttools/ld /tmp/ramdisk/build/work/core2-64-oe-linux/openjre-8/242-r0/build/jdk/btnative/genSocketOptionRegistry/genSocketOptionRegistry.o -o /tmp/ramdisk/build/work/core2-64-oe-linux/openjre-8/242-r0/build/jdk/btnative/genSocketOptionRegistry/genSocket OptionRegistry | /tmp/ramdisk/build/hosttools/ld: warning: cannot find entry symbol _start; defaulting to 00401000 | /tmp/ramdisk/build/hosttools/ld: /tmp/ramdisk/build/work/core2-64-oe-linux/openjre-8/242-r0/build/jdk/btnative/genSocketOptionRegistry/genSocketOptionRegistry.o: in function `out': | genSocketOptionRegistry.c:(.text+0x14): undefined reference to `puts' | /tmp/ramdisk/build/hosttools/ld: /tmp/ramdisk/build/work/core2-64-oe-linux/openjre-8/242-r0/build/jdk/btnative/genSocketOptionRegistry/genSocketOptionRegistry.o: in function `emit': | genSocketOptionRegistry.c:(.text+0x47): undefined reference to `printf' | /tmp/ramdisk/build/hosttools/ld: genSocketOptionRegistry.c:(.text+0x5e): undefined reference to `printf' | make[2]: *** [gensrc/GensrcMisc.gmk:75: /tmp/ramdisk/build/work/core2-64-oe-linux/openjre-8/242-r0/build/jdk/btnative/genSocketOptionRegistry/genSocketOptionRegistry] Error 1 | make[2]: *** Waiting for unfinished jobs | Note: Some input files use unchecked or unsafe operations. | Note: Recompile with -Xlint:unchecked for details. | + /tmp/ramdisk/build/hosttools/mv /tmp/ramdisk/build/work/core2-64-oe-linux/openjre-8/242-r0/build/jdk/btclasses/_the.BUILD_TOOLS_batch.tmp /tmp/ramdisk/build/work/core2-64-oe-linux/openjre-8/242-r0/build/jdk/btclasses/_the.BUILD_TOOLS_batch | make[1]: *** [BuildJdk.gmk:55: gensrc-only] Error 2 | make: *** [/tmp/ramdisk/build/work/core2-64-oe-linux/openjre-8/242-r0/jdk8u-jdk8u242-ga//make/Main.gmk:117: jdk-only] Error 2 | WARNING: /tmp/ramdisk/build/work/core2-64-oe-linux/openjre-8/242-r0/temp/run.do_compile.384980:1 exit 1 from 'exit 1' -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-java] openjre-8: java from master/master-next crashes on startup
Hi Richard, On 1/24/20 11:37 AM, Richard Leitner wrote: Hi Jacob, On Tue, Jan 21, 2020 at 09:01:01AM +0100, Jacob Kroon wrote: On 1/21/20 8:12 AM, Richard Leitner wrote: On Tue, Jan 21, 2020 at 08:00:48AM +0100, Jacob Kroon wrote: Hi Richard, I still can't use openjre-8 in either warrior, zeus or master builds, due to the crash below. I wonder, is anyone successfully using openjre-8 in either of these releases ? I'm doing tests for the machine/distro combinations stated in the README before pushing to master. Therefore it should have worked back then. Nonetheless I'm currently working on updating openjdk-8 to the latest ga and implementing continous (e.g. nightly/weekly) builds/tests on qemu to make sure nothing broke over time. So maybe I could get something out of this. Thanks for doing this work. I've just pushed a temporary branch named "g0hl1n/wip-jdk8u242ga" to the meta-java repository [1]. It contains (as you might have guessed) the update to openjdk 8u242ga. Can you please give it a try if it works for you? I tried but I get a linker error during openjre-8_242.bb:do_compile: Making adlc ... /tmp/ramdisk/build/hosttools/ld: cannot find -lstdc++ collect2: error: ld returned 1 exit status This is using MACHINE=qemux86-64. Gonna try and fix and get the build going. Cheers, Jacob -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-java] openjre-8: java from master/master-next crashes on startup
On 1/21/20 12:38 PM, Richard Leitner wrote: On Tue, Jan 21, 2020 at 09:01:01AM +0100, Jacob Kroon wrote: On 1/21/20 8:12 AM, Richard Leitner wrote: On Tue, Jan 21, 2020 at 08:00:48AM +0100, Jacob Kroon wrote: ... I see the crash both in my custom machine and in qemux86-64. If it makes a difference, I'm using PREFERRED_PROVIDER_virtual/java-initial-native = "cacao-initial-native" PREFERRED_PROVIDER_virtual/java-native = "cacao-native" since building with jamvm has always been broken for me. Any hints ? Nothing that comes to my mind directly, but I'll take a look into this. So your "custom machine" is also x86-64? The custom machine .conf file is just: -- require conf/machine/include/tune-i686.inc require conf/machine/include/x86-base.inc -- So 32bit x86. Is your JVM crashing in interpreted and mixed mode? It only crashes in mixed mode. If I pass -Xint I don't see any issues. -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-java] openjre-8: java from master/master-next crashes on startup
Hi Richard, On 1/21/20 8:12 AM, Richard Leitner wrote: Hi Jacob, On Tue, Jan 21, 2020 at 08:00:48AM +0100, Jacob Kroon wrote: Hi Richard, I still can't use openjre-8 in either warrior, zeus or master builds, due to the crash below. I wonder, is anyone successfully using openjre-8 in either of these releases ? I'm doing tests for the machine/distro combinations stated in the README before pushing to master. Therefore it should have worked back then. Nonetheless I'm currently working on updating openjdk-8 to the latest ga and implementing continous (e.g. nightly/weekly) builds/tests on qemu to make sure nothing broke over time. So maybe I could get something out of this. Thanks for doing this work. I see the crash both in my custom machine and in qemux86-64. If it makes a difference, I'm using PREFERRED_PROVIDER_virtual/java-initial-native = "cacao-initial-native" PREFERRED_PROVIDER_virtual/java-native = "cacao-native" since building with jamvm has always been broken for me. Any hints ? Nothing that comes to my mind directly, but I'll take a look into this. So your "custom machine" is also x86-64? The custom machine .conf file is just: -- require conf/machine/include/tune-i686.inc require conf/machine/include/x86-base.inc -- So 32bit x86. Cheers, Jacob -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-java] openjre-8: java from master/master-next crashes on startup
Hi Richard,
I still can't use openjre-8 in either warrior, zeus or master builds,
due to the crash below. I wonder, is anyone successfully using openjre-8
in either of these releases ?
I see the crash both in my custom machine and in qemux86-64. If it makes
a difference, I'm using
PREFERRED_PROVIDER_virtual/java-initial-native = "cacao-initial-native"
PREFERRED_PROVIDER_virtual/java-native = "cacao-native"
since building with jamvm has always been broken for me.
Any hints ?
Cheers,
Jacob
On 10/13/19 11:33 AM, Jacob Kroon wrote:
Hi,
For me, this has been the status for quite a while now. First of all,
openjre-8 from master/master-next doesn't build on my Fedora 30 box,
I've been using the patch attached to this mail to get it to build.
Secondly, java always crashes on startup here:
#0 Node::Node (this=0x805fa88, req=3)
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/opto/node.cpp:327
#1 0xb7499b63 in RegionNode::RegionNode (required=3, this=0x805fa88)
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/opto/cfgnode.hpp:73
#2 LoopNode::LoopNode (backedge=0x0, entry=0x0, this=0x805fa88)
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/opto/loopnode.hpp:88
#3 RootNode::RootNode (this=0x805fa88)
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/opto/rootnode.hpp:37
#4 Compile::Init (this=0x977feaf8, aliaslevel=0)
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/opto/compile.cpp:1079
#5 0xb74a02c5 in Compile::Compile (this=0x977feaf8,
ci_env=0x977ff188, generator=0xb7895270
, stub_function=0xb7897070
"U\211\345WVS\350\245+\237\377\201\303\205\217\062",
stub_name=0xb7a0a62f "_new_instance_Java", is_fancy_jump=0,
pass_tls=true, save_arg_registers=false, return_pc=false)
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/opto/compile.cpp:1012
#6 0xb789587b in OptoRuntime::generate_stub (env=0x977ff188,
gen=0xb7895270 ,
C_function=0xb7897070 JavaThread*)> "U\211\345WVS\350\245+\237\377\201\303\205\217\062",
name=0xb7a0a62f "_new_instance_Java", is_fancy_jump=0,
pass_tls=true, save_argument_registers=false, return_pc=false)
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/opto/runtime.cpp:181
#7 0xb7895a5f in OptoRuntime::generate (env=0x977ff188)
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/opto/runtime.cpp:147
#8 0xb7417de9 in C2Compiler::init_c2_runtime ()
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/runtime/thread.hpp:1854
#9 0xb7417e35 in C2Compiler::initialize (this=0xb706b180)
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/opto/c2compiler.cpp:102
#10 0xb74a508c in CompileBroker::init_compiler_runtime ()
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/compiler/compileBroker.cpp:1674
#11 0xb74ab6ed in CompileBroker::compiler_thread_loop ()
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/compiler/compileBroker.cpp:1772
#12 0xb79238d5 in compiler_thread_entry (thread=0xb7079400,
__the_thread__=0xb7079400)
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/runtime/thread.cpp:3230
#13 0xb792d782 in JavaThread::thread_main_inner (this=0xb7079400)
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/runtime/thread.hpp:1035
#14 0xb7820ce4 in java_start (thread=0xb7079400)
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/os/linux/vm/os_linux.cpp:790
#15 0x4121244c in start_thread (arg=) at
pthread_create.c:479
#16 0x4113bd56 in clone () at
../sysdeps/unix/sysv/linux/i386/clone.S:108
The Node::Node() constructor uses an IDX_INIT() macro:
#define IDX_INIT(req) this->Init((req), (Compile*) this->_out)
and Node::Init() does:
297 inline int Node::Init(int req, Compile* C) {
298 assert(Compile::current() == C, "must use operator
new(Compile*)");
299 int idx = C->next_unique();
and here, this->_out = C = null, which I think leads to a segfault on
line 299.
I've tried looking at upstream jdk8u but can't find anything. Does
anyone have an idea of whats going on here, or if I should raise the
issue to openjdk ml ?
Cheers,
Jacob
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-java][PATCH] cacao: Use python3 in java wrapper script
OE-Core has dropped python2 so switch to using python3.
Signed-off-by: Jacob Kroon
---
.../cacao-1.6.1-do-not-rely-on-absolute-paths.patch | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git
a/recipes-core/cacao/cacao-2d6f6c14daf9/cacao-1.6.1-do-not-rely-on-absolute-paths.patch
b/recipes-core/cacao/cacao-2d6f6c14daf9/cacao-1.6.1-do-not-rely-on-absolute-paths.patch
index 2b046cf..665377b 100644
---
a/recipes-core/cacao/cacao-2d6f6c14daf9/cacao-1.6.1-do-not-rely-on-absolute-paths.patch
+++
b/recipes-core/cacao/cacao-2d6f6c14daf9/cacao-1.6.1-do-not-rely-on-absolute-paths.patch
@@ -27,8 +27,8 @@ index 0790f02..2118de7 100644
+
+# if this shell script is invoked through symlink, we get invalid paths. As
getting the real path to this file is
+# very painful using only /bin/sh, python is used instead.
-+PYTHON_STR="import os; print
os.path.dirname(os.path.realpath(os.readlink(\"${CURRENT_DIR}/java\"))) if
os.path.islink(\"${CURRENT_DIR}/java\") else \"${CURRENT_DIR}\""
-+REAL_DIR=`python -c "${PYTHON_STR}"`
++PYTHON_STR="import os;
print(os.path.dirname(os.path.realpath(os.readlink(\"${CURRENT_DIR}/java\")))
if os.path.islink(\"${CURRENT_DIR}/java\") else \"${CURRENT_DIR}\")"
++REAL_DIR=`python3 -c "${PYTHON_STR}"`
+
+PARENT_DIR=`dirname ${REAL_DIR}`
+
--
2.23.0
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-java] openjre-8: java from master/master-next crashes on startup
Hi,
For me, this has been the status for quite a while now. First of all,
openjre-8 from master/master-next doesn't build on my Fedora 30 box,
I've been using the patch attached to this mail to get it to build.
Secondly, java always crashes on startup here:
#0 Node::Node (this=0x805fa88, req=3)
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/opto/node.cpp:327
#1 0xb7499b63 in RegionNode::RegionNode (required=3, this=0x805fa88)
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/opto/cfgnode.hpp:73
#2 LoopNode::LoopNode (backedge=0x0, entry=0x0, this=0x805fa88)
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/opto/loopnode.hpp:88
#3 RootNode::RootNode (this=0x805fa88)
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/opto/rootnode.hpp:37
#4 Compile::Init (this=0x977feaf8, aliaslevel=0)
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/opto/compile.cpp:1079
#5 0xb74a02c5 in Compile::Compile (this=0x977feaf8, ci_env=0x977ff188,
generator=0xb7895270 ,
stub_function=0xb7897070 "U\211\345WVS\350\245+\237\377\201\303\205\217\062",
stub_name=0xb7a0a62f "_new_instance_Java", is_fancy_jump=0, pass_tls=true,
save_arg_registers=false, return_pc=false)
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/opto/compile.cpp:1012
#6 0xb789587b in OptoRuntime::generate_stub (env=0x977ff188,
gen=0xb7895270 ,
C_function=0xb7897070 "U\211\345WVS\350\245+\237\377\201\303\205\217\062", name=0xb7a0a62f "_new_instance_Java",
is_fancy_jump=0, pass_tls=true, save_argument_registers=false, return_pc=false)
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/opto/runtime.cpp:181
#7 0xb7895a5f in OptoRuntime::generate (env=0x977ff188)
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/opto/runtime.cpp:147
#8 0xb7417de9 in C2Compiler::init_c2_runtime ()
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/runtime/thread.hpp:1854
#9 0xb7417e35 in C2Compiler::initialize (this=0xb706b180)
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/opto/c2compiler.cpp:102
#10 0xb74a508c in CompileBroker::init_compiler_runtime ()
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/compiler/compileBroker.cpp:1674
#11 0xb74ab6ed in CompileBroker::compiler_thread_loop ()
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/compiler/compileBroker.cpp:1772
#12 0xb79238d5 in compiler_thread_entry (thread=0xb7079400,
__the_thread__=0xb7079400)
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/runtime/thread.cpp:3230
#13 0xb792d782 in JavaThread::thread_main_inner (this=0xb7079400)
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/share/vm/runtime/thread.hpp:1035
#14 0xb7820ce4 in java_start (thread=0xb7079400)
at
/usr/src/debug/openjre-8/172b11-r0/jdk8u-33d274a7dda0/hotspot/src/os/linux/vm/os_linux.cpp:790
#15 0x4121244c in start_thread (arg=) at pthread_create.c:479
#16 0x4113bd56 in clone () at ../sysdeps/unix/sysv/linux/i386/clone.S:108
The Node::Node() constructor uses an IDX_INIT() macro:
#define IDX_INIT(req) this->Init((req), (Compile*) this->_out)
and Node::Init() does:
297 inline int Node::Init(int req, Compile* C) {
298 assert(Compile::current() == C, "must use operator new(Compile*)");
299 int idx = C->next_unique();
and here, this->_out = C = null, which I think leads to a segfault on
line 299.
I've tried looking at upstream jdk8u but can't find anything. Does
anyone have an idea of whats going on here, or if I should raise the
issue to openjdk ml ?
Cheers,
Jacob
From: Jacob Kroon
Date: Fri, 2 Aug 2019 15:59:52 +0200
Subject: [PATCH] Fix build
---
recipes-core/icedtea/icedtea7-native.inc | 2 ++
.../icedtea/openjdk-7-03b147/fix-build.patch | 22 +++
2 files changed, 24 insertions(+)
create mode 100644 recipes-core/icedtea/openjdk-7-03b147/fix-build.patch
diff --git a/recipes-core/icedtea/icedtea7-native.inc b/recipes-core/icedtea/icedtea7-native.inc
index b578cb2..522244e 100644
--- a/recipes-core/icedtea/icedtea7-native.inc
+++ b/recipes-core/icedtea/icedtea7-native.inc
@@ -49,6 +49,7 @@ SRC_URI = " \
file://timezoneszip.patch \
file://sigsegv.patch;apply=no \
file://vframeArray_Fix_sigsegv.patch;apply=no \
+ file://fix-build.patch;apply=no \
"
S = "${WORKDIR}/${ICEDTEA}"
@@ -161,6 +162,7 @@ do_configure_append() {
patch -p1 < ${WORKDIR}/jaxws_fix_NullPointerException.patch
patch -p1 < ${WORKDIR}/sigsegv.patch
patch -p1 < ${WORKDIR}/vframeArray_Fix_sigsegv.patch
+ patch -p1 <
Re: [oe] [meta-java][PATCH] openjdk-8: fix build on machines with linux kernel 5.0+
On 8/21/19 3:46 PM, Matthias Schoepfer via Openembedded-devel wrote:
Signed-off-by: Matthias Schoepfer
---
.../openjdk-8-release-172b11-common.inc | 1 +
...fixing-build-for-kernel-versions-5.x.patch | 27 +++
2 files changed, 28 insertions(+)
create mode 100644
recipes-core/openjdk/patches-openjdk-8/0001-fixing-build-for-kernel-versions-5.x.patch
diff --git a/recipes-core/openjdk/openjdk-8-release-172b11-common.inc
b/recipes-core/openjdk/openjdk-8-release-172b11-common.inc
index ad20585..eb7e38a 100644
--- a/recipes-core/openjdk/openjdk-8-release-172b11-common.inc
+++ b/recipes-core/openjdk/openjdk-8-release-172b11-common.inc
@@ -21,6 +21,7 @@ PATCHES_URI = "\
${LIBPNG_NEON_PATCH} \
file://handle_extra_output.patch \
file://hotspot-handle-gcc7-format-overflow.patch \
+file://0001-fixing-build-for-kernel-versions-5.x.patch \
"
HOTSPOT_UB_PATCH = "\
file://0014-hotspot-zero-fix-undefined-behaviour-gcc-v8-fix.patch \
diff --git
a/recipes-core/openjdk/patches-openjdk-8/0001-fixing-build-for-kernel-versions-5.x.patch
b/recipes-core/openjdk/patches-openjdk-8/0001-fixing-build-for-kernel-versions-5.x.patch
new file mode 100644
index 000..9b17b66
--- /dev/null
+++
b/recipes-core/openjdk/patches-openjdk-8/0001-fixing-build-for-kernel-versions-5.x.patch
@@ -0,0 +1,27 @@
+From 51b022bcc596aa85e72e8d8bb93ffc3be843b8b9 Mon Sep 17 00:00:00 2001
+From: Matthias Schoepfer
+Date: Wed, 21 Aug 2019 14:33:02 +0200
+Subject: [PATCH] fixing build for kernel versions 5.x
+
+This simply extends the possibility of having a version 5+ linux kernel
+
+
+Signed-off-by: Matthias Schoepfer
+
+---
+ hotspot/make/linux/Makefile | 2 +-
+ 1 file changed, 1 insertion(+), 1 deletion(-)
+
+diff --git a/hotspot/make/linux/Makefile b/hotspot/make/linux/Makefile
+index 17c14097..02cb0b33 100644
+--- a/hotspot/make/linux/Makefile
b/hotspot/make/linux/Makefile
+@@ -233,7 +233,7 @@ checks: check_os_version check_j2se_version
+ # Solaris 2.5.1, 2.6).
+ # Disable this check by setting DISABLE_HOTSPOT_OS_VERSION_CHECK=ok.
+
+-SUPPORTED_OS_VERSION = 2.4% 2.5% 2.6% 3% 4%
++SUPPORTED_OS_VERSION = 2.4% 2.5% 2.6% 3% 4% 5%
+ OS_VERSION := $(shell uname -r)
+ EMPTY_IF_NOT_SUPPORTED = $(filter $(SUPPORTED_OS_VERSION),$(OS_VERSION))
+
This is already fixed in master-next.
/me wonders when those patches will be merged to master.
/Jacob
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-java] pyro, rocko and per-recipe sysroots
Hi Otavio, On Fri, Dec 22, 2017 at 1:26 PM, Otavio Salvador <otavio.salva...@ossystems.com.br> wrote: > On Fri, Dec 22, 2017 at 7:30 AM, Jacob Kroon <jacob.kr...@gmail.com> wrote: >> Hi, >> >> On Fri, Dec 22, 2017 at 8:55 AM, Jacob Kroon <jacob.kr...@gmail.com> wrote: >>> Hi all, >>> >>> I'm trying to get a recent yocto build together with meta-java up and >>> running, but ran into some problems: >>> >>> 1. For rocko, I don't see the corresponding branch in meta-java, >>> should I use "master" ? >>> 2. For pyro, there is a branch in meta-java, but the build fails >>> unless I backport the patch in commit >>> c4b9b97c7c60ad5c0c3392ccf143b2e4e6c7df24: >>> >>> "java-library.bbclass: compatibility with per-recipe sysroots" >>> >>> Could we get this patch officially backported to pyro ? >>> >> >> One more patch I needed backported for pyro to build: >> >> 2a706cd2d656a203ecb83bef1c687703c969732e >> "icedtea-native: Fix segmentation build during build" >> >> since I'm on Fedora with a recent gcc7 I assume. > > I backported both for pyro; please try rocko using master ... I've > been using it without problem for few customers. Thanks, with those pyro is building fine now, will try rocko with meta-java/master next. /Jacob -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-java] pyro, rocko and per-recipe sysroots
Hi, On Fri, Dec 22, 2017 at 8:55 AM, Jacob Kroon <jacob.kr...@gmail.com> wrote: > Hi all, > > I'm trying to get a recent yocto build together with meta-java up and > running, but ran into some problems: > > 1. For rocko, I don't see the corresponding branch in meta-java, > should I use "master" ? > 2. For pyro, there is a branch in meta-java, but the build fails > unless I backport the patch in commit > c4b9b97c7c60ad5c0c3392ccf143b2e4e6c7df24: > > "java-library.bbclass: compatibility with per-recipe sysroots" > > Could we get this patch officially backported to pyro ? > One more patch I needed backported for pyro to build: 2a706cd2d656a203ecb83bef1c687703c969732e "icedtea-native: Fix segmentation build during build" since I'm on Fedora with a recent gcc7 I assume. /Jacob -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-java] pyro, rocko and per-recipe sysroots
Hi all, I'm trying to get a recent yocto build together with meta-java up and running, but ran into some problems: 1. For rocko, I don't see the corresponding branch in meta-java, should I use "master" ? 2. For pyro, there is a branch in meta-java, but the build fails unless I backport the patch in commit c4b9b97c7c60ad5c0c3392ccf143b2e4e6c7df24: "java-library.bbclass: compatibility with per-recipe sysroots" Could we get this patch officially backported to pyro ? Thanks Jacob -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] Cross canadian gcc configuration
Hi Alfonso,
On Sun, Dec 14, 2014 at 8:46 AM, Alfonso Tamés alfo...@tames.com wrote:
When I try to compile a hello world program with the
arm-poky-linux-gnueabi-gcc compiler, first I get an error not finding
stdio.h, and when I pass the directory through the -I flag, I get the
classicstubs.h:7:29: fatal error: gnu/stubs-soft.h: No such file or
directory error which I supposed is being thrown since gcc cross canadian
is not being configured by default (as my image) hard float.
I noticed in the gcc-cross-canadian.inc that the --with-gxx-include-dir
and --with-sysroot in EXTRA_OECONF_PATHS have a /not/exist string:
EXTRA_OECONF_PATHS = \
--with-gxx-include-dir=/not/exist${target_includedir}/c++/${BINV} \
--with-build-time-tools=${STAGING_DIR_NATIVE}${prefix_native}/${TARGET_SYS}/bin
\
--with-sysroot=/not/exist \
--with-build-sysroot=${STAGING_DIR_TARGET} \
The sysroot is poisoned by intention. Try sourcing the environment script
from the SDK, and use $CC instead of invoking the compiler binary directly.
See git commit 04b725511a505c582a3abdf63d096967f0320779 in
openembedded-core.
Jacob
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-musl][PATCH] README.md: Fix typo in command-line example
Chosen C library is set with TCLIBC, not LIBC. Signed-off-by: Jacob Kroon jacob.kr...@mikrodidakt.se --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 7db8c8d..6c74799 100644 --- a/README.md +++ b/README.md @@ -28,7 +28,7 @@ BBLAYERS ?= \ Below we build for qemuarm machine as an example ```shell -$ MACHINE=qemuarm LIBC=musl bitbake core-image-minimal +$ MACHINE=qemuarm TCLIBC=musl bitbake core-image-minimal ``` # Running -- 1.9.3 -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-qt5][PATCH] qtbase: fix 0003-Add-external-hostbindir-option.patch
On Thu, Jul 10, 2014 at 6:08 AM, Jonathan Liu net...@gmail.com wrote: The isEmpty function takes a variable name instead of a value. As $$[QT_EXTERNAL_HOST_BINS] is a built-in we need to assign it to a variable to check if it is empty. If QT_EXTERNAL_HOST_BINS is empty (as is the case for generated Qt 5 toolchains), we need to fallback to QT_HOST_BINS when determining the path for tools such as moc otherwise you get the following error when building a Qt project using the generated toolchain: make[1]: /moc: Command not found Signed-off-by: Jonathan Liu net...@gmail.com --- .../0003-Add-external-hostbindir-option.patch | 65 +- .../0003-Add-external-hostbindir-option.patch | 65 +- 2 files changed, 78 insertions(+), 52 deletions(-) On my system, this change seems to cause syncqt.pl being removed from the qtbase-tools package. From buildhistory: -FILELIST = /usr/bin/qt5/moc /usr/bin/qt5/qdoc /usr/bin/qt5/qlalr /usr/bin/qt5/rcc /usr/bin/qt5/syncqt.pl +FILELIST = /usr/bin/qt5/moc /usr/bin/qt5/qdoc /usr/bin/qt5/qlalr /usr/bin/qt5/rcc Is this by intention ? The package still RDEPENDS on perl. /Jacob -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH] gnokii: Add missing intltool-native dependency
Building from sstate-cache fails otherwise.
Fixes:
| NOTE: Executing intltoolize --copy --force --automake
| .../gnokii/0.6.31-r0/temp/run.do_configure.6895: line 181:
intltoolize: command not found
Signed-off-by: Jacob Kroon jacob.kr...@mikrodidakt.se
---
meta-oe/recipes-connectivity/gnokii/gnokii_0.6.31.bb | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/meta-oe/recipes-connectivity/gnokii/gnokii_0.6.31.bb
b/meta-oe/recipes-connectivity/gnokii/gnokii_0.6.31.bb
index 4826452..de6e86c 100644
--- a/meta-oe/recipes-connectivity/gnokii/gnokii_0.6.31.bb
+++ b/meta-oe/recipes-connectivity/gnokii/gnokii_0.6.31.bb
@@ -5,7 +5,7 @@ LIC_FILES_CHKSUM =
file://COPYING;md5=0636e73ff0215e8d672dc4c32c317bb3
SRC_URI = http://www.gnokii.org/download/gnokii/gnokii-${PV}.tar.bz2;
-DEPENDS = glib-2.0
+DEPENDS = glib-2.0 intltool-native
X11DEPENDS = libxpm gtk+
PACKAGECONFIG ??= ${@base_contains('DISTRO_FEATURES', 'x11', 'x11', '', d)}
--
1.9.3
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 1/2] ccid: Fix floating runtime dependency on flex
Signed-off-by: Jacob Kroon jacob.kr...@mikrodidakt.se
---
.../recipes-support/ccid/ccid/no-dep-on-libfl.patch| 18 ++
meta-oe/recipes-support/ccid/ccid_1.4.8.bb | 4 +++-
2 files changed, 21 insertions(+), 1 deletion(-)
create mode 100644 meta-oe/recipes-support/ccid/ccid/no-dep-on-libfl.patch
diff --git a/meta-oe/recipes-support/ccid/ccid/no-dep-on-libfl.patch
b/meta-oe/recipes-support/ccid/ccid/no-dep-on-libfl.patch
new file mode 100644
index 000..477cc31
--- /dev/null
+++ b/meta-oe/recipes-support/ccid/ccid/no-dep-on-libfl.patch
@@ -0,0 +1,18 @@
+No need to link with libfl.
+
+Upstream-Status: Pending
+Signed-off-by: Jacob Kroon jacob.kr...@mikrodidakt.se
+
+Index: ccid-1.4.8/src/Makefile.am
+===
+--- ccid-1.4.8.orig/src/Makefile.am
ccid-1.4.8/src/Makefile.am
+@@ -53,7 +53,7 @@ PROVIDED_BY_PCSC = debug.c
+ endif
+
+ libccid_la_SOURCES = $(COMMON) $(USB) $(TOKEN_PARSER) $(PROVIDED_BY_PCSC)
$(T1)
+-libccid_la_LIBADD = $(LEXLIB) $(LIBUSB_LIBS) $(PTHREAD_LIBS)
++libccid_la_LIBADD = $(LIBUSB_LIBS) $(PTHREAD_LIBS)
+ libccid_la_CFLAGS = $(PCSC_CFLAGS) $(LIBUSB_CFLAGS) $(PTHREAD_CFLAGS) \
+ $(SYMBOL_VISIBILITY) -D$(CCID_VERSION) -DSIMCLIST_NO_DUMPRESTORE
+ libccid_la_LDFLAGS = -avoid-version
diff --git a/meta-oe/recipes-support/ccid/ccid_1.4.8.bb
b/meta-oe/recipes-support/ccid/ccid_1.4.8.bb
index 14f8875..078f785 100644
--- a/meta-oe/recipes-support/ccid/ccid_1.4.8.bb
+++ b/meta-oe/recipes-support/ccid/ccid_1.4.8.bb
@@ -6,7 +6,9 @@ LIC_FILES_CHKSUM =
file://COPYING;md5=2d5025d4aa3495befef8f17206a5b0a1
DEPENDS = virtual/libusb0 pcsc-lite
RDEPENDS_${PN} = pcsc-lite
-SRC_URI =
https://alioth.debian.org/frs/download.php/file/3768/ccid-${PV}.tar.bz2;
+SRC_URI =
https://alioth.debian.org/frs/download.php/file/3768/ccid-${PV}.tar.bz2 \
+ file://no-dep-on-libfl.patch
+
SRC_URI[md5sum] = b6c37110f50b059a8ba94f118cfd679a
SRC_URI[sha256sum] =
7aca09fa4b9099e423402fd5df424adbafa502888710d1fda6015b6c1ff637ef
--
1.9.3
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 2/2] gnokii: Fix floating runtime dependencies
Add PACKAGECONFIG for bluez, libical, pcsc-lite, readline and usb.
Signed-off-by: Jacob Kroon jacob.kr...@mikrodidakt.se
---
meta-oe/recipes-connectivity/gnokii/gnokii_0.6.31.bb | 5 +
1 file changed, 5 insertions(+)
diff --git a/meta-oe/recipes-connectivity/gnokii/gnokii_0.6.31.bb
b/meta-oe/recipes-connectivity/gnokii/gnokii_0.6.31.bb
index 356d5bd..4826452 100644
--- a/meta-oe/recipes-connectivity/gnokii/gnokii_0.6.31.bb
+++ b/meta-oe/recipes-connectivity/gnokii/gnokii_0.6.31.bb
@@ -9,6 +9,11 @@ DEPENDS = glib-2.0
X11DEPENDS = libxpm gtk+
PACKAGECONFIG ??= ${@base_contains('DISTRO_FEATURES', 'x11', 'x11', '', d)}
+PACKAGECONFIG[bluez] = --enable-bluetooth,--disable-bluetooth,bluez4
+PACKAGECONFIG[libical] = --enable-libical,--disable-libical,libical
+PACKAGECONFIG[pcsc-lite] =
--enable-libpcsclite,--disable-libpcsclite,pcsc-lite
+PACKAGECONFIG[readline] = --with-readline,--without-readline,readline
+PACKAGECONFIG[usb] = --enable-libusb,--disable-libusb,virtual/libusb0
PACKAGECONFIG[x11] = ,--without-x,${X11DEPENDS}
inherit autotools pkgconfig
--
1.9.3
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 2/2] gammu: Add missing glib-2.0 and udev build dependencies
Since they are autodetected they introduce floating dependencies, so for now just require them at build time. Signed-off-by: Jacob Kroon jacob.kr...@mikrodidakt.se --- meta-oe/recipes-connectivity/gammu/gammu_1.32.0.bb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/meta-oe/recipes-connectivity/gammu/gammu_1.32.0.bb b/meta-oe/recipes-connectivity/gammu/gammu_1.32.0.bb index d123d38..0dac7cb 100644 --- a/meta-oe/recipes-connectivity/gammu/gammu_1.32.0.bb +++ b/meta-oe/recipes-connectivity/gammu/gammu_1.32.0.bb @@ -1,6 +1,6 @@ SUMMARY = GNU All Mobile Managment Utilities SECTION = console/network -DEPENDS = cmake-native virtual/libiconv libdbi mysql5 +DEPENDS = cmake-native virtual/libiconv libdbi mysql5 glib-2.0 udev LICENSE = GPLv2 LIC_FILES_CHKSUM = file://COPYING;md5=a17cb0a873d252440acfdf9b3d0e7fbf HOMEPAGE = http://www.gammu.org/; -- 1.9.3 -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 1/2] liboauth: Add missing OpenSSL build dependency
Fixes:
| liboauth-1.0.3/src/hash.c:363:26: fatal error: openssl/hmac.h:
| No such file or directory
Signed-off-by: Jacob Kroon jacob.kr...@mikrodidakt.se
---
meta-oe/recipes-support/liboauth/liboauth_1.0.3.bb | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/meta-oe/recipes-support/liboauth/liboauth_1.0.3.bb
b/meta-oe/recipes-support/liboauth/liboauth_1.0.3.bb
index 2535b8e..3311da5 100644
--- a/meta-oe/recipes-support/liboauth/liboauth_1.0.3.bb
+++ b/meta-oe/recipes-support/liboauth/liboauth_1.0.3.bb
@@ -7,7 +7,7 @@ LICENSE = MIT|GPL-2.0
LIC_FILES_CHKSUM = file://COPYING.MIT;md5=6266718a5241c045c8099d9be48817df \
file://COPYING.GPL;md5=0636e73ff0215e8d672dc4c32c317bb3
SECTION = libs
-DEPENDS = curl
+DEPENDS = curl openssl
SRC_URI = ${SOURCEFORGE_MIRROR}/project/${BPN}/${BP}.tar.gz
SRC_URI[md5sum] = 689b46c2b3ab1a39735ac33f714c4f7f
--
1.9.3
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH] glade3: Add PACKAGECONFIG for Gnome support
./configure checks for libbonoboui and libgnomeui.
Add a PACKAGECONFIG for it, enabled by default.
Signed-off-by: Jacob Kroon jacob.kr...@mikrodidakt.se
---
meta-oe/recipes-devtools/glade/glade3_3.8.2.bb | 3 +++
1 file changed, 3 insertions(+)
diff --git a/meta-oe/recipes-devtools/glade/glade3_3.8.2.bb
b/meta-oe/recipes-devtools/glade/glade3_3.8.2.bb
index 3f93eba..d0657e4 100644
--- a/meta-oe/recipes-devtools/glade/glade3_3.8.2.bb
+++ b/meta-oe/recipes-devtools/glade/glade3_3.8.2.bb
@@ -16,6 +16,9 @@ SRC_URI[sha256sum] =
f180a5018eee6e3fe574854cb025af897dd9962b01d17d5752e626876d
EXTRA_OECONF += --disable-scrollkeeper
+PACKAGECONFIG ??= gnome
+PACKAGECONFIG[gnome] = --enable-gnome,--disable-gnome,libbonoboui libgnomeui
+
do_configure_prepend() {
sed -i '/^if HAVE_GNOME_DOC_UTILS/,/^endif/d' ${S}/Makefile.am
}
--
1.9.3
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-qt5] moc.exe in win32 generated SDK
Hi, Continuing my adventurous road in getting Yocto to create SDK's for the Win32 platform, I've now stopped to think about what to do with moc.exe (and perhaps qmake.exe). I did try building nativesdk-qtbase with SDKMACHINE=x86_64-mingw32, however, not to my surprise, the build failed. Right now I just need moc.exe in the SDK. I guess I could try just downloading an official Qt build for win32, and putting moc.exe in the SDK, but I thought I'd ask the mailing list if anyone has a better idea ? Regards Jacob -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-qt5] Error during incremental builds, qtserialport
Hi, I've ran into this one a couple of times. I've never seen this problem when I do a build from scratch, only during incremental builds. When I create the SDK for my image, I see errors like this: Collected errors: * satisfy_dependencies_for: Cannot satisfy the following dependencies for packagegroup-qt5-toolchain-target: * libqt5serialport-dev * libqt5serialport-mkspecs * * opkg_install_cmd: Cannot install package packagegroup-qt5-toolchain-target. Sometimes it's not only qtserialport, but other qt-packages. If I check log.do_package for qtserialport I see: WARNING: QA Issue: qtserialport: Files/directories were installed but not shipped /home/jkroon/Projects/oe-devel/build/tmp-eglibc/sysroots/wandboard-dual/usr/include/qt5/QtSerialPort/qserialport.h ... and a whole bunch of other files. These files actually don't exist in my system when I check for them. It's strange because I'm not building the SDK for wandboard-dual, but for the wandboard-solo MACHINE. Am I confusing the build-system since I am jumping back and forth between building for wandboard-dual/wandboard-solo ? /Jacob -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-qt5] Qt5 SDK problem when building for Wandboard
On Fri, May 2, 2014 at 10:25 AM, Jonathan Liu net...@gmail.com wrote: On 2/05/2014 6:17 PM, Jacob Kroon wrote: On Fri, May 2, 2014 at 8:34 AM, Jonathan Liu net...@gmail.com wrote: On 2/05/2014 4:28 PM, Jacob Kroon wrote: Hi Jonathan, On Fri, May 2, 2014 at 3:01 AM, Jonathan Liu net...@gmail.com wrote: On 1/05/2014 6:29 PM, Jacob Kroon wrote: Hi, For a while I've been working around a problem with using populate_sdk_qt5.bbclass in my image recipe, I've had to use a packagegroup-qt5-toolchain-target.bbappend to remove the following dependant packages: qt3d-dev qt3d-mkspecs qt3d-qmlplugins qtlocation-dev qtlocation-mkspecs qtlocation-plugins qtlocation-qmlplugins I'm building for wandboard-* machines, and if I try to bitbake qt3d or qtlocation, I get the following error: Project ERROR: Unknown module(s) in QT: opengl In my distro.conf I'm using: DISTRO_FEATURES_append = opengl ptest systemd DISTRO_FEATURES_remove = pulseaudio sysvinit x11 Could anyone hint at what a proper solution would be ? I'm using master branches. Build Configuration: BB_VERSION= 1.23.0 BUILD_SYS = x86_64-linux NATIVELSBSTRING = Fedora-20 TARGET_SYS= arm-oe-linux-gnueabi MACHINE = wandboard-dual DISTRO= monkey DISTRO_VERSION= 0.1 TUNE_FEATURES = arm armv7a vfp neon callconvention-hard cortexa9 TARGET_FPU= vfp-neon meta-qt5 = master:485c277940d26e0c96953d8a649a9c84eed0afb8 meta = master:4c2ffce2ae32447a1e39877b9bcbc2cc11ddd2fa meta-fsl-arm = master:27b0ffc33d75f0c37a876849a0db9c0277d5f036 meta-fsl-arm-extra = master:a7515d4b20e43cfdd8654249fc16ac6b8c5ec3bc -- Jacob Try adding the following to your distro.conf if you are using OpenGL ES 2.0: PACKAGECONFIG_GL_pn-qtbase = gles2 I think this is already done by a qtbase_%.bbappend in the meta-fsl-arm layer. I do have: [OE jkroon@localhost build]$ bitbake -e qtbase | grep -e ^PACKAGECONFIG= PACKAGECONFIG= release udev evdev tools libs openssl gles2jpeg libpng zlib accessibility tslib icu Could it be that Qt3D and QtLocation requires full OpenGL support, and that gles2 is not enough ? -- Jacob I have built all Qt modules fine with gles2. However, your PACKAGECONFIG doesn't have widgets. I think the QtOpenGL module depends on widgets. Indeed, adding widgets solves the issue, many thanks. Can we describe this dependency in the recipes somehow ? -- Jacob widgets is added to PACKAGECONFIG by default in meta-qt5. You could submit a patch to add a comment for PACKAGECONFIG[widgets] in qtbase.inc. Something like this ? +# needed for qt3d and qtlocation (otherwise do_configure fails to find 'opengl' module, which in turn requires 'widgets') PACKAGECONFIG[widgets] = -widgets,-no-widgets -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-qt5] Qt5 SDK problem when building for Wandboard
On Fri, May 2, 2014 at 5:06 PM, Jacob Kroon jacob.kr...@gmail.com wrote: On Fri, May 2, 2014 at 10:25 AM, Jonathan Liu net...@gmail.com wrote: On 2/05/2014 6:17 PM, Jacob Kroon wrote: On Fri, May 2, 2014 at 8:34 AM, Jonathan Liu net...@gmail.com wrote: On 2/05/2014 4:28 PM, Jacob Kroon wrote: Hi Jonathan, On Fri, May 2, 2014 at 3:01 AM, Jonathan Liu net...@gmail.com wrote: On 1/05/2014 6:29 PM, Jacob Kroon wrote: Hi, For a while I've been working around a problem with using populate_sdk_qt5.bbclass in my image recipe, I've had to use a packagegroup-qt5-toolchain-target.bbappend to remove the following dependant packages: qt3d-dev qt3d-mkspecs qt3d-qmlplugins qtlocation-dev qtlocation-mkspecs qtlocation-plugins qtlocation-qmlplugins I'm building for wandboard-* machines, and if I try to bitbake qt3d or qtlocation, I get the following error: Project ERROR: Unknown module(s) in QT: opengl In my distro.conf I'm using: DISTRO_FEATURES_append = opengl ptest systemd DISTRO_FEATURES_remove = pulseaudio sysvinit x11 Could anyone hint at what a proper solution would be ? I'm using master branches. Build Configuration: BB_VERSION= 1.23.0 BUILD_SYS = x86_64-linux NATIVELSBSTRING = Fedora-20 TARGET_SYS= arm-oe-linux-gnueabi MACHINE = wandboard-dual DISTRO= monkey DISTRO_VERSION= 0.1 TUNE_FEATURES = arm armv7a vfp neon callconvention-hard cortexa9 TARGET_FPU= vfp-neon meta-qt5 = master:485c277940d26e0c96953d8a649a9c84eed0afb8 meta = master:4c2ffce2ae32447a1e39877b9bcbc2cc11ddd2fa meta-fsl-arm = master:27b0ffc33d75f0c37a876849a0db9c0277d5f036 meta-fsl-arm-extra = master:a7515d4b20e43cfdd8654249fc16ac6b8c5ec3bc -- Jacob Try adding the following to your distro.conf if you are using OpenGL ES 2.0: PACKAGECONFIG_GL_pn-qtbase = gles2 I think this is already done by a qtbase_%.bbappend in the meta-fsl-arm layer. I do have: [OE jkroon@localhost build]$ bitbake -e qtbase | grep -e ^PACKAGECONFIG= PACKAGECONFIG= release udev evdev tools libs openssl gles2jpeg libpng zlib accessibility tslib icu Could it be that Qt3D and QtLocation requires full OpenGL support, and that gles2 is not enough ? -- Jacob I have built all Qt modules fine with gles2. However, your PACKAGECONFIG doesn't have widgets. I think the QtOpenGL module depends on widgets. Indeed, adding widgets solves the issue, many thanks. Can we describe this dependency in the recipes somehow ? -- Jacob widgets is added to PACKAGECONFIG by default in meta-qt5. You could submit a patch to add a comment for PACKAGECONFIG[widgets] in qtbase.inc. Something like this ? +# needed for qt3d and qtlocation (otherwise do_configure fails to find 'opengl' module, which in turn requires 'widgets') PACKAGECONFIG[widgets] = -widgets,-no-widgets Would it not be possible to write the recipe so that it depends on just the mkspec/dev packages that are actually used in the target image ? -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-qt5]remove -reduce-relocations on arm platform
On Sat, Apr 26, 2014 at 9:44 PM, Khem Raj raj.k...@gmail.com wrote: On Fri, Apr 25, 2014 at 6:02 AM, Jacob Kroon jacob.kr...@gmail.com wrote: I'm building for Wandboard, ARM Cortex-A9, but I haven't had any problems building qtbase. Is this more of a runtime bug ? are you using Linaro toolchain ? No, I've been using the vanilla gcc toolchain that is built by OE, without meta-linaro. But I may not have exercised Qt5 enough in my application to have noticed this bug. -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-qt5]remove -reduce-relocations on arm platform
On Fri, Apr 25, 2014 at 8:16 AM, Neuer User auslands...@gmx.de wrote: On arm platform the compiler does not implement -reduce-relocations, see here: https://bugreports.qt-project.org/browse/QTBUG-36129 The meta-qt5 layer, however, uses this option, so configure will fail for qtbase. I'm building for Wandboard, ARM Cortex-A9, but I haven't had any problems building qtbase. Is this more of a runtime bug ? /Jacob -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-qt5][PATCH] qtbase: Add libxi to depends.
Hi Søren,
On Wed, Apr 23, 2014 at 9:00 AM, Søren Holm s...@sgh.dk wrote:
Hi MArtin
This fix is not in master yet. Is it lost between two chairs ?
Onsdag den 19. marts 2014 18:40:24 skrev Martin Jansa:
On Wed, Mar 19, 2014 at 06:02:57PM +0100, Søren Holm wrote:
yes problably. I did not investigate the whole structure that thorough
apparently.
Could that change get into dora?
Yes, but first it needs to go to master.
Onsdag den 19. marts 2014 15:15:01 skrev Martin Jansa:
On Wed, Mar 19, 2014 at 02:56:18PM +0100, Jacob Kroon wrote:
Maybe do something this instead ? This is only necessary for X11
right
?
-DEPENDS += qtbase-native
+DEPENDS += \
+qtbase-native \
+${@base_contains('DISTRO_FEATURES', 'x11', 'libxi', '', d)} \
+
/Jacob
I think it belongs to
PACKAGECONFIG[xinput2] = -xinput2,-no-xinput2,libxi
Please test the solution that Martin proposed, and if everything looks
good, submit a patch using the instructions found in meta-qt5/README
and http://www.openembedded.org/wiki/How_to_submit_a_patch_to_OpenEmbedded.
/Jacob
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-qt5][PATCH] populate_sdk_qt5.bbclass: Fix PACKAGE_GROUP deprecation warning
Hi Martin, On Sun, Apr 20, 2014 at 12:34 PM, Martin Jansa martin.ja...@gmail.comwrote: On Tue, Apr 15, 2014 at 07:12:45PM +0200, Jacob Kroon wrote: Fixes the following warning: WARNING: PACKAGE_GROUP is deprecated, please use FEATURE_PACKAGES instead It's correct patch, but I'm a bit tempted to take this patch after branching daisy, because some people are using latest meta-qt5 with older oe-core releases and this undoing this change is easy to miss (unlike e.g. cmake .bbappends which fail when not updated for correct oe-core releases). What do you think? I think I follow you. So the deprecation happened in commit d703e07ea5056624101e1bc48f10d25b602635e9, 2014-03-07, so I guess only daisy/master will give the warning. For older oe-core releases with meta-qt5 master, this patch could silently break stuff I can imagine. So like you propose, keeping this one for master only, after branching daisy, seems like a good idea to me. /Jacob -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [OE-core] State of bitbake world, test-dependencies 2014-04-18
On Sat, Apr 19, 2014 at 1:03 PM, Jacob Kroon jacob.kr...@gmail.com wrote: On Fri, Apr 18, 2014 at 8:36 PM, Martin Jansa martin.ja...@gmail.comwrote: And more importantly floating dependencies, reply here when you start working on some of these. Found differences: ccid/ccid/latest lost dependency on flex console-tools/console-tools/latest lost dependency on flex I'll have a go at resolving the console-tools/flex floating dep. If I understand things correctly: libfl.so from flex provides default implementations for yylex() and yywrap(). console-tools/loadkeys provides its own implementations for these, so linking with libfl.so is not necessary. In order to remove the floating dependency I can patch to either remove @LEXLIB@ from LDADD, or I can pass -Wl,--as-needed in LDFLAGS, both ways seems to get rid of the dependency on libfl. Any preferred way of handling this ? /Jacob -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [OE-core] State of bitbake world, test-dependencies 2014-04-18
On Fri, Apr 18, 2014 at 8:36 PM, Martin Jansa martin.ja...@gmail.comwrote: And more importantly floating dependencies, reply here when you start working on some of these. Found differences: ccid/ccid/latest lost dependency on flex console-tools/console-tools/latest lost dependency on flex I'll have a go at resolving the console-tools/flex floating dep. consolekit/consolekit/latest lost dependency on polkit daq/daq/latest lost dependency on libdnet libmnl libnetfilter-queue libnfnetlink directfb/directfb/latest lost dependency on libdrm libdrm-kms liblzma tiff dovecot/dovecot/latest lost dependency on libcap engrave/engrave/latest lost dependency on flex epiphany/epiphany/latest lost dependency on libavahi-client libavahi-common libavahi-gobject libnotify expedite/expedite/latest lost dependency on libsdl fbida/fbida/latest lost dependency on libx11 libxext libxpm libxt openmotif flashrom/flashrom/latest lost dependency on libftdi gammu/gammu/latest lost dependency on glib-2.0 libgudev-1.0 gammu/libgammu/latest lost dependency on glib-2.0 gimp/gimp/latest lost dependency on jasper libxmu libxpm glade3/glade3/latest lost dependency on dbus-glib dbus-lib gconf gnome-vfs libart-lgpl libavahi-client libavahi-common libavahi-glib libbonobo libbonoboui libcanberra libcap libgcrypt libgnome libgnome-keyring libgnomecanvas libgnomeui libgpg-error libice libltdl libogg libsm libvorbis libxcomposite libxcursor libxdamage libxfixes libxrandr orbit2 popt util-linux-libuuid gmtk/gmtk/latest lost dependency on cairo-gobject dbus-lib gtk+3 json-c libcap libice libpulse libpulse-mainloop-glib libpulsecommon libsm libsndfile1 libx11-xcb libxi libxtst util-linux-libuuid gnokii/libgnokii/latest lost dependency on bluez4 libical libusb-compat libusb1 pcsc-lite-lib gnome-mplayer/gnome-mplayer/latest lost dependency on cairo gtk+3 gnome-mplayer/gnome-mplayer-nautilus-extension/latest lost dependency on eglibc glib-2.0 gtk+ gnome-panel/gnome-panel/latest lost dependency on dbus-lib gst-plugins-bad/gst-plugins-bad-meta/latest lost dependency on gst-plugins-bad-dc1394 gst-plugins-bad-faac gst-plugins-bad-schro gst-plugins-good/gst-plugins-good-ximagesrc/latest lost dependency on libxdamage gstreamer1.0-plugins-good/gstreamer1.0-plugins-good-ximagesrc/latest lost dependency on libxdamage guile/guile/latest lost dependency on libgc ncurses-libncurses readline libarchive/libarchive/latest lost dependency on lzo libarchive/libarchive-bin/latest lost dependency on lzo libdc1394/libdc1394/latest lost dependency on libx11 libxv libmicrohttpd/libmicrohttpd-bin/latest lost dependency on file libungif/libungif/latest lost dependency on libice libsm libx11 libxau libxcb libxdmcp util-linux-libuuid libungif/libungif-utils/latest lost dependency on libx11 mesa-demos/mesa-demos/latest lost dependency on libhybris midori/midori/latest lost dependency on libxscrnsaver mpd/mpd/latest lost dependency on fluidsynth libao libcdio libcdio-cdda libcdio-paranoia libjack libmms libmodplug libopus mpg123 wavpack yajl mpeg2dec/mpeg2dec/latest lost dependency on libx11 libxext libxv mpg123/mpg123/latest lost dependency on audiofile esound libjack libsdl portaudio-v19 tslib mplayer2/mplayer2/latest lost dependency on libpostproc libxscrnsaver mpg123 mx-1.0/mx-1.0/latest lost dependency on libhybris navit/navit/latest lost dependency on fribidi navit/navit-graphics-gtk/latest lost dependency on imlib2 navit/navit-graphics-sdl/latest lost dependency on eglibc freetype libjpeg-turbo liblzma libpng libsdl libsdl-image tiff tslib zlib opencv/libopencv-highgui/latest lost dependency on jasper libdc1394 oscam/oscam/latest lost dependency on pcsc-lite-lib piglit/piglit/latest lost dependency on libhybris libxrender protobuf/protobuf/latest lost dependency on zlib python-efl/python-elementary/latest lost dependency on eglibc eina elementary evas libpython2 qtbase/qtbase-plugins/latest lost dependency on libxcb-render libxkbcommon xcb-util-renderutil qtlocation/qtlocation-plugins/latest lost dependency on gconf glib-2.0 libgypsy qtwayland/qtwayland-plugins/latest lost dependency on libhybris qtwebkit/qtwebkit/latest lost dependency on qtmultimedia sox/sox/latest lost dependency on libao stunnel/stunnel/latest lost dependency on libwrap zlib synergy/synergy/latest lost dependency on libxinerama tracker/tracker/latest lost dependency on dbus-glib enca gnutls gtk+3 libgee libgsf libgudev-1.0 libnmglib libnmutil libtasn1 libudev libunistring upower tracker/tracker-dev/latest lost dependency on cairo enca expat fontconfig freetype giflib gstreamer lcms libbz2 libgsf libgstpbutils-0.10 libgsttag-0.10 liblzma libpoppler libpoppler-glib libunistring libx11 libxau libxcb libxcb-render libxcb-shm libxdmcp libxext libxrender pixman taglib taglib-c tiff tracker/tracker-nautilus-extension/latest lost dependency on libunistring
[oe] [meta-qt5][PATCH] populate_sdk_qt5.bbclass: Fix PACKAGE_GROUP deprecation warning
Fixes the following warning:
WARNING: PACKAGE_GROUP is deprecated, please use FEATURE_PACKAGES instead
Signed-off-by: Jacob Kroon jacob.kr...@mikrodidakt.se
---
classes/populate_sdk_qt5.bbclass | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/classes/populate_sdk_qt5.bbclass b/classes/populate_sdk_qt5.bbclass
index c921983..7b31bbb 100644
--- a/classes/populate_sdk_qt5.bbclass
+++ b/classes/populate_sdk_qt5.bbclass
@@ -53,4 +53,4 @@ toolchain_create_sdk_env_script_append () {
echo 'HostBinaries = ${SDKPATHNATIVE}${OE_QMAKE_PATH_HOST_BINS}' $qtconf
}
-PACKAGE_GROUP_qtcreator-debug = packagegroup-qt5-qtcreator-debug
+FEATURE_PACKAGES_qtcreator-debug = packagegroup-qt5-qtcreator-debug
--
1.9.0
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-qt5][PATCH] qtbase: Add libxi to depends.
Maybe do something this instead ? This is only necessary for X11 right ?
-DEPENDS += qtbase-native
+DEPENDS += \
+qtbase-native \
+${@base_contains('DISTRO_FEATURES', 'x11', 'libxi', '', d)} \
+
/Jacob
On Wed, Mar 19, 2014 at 2:45 PM, Søren Holm s...@sgh.dk wrote:
I'd like this to go into dora if possible.
Sorry for the double posts :)
--
Søren Holm
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
--
-- Jacob
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-qt5][PATCH v2] Explicitly disable Debian-style package renaming for nativesdk-qtbase-tools
Since commit 8da5017712a7ddcb8488e095b9aa03b0273f23e9, libQt5*.so files
are included in the nativesdk-qtbase-tools package. However, this together
with the fact that the binaries are placed in a subdir qt5 of bin will
confuse the Debian-style package renaming logic in debian.bbclass, causing the
resulting ipk package to be created as nativesdk-libqt5core5.
Signed-off-by: Jacob Kroon jacob.kr...@mikrodidakt.se
---
recipes-qt/qt5/nativesdk-qtbase.inc | 6 ++
1 file changed, 6 insertions(+)
diff --git a/recipes-qt/qt5/nativesdk-qtbase.inc
b/recipes-qt/qt5/nativesdk-qtbase.inc
index 3cb9b0e..884b1a6 100644
--- a/recipes-qt/qt5/nativesdk-qtbase.inc
+++ b/recipes-qt/qt5/nativesdk-qtbase.inc
@@ -60,6 +60,12 @@ FILES_${PN}-tools = \
${OE_QMAKE_PATH_BINS}/* \
+# qttools binaries are placed in a subdir of bin in order to avoid
+# collisions with qt4. This would trigger debian.bbclass to rename the
+# package, since it doesn't detect binaries in subdirs. Explicitly
+# disable package auto-renaming for the tools-package.
+DEBIAN_NOAUTONAME_${PN}-tools = 1
+
QT_CONFIG_FLAGS += \
-reduce-relocations \
-shared \
--
1.8.5.3
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-qt5][PATCH v3] nativesdk-qtbase-tools: disable Debian-style package renaming
Since commit 8da5017712a7ddcb8488e095b9aa03b0273f23e9, libQt5*.so files
are included in the nativesdk-qtbase-tools package. However, this together
with the fact that the binaries are placed in a subdir qt5 of bin will
confuse the Debian-style package renaming logic in debian.bbclass, causing the
resulting ipk package to be created as nativesdk-libqt5core5.
Signed-off-by: Jacob Kroon jacob.kr...@mikrodidakt.se
Acked-by: Otavio Salvador ota...@ossystems.com.br
---
recipes-qt/qt5/nativesdk-qtbase.inc | 6 ++
1 file changed, 6 insertions(+)
diff --git a/recipes-qt/qt5/nativesdk-qtbase.inc
b/recipes-qt/qt5/nativesdk-qtbase.inc
index 3cb9b0e..884b1a6 100644
--- a/recipes-qt/qt5/nativesdk-qtbase.inc
+++ b/recipes-qt/qt5/nativesdk-qtbase.inc
@@ -60,6 +60,12 @@ FILES_${PN}-tools = \
${OE_QMAKE_PATH_BINS}/* \
+# qttools binaries are placed in a subdir of bin in order to avoid
+# collisions with qt4. This would trigger debian.bbclass to rename the
+# package, since it doesn't detect binaries in subdirs. Explicitly
+# disable package auto-renaming for the tools-package.
+DEBIAN_NOAUTONAME_${PN}-tools = 1
+
QT_CONFIG_FLAGS += \
-reduce-relocations \
-shared \
--
1.8.5.3
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-qt5][PATCH] Disable Debian-style package renaming for nativesdk-qtbase-tools
On Fri, Feb 21, 2014 at 11:29 PM, Jacob Kroon jacob.kr...@mikrodidakt.sewrote:
Since commit 8da5017712a7ddcb8488e095b9aa03b0273f23e9, libQt5*.so files
are included in the nativesdk-qtbase-tools package. However this
seems to enable Debian-style renaming, causing the resulting ipk package
to be created as nativesdk-libqt5core5.
Explicitly disable renaming for nativesdk-qtbase-tools package.
Signed-off-by: Jacob Kroon jacob.kr...@mikrodidakt.se
---
recipes-qt/qt5/nativesdk-qtbase.inc | 2 ++
1 file changed, 2 insertions(+)
diff --git a/recipes-qt/qt5/nativesdk-qtbase.inc
b/recipes-qt/qt5/nativesdk-qtbase.inc
index 3cb9b0e..e87624b 100644
--- a/recipes-qt/qt5/nativesdk-qtbase.inc
+++ b/recipes-qt/qt5/nativesdk-qtbase.inc
@@ -60,6 +60,8 @@ FILES_${PN}-tools = \
${OE_QMAKE_PATH_BINS}/* \
+DEBIAN_NOAUTONAME_${PN}-tools = 1
+
QT_CONFIG_FLAGS += \
-reduce-relocations \
-shared \
--
1.8.5.3
Now that I think of it, would it be a better idea to somehow handle this in
debian.bbclass ? Or would that break too many existing package names..
-- Jacob
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-qt5][PATCH] Disable Debian-style package renaming for nativesdk-qtbase-tools
On Mon, Feb 24, 2014 at 4:19 PM, Jacob Kroon jacob.kr...@gmail.com wrote:
On Fri, Feb 21, 2014 at 11:29 PM, Jacob Kroon
jacob.kr...@mikrodidakt.sewrote:
Since commit 8da5017712a7ddcb8488e095b9aa03b0273f23e9, libQt5*.so files
are included in the nativesdk-qtbase-tools package. However this
seems to enable Debian-style renaming, causing the resulting ipk package
to be created as nativesdk-libqt5core5.
Explicitly disable renaming for nativesdk-qtbase-tools package.
Signed-off-by: Jacob Kroon jacob.kr...@mikrodidakt.se
---
recipes-qt/qt5/nativesdk-qtbase.inc | 2 ++
1 file changed, 2 insertions(+)
diff --git a/recipes-qt/qt5/nativesdk-qtbase.inc
b/recipes-qt/qt5/nativesdk-qtbase.inc
index 3cb9b0e..e87624b 100644
--- a/recipes-qt/qt5/nativesdk-qtbase.inc
+++ b/recipes-qt/qt5/nativesdk-qtbase.inc
@@ -60,6 +60,8 @@ FILES_${PN}-tools = \
${OE_QMAKE_PATH_BINS}/* \
+DEBIAN_NOAUTONAME_${PN}-tools = 1
+
QT_CONFIG_FLAGS += \
-reduce-relocations \
-shared \
--
1.8.5.3
Now that I think of it, would it be a better idea to somehow handle this
in debian.bbclass ? Or would that break too many existing package names..
Looking closer at debian.bbclass, it seems that the function
auto_libname() fails to detect that nativesdk-qtbase-tools contains
binaries,
since the actual binaries are installed under bin/qt5/ not bin/, and
the regexp used for detecteding binaries is .*/s?bin$, which won't
produce any matches.
-- Jacob
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-qt5][PATCH 2/2] nativesdk-qtbase.inc: Remove unused contents
On Mon, Feb 17, 2014 at 6:45 PM, Otavio Salvador ota...@ossystems.com.brwrote:
On Mon, Feb 17, 2014 at 2:36 PM, Denys Dmytriyenko de...@denix.org
wrote:
On Fri, Feb 14, 2014 at 07:02:34PM -0200, Otavio Salvador wrote:
Fixes:
ERROR: QA Issue: nativesdk-qtbase: Files/directories were installed but
not shipped
/opt/oel/13.0+snapshot/sysroots/i686-oelsdk-linux/usr/lib
/opt/oel/13.0+snapshot/sysroots/i686-oelsdk-linux/usr/include
/opt/oel/13.0+snapshot/sysroots/i686-oelsdk-linux/usr/share
/opt/oel/13.0+snapshot/sysroots/i686-oelsdk-linux/usr/lib/libQt5Test.so
/opt/oel/13.0+snapshot/sysroots/i686-oelsdk-linux/usr/lib/libQt5Network.so
...
Signed-off-by: Otavio Salvador ota...@ossystems.com.br
Acked-by: Denys Dmytriyenko de...@ti.com
I applied it and found it broke the build afterwards. I fixed this one
(will send the patch in some minutes) but this uncovered another issue
I am working on now.
I may be wrong, but I think
7d8f8d6b5a00af7a212cce41350f4e1c5989aabd
or
8da5017712a7ddcb8488e095b9aa03b0273f23e9
screws up the package name for nativesdk-qtbase-tools. On my system now, no
nativesdk-qtbase-tools ipk is created, instead the moc program is placed
in nativesdk-libqt5core5 ipk.
Adding
DEBIAN_NOAUTONAME_${PN}-tools = 1
in nativesdk-qtbase.inc restores the created ipk package name to
nativesdk-qtbase-tools.
-- Jacob
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-qt5][PATCH 2/2] nativesdk-qtbase.inc: Remove unused contents
On Fri, Feb 21, 2014 at 10:46 PM, Otavio Salvador
ota...@ossystems.com.brwrote:
Hello Jacob,
On Fri, Feb 21, 2014 at 6:31 PM, Jacob Kroon jacob.kr...@gmail.com
wrote:
On Mon, Feb 17, 2014 at 6:45 PM, Otavio Salvador
ota...@ossystems.com.brwrote:
On Mon, Feb 17, 2014 at 2:36 PM, Denys Dmytriyenko de...@denix.org
wrote:
On Fri, Feb 14, 2014 at 07:02:34PM -0200, Otavio Salvador wrote:
Fixes:
ERROR: QA Issue: nativesdk-qtbase: Files/directories were installed
but
not shipped
/opt/oel/13.0+snapshot/sysroots/i686-oelsdk-linux/usr/lib
/opt/oel/13.0+snapshot/sysroots/i686-oelsdk-linux/usr/include
/opt/oel/13.0+snapshot/sysroots/i686-oelsdk-linux/usr/share
/opt/oel/13.0+snapshot/sysroots/i686-oelsdk-linux/usr/lib/libQt5Test.so
/opt/oel/13.0+snapshot/sysroots/i686-oelsdk-linux/usr/lib/libQt5Network.so
...
Signed-off-by: Otavio Salvador ota...@ossystems.com.br
Acked-by: Denys Dmytriyenko de...@ti.com
I applied it and found it broke the build afterwards. I fixed this one
(will send the patch in some minutes) but this uncovered another issue
I am working on now.
I may be wrong, but I think
7d8f8d6b5a00af7a212cce41350f4e1c5989aabd
or
8da5017712a7ddcb8488e095b9aa03b0273f23e9
screws up the package name for nativesdk-qtbase-tools. On my system now,
no
nativesdk-qtbase-tools ipk is created, instead the moc program is
placed
in nativesdk-libqt5core5 ipk.
Adding
DEBIAN_NOAUTONAME_${PN}-tools = 1
in nativesdk-qtbase.inc restores the created ipk package name to
nativesdk-qtbase-tools.
Indeed; are you going to send a patch for it or want me to handle this?
Sure, I can send a patch, I just wasn't sure wether this fix was correct or
not.
-- Jacob
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-qt5][PATCH] Disable Debian-style package renaming for nativesdk-qtbase-tools
Since commit 8da5017712a7ddcb8488e095b9aa03b0273f23e9, libQt5*.so files
are included in the nativesdk-qtbase-tools package. However this
seems to enable Debian-style renaming, causing the resulting ipk package
to be created as nativesdk-libqt5core5.
Explicitly disable renaming for nativesdk-qtbase-tools package.
Signed-off-by: Jacob Kroon jacob.kr...@mikrodidakt.se
---
recipes-qt/qt5/nativesdk-qtbase.inc | 2 ++
1 file changed, 2 insertions(+)
diff --git a/recipes-qt/qt5/nativesdk-qtbase.inc
b/recipes-qt/qt5/nativesdk-qtbase.inc
index 3cb9b0e..e87624b 100644
--- a/recipes-qt/qt5/nativesdk-qtbase.inc
+++ b/recipes-qt/qt5/nativesdk-qtbase.inc
@@ -60,6 +60,8 @@ FILES_${PN}-tools = \
${OE_QMAKE_PATH_BINS}/* \
+DEBIAN_NOAUTONAME_${PN}-tools = 1
+
QT_CONFIG_FLAGS += \
-reduce-relocations \
-shared \
--
1.8.5.3
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-qt5][jansa/qt5-5.2.0] libQt5*.la references OE build directories
Hi Andreas and Martin,
On Tue, Jan 14, 2014 at 1:03 AM, Jacob Kroon jacob.kr...@gmail.com wrote:
On Mon, 13 Jan 2014, Martin Jansa wrote:
On Mon, Jan 13, 2014 at 01:58:25AM +0100, Jacob Kroon wrote:
Hi,
I'm seeing what I believe is a problem with the *.la files in the Qt5
dev-packages.
The files references paths in my OE build environment, like so:
...
# Libraries that this one depends upon.
dependency_libs='-lQt5Gui
-L/home/jkroon/Projects/oe-devel/build/tmp-eglibc/
sysroots/wandboard-solo/usr/lib
-lQt5Core -lpthread '
...
# Directory that this library needs to be installed in:
libdir='/home/jkroon/Projects/oe-devel/build/tmp-eglibc/
sysroots/wandboard-solo/usr'
...
I noticed this since my Qt5 autotools project wouldn't link properly when
using the SDK, unless I first removed all libQt5*.la files.
I'm not quite sure where to look for the problem so any pointers would be
appreciated.
Probably bad rebase on my side, please compare
mkspecs/features/qt_module.prf in 5.1.1 and 5.2.0
changed in:
https://github.com/meta-qt5/meta-qt5/blob/master/recipes-
qt/qt5/qtbase-5.1.1/0008-qt_module-Fix-pkgconfig-replacement.patch
and
https://github.com/meta-qt5/meta-qt5/blob/jansa/qt5-5.2.0/
recipes-qt/qt5/qtbase/0005-qt_module-Fix-pkgconfig-replacement.patch
Or better in meta-qt5's qtbase repo:
https://github.com/meta-qt5/qtbase/blob/b5.1.1/mkspecs/
features/qt_module.prf
https://github.com/meta-qt5/qtbase/blob/b5.2.0/mkspecs/
features/qt_module.prf
I tried building Qt 5.1.1 from master again, the .la-files do look better,
libdir looks correct (taken from libQt5Svg.la):
snip
# Directory that this library needs to be installed in:
libdir='/usr/lib'
/snip
however
snip
dependency_libs='-lQt5Gui -L/home/jkroon/Projects/oe-
devel/build/tmp-eglibc/sysroots/wandboard-solo/usr/lib -lQt5Core
-lpthread '
/snip
Dunno wether this is a problem or not, I never tried building the project
with 5.1.1 using the SDK.
Will hopefully have some time tomorrow to look at the links you sent.
Thanks,
Jacob
I'm trying to get the .la-files for Qt5 dev-packages correct.
What I've come up with so far is attached in the patch (to be applied in
top of jansa-qt5-5.2.0, b8a236df79c25802c1f2eb2525e8a534e73e608e)
With the patch applied I can still build both an image and an SDK, and also
the resulting SDK can be used to build my QT5 autotools projects. But I get
these warnings when building for instance qtxmlpatterns:
WARNING: QA Issue: qtxmlpatterns: The compile log indicates that host
include and/or library paths were used.
With my patch applied,
recipes-qt/qt5/qtbase/0005-qt_module-Fix-pkgconfig-replacement.patch looks
like this:
--- qtbase-opensource-src-5.2.0.orig/mkspecs/features/qt_module.prf
+++ qtbase-opensource-src-5.2.0/mkspecs/features/qt_module.prf
@@ -135,30 +135,36 @@ load(qt_installs)
rplbase = $$dirname(_QMAKE_SUPER_CACHE_)/[^/][^/]*
else: \
rplbase = $$MODULE_BASE_OUTDIR
-include_replace.match = $$rplbase/include
-include_replace.replace = $$[QT_INSTALL_HEADERS/raw]
-include_replace.CONFIG = path
+pkgconfig_include_replace.match = $$rplbase/include
+pkgconfig_include_replace.replace = \$$\\{includedir}
+pkgconfig_include_replace.CONFIG = path
+pkgconfig_lib_replace.match = $$rplbase/lib
+pkgconfig_lib_replace.replace = \$$\\{libdir}
+pkgconfig_lib_replace.CONFIG = path
lib_replace.match = $$rplbase/lib
host_build: \
lib_replace.replace = $$[QT_HOST_LIBS]
else: \
lib_replace.replace = $$[QT_INSTALL_LIBS/raw]
lib_replace.CONFIG = path
-QMAKE_PRL_INSTALL_REPLACE += include_replace lib_replace
+QMAKE_PRL_INSTALL_REPLACE += lib_replace
unix|win32-g++* {
CONFIG += create_pc
QMAKE_PKGCONFIG_LIBDIR = $$lib_replace.replace
- QMAKE_PKGCONFIG_INCDIR = $$include_replace.replace
+ QMAKE_PKGCONFIG_INCDIR = $$[QT_INSTALL_HEADERS/raw]
QMAKE_PKGCONFIG_CFLAGS = -I${includedir}/$$MODULE_INCNAME
QMAKE_PKGCONFIG_DESTDIR = pkgconfig
- QMAKE_PKGCONFIG_INSTALL_REPLACE += include_replace lib_replace
+ QMAKE_PKGCONFIG_INSTALL_REPLACE += pkgconfig_include_replace
pkgconfig_lib_replace
}
unix {
+ lafile_replace.match = -L$$[QT_INSTALL_LIBS/raw]
+ lafile_replace.replace =
+ lafile_replace.CONFIG = path
CONFIG += create_libtool explicitlib
QMAKE_LIBTOOL_LIBDIR = $$lib_replace.replace
- QMAKE_LIBTOOL_INSTALL_REPLACE += include_replace lib_replace
+ QMAKE_LIBTOOL_INSTALL_REPLACE += lib_replace lafile_replace
}
The lafile_replace does the substitution -L/usr/lib - . Maybe this is
needed for the .prl-files aswell ? Is the QMAKE_PRL_BUILD_DIR needed at all
? With all the sed:ing going on, that path looks phony to me...
From native sysroots/x86_64-linux/usr/lib/libQt5Xml.prl:
QMAKE_PRL_BUILD_DIR =
/home/jkroon/Projects/oe-devel/build/tmp-eglibc/work/x86_64-linux/qtbase-native/5.2.0-r0/build/src/xml
QMAKE_PRL_LIBS =
-L/home/jkroon/Projects/oe-devel/build/tmp-eglibc/sysroots/x86_64-linux/usr/lib
-lQt5Core -lpthread
Re: [oe] [meta-qt5][jansa/qt5-5.2.0] libQt5*.la references OE build directories
On Sat, Jan 18, 2014 at 11:03 AM, Jacob Kroon jacob.kr...@gmail.com wrote:
Hi Andreas and Martin,
On Tue, Jan 14, 2014 at 1:03 AM, Jacob Kroon jacob.kr...@gmail.comwrote:
On Mon, 13 Jan 2014, Martin Jansa wrote:
On Mon, Jan 13, 2014 at 01:58:25AM +0100, Jacob Kroon wrote:
Hi,
I'm seeing what I believe is a problem with the *.la files in the Qt5
dev-packages.
The files references paths in my OE build environment, like so:
...
# Libraries that this one depends upon.
dependency_libs='-lQt5Gui
-L/home/jkroon/Projects/oe-devel/build/tmp-eglibc/
sysroots/wandboard-solo/usr/lib
-lQt5Core -lpthread '
...
# Directory that this library needs to be installed in:
libdir='/home/jkroon/Projects/oe-devel/build/tmp-eglibc/
sysroots/wandboard-solo/usr'
...
I noticed this since my Qt5 autotools project wouldn't link properly
when
using the SDK, unless I first removed all libQt5*.la files.
I'm not quite sure where to look for the problem so any pointers would
be
appreciated.
Probably bad rebase on my side, please compare
mkspecs/features/qt_module.prf in 5.1.1 and 5.2.0
changed in:
https://github.com/meta-qt5/meta-qt5/blob/master/recipes-
qt/qt5/qtbase-5.1.1/0008-qt_module-Fix-pkgconfig-replacement.patch
and
https://github.com/meta-qt5/meta-qt5/blob/jansa/qt5-5.2.0/
recipes-qt/qt5/qtbase/0005-qt_module-Fix-pkgconfig-replacement.patch
Or better in meta-qt5's qtbase repo:
https://github.com/meta-qt5/qtbase/blob/b5.1.1/mkspecs/
features/qt_module.prf
https://github.com/meta-qt5/qtbase/blob/b5.2.0/mkspecs/
features/qt_module.prf
I tried building Qt 5.1.1 from master again, the .la-files do look
better, libdir looks correct (taken from libQt5Svg.la):
snip
# Directory that this library needs to be installed in:
libdir='/usr/lib'
/snip
however
snip
dependency_libs='-lQt5Gui -L/home/jkroon/Projects/oe-
devel/build/tmp-eglibc/sysroots/wandboard-solo/usr/lib -lQt5Core
-lpthread '
/snip
Dunno wether this is a problem or not, I never tried building the project
with 5.1.1 using the SDK.
Will hopefully have some time tomorrow to look at the links you sent.
Thanks,
Jacob
I'm trying to get the .la-files for Qt5 dev-packages correct.
What I've come up with so far is attached in the patch (to be applied in
top of jansa-qt5-5.2.0, b8a236df79c25802c1f2eb2525e8a534e73e608e)
With the patch applied I can still build both an image and an SDK, and
also the resulting SDK can be used to build my QT5 autotools projects. But
I get these warnings when building for instance qtxmlpatterns:
WARNING: QA Issue: qtxmlpatterns: The compile log indicates that host
include and/or library paths were used.
With my patch applied,
recipes-qt/qt5/qtbase/0005-qt_module-Fix-pkgconfig-replacement.patch looks
like this:
--- qtbase-opensource-src-5.2.0.orig/mkspecs/features/qt_module.prf
+++ qtbase-opensource-src-5.2.0/mkspecs/features/qt_module.prf
@@ -135,30 +135,36 @@ load(qt_installs)
rplbase = $$dirname(_QMAKE_SUPER_CACHE_)/[^/][^/]*
else: \
rplbase = $$MODULE_BASE_OUTDIR
-include_replace.match = $$rplbase/include
-include_replace.replace = $$[QT_INSTALL_HEADERS/raw]
-include_replace.CONFIG = path
+pkgconfig_include_replace.match = $$rplbase/include
+pkgconfig_include_replace.replace = \$$\\{includedir}
+pkgconfig_include_replace.CONFIG = path
+pkgconfig_lib_replace.match = $$rplbase/lib
+pkgconfig_lib_replace.replace = \$$\\{libdir}
+pkgconfig_lib_replace.CONFIG = path
lib_replace.match = $$rplbase/lib
host_build: \
lib_replace.replace = $$[QT_HOST_LIBS]
else: \
lib_replace.replace = $$[QT_INSTALL_LIBS/raw]
lib_replace.CONFIG = path
-QMAKE_PRL_INSTALL_REPLACE += include_replace lib_replace
+QMAKE_PRL_INSTALL_REPLACE += lib_replace
unix|win32-g++* {
CONFIG += create_pc
QMAKE_PKGCONFIG_LIBDIR = $$lib_replace.replace
- QMAKE_PKGCONFIG_INCDIR = $$include_replace.replace
+ QMAKE_PKGCONFIG_INCDIR = $$[QT_INSTALL_HEADERS/raw]
QMAKE_PKGCONFIG_CFLAGS = -I${includedir}/$$MODULE_INCNAME
QMAKE_PKGCONFIG_DESTDIR = pkgconfig
- QMAKE_PKGCONFIG_INSTALL_REPLACE += include_replace lib_replace
+ QMAKE_PKGCONFIG_INSTALL_REPLACE += pkgconfig_include_replace
pkgconfig_lib_replace
}
unix {
+ lafile_replace.match = -L$$[QT_INSTALL_LIBS/raw]
+ lafile_replace.replace =
+ lafile_replace.CONFIG = path
CONFIG += create_libtool explicitlib
QMAKE_LIBTOOL_LIBDIR = $$lib_replace.replace
- QMAKE_LIBTOOL_INSTALL_REPLACE += include_replace lib_replace
+ QMAKE_LIBTOOL_INSTALL_REPLACE += lib_replace lafile_replace
}
The lafile_replace does the substitution -L/usr/lib - . Maybe this is
needed for the .prl-files aswell ? Is the QMAKE_PRL_BUILD_DIR needed at all
? With all the sed:ing going on, that path looks phony to me...
From native sysroots/x86_64-linux/usr/lib/libQt5Xml.prl:
QMAKE_PRL_BUILD_DIR =
/home/jkroon/Projects/oe-devel/build/tmp-eglibc/work/x86_64-linux/qtbase-native
Re: [oe] [meta-qt5][jansa/qt5-5.2.0] libQt5*.la references OE build directories
On Sat, 18 Jan 2014, Martin Jansa wrote:
On Sat, Jan 18, 2014 at 11:03:09AM +0100, Jacob Kroon wrote:
Hi Andreas and Martin,
On Tue, Jan 14, 2014 at 1:03 AM, Jacob Kroon jacob.kr...@gmail.com wrote:
On Mon, 13 Jan 2014, Martin Jansa wrote:
On Mon, Jan 13, 2014 at 01:58:25AM +0100, Jacob Kroon wrote:
Hi,
I'm seeing what I believe is a problem with the *.la files in the Qt5
dev-packages.
The files references paths in my OE build environment, like so:
...
# Libraries that this one depends upon.
dependency_libs='-lQt5Gui
-L/home/jkroon/Projects/oe-devel/build/tmp-eglibc/
sysroots/wandboard-solo/usr/lib
-lQt5Core -lpthread '
...
# Directory that this library needs to be installed in:
libdir='/home/jkroon/Projects/oe-devel/build/tmp-eglibc/
sysroots/wandboard-solo/usr'
...
I noticed this since my Qt5 autotools project wouldn't link properly when
using the SDK, unless I first removed all libQt5*.la files.
I'm not quite sure where to look for the problem so any pointers would be
appreciated.
Probably bad rebase on my side, please compare
mkspecs/features/qt_module.prf in 5.1.1 and 5.2.0
changed in:
https://github.com/meta-qt5/meta-qt5/blob/master/recipes-
qt/qt5/qtbase-5.1.1/0008-qt_module-Fix-pkgconfig-replacement.patch
and
https://github.com/meta-qt5/meta-qt5/blob/jansa/qt5-5.2.0/
recipes-qt/qt5/qtbase/0005-qt_module-Fix-pkgconfig-replacement.patch
Or better in meta-qt5's qtbase repo:
https://github.com/meta-qt5/qtbase/blob/b5.1.1/mkspecs/
features/qt_module.prf
https://github.com/meta-qt5/qtbase/blob/b5.2.0/mkspecs/
features/qt_module.prf
I tried building Qt 5.1.1 from master again, the .la-files do look better,
libdir looks correct (taken from libQt5Svg.la):
snip
# Directory that this library needs to be installed in:
libdir='/usr/lib'
/snip
however
snip
dependency_libs='-lQt5Gui -L/home/jkroon/Projects/oe-
devel/build/tmp-eglibc/sysroots/wandboard-solo/usr/lib -lQt5Core
-lpthread '
/snip
Dunno wether this is a problem or not, I never tried building the project
with 5.1.1 using the SDK.
Will hopefully have some time tomorrow to look at the links you sent.
Thanks,
Jacob
I'm trying to get the .la-files for Qt5 dev-packages correct.
What I've come up with so far is attached in the patch (to be applied in
top of jansa-qt5-5.2.0, b8a236df79c25802c1f2eb2525e8a534e73e608e)
With the patch applied I can still build both an image and an SDK, and also
the resulting SDK can be used to build my QT5 autotools projects. But I get
these warnings when building for instance qtxmlpatterns:
WARNING: QA Issue: qtxmlpatterns: The compile log indicates that host
include and/or library paths were used.
This warning is the reason why I've originally created this patch, so
something is definitely wrong in new version.
I got rid of this warning by adding the -L/usr/lib - substitution
for the .prl file aswell. But it's an ugly hack I guess..
I know it's very confusing .prf (and last upstream changes didn't make
it any easier), I'll try to look at it this weekend.
It would be great to prepare some nice reproducer which we can use in
upstream gerrit to show that this really doesn't work well in more
complicated case (like our cross-compile with sstate-relocateable
.prl,.la,.pc files, where we really don't want to include host paths).
Do you have something like smaller test case or were you testing it on
big qtbase build?
Unfortunately no, I did all testing on the big
qtbase-native/qtbase/nativesdk-qtbase packages (pats my poor laptop
harddrive). Do you have an idea of where to start for creating such a test
case ? Something like a recipe that uses qtbase but only builds qmake + a
minimal Qt module ?
With my patch applied,
recipes-qt/qt5/qtbase/0005-qt_module-Fix-pkgconfig-replacement.patch looks
like this:
--- qtbase-opensource-src-5.2.0.orig/mkspecs/features/qt_module.prf
+++ qtbase-opensource-src-5.2.0/mkspecs/features/qt_module.prf
@@ -135,30 +135,36 @@ load(qt_installs)
rplbase = $$dirname(_QMAKE_SUPER_CACHE_)/[^/][^/]*
else: \
rplbase = $$MODULE_BASE_OUTDIR
-include_replace.match = $$rplbase/include
-include_replace.replace = $$[QT_INSTALL_HEADERS/raw]
-include_replace.CONFIG = path
+pkgconfig_include_replace.match = $$rplbase/include
+pkgconfig_include_replace.replace = \$$\\{includedir}
+pkgconfig_include_replace.CONFIG = path
+pkgconfig_lib_replace.match = $$rplbase/lib
+pkgconfig_lib_replace.replace = \$$\\{libdir}
+pkgconfig_lib_replace.CONFIG = path
lib_replace.match = $$rplbase/lib
host_build: \
lib_replace.replace = $$[QT_HOST_LIBS]
else: \
lib_replace.replace = $$[QT_INSTALL_LIBS/raw]
lib_replace.CONFIG = path
-QMAKE_PRL_INSTALL_REPLACE += include_replace lib_replace
+QMAKE_PRL_INSTALL_REPLACE += lib_replace
unix|win32-g++* {
CONFIG += create_pc
QMAKE_PKGCONFIG_LIBDIR = $$lib_replace.replace
- QMAKE_PKGCONFIG_INCDIR = $$include_replace.replace
+ QMAKE_PKGCONFIG_INCDIR = $$[QT_INSTALL_HEADERS/raw]
QMAKE_PKGCONFIG_CFLAGS
Re: [oe] [meta-qt5][jansa/qt5-5.2.0] libQt5*.la references OE build directories
On Mon, 13 Jan 2014, Martin Jansa wrote: On Mon, Jan 13, 2014 at 01:58:25AM +0100, Jacob Kroon wrote: Hi, I'm seeing what I believe is a problem with the *.la files in the Qt5 dev-packages. The files references paths in my OE build environment, like so: ... # Libraries that this one depends upon. dependency_libs='-lQt5Gui -L/home/jkroon/Projects/oe-devel/build/tmp-eglibc/sysroots/wandboard-solo/usr/lib -lQt5Core -lpthread ' ... # Directory that this library needs to be installed in: libdir='/home/jkroon/Projects/oe-devel/build/tmp-eglibc/sysroots/wandboard-solo/usr' ... I noticed this since my Qt5 autotools project wouldn't link properly when using the SDK, unless I first removed all libQt5*.la files. I'm not quite sure where to look for the problem so any pointers would be appreciated. Probably bad rebase on my side, please compare mkspecs/features/qt_module.prf in 5.1.1 and 5.2.0 changed in: https://github.com/meta-qt5/meta-qt5/blob/master/recipes-qt/qt5/qtbase-5.1.1/0008-qt_module-Fix-pkgconfig-replacement.patch and https://github.com/meta-qt5/meta-qt5/blob/jansa/qt5-5.2.0/recipes-qt/qt5/qtbase/0005-qt_module-Fix-pkgconfig-replacement.patch Or better in meta-qt5's qtbase repo: https://github.com/meta-qt5/qtbase/blob/b5.1.1/mkspecs/features/qt_module.prf https://github.com/meta-qt5/qtbase/blob/b5.2.0/mkspecs/features/qt_module.prf I tried building Qt 5.1.1 from master again, the .la-files do look better, libdir looks correct (taken from libQt5Svg.la): snip # Directory that this library needs to be installed in: libdir='/usr/lib' /snip however snip dependency_libs='-lQt5Gui -L/home/jkroon/Projects/oe-devel/build/tmp-eglibc/sysroots/wandboard-solo/usr/lib -lQt5Core -lpthread ' /snip Dunno wether this is a problem or not, I never tried building the project with 5.1.1 using the SDK. Will hopefully have some time tomorrow to look at the links you sent. Thanks, Jacob ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
