[yocto] Yocto Dizzy + Beaglebone + Qt5 (Internal compiler error)

[Martin Beaudet]

Hi,

I'm building an image for the beaglebone. I would like to use (at least) the version 5.4 of QT with elgfs platform. 
Also the hardware accelerated support on Beaglebone Black.

Here is what I've done yet :

	# SHELL COMMAND :
	$ mkdir poky
	$ git clone -b dizzy git://git.yoctoproject.org/poky
	$ cd poky
	$ git clone git://git.yoctoproject.org/meta-ti
	$ git clone -b dizzy git://github.com/meta-qt5/meta-qt5.git
	$ git clone -b dizzy git://git.openembedded.org/meta-openembedded
	$ git clone -b daisy https://bitbucket.org/embarcados/meta-embarcados.git
	$ source oe-init-build-env beaglebone

	# FILE : poky/beaglebone/conf/local.conf
	DISTRO_FEATURES_remove = "x11 wayland"

	 # FILE : poky/beaglebone/conf/bblayers.conf
	BBLAYERS ?= \
	  ${TOPDIR}/../meta \
	  ${TOPDIR}/../meta-yocto \
	  ${TOPDIR}/../meta-openembedded/meta-oe \
	  ${TOPDIR}/../meta-ti \
	  ${TOPDIR}/../meta-qt5 \
	  ${TOPDIR}/../meta-embarcados \
	  ${TOPDIR}/../meta-embarcados/meta-bbb \

	# SHELL COMMAND :
	$ bitbake qt5-image-demo

Now I'm getting a compiler error : 

	/home/test/poky/beaglebone/tmp/work/cortexa8t2hf-vfp-neon-poky-linux-gnueabi/qttools/5.3.2-r0/qttools-opensource-src-5.3.2/src/linguist/shared/po.cpp: In function 'bool loadPO(Translator&, QIODevice&, ConversionData&)':
	/home/test/poky/beaglebone/tmp/work/cortexa8t2hf-vfp-neon-poky-linux-gnueabi/qttools/5.3.2-r0/qttools-opensource-src-5.3.2/src/linguist/shared/po.cpp:717:1: internal compiler error: in add_stores, at var-tracking.c:6000
	}
	^
	Please submit a full bug report,
	with preprocessed source if appropriate.
	See <http://gcc.gnu.org/bugs.html> for instructions.
	Makefile:934: recipe for target '.obj/po.o' failed
	make[3]: *** [.obj/po.o] Error 1
	make[3]: Leaving directory '/home/test/poky/beaglebone/tmp/work/cortexa8t2hf-vfp-neon-poky-linux-gnueabi/qttools/5.3.2-r0/build/src/linguist/lupdate'
	Makefile:66: recipe for target 'sub-lupdate-make_first' failed
	make[2]: *** [sub-lupdate-make_first] Error 2
	make[2]: *** Waiting for unfinished jobs....
	compiling /home/test/poky/beaglebone/tmp/work/cortexa8t2hf-vfp-neon-poky-linux-gnueabi/qttools/5.3.2-r0/qttools-opensource-src-5.3.2/src/linguist/shared/po.cpp
	compiling /home/test/poky/beaglebone/tmp/work/cortexa8t2hf-vfp-neon-poky-linux-gnueabi/qttools/5.3.2-r0/qttools-opensource-src-5.3.2/src/linguist/shared/ts.cpp
	compiling /home/test/poky/beaglebone/tmp/work/cortexa8t2hf-vfp-neon-poky-linux-gnueabi/qttools/5.3.2-r0/qttools-opensource-src-5.3.2/src/linguist/shared/ts.cpp
	/home/test/poky/beaglebone/tmp/work/cortexa8t2hf-vfp-neon-poky-linux-gnueabi/qttools/5.3.2-r0/qttools-opensource-src-5.3.2/src/linguist/shared/po.cpp: In function 'bool loadPO(Translator&, QIODevice&, ConversionData&)':
	/home/test/poky/beaglebone/tmp/work/cortexa8t2hf-vfp-neon-poky-linux-gnueabi/qttools/5.3.2-r0/qttools-opensource-src-5.3.2/src/linguist/shared/po.cpp:717:1: internal compiler error: in add_stores, at var-tracking.c:6000
	}
	^
	Please submit a full bug report,
	with preprocessed source if appropriate.
	See <http://gcc.gnu.org/bugs.html> for instructions.
	Makefile:1022: recipe for target '.obj/po.o' failed
	make[3]: *** [.obj/po.o] Error 1
	make[3]: *** Waiting for unfinished jobs....
	compiling /home/test/poky/beaglebone/tmp/work/cortexa8t2hf-vfp-neon-poky-linux-gnueabi/qttools/5.3.2-r0/qttools-opensource-src-5.3.2/src/linguist/shared/ts.cpp
	/home/test/poky/beaglebone/tmp/work/cortexa8t2hf-vfp-neon-poky-linux-gnueabi/qttools/5.3.2-r0/qttools-opensource-src-5.3.2/src/linguist/shared/po.cpp: In function 'bool loadPO(Translator&, QIODevice&, ConversionData&)':
	/home/test/poky/beaglebone/tmp/work/cortexa8t2hf-vfp-neon-poky-linux-gnueabi/qttools/5.3.2-r0/qttools-opensource-src-5.3.2/src/linguist/shared/po.cpp:717:1: internal compiler error: in add_stores, at var-tracking.c:6000
	}
	^

I read about similar error here (https://bugs.launchpad.net/gcc-linaro/+bug/1312931) but they seem to indicate that this error has been fixed in new release of GCC and they are talking about 4.8 so I supposed it has been fixed in 4.9 ?

I hope somebody can point me out the problem because I'm out of idea.

Best regards,
Martin