@Kanni1303

I guess you might have run into the same problem:

/home/myhomeismycastle/raspi/tools/arm-bcm2708/gcc-linaro-arm-linux-gnueabihf-raspbian-x64/bin/arm-linux-gnueabihf-g++ -c -include .pch/Qt5Core -march=armv8-a -mtune=cortex-a53 -mfpu=crypto-neon-fp-armv8 -mfloat-abi=hard --sysroot=/home/myhomeismycastle/raspi/sysroot -O3 -std=c++11 -fvisibility=hidden -fvisibility-inlines-hidden -Wall -W -Wvla -Wno-missing-field-initializers -D_REENTRANT -fPIC -DQT_NO_USING_NAMESPACE -DQT_NO_FOREACH -DQT_NO_NARROWING_CONVERSIONS_IN_CONNECT -DQT_BUILD_CORE_LIB -DQT_BUILDING_QT -DQT_NO_CAST_TO_ASCII -DQT_ASCII_CAST_WARNINGS -DQT_MOC_COMPAT -DQT_USE_QSTRINGBUILDER -DQT_DEPRECATED_WARNINGS -DQT_DISABLE_DEPRECATED_BEFORE=0x050000 -D_LARGEFILE64_SOURCE -D_LARGEFILE_SOURCE -DQT_NO_DEBUG -DPCRE2_CODE_UNIT_WIDTH=16 -I. -Iglobal -I../3rdparty/harfbuzz/src -I../3rdparty/md5 -I../3rdparty/md4 -I../3rdparty/sha3 -I../3rdparty/forkfd -I../3rdparty/tinycbor/src -I../../include -I../../include/QtCore -I../../include/QtCore/5.12.0 -I../../include/QtCore/5.12.0/QtCore -I.moc -I.tracegen -I../3rdparty/pcre2/src -I/home/myhomeismycastle/raspi/sysroot/usr/include/glib-2.0 -I/home/myhomeismycastle/raspi/sysroot/usr/lib/arm-linux-gnueabihf/glib-2.0/include -I../../mkspecs/devices/linux-rasp-pi3-g++ -o .obj/qendian.o global/qendian.cpp In file included from global/qendian.cpp:41:0: global/qendian.h:165:61: error: redeclaration ‘constexpr T qbswap(T) [with T = qfloat16]’ differs in ‘constexpr’ template <> inline qfloat16 qbswap<qfloat16>(qfloat16 source) ^ global/qendian.h:96:42: error: from previous declaration ‘constexpr T qbswap(T) [with T = qfloat16]’ template <typename T> Q_DECL_CONSTEXPR T qbswap(T source); ^ global/qendian.h:170:52: error: redeclaration ‘constexpr T qbswap(T) [with T = float]’ differs in ‘constexpr’ template <> inline float qbswap<float>(float source) ^ global/qendian.h:96:42: error: from previous declaration ‘constexpr T qbswap(T) [with T = float]’ template <typename T> Q_DECL_CONSTEXPR T qbswap(T source); ^ global/qendian.h:175:55: error: redeclaration ‘constexpr T qbswap(T) [with T = double]’ differs in ‘constexpr’ template <> inline double qbswap<double>(double source) ^ global/qendian.h:96:42: error: from previous declaration ‘constexpr T qbswap(T) [with T = double]’ template <typename T> Q_DECL_CONSTEXPR T qbswap(T source); ^ Makefile:11849: recipe for target '.obj/qendian.o' failed make[2]: *** [.obj/qendian.o] Error 1 make[2]: Leaving directory '/home/myhomeismycastle/raspi/qt-everywhere-src-5.12.0/qtbase/src/corelib' Makefile:224: recipe for target 'sub-corelib-make_first' failed make[1]: *** [sub-corelib-make_first] Error 2 make[1]: Leaving directory '/home/myhomeismycastle/raspi/qt-everywhere-src-5.12.0/qtbase/src' Makefile:48: recipe for target 'sub-src-make_first' failed make: *** [sub-src-make_first] Error 2

I have tried to cross-compile the complete source of Qt 5.12.0 from archives, which did you cross-compile?

I saw that you actually skip wayland, which is my reason for trying Qt 5.12.0. I still have issues using openGL on raspberry pi3 and I thought that I can fix only with latest Qt 5.12. Did you get as far yet?

May I suggest to change for Python 3 ?
Python 2.7 will reached end of life next year so it's not a good idea to start a new project with it.

I tested it with Qt 5.11.0, but more deficiency detected.
The General Attribute also undetected and characteristic write is not working :(

No, because they were neved emitted. I just found this: https://bugreports.qt.io/browse/QTBUG-38401, it seems that they are not implemented yet for Bluez 5. However, I would like to set the pin in only one-way, the raspberry shouldn't accept any pairing if the pin code (probably hardcoded in the raspberry) is different from the one entered by the user. Anyway, I'm not sure if this is possible.

I have found a solution here.
Looks like Ubuntu couldn't run 32bit binary, but it's working now.

@ronpetit said in Qt on Raspberry pi3:

@sierdzio I'm running it without X server, as I should

Now, I'm curious about this, what libs are you using for glesv2 and egl?, I'm using the ones in /opt/vc/lib folder, and I also cross compiled the Qt with those

Oh I don't remember to be honest. I think I didn't change anything from stock settings. I also didn't specify the OpenGL libs manually anywhere when building Qt or apps.

@Worly said in Qt cross platform compilation with QtRpi to rpi3 - could not find "eglfs":

QtRpi has only up to 5.7.0 version of Qt, does that mean that I have to use v5.7.0 plugins or I can use the latest?

Would you consider going through the steps from this guide? It allows you to cross compile Qt for RPi and it's been used with newer Qt versions and several RPi models.

@SGaist

I found the solution in the mplayer's docs, the -wid not working in linuxfb environment. The qprocess started the mplayer who writes the framebuffer directly, it's working.

@Jerwinprabu Thank You. It is working fine now.

Not sure, but it sounds like a broken toolchain. Make sure your cross compilation toolchain and sysroot are valid.

@Xenoshell If a command/tool needs root rights it is better to set http://linuxg.net/how-to-set-the-setuid-and-setgid-bit-for-files-in-linux-and-unix/ instead of running whole IDE as root! I never saw a need to run an IDE as root and I'm software developer for quite some time.

Ok, I've missed that article.

From the looks of it, you seem to be currently essentially interested in building qtbase, so you could call configure directly in it to shorten the list of options.

Intuitively, I'd avoid the reduce-exports and sysroot parameters.

Thank you a lot , i will looking for it.

@moldovan.catalin You can create symbolic link in /opt/vc/lib pointing to these libraries but having old names.

thanks a lot for help i already solved it. I just deleted the old compiler added new set same path and restart QT finally it works :)

qtscript

ARM is an architecture

Hi,

I'd recommend posting that question to the the QtWebEngine mailing list. You'll find there QtWebEngine's developers/maintainers. This forum is more user oriented.

@micland I thought of that but discarded the idea because Raspbian seems to be the de facto standard distribution for RPi, and IMHO the mkspec should work out-of-the-box with the Rasbian library paths. Or at least there should be a mkspec for Raspbian.

Also, if I understand it right, the symlinks won't fix the run time error. (But I have to admit I am not at all sure I understand it right.)

One thing that puzzles me is the big difference between RPi3 and RPi2 mkspecs. Wouldn't it be enough to change the processor optimization flags?

hi, i has same problem, i refresh raspberry, but issue also happen. please help me to solved issue. thanks u.

@pinballsp I am having the same problem. Did you find a way to solve it?