Experimental cross compiling Qt in Debian packages

Some time ago we the Qt/KDE team were contacted by Helmut Grohne. He was trying to cross compile Debian packages in general thanks to Ubuntu/Debian's multi-arch support, and he was having problems with Qt-based ones.

As far as we understand Qt upstreams only support cross compiling by having a toolchain for each pair of architectures involved. In Debian terms, and only considering current official architectures, that would mean building 90 cross toolchains. It clearly doesn't scale.

So we set up to discuss if somehow we could use multiarch to let debian packages using Qt to cross compile.

In the meantime Enrico Zini had the same idea. He wrote a nice summary of the situation at that time in his blog.

After many thinking some ideas were tested and we've got to the point of solving/hacking the issue. As this is not something directly supported by upstream you should take care, and file bugs whenever necessary.

Dmitry Schachnev from our team's side and Helmut from the debian-cross side worked a lot on it, and I would like to present what they have done. To be fair it's mostly described in our team's gobby qt-cross page, but I would like to give it some publicity in order to let people know about it and why not, find and help solving bugs.

General stuff

The first thing that was done was to move Qt binaries from their (Debian original) multi-arch path to a non multi-arch one, providing symlinks for compatibility. In this way the path of the binaries is the same for any arch (why they were not there is a long story, but nothing to worry now).

This move needed some other touches, like qtchooser being updated with the new paths.

The other changes where related to how we do our packaging:

• All packages containing binaries are now M-A:foreign.
• Some packages (qt3d, qtwayland) had binaries split to allow that.
• qttools5-dev-tools now depends on libqt5sql5-sqlite (not uploaded yet)

qmake related changes

We also needed to address qmake. To begin with we splitted the package containing it into qt5-qmake-bin (M-A:foreign) and qt5-qmake (M-A:same). The first one has the binaries and the second the relevant mkspecs for some arch.

The rest of the "magic" comes from debhelper. It generates a qt.conf file with the right paths for each cross compilation and also passes cross QMAKE_CC and QMAKE_CXX to qmake when needed.

autotools

qt5-qmake will ship /usr/bin/$(DEB_HOST_GNU_TYPE)-qmake executable for use with AC_CHECK_TOOL (not uploaded yet).

There is still work to be done, but so far we have been able to cross compile packages using for example sbuild.

Edit 20171129 11:43 ARST: You should really look at the new Enrico's post.

Removing Qt 4 from Debian testing (aka Buster): some statistics

I have just looked upon our Qt 4 removal wiki page to see how we are doing. Out of 438 bugs filled:

• 88 (20.09%) have been already fixed by either porting the app/library to Qt 5 or a removal from the archive has happened. On most cases the code has been ported and most of the deletions are due to Qt 5 replacements already available in the archive and a few due to dead upstreams.
• 3 (0.68%) packages are marked as fixed an an upload is pending.
• 65 (14.84%) of the open bugs are maintained inside the Qt/KDE team. Many of them should get a Qt 5 version with the next KF5 uploads.

We started filing bugs around September 9.  That means roughly 11 weeks, which gives us around 8 packages fixed a week, aka 1.14 packages per day. Not bad at all!

So, how can you help?

If you are a maintainer of any of the packages still affected try to get upstream to make a port and package it.

If you are not a maintainer you might want to take a look at the list of packages in our wiki page and try to create a patch for them. If you can submit it directly to upstream, the better.

Qt 4 and 5 and OpenSSL1.0 removal

Today we received updates on the OpenSSL 1.0 removal status:

<https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=828522#206>
<https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=859671#19>

So those removal bugs' severities will be raised to RC in aproximately a month.

We still don't have any solutions for Qt 4 or 5.

For the Qt 5 case we will probably keep the bug open until Qt 5.10 is in the archive which should bring OpenSSL 1.1 support *or* FTP masters decide to remove OpenSSL1.0. In this last case the fate will be the same as with Qt4, below.

For Qt4 we do not have patches available and there will probably be none in time (remember we do not have upstream support). That plus the fact that we are actively trying to remove it from the archive it means we will remove openssl support. This might mean that apps using Qt4:

- Might cease to work.
- Might keep working:
  - Informing their users that no SSL support is available → programmer did a good job.
  - Not informing their users that no SSL support is available and establishing connections non the less → programmer might have not done a good job.

Trying to inform users as soon as possible,

Lisandro for the Qt/KDE team.

Qt 4 removal in Debian testing (Buster)/unstable

We have been announcing it: we are going to remove Qt 4 during the Buster cycle.

Or at least that's the best outcome we can expect. Removing a very highly used library is hard, as Qt4's Webkit has proved. Qt 4 is long dead upstream and we have already started to need to patch it with untested patches as in the OpenSSL 1.1 case (will be in experimental in a few hours after this post).

We will try to put as less effort as possible in keeping it alive meaning that from now on if we need to patch it to make it support a newer lib or alike we will simply remove its support if possible. Using the OpenSSL case as an example, if we need to support any version > 1.1 we will simply remove the SSL support. That means things will break.

So, if you depend on FLOSS which is still based on Qt 4 be sure to try to port it. If you depend on a proprietary vendor software which uses Qt 4 then you better start telling them it's really time to update it. Really.

We will soon start filing bugs against packages using Qt 4. I'll update this blog post later to add that info.

For the Qt/KDE team, Lisandro.

Qt 5.7 submodules that didn't make it to Stretch but will be in testing

There are two Qt 5.7 submodules that we could not package in time for Strech but are/will be available in their 5.7 versions in testing. This are qtdeclarative-render2d-plugin and qtvirtualkeyboard.

declarative-render2d-plugin makes use of the Raster paint engine instead of OpenGL to render the  contents of a scene graph, thus making it useful when Qt Quick2 applications  are run in a system without OpenGL 2  enabled hardware. Using it might require tweaking Debian's /etc/X11/Xsession.d/90qt5-opengl. On Qt 5.9 and newer this plugin is merged in Qt GUI so there should be no need to perform any action on the user's behalf.

Debian's VirtualKeyboard currently has a gotcha: we are not building it with the embedded code it ships. Upstream ships 3rd party code but lacks a way to detect and use the system versions of them. See QTBUG-59594, patches are welcomed. Please note that we prefer patches sent directly upstream to the current dev revision, we will be happy to backport patches if necessary.
Yes, this means no hunspell, openwnn, pinyin, tcime nor lipi-toolkit/t9write support.

Note 2017-07-16: please use 5.7.1+dfsg-2, it fixes a wrong dependency and adds use of the system's hunspell.

Developing an nrf51822 based embedded device with Qt Creator and Debian

I'm currently developing an nRF51822-based embedded device. Being one the Qt/Qt Creator maintainers in Debian I would of course try to use it for the development. Turns out it works pretty good... with some caveats.

There are already two quite interesting blog posts about using Qt Creator on MAC and on Windows, so I will not repeat the basics, as they are there. Both use qbs, but I managed to use CMake.

Instead I'll add some tips on the stuff that I needed to solve in order to make this happen on current Debian Sid.

• The required toolchain is already in Debian, just install binutils-arm-none-eabi, gcc-arm-none-eabi and gdb-arm-none-eabi.
• You will not find arm-none-eabi-gdb-py on the gdb-arm-none-eabi package. Fear not, the provided gdb binary is compiled against python so it will work.
• To enable proper debugging be sure to follow this flag setup. If you are using CMake like in this example be sure to modify CMake/toolchain_gcc.cmake as necessary.
• In Qt Creator you might find that, while try to run or debug your app, you are greated with a message box that says "Cannot debug: Local executable is not set." Just go to Projects →Run and change "Run configuration" until you get a valid path (ie, a path to the .elf or .out file) in the "Executable" field.

Cheers!