Can someone help me understand what are the problems in running multiple QApplications on Qt for Embedded Linux? Please point me to some documentation of mailing-list threads.
While going through some of the mails in mailing lists, I have seen some comments which say that, running multiple QApplications in Qt is not supported by design and why at all it is required? How can I understand this more clearly?
However, while reading the document "Qt for Embedded Linux Architecture" I did not find anything which says that we should not run multiple QApplication instances at the same time.
I am executing two QApplications on a Embedded Linux platform (not a PC) and one of them in full screen mode. The one which is in fullScreen mode it is not getting keyboard focus, even though it receives mousePress events. If same app is run in normalMode, it gets the mousePress event followed by focusInEvent`.
Can somebody provide pointers on it?
You can run multiple processes each with one QApplication just fine.
However, with Qt for Embedded Linux, only one of these must be the QWS server. In other words, you should start the first process with `-qws', and all other processes without.
QApplication is a singleton class, so its "single" by design. You can have only a single QApplication object per program.
But in Qt there is no inherent limitation of the number of qt programs using the QApplication class you can run parallel. You can have more than one program using qt (and thus very likely QApplication) at the same time.
Probably this got somehow confused in your mailing lists.
My guess is that one QApplication would accept the mouse event or keyboard event and therefore the other would not get it.
Its probably a little random as to which QApplication accepts which events based on having so many QApplications in a single process.
I cannot imagine the use case as to why you would want multiple QApplications within a process. Could you expand on what you are trying to do?
Related
I am attempting to use QTCreator's Memcheck tool to analyze the memory of a QT project, which is a dynamic library. It uses a QT Gui (QMainWindow) to allow the user to select a file, which is then processed, and then eventually returns to the mainwindow.
However, I cannot seem to use Memcheck properly.
When I select "Memcheck" and hit run, it instantly goes to "Analyzing memory" without ever letting the Gui pop up.
This is problematic. How can I get memcheck to work with this program?
I had two main issues:
1: Valgrind does not seem to play nice with QT Gui applications. It generates logs that are thousands of entries long for all the work QT is doing before it even gets to my application.
I had to make a separate, small non-GUI C++ program that would drive instead of the GUI application.
2: When trying to run from the command line, I needed to set an environment variable by using export. This needs to be the same as LD_LIBRARY_PATHS in QT Creator.
So I ran:
export LD_LIBRARY_PATH=X where X was the exact value I copied from LD_LIBRARY_PATHS in the variable from the QT Project.
Note: Running from the command line may have not be neccesary now that it isn't a GUI Application, memcheck might have passed just fine. Haven't tested since.
I have some legacy code which has been running on an older Qt version where QList still had an autodelete feature on pointers (at least that's what the in-code comments said). So the GUI was passed pointers to data objects, stored them inside some Qt container and was responsible for deleting them appropriately when sth (user or internal signal) triggered a Qt event for deleting one of the data objects. With the auto deletion feature this seems to have worked smoothly but now that it's gone I'm running into memory problems under Qt5.
It seems like I somehow have to synchronize the deletion with the Qt message loop because it's every time triggered from within a slot. If I just hard delete it by myself, the program will crash at some point because pending Qt events try to access the deleted data when it's already gone.
I think deleteLater would already do the trick, however the objects in question are derived from QTableWidgetItem and thus don't own a deleteLater slot.
I am working on a QT application for which I've integrated DirectX 11 into a custom widget. The application renders a scrolling display - a graphical representation of data being read from a file. The user can speed up and slow down the scrolling speed.
For the most part, this is working great. The DirectX 11 rendering is presented to my custom widget just as I'd expect. The problem is that the graphics driver randomly hangs and crashes my program. I say "random" because I have been testing this with the same data file and it never seems to crash at the same point in the file, after a specific amount of time, or at a specific scrolling speed (the faster the scrolling speed, the more work being done by the GPU per frame).
When the application hangs, my screen freezes for a moment, goes black, then returns with a nice message from NVidia that it has recovered from a driver crash. The Debug Output in Visual Studio contains the following:
D3D11: Removing Device.
D3D11 ERROR: ID3D11Device::RemoveDevice:
Device removal has been triggered for the following reason
(DXGI_ERROR_DEVICE_HUNG: The Device took an unreasonable amount of
time to execute its commands, or the hardware crashed/hung. As a
result, the TDR (Timeout Detection and Recovery) mechanism has been
triggered. The current Device Context was executing commands when the
hang occurred. The application may want to respawn and fallback to
less aggressive use of the display hardware). [ EXECUTION ERROR #378:
DEVICE_REMOVAL_PROCESS_AT_FAULT]
I have discovered that by simply commenting out the IDXGISwapChain1::Present call, the application will run through the file at blazing speed. Graphics-wise it is still pushing data to the GPU and drawing to render targets, it just never gets displayed to my window.
What I'm hoping for is help with ideas of what types of things cause driver hangs. My shaders are incredibly simple - basically just positioning my vertices using a projection matrix. And considering what I described in the above paragraph, shaders should still be cranking through vertices and pixels even when Present isn't being called, yes?
I was suspicious that this could be a compatibility issue with Qt - I know DirectX isn't officially supported by Qt. So I tried creating a separate window using CreateWindowEx and using that for my swap chain instead of the custom Qt widget. It rendered to that window but also hung the driver just like before.
I was also suspicious of a driver bug in my laptop, so I tried running the application on a beefier desktop PC that regularly runs another DirectX 11 application (non-Qt) without a hitch (worth mentioning that this other application renders similar data in a scrolling display as well, using shaders that are a lot more complex). But my QT application hangs the driver on that PC as well.
Anyone know of a way I can get a more detailed description of what is causing the driver to hang?
Thank you in advance for any help you can offer.
UPDATE: 2013-08-01 17:16 CST
I am currently investigating a possible thread syncing issue that may be the culprit. Will continue tomorrow morning and post if I solve this on my own.
After some testing today, it appears to have been a threading issue. I have run several times today with no graphics crash. So my problem must be fixed, unless I've just been getting lucky with my tests today (or unlucky, rather - if this shows its ugly face again in a day or two).
I was aware that the immediate device context is not thread safe. Rather than using deferred contexts, though, I was using critical sections to sync my threads and coordinate use of the device context. What I did not realize is that it is not safe to call IDXGISwapChain1::Present while another thread is using the device context. Makes sense, but since it is not call directly from the device context itself, I overlooked it. I literally moved my Present() call a few lines up into my critical section block, and it hasn't given me a crash since.
Is there a framework to simulate a deterministically pseudorandom series of clicks imposed upon a Qt application - in order to try to trigger any memory leaks, threading errors etc - typical monkey-testing?
Some exotic, monkey use for QTestLib?
Yes, you can use the useful unit test module. Look in particular at the class QTestEventList. Just provide the QWidget you want to test, or the QMainWindow or whatever subclass you want and add the list of events you want to generate. If you want to generate a sequence of points so that you can reproduce in case of failure, use qsrand() and qrand().
The Squish automated GUI testing framework can be used to do monkey clicking in your application. There are a couple of nice things about using Squish for this purpose:
Squish runs on Windows, Mac, Linux and Android
It logs whatever random steps it happens to perform as a script that can be re-run (monkey testing docs)
It has an option to take a screenshot on application crash (screenshot on failure settings docs)
Disadvantages: Squish is not free. Setting up Squish to run your application can be super-irritating, especially if your application requires specific build characteristics of Qt or Python.
I was wondering is there is a fool-proof way to run a program on windows such that I'm guaranteed that no interactive dialogs of any kind are displayed.
I've tried the registry ErrorMode hack, calling _CrtSetReportMode(), etc., but they all have holes in them or require you to modify the program.
I need a way to run an arbitrary program and practically force Windows to execute them such that there is no possibility for them to open a window. It is perfectly ok for the program to crash if it attempts to open a window.
Would running the program as a service solve the problem?
I'm not aware of any other ways that might possibly work. Running as a service won't solve the problem either as the dialog will be displayed on the service's desktop, which you generally don't have access to.
You could use a library such as Detours to intercept all calls to functions that might display a dialog box (this might in fact be nearly everything in user32.dll).