Re: [maemo-developers] weird crashing... (X server badmatch)
Hi, why does it work on the scratchbox/xvnc/x86 through ? (i know there are many diffences in pratical terms, but in theory the behaviour should be the same, right ?) If you are starting the program form the 'scratchbox console' (and not from the xterm running inside maemo) If people run applications from Sbox console, they should use: run-standalone.sh app This sets up all the environment variables etc. required by D-BUS, Gtk themeing etc. the issue may as well be the osso theme bug described here : https://maemo.org/bugzilla/show_bug.cgi?id=179 Ah, that's the most likely problem. The BadMatch error can be also a resouce mismatch and not just a missing resource. :-) - Eero ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] weird crashing... (X server badmatch)
If you are starting the program form the 'scratchbox console' (and not from the xterm running inside maemo) Actually, I'm starting the program from the x-console app, from the device directly. On the scratchbox env, it works perfectly from the console (regardless the --sync option). And I do use the 'run-standalone.sh' script for both cases. Well, it is not an GTK/Hildon app. Its UI is built using the XUL [1] toolkit), as I mentioned before ! I am just trying this for experimental performance issues on the device. If people run applications from Sbox console, they should use: run-standalone.sh app This sets up all the environment variables etc. required by D-BUS, Gtk themeing etc. Humm.. good point once I'm not using GTK :S the issue may as well be the osso theme bug described here : https://maemo.org/bugzilla/show_bug.cgi?id=179 I hope to go deeper on this thread soon. The BadMatch error can be also a resouce mismatch and not just a missing resource. :-) yep, it is a really heavy app. regards -- Jorge Souza ___ Yahoo! Acesso Grátis: Internet rápida e grátis. Instale o discador agora! http://br.acesso.yahoo.com/ ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
[maemo-developers] weird crashing... (X server badmatch)
Hi, The program 'testing-xul' received an X Window Systemerror. This probably reflects a bug in the program. The error was 'BadMatch'. (Details: serial 2527 error_code 8 request_code 56 minor_code 0) Request code 56 is X_ChangeGC. (Request codes are listed in X books like Introduction to X Window System by Oliver Jones.) (Note to programmers: normally, X errors are reported asynchronously; that is, you will receive the error a while after causing it. To debug your program, run it with the --sync command line option to change this behavior. You can then get ameaningful backtrace from your debugger if you break on thegdk_x_error() function.) This error means that your application tried to access an X server resource that doesn't exist (anymore), either directly or through the libraries it used. For example you removed a window as a response to a button click, but you had an event callback for the window that handled some event for the window only after you had removed the window. In this case the resource was (down below) an X server graphics context, not a window. It starts up, but after some time, it just crash, printing as above. The '--sync' suggested option has no effect. Your stacktrace in GDB is not more readable? Normally --sync just forces the error to manifest immediately when it's done instead of later. If --sync option helps to fix the problem, in the above example it would mean that the two operations were not properly synchronized (X server works asynchronously to your program and events requests are buffered) and you can fix the problem with correctly placed XSync(). Or preferably, with the corresponding Gdk alternative. Any special reason for that ? The code is broken. - Eero PS. In some very rare cases (if you're writing a window manager), application needs to access X server resources which it doesn't control itself. In these cases it can trap the X error while accessing such a resource. ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] weird crashing... (X server badmatch)
Hummm ... very clarifying, Thanks. But the point is: why does it work on the scratchbox/xvnc/x86 through ? (i know there are many diffences in pratical terms, but in theory the behaviour should be the same, right ?) regards The program 'testing-xul' received an X Window Systemerror. This probably reflects a bug in the program. The error was 'BadMatch'. (Details: serial 2527 error_code 8 request_code 56 minor_code 0) Request code 56 is X_ChangeGC. (Request codes are listed in X books like Introduction to X Window System by Oliver Jones.) (Note to programmers: normally, X errors are reported asynchronously; that is, you will receive the error a while after causing it. To debug your program, run it with the --sync command line option to change this behavior. You can then get ameaningful backtrace from your debugger if you break on thegdk_x_error() function.) This error means that your application tried to access an X server resource that doesn't exist (anymore), either directly or through the libraries it used. For example you removed a window as a response to a button click, but you had an event callback for the window that handled some event for the window only after you had removed the window. In this case the resource was (down below) an X server graphics context, not a window. It starts up, but after some time, it just crash, printing as above. The '--sync' suggested option has no effect. Your stacktrace in GDB is not more readable? Normally --sync just forces the error to manifest immediately when it's done instead of later. If --sync option helps to fix the problem, in the above example it would mean that the two operations were not properly synchronized (X server works asynchronously to your program and events requests are buffered) and you can fix the problem with correctly placed XSync(). Or preferably, with the corresponding Gdk alternative. Any special reason for that ? The code is broken. - Eero PS. In some very rare cases (if you're writing a window manager), application needs to access X server resources which it doesn't control itself. In these cases it can trap the X error while accessing such a resource. ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] weird crashing... (X server badmatch)
Hi, But the point is: - why does it work on the scratchbox/xvnc/x86 through ? (i know there are many diffences in pratical terms, but in theory the behaviour should be the same, right ?) When it's a question about asynchronous operations, timings can matter: - N770 is much slower than Desktop - If you're on x86 using DISPLAY variable value with host part (as is erronously mentioned in some Maemo docs, see also Maemo bug 22), and/or remote display, X server will use inet socket instead of a unix one and assumably also disables the use of shared memory. This will slow client image operations (like drawing the widgets) a lot. Of course the problem could be something else too, like different SW versions on the target and x86, but I assume you've already checked that? - Eero ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] weird crashing... (X server badmatch)
On 22.11.2005 16:56, Antônio Gomes wrote: Hummm ... very clarifying, Thanks. But the point is: why does it work on the scratchbox/xvnc/x86 through ? (i know there are many diffences in pratical terms, but in theory the behaviour should be the same, right ?) If you are starting the program form the 'scratchbox console' (and not from the xterm running inside maemo) the issue may as well be the osso theme bug described here : https://maemo.org/bugzilla/show_bug.cgi?id=179 To check run export GTK2_RC_FILES= before starting on 770 or set it to the osso/sapwood theme file in your 'scratchbox console' to reproduce the crash on x86. Hans at Breuer dot Org --- Tell me what you need, and I'll tell you how to get along without it.-- Dilbert ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
