Inline... On Wed, 2009-10-14 at 16:41 +0200, Mirko Vogt wrote: > Hey, thanks for your suggestions! > > One step closer - there was an uninitialized variable within the > framebuffer code :) > > Now I can run: > > ==== > r...@bennanonote:/# dfbg --dfb:debug -c ffffff > > ~~~~~~~~~~~~~~~~~~~~~~~~~~| DirectFB 1.4.2 | > ~~~~~~~~~~~~~~~~~~~~~~~~~~ > (c) 2001-2009 The world wide DirectFB Open Source Community > (c) 2000-2004 Convergence (integrated media) GmbH > ---------------------------------------------------------------- > > (*) DirectFB/Core: Single Application Core. (2009-10-14 02:42) > (*) Direct/Thread: Started 'VT Switcher' (633) [CRITICAL OTHER/OTHER > 0/0] <2093056>... > (*) Direct/Thread: Started 'VT Flusher' (634) [DEFAULT OTHER/OTHER 0/0] > <2093056>... > (*) DirectFB/FBDev: Found 'JZ4740 FB' (ID 0) with frame buffer at > 0x01980000, 300k (MMIO 0x13050000, 64k) > (*) Direct/Thread: Started 'Keyboard Input' (635) [INPUT OTHER/OTHER > 0/0] <2093056>... > (*) DirectFB/Input: Keyboard 0.9 (directfb.org) > (*) DirectFB/Graphics: Generic Software Rasterizer 0.6 (directfb.org) > (*) DirectFB/Core/WM: Default 0.3 (directfb.org) > ==== > > without freezing the device. But the display is only flickering but the > bg isn't changing.
Okay, it seems the given color is flickering over the display. Tried it with a well-known-working device and it's the same here - calling "dfbg -c 222ccc" "flashes" the display blue - on both devices, so it seems to work! > > Running the gtk2-app "gtkperf" (gtk2 linked against directfb) prints the > following: > > ==== > r...@bennanonote:/# gtkperf > commandline read: gtkperf > > ~~~~~~~~~~~~~~~~~~~~~~~~~~| DirectFB 1.4.2 | > ~~~~~~~~~~~~~~~~~~~~~~~~~~ > (c) 2001-2009 The world wide DirectFB Open Source Community > (c) 2000-2004 Convergence (integrated media) GmbH > ---------------------------------------------------------------- > > (*) DirectFB/Core: Single Application Core. (2009-10-14 02:42) > (*) Direct/Thread: Started 'VT Switcher' (638) [CRITICAL OTHER/OTHER > 0/0] <2093056>... > (*) Direct/Thread: Started 'VT Flusher' (639) [DEFAULT OTHER/OTHER 0/0] > <2093056>... > (*) DirectFB/FBDev: Found 'JZ4740 FB' (ID 0) with frame buffer at > 0x01980000, 300k (MMIO 0x13050000, 64k) > (*) Direct/Thread: Started 'Keyboard Input' (640) [INPUT OTHER/OTHER > 0/0] <2093056>... > (*) DirectFB/Input: Keyboard 0.9 (directfb.org) > (*) DirectFB/Graphics: Generic Software Rasterizer 0.6 (directfb.org) > (*) DirectFB/Core/WM: Default 0.3 (directfb.org) > > (gtkperf:636): Gdk-CRITICAL **: gdk_drawable_get_colormap: assertion > `GDK_IS_DRAWABLE (drawable)' failed > (*) Direct/Thread: Started 'EventBufferFeed' (641) [MESSAGING > OTHER/OTHER 0/0] <2093056>... > (!) [ 636: 0.000] --> Caught signal 11 (at 0x28, invalid address) > <-- > (!!!) *** WARNING [still objects in 'Window Pool'] *** [object.c:241 in > fusion_object_pool_destroy()] > (!!!) *** WARNING [still objects in 'Layer Region Pool'] *** > [object.c:241 in fusion_object_pool_destroy()] > (!!!) *** WARNING [still objects in 'Layer Context Pool'] *** > [object.c:241 in fusion_object_pool_destroy()] > (!!!) *** WARNING [still objects in 'Surface Pool'] *** [object.c:241 > in fusion_object_pool_destroy()] > Aborted > ==== I tried this also with a well-known-working device - same problem. Seems like an gtk/gdk-error to me. I found much people, while looking around in the webm who have the same problem with other gtk-applications. However I found no solution or even suggestion at all that might help :/ Thanks again, mirko > > Any kind of help is highly appreciated - thanks a lot in advance! > > Mirko > > > On Fri, 2009-10-09 at 13:18 +0200, Niels Roest wrote: > > Hi Mirko, > > assuming you are using remote debugging.. > > > > the easiest way is to rebuild directfb from sources, with the > > "./configure" option "--enable-debug", and then run your program as > > "dfbg --dfb:debug -c ffffff". This will print out lots of debugging, and > > should give you an indication after which command it crashes. If your > > link doesn't cope with the traffic you can dump to a file with the extra > > option "--dfb:log-file=filename.txt". I expect the problem to be in the > > area of frame buffer ioctls, which is normally a pain to debug - think > > kernel frame buffer driver. > > > > hth > > Niels > > > > > > Mirko Vogt wrote: > > > Hey, > > > > > > I wanna run DirectFB on the "Ben NanoNote" by qi-hardware > > > (http://www.qi-hardware.com/products/ben-nanonote/) so I packaged the > > > library for the embedded distribution OpenWrt which I'm running on the > > > Ben. > > > Unfortunately running the shipped tools (e.g. dfbg) cause the device to > > > freeze reproducible: > > > > > > r...@openwrt:~# dfbg -c ffffff > > > > > > ~~~~~~~~~~~~~~~~~~~~~~~~~~| DirectFB 1.4.2 | > > > ~~~~~~~~~~~~~~~~~~~~~~~~~~ > > > (c) 2001-2009 The world wide DirectFB Open Source Community > > > (c) 2000-2004 Convergence (integrated media) GmbH > > > ---------------------------------------------------------------- > > > > > > (*) DirectFB/Core: Single Application Core. (2009-10-08 01:40) > > > (*) Direct/Thread: Started 'VT Switcher' (547) [CRITICAL OTHER/OTHER > > > 0/0] <2093056>... > > > (*) Direct/Thread: Started 'VT Flusher' (548) [DEFAULT OTHER/OTHER 0/0] > > > <2093056>... > > > (*) DirectFB/FBDev: Found 'JZ4740 FB' (ID 0) with frame buffer at > > > 0x00000000, 300k (MMIO 0x13050000, 64k) > > > (*) Direct/Thread: Started 'Keyboard Input' (549) [INPUT OTHER/OTHER > > > 0/0] <2093056>... > > > (*) DirectFB/Input: Keyboard 0.9 (directfb.org) > > > (*) DirectFB/Graphics: Generic Software Rasterizer 0.6 (directfb.org) > > > Connection to 192.168.1.1 closed. > > > > > > The display turns black and it does not react anymore at all (e.g. no > > > ICMP-responses, etc.) > > > > > > I know that's not much information - how to get some more useful debug > > > info? > > > How to start debugging such issues? > > > > > > Thanks a lot in advance! > > > > > > Regards > > > > > > mirko > > > > > > > > > > > > > > > > > > > > -- This email address is used for mailinglist purposes only. Non-mailinglist emails will be dropped automatically. If you want to get in contact with me personally, please mail to: mirko.vogt <at> nanl <dot> de _______________________________________________ directfb-dev mailing list directfb-dev@directfb.org http://mail.directfb.org/cgi-bin/mailman/listinfo/directfb-dev
