On Mon, Mar 21, 2016 at 2:50 AM, Jasvir Singh wrote:
> It's working fine now.
> http://screencloud.net/v/528K
I have submitted a patch: https://sourceforge.net/p/brlcad/patches/437/
Please review this one.
--
Jasvir Singh Grewal
https://github.com/jasvir99
-
On Mon, Mar 21, 2016 at 4:22 AM, Clifford Yapp wrote:
> Does the red line in the lower left indicate you are doing drawing?
> If so, that's probably activating/refreshing the view.
>
I tried drawing a simple line using GL Lines.
> If I were going to suggest "first steps" for work with the Qt dis
On Mon, Mar 21, 2016 at 5:50 AM, Jasvir Singh wrote:
> On Fri, Mar 11, 2016 at 11:06 AM, Clifford Yapp wrote:
>>
>> That looks like it's just the placeholder OpenGL Qt widget not getting
>> updated - remember, there's no drawing activity going on so the canvas
>> area is not refreshing itself.
>
On Fri, Mar 11, 2016 at 11:06 AM, Clifford Yapp wrote:
>> Can anyone please help me to figure it out?
>
> That looks like it's just the placeholder OpenGL Qt widget not getting
> updated - remember, there's no drawing activity going on so the canvas
> area is not refreshing itself.
It's working f
On Fri, Mar 11, 2016 at 11:06 AM, Clifford Yapp wrote:
> http://screencloud.net/v/ori3
>>
>> Can anyone please help me to figure it out?
>
> That looks like it's just the placeholder OpenGL Qt widget not getting
> updated - remember, there's no drawing activity going on so the canvas
> area is not
On Fri, Mar 11, 2016 at 1:37 PM, Jasvir Singh wrote:
> On Thu, Mar 10, 2016 at 10:08 PM, Jasvir Singh
> wrote:
>>> Cheers!
>>> Sean
>>
>> As suggested by Clifford, previous code is working fine.
>
> Everything is working fine except that when I star qged, draw area
> opens with a screenshot of b
On Thu, Mar 10, 2016 at 10:08 PM, Jasvir Singh wrote:
>> Cheers!
>> Sean
>
> As suggested by Clifford, previous code is working fine.
Everything is working fine except that when I star qged, draw area
opens with a screenshot of background.
http://screencloud.net/v/ori3
Can anyone please help me
On Tue, Mar 8, 2016 at 2:43 PM, Christopher Sean Morrison
wrote:
> So first step is probably to get familiarized with the old code, get it up
> and running, perhaps submit a patch if changes are needed to get it working
> again.
>
> Cheers!
> Sean
As suggested by Clifford, previous code is work
On Tue, Mar 8, 2016 at 5:43 PM, Christopher Sean Morrison
wrote:
>
>> On Mar 8, 2016, at 3:58 PM, Jasvir Singh wrote:
>>
>> Further I wish to know about whether someone is working on qtged or
>> not. I am trying to understand working of MGED and I wish to
>> contribute into qtged. I'll really app
> On Mar 8, 2016, at 3:58 PM, Jasvir Singh wrote:
>
> Further I wish to know about whether someone is working on qtged or
> not. I am trying to understand working of MGED and I wish to
> contribute into qtged. I'll really appreciate if someone can help me
> to start with it.
Nobody is working o
2013/9/22 Vlad Bogolin :
>> PS: An idea for a fix of the segmentation fault bug: This could be
>> caused by mixing the single- and multithread mode of libX11. I.e. the
>> main window will be opened from Tk in singlethread mode. The Qt
>> switches to multithread and closing the main window will le
Hi,
As a conclusion to your GSoC project you should write down what was
> achieved and what's still open ("not implemented" functions, frame
> buffer, ...)
>
> I have made a summary wiki page of my project that can be found here:
http://brlcad.org/wiki/User:Vladbogolin/GSoC2013/qt-display-manager
Vlad,
You should make a break in searching for the reason of the
segmentation fault and polish your work a little bit before GSoC ends.
There are still some functions "not implemented". Maybe some of them
are trivial.
In qt_close() you write privars->win->~QTkMainWindow();. That's
rather unusu
Hi,
First of all thanks for the hints.
>
> > Next you could put break points where the resource is allocated
> > (NewDatabase) and destroyed (XrmDestroyDatabase). In case of the Qt
> > display manager the database resource will be destroyed without
> > creating it before. That's probable not op
2013/9/18 Daniel Roßberg :
> Next you could put break points where the resource is allocated
> (NewDatabase) and destroyed (XrmDestroyDatabase). In case of the Qt
> display manager the database resource will be destroyed without
> creating it before. That's probable not optimal.
Ups, that's not
2013/9/18 Vlad Bogolin :
>> You really need a stack trace or debugging information of some sort to go
>> on. You can also try to compile BRL-CAD as single-threaded to see if the
>> problem is specific to something Qt's doing or our parallelism. (cmake
>> -DBRLCAD_ENABLE_SMP=NO)
>>
> I've did a ba
Hi,
> Can you reproduce the segfault in the debugger?
The segfault can be reproduced in the debugger but I've managed to catch it
just when not using classic mged (without the -c option).
You really need a stack trace or debugging information of some sort to go
> on. You can also try to comp
I can reproduce the segmentation fault but I have no hint how to solve
this issue.
It looks like the function pthread_mutex_lock() is called with an
invalid (NULL) mutex. This happens deep inside libX11 called by Tk.
So my assumption was that Qt and Tk disagree on the ownership of a X11
resource.
On Sep 15, 2013, at 2:15 PM, Vlad Bogolin wrote:
> When closing the qt display manager sometimes a segfault occurs. I've seen
> that this occurs after the close function is called and gdb says the
> fallowing:
>
> Program received signal SIGSEGV, Segmentation fault.
> __GI___pthread_mutex_lock
Hi all,
When closing the qt display manager sometimes a segfault occurs. I've seen
that this occurs after the close function is called and gdb says the
fallowing:
Program received signal SIGSEGV, Segmentation fault.
__GI___pthread_mutex_lock (mutex=0x0) at pthread_mutex_lock.c:50
50pthread_mu
On Sep 11, 2013, at 4:16 AM, Vlad Bogolin wrote:
> I modified the code so that every time Tcl_DoOneEvent is called
> QApplication::processEvents is also called so that there aren't any
> modification to the code that have to be done besides checking for Tk events
> non-blocking. I couldn't fig
Hi,
I found another solution, actually more like a hack that simulates
non-blocking behavior. I use Tcl_DoWhenIdle to call a procedure and then
reschedule the procedure using Tcl_CreateTimerHandler but I don't think is
the best solution so I'll make a little more research. What do you think?
Do y
Vlad,
Unfortunately I won't be able to look at it before next Monday (I'm
currently not at home).
Daniel
Am 11.09.2013 10:16 schrieb "Vlad Bogolin" :
> Hi,
>
> I saw that. Can you think of a way to register your Qt event loop
>> processing without directly modifying bwish?
>>
>
> I modified the
Hi,
I saw that. Can you think of a way to register your Qt event loop
> processing without directly modifying bwish?
>
I modified the code so that every time Tcl_DoOneEvent is called
QApplication::processEvents is also called so that there aren't any
modification to the code that have to be done
On Sep 08, 2013, at 03:01 PM, Vlad Bogolin wrote:I have finished the Qt event processing in archer. Now everything should be displayed properly, but I'm not sure if I used the best approach as I had troubles finding the connection between bwish and archer so I am open to suggestions or improvement
Hi,
I have finished the Qt event processing in archer. Now everything should be
displayed properly, but I'm not sure if I used the best approach as I had
troubles finding the connection between bwish and archer so I am open to
suggestions or improvements.
Cheers,
Vlad
Hi,
The version of the Qt display manager at SourceForge still crashes
> when used in archer. What stops you from checking in your changes so
> we all could have a look at it?
>
Sorry about that. I was working in processing the events in a better way
and I forgot to commit the changes. Now, arch
Mesut,
Some thought about your patch:
First of all it looks extensive as if it could work. But I couldn't
compile it because of many unused-parameter errors. There are
differences between the gcc versions. If you use a more current one
you should see these errors too.
Some smaller things:
- Th
2012/6/5 "Mesut Özdogan"
>
> Hi Daniel,
>
> thank you for your help. You're right, i have done it the wrong way. I
> wanted to add the functions step by step, which could only compiled with
> brlcad after all was finished...
I recommend to start with the task described in my mail i.e. dm-qt.c.
>
29 matches
Mail list logo
