[Paraview-developers] RenderView management python segfault

[Mathieu Westphal]

Hello All

I'm experiencing a systematic segfault with the following code in pvpython :

from paraview.simple import *
import gc

v=CreateRenderView()
Render(v)
w=CreateRenderView()
Render(w)

Delete(v)
del v
gc.collect()  # not absolutelly necessary, the destructor will be called
eventually, but this ensure the bug to happen at the next call.

Render(w)


[xcb] Unknown sequence number while appending request
[xcb] Most likely this is a multi-threaded client and XInitThreads has not
been called
[xcb] Aborting, sorry about that.
pvpython: xcb_io.c:161: append_pending_request: Assertion
`!xcb_xlib_unknown_seq_number' failed.
Aborted (core dumped)

Basically,
 * Creating and rendering a first view,
 * then Creating and rendering a second render view,
 * then destroying the first view, will put the second in a unstable state
that will lead to a segfault.

Does not reproduce in paraview, pvpython only.

I've added a bug in paraview bugtracker :
http://www.paraview.org/Bug/view.php?id=16767

If you have any idea why does it happen and how to correct it it would be
great !

Mathieu Westphal

On Fri, Jul 1, 2016 at 5:19 PM, Mathieu Westphal <
mathieu.westphal at kitware.com> wrote:

> Hello
>
> I'm experiencing a systematic segfault with the following code in pvpython
> :
>
> from paraview.simple import *
> import gc
>
> v=CreateRenderView()
> Render(v)
> w=CreateRenderView()
> Render(w)
>
> Delete(v)
> del v
> gc.collect()  # not absolutally necessary, the destructor will be called
> eventually
>
> Render(w)
>
>
> [xcb] Unknown sequence number while appending request
> [xcb] Most likely this is a multi-threaded client and XInitThreads has not
> been called
> [xcb] Aborting, sorry about that.
> pvpython: xcb_io.c:161: append_pending_request: Assertion
> `!xcb_xlib_unknown_seq_number' failed.
> Aborted (core dumped)
>
> Basically, Creating and rendering a first view, then Creating and
> rendering a second render view, then destroying the first view, will put
> the second in a unstable state that will lead to a segfault.
> Does not reproduce in paraview.
>
> I've added a bug in paraview bugtracker :
> http://www.paraview.org/Bug/view.php?id=16767
>
> If you have any idea why does it happen and how to correct it it would be
> great !
>
> Mathieu Westphal
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20160810/cdef197e/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bugRender.py
Type: text/x-python
Size: 144 bytes
Desc: not available
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20160810/cdef197e/attachment.py>