[Paraview] Paraview 4.3.1 OSMesa build - QWidget::repaint: Recursive repaint detected
Utkarsh Ayachit
utkarsh.ayachit at kitware.com
Wed Sep 9 10:21:40 EDT 2015
I have a feeling something's wrong with the OSMesa build. Try running
pvpython/pvbatch on that build. Are you able to render/save images from the
Python clients?
Utkarsh
On Wed, Sep 9, 2015 at 9:52 AM, David Trudgian <
David.Trudgian at utsouthwestern.edu> wrote:
> Have sent the CMakeCache.txt off list as it’s a couple hundred KB.
>
>
>
> I still get the recursive paint error on the client – it immediately
> precedes the warning about the 3/4 component images from the client, and
> the crash on the server. In the non-debug server I don’t get a segfault –
> but it becomes completely unresponsive to further connection attempts.
>
>
>
> The recursive paint error is server dependent. If I use the client to
> connect to a standard binary, or a self-built OpenGL pvserver then no
> issues. Only get the recursive paint issue when I try to connect to the
> self-built OSMesa pvserver.
>
>
>
> Thanks again,
>
>
>
> DT
>
>
>
>
>
> --
> David Trudgian Ph.D.
> Computational Scientist, BioHPC
> UT Southwestern Medical Center
> Dallas, TX 75390-9039
> Tel: (214) 648-4833
>
>
>
> *From:* Utkarsh Ayachit [mailto:utkarsh.ayachit at kitware.com]
> *Sent:* Tuesday, September 8, 2015 4:21 PM
>
> *To:* David Trudgian <David.Trudgian at UTSouthwestern.edu>
> *Cc:* paraview at paraview.org
> *Subject:* Re: [Paraview] Paraview 4.3.1 OSMesa build - QWidget::repaint:
> Recursive repaint detected
>
>
>
> Mind attaching the CMakeCache.txt from the server build? BTW, this is
> totally different error than what we started with. The original issue was
> that the client was complaining about recursive paint. Seems to me the
> pvserver build is a little messed up right now.
>
>
>
> On Tue, Sep 8, 2015 at 4:24 PM, David Trudgian <
> David.Trudgian at utsouthwestern.edu> wrote:
>
> Hi,
>
>
>
> Built a debug version. On the client side I now noticed the following
> warning on the console after I acknowledge the error dialog:
>
>
>
> Generic Warning: In
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/VTK/Rendering/Parallel/vtkSynchronizedRenderers.cxx,
> line 730
>
> Only 3 or 4 component images are handled.
>
>
>
> The pvserver stack trace is below. I guess this is something to do with
> the OSMesa config?
>
>
>
> Program received signal SIGSEGV, Segmentation fault.
>
> try_update_scene_state (setup=0x1d66380) at lp_setup.c:1050
>
> 1050 fstored[i] = setup->blend_color.current.color[i % 4];
>
> Missing separate debuginfos, use: debuginfo-install
> glibc-2.12-1.166.el6_7.1.x86_64 infinipath-psm-3.1-364.1140_open.x86_64
> libcxgb3-1.3.1mlnx1-OFED.2.0.2.6.2.28.g58e86c2.x86_64
> libcxgb4-1.3.0mlnx1-OFED.2.0.2.6.2.28.g3287317.x86_64
> libgcc-4.4.7-11.el6.x86_64 libibmad-1.3.9.MLNX20130522.1e79ec6-0.1.x86_64
> libibumad-1.3.8.MLNX20130522.da65ddf-0.1.x86_64
> libibverbs-1.1.6mlnx2-OFED.2.0.2.6.8.10.g4c4c3f1.x86_64
> libipathverbs-1.2mlnx-1.x86_64
> libmlx4-1.0.4mlnx2-OFED.2.0.2.6.7.8.g5ddd0ea.x86_64
> libmlx5-1.0.0-OFED.2.0.3.1.0.3.g00bb19b.x86_64
> libnes-1.1.3mlnx1-OFED.2.0.2.6.3.1.gae2c5c5.x86_64
> librdmacm-1.0.17mlnx2-OFED.2.0.2.6.2.22.g13cb0e8.x86_64
> libstdc++-4.4.7-11.el6.x86_64 ncurses-libs-5.7-3.20090208.el6.x86_64
> nss-pam-ldapd-0.7.5-20.el6_6.3.x86_64 numactl-2.0.9-2.el6.x86_64
> opensm-libs-4.0.5.MLNX20130808.c2b40b1-0.1.x86_64 zlib-1.2.3-29.el6.x86_64
>
> (gdb) set loggin on
>
> Copying output to gdb.txt.
>
> (gdb) bt
>
> #0 try_update_scene_state (setup=0x1d66380) at lp_setup.c:1050
>
> #1 0x00002aaabe71d05d in begin_binning (setup=0x1d66380) at lp_setup.c:207
>
> #2 0x00002aaabe71d6b5 in execute_clears (setup=0x1d66380, new_state=224,
> reason=0x1 <Address 0x1 out of bounds>)
>
> at lp_setup.c:279
>
> #3 set_scene_state (setup=0x1d66380, new_state=224, reason=0x1 <Address
> 0x1 out of bounds>) at lp_setup.c:327
>
> #4 0x00002aaabe71d4b0 in lp_setup_flush (setup=0x1d66380, fence=0xe0,
> reason=0x1 <Address 0x1 out of bounds>)
>
> at lp_setup.c:359
>
> #5 0x00002aaabe6fa7d0 in llvmpipe_flush (pipe=0x1d66380, resource=0xe0,
> level=1, read_only=0 '\000',
>
> cpu_access=0 '\000', do_not_block=128 '\200', reason=0x2aaabe7bfe30
> "llvmpipe_transfer_map") at lp_flush.c:55
>
> #6 llvmpipe_finish (pipe=0x1d66380, resource=0xe0, level=1, read_only=0
> '\000', cpu_access=0 '\000',
>
> do_not_block=128 '\200', reason=0x2aaabe7bfe30
> "llvmpipe_transfer_map") at lp_flush.c:82
>
> #7 llvmpipe_flush_resource (pipe=0x1d66380, resource=0xe0, level=1,
> read_only=0 '\000', cpu_access=0 '\000',
>
> do_not_block=128 '\200', reason=0x2aaabe7bfe30
> "llvmpipe_transfer_map") at lp_flush.c:120
>
> #8 0x00002aaabe734cdf in llvmpipe_transfer_map (pipe=0x1d66380,
> resource=0xe0, level=1, usage=0, box=0x2200100,
>
> transfer=0x2200080) at lp_texture.c:521
>
> #9 0x00002aaabe4399fa in pipe_transfer_map (ctx=0x1d66380, rb=0xe0, x=1,
> y=0, w=35651840, h=35651712, mode=1,
>
> mapOut=0x7fffffff8eb8, rowStrideOut=0x7fffffff8ee0) at
> ../../src/gallium/auxiliary/util/u_inlines.h:447
>
> #10 st_MapRenderbuffer (ctx=0x1d66380, rb=0xe0, x=1, y=0, w=35651840,
> h=35651712, mode=1, mapOut=0x7fffffff8eb8,
>
> rowStrideOut=0x7fffffff8ee0) at state_tracker/st_cb_fbo.c:796
>
> #11 0x00002aaabe365d22 in readpixels_memcpy (ctx=0x1d66380, x=0, y=0,
> width=0, height=30864256, format=35651712,
>
> type=1396, packing=0x1908, pixels=0x22f1040) at main/readpix.c:226
>
> #12 _mesa_readpixels (ctx=0x1d66380, x=0, y=0, width=0, height=30864256,
> format=35651712, type=1396, packing=0x1908,
>
> pixels=0x22f1040) at main/readpix.c:872
>
> #13 0x00002aaabe43bb87 in st_readpixels (ctx=0x1d66380, x=224, y=1,
> width=0, height=35651840, format=35651712,
>
> type=6408, pack=0x574, pixels=0x2e6) at
> state_tracker/st_cb_readpixels.c:227
>
> #14 0x00002aaabe366a25 in _mesa_ReadnPixelsARB (x=30827392, y=224,
> width=1, height=0, format=35651840,
>
> type=35651712, pixels=0x0) at main/readpix.c:1088
>
> #15 _mesa_ReadPixels (x=30827392, y=224, width=1, height=0,
> format=35651840, type=35651712, pixels=0x0)
>
> at main/readpix.c:1096
>
> #16 0x00002aaabdd33cab in vtkOpenGLRenderWindow::GetRGBACharPixelData
> (this=0x1d56080, x1=0, y1=0, x2=741, y2=1395,
>
> front=1, data=0x22f1040 "")
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/VTK/Rendering/OpenGL/vtkOpenGLRenderWindow.cxx:1225
>
> #17 0x00002aaabdd30f25 in vtkOpenGLRenderWindow::GetRGBACharPixelData
> (this=0x1d56080, x1=0, y1=0, x2=741, y2=1395,
>
> front=1, data=0x1d7a2e0)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/VTK/Rendering/OpenGL/vtkOpenGLRenderWindow.cxx:1159
>
> #18 0x00002aaabac03127 in vtkSynchronizedRenderers::vtkRawImage::Capture
> (this=0x1d87c80, ren=0x1d562a0)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/VTK/Rendering/Parallel/vtkSynchronizedRenderers.cxx:796
>
> #19 0x00002aaabac01e5b in vtkSynchronizedRenderers::CaptureRenderedImage
> (this=0x1d87c20)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/VTK/Rendering/Parallel/vtkSynchronizedRenderers.cxx:269
>
> #20 0x00002aaab703249c in
> vtkPVClientServerSynchronizedRenderers::SlaveEndRender (this=0x1d87c20)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/ParaViewCore/ClientServerCore/Rendering/vtkPVClientServerSynchronizedRenderers.cxx:80
>
> #21 0x00002aaabac01b27 in vtkSynchronizedRenderers::HandleEndRender
> (this=0x1d87c20)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/VTK/Rendering/Parallel/vtkSynchronizedRenderers.cxx:227
>
> #22 0x00002aaabac051f9 in vtkSynchronizedRenderers::vtkObserver::Execute
> (this=0x1cfd720, eventId=4)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/VTK/Rendering/Parallel/vtkSynchronizedRenderers.cxx:61
>
> #23 0x00002aaab2a97906 in vtkSubjectHelper::InvokeEvent (this=0x1d6ab60,
> event=4, callData=0x0, self=0x1d562a0)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/VTK/Common/Core/vtkObject.cxx:616
>
> #24 0x00002aaab2a950be in vtkObject::InvokeEvent (this=0x1d562a0, event=4,
> callData=0x0)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/VTK/Common/Core/vtkObject.cxx:785
>
> #25 0x00002aaaad499189 in vtkRenderer::Render (this=0x1d562a0)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/VTK/Rendering/Core/vtkRenderer.cxx:344
>
> #26 0x00002aaaad4957dc in vtkRendererCollection::Render (this=0x1d3c340)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/VTK/Rendering/Core/vtkRendererCollection.cxx:51
>
> #27 0x00002aaaad4bcde0 in vtkRenderWindow::DoStereoRender (this=0x1d56080)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/VTK/Rendering/Core/vtkRenderWindow.cxx:771
>
> #28 0x00002aaaad4bd728 in vtkRenderWindow::DoFDRender (this=0x1d56080)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/VTK/Rendering/Core/vtkRenderWindow.cxx:740
>
> #29 0x00002aaaad4be23b in vtkRenderWindow::DoAARender (this=0x1d56080)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/VTK/Rendering/Core/vtkRenderWindow.cxx:619
>
> #30 0x00002aaaad4b7e14 in vtkRenderWindow::Render (this=0x1d56080)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/VTK/Rendering/Core/vtkRenderWindow.cxx:435
>
> #31 0x00002aaab70aa249 in vtkPVRenderView::Render (this=0x1d10440,
> interactive=false, skip_rendering=false)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/ParaViewCore/ClientServerCore/Rendering/vtkPVRenderView.cxx:1232
>
> #32 0x00002aaab70a35d8 in vtkPVRenderView::StillRender (this=0x1d10440)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/ParaViewCore/ClientServerCore/Rendering/vtkPVRenderView.cxx:1082
>
> #33 0x00002aaaaba412e1 in vtkPVRenderViewCommand (arlu=0x7b9fc0,
> ob=0x1d10440, method=0xd60699 "StillRender",
>
> msg=..., resultStream=...)
>
> at
> /home2/dtrudgian/paraview/buildosmesa2-debug/Wrapping/ClientServer/vtkPVRenderViewClientServer.cxx:201
>
> #34 0x00002aaaafd2df55 in vtkClientServerInterpreter::CallCommandFunction
> (this=0x7b9fc0,
>
> cname=0x2aaab7121580 "vtkPVRenderView", ptr=0x1d10440, method=0xd60699
> "StillRender", msg=..., result=...)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/ParaViewCore/ClientServerStream/vtkClientServerInterpreter.cxx:848
>
> #35 0x00002aaaafd3035a in vtkClientServerInterpreter::ProcessCommandInvoke
> (this=0x7b9fc0, css=..., midx=0)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/ParaViewCore/ClientServerStream/vtkClientServerInterpreter.cxx:436
>
> #36 0x00002aaaafd2bd95 in vtkClientServerInterpreter::ProcessOneMessage
> (this=0x7b9fc0, css=..., message=0)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/ParaViewCore/ClientServerStream/vtkClientServerInterpreter.cxx:269
>
> #37 0x00002aaaafd2bbf3 in vtkClientServerInterpreter::ProcessStream
> (this=0x7b9fc0, css=...)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/ParaViewCore/ClientServerStream/vtkClientServerInterpreter.cxx:238
>
> #38 0x00002aaaac6dd65d in vtkPVSessionCore::ExecuteStreamInternal
> (this=0x7b9f00, stream=..., ignore_errors=false)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/ParaViewCore/ServerImplementation/Core/vtkPVSessionCore.cxx:669
>
> #39 0x00002aaaac6d931b in vtkPVSessionCore::ExecuteStream (this=0x7b9f00,
> location=21, stream=...,
>
> ignore_errors=false)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/ParaViewCore/ServerImplementation/Core/vtkPVSessionCore.cxx:639
>
> #40 0x00002aaaac6d6434 in vtkPVSessionBase::ExecuteStream (this=0x7b9b80,
> location=21, stream=...,
>
> ignore_errors=false)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/ParaViewCore/ServerImplementation/Core/vtkPVSessionBase.cxx:16---Type
> <return> to continue, or q <return> to quit---
>
> 8
>
> #41 0x00002aaaac6ed1e0 in vtkPVSessionServer::OnClientServerMessageRMI
> (this=0x7b9b80, message=0xd58940,
>
> message_length=16)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:552
>
> #42 0x00002aaaac6edc50 in (anonymous namespace)::RMICallback
> (localArg=0x7b9b80, remoteArg=0xd58940,
>
> remoteArgLength=16)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:56
>
> #43 0x00002aaab0bd1127 in vtkMultiProcessController::ProcessRMI
> (this=0x1cd1460, remoteProcessId=1, arg=0xd58940,
>
> argLength=16, rmiTag=55625)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/VTK/Parallel/Core/vtkMultiProcessController.cxx:774
>
> #44 0x00002aaab0bd02e4 in vtkMultiProcessController::ProcessRMIs
> (this=0x1cd1460, reportErrors=0, dont_loop=1)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/VTK/Parallel/Core/vtkMultiProcessController.cxx:719
>
> #45 0x00002aaaacad0970 in
> vtkTCPNetworkAccessManager::ProcessEventsInternal (this=0x7aaf40,
> timeout_msecs=0,
>
> do_processing=true)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:268
>
> #46 0x00002aaaacacfff8 in vtkTCPNetworkAccessManager::ProcessEvents
> (this=0x7aaf40, timeout_msecs=0)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:165
>
> #47 0x00000000004032a4 in RealMain(int, char **, enum
> vtkProcessModule::ProcessTypes) (argc=1, argv=0x7fffffffd5d8,
>
> type=vtkProcessModule::PROCESS_SERVER)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/CommandLineExecutables/pvserver_common.h:91
>
> #48 0x000000000040337e in main (argc=1, argv=0x7fffffffd5d8)
>
> at
> /home2/dtrudgian/paraview/ParaView-v4.3.1-source/CommandLineExecutables/pvserver.cxx:27
>
>
>
> --
> David Trudgian Ph.D.
> Computational Scientist, BioHPC
> UT Southwestern Medical Center
> Dallas, TX 75390-9039
> Tel: (214) 648-4833
>
>
>
> *From:* David Trudgian
> *Sent:* Tuesday, September 8, 2015 2:09 PM
> *To:* 'Utkarsh Ayachit' <utkarsh.ayachit at kitware.com>
> *Cc:* paraview at paraview.org
> *Subject:* RE: [Paraview] Paraview 4.3.1 OSMesa build - QWidget::repaint:
> Recursive repaint detected
>
>
>
> Utkarsh,
>
>
>
> Thanks for the note. Glad to know that we might not be the only ones with
> the issue. I’m working on getting a debug build and callstack now. Just
> thought I’d post first in case anyone had any ideas.
>
>
>
> --
> David Trudgian Ph.D.
> Computational Scientist, BioHPC
> UT Southwestern Medical Center
> Dallas, TX 75390-9039
> Tel: (214) 648-4833
>
>
>
> *From:* Utkarsh Ayachit [mailto:utkarsh.ayachit at kitware.com
> <utkarsh.ayachit at kitware.com>]
> *Sent:* Tuesday, September 8, 2015 1:56 PM
> *To:* David Trudgian <David.Trudgian at UTSouthwestern.edu>
> *Cc:* paraview at paraview.org
> *Subject:* Re: [Paraview] Paraview 4.3.1 OSMesa build - QWidget::repaint:
> Recursive repaint detected
>
>
>
> Oddly, I was communicating with someone about a similar issue, but haven't
> been able to get to the root cause. Do you have a debug build of ParaView?
> Would it be possible to get a callstack to when than error message is
> popped up?
>
>
>
> Utkarsh
>
>
>
> On Tue, Sep 8, 2015 at 2:26 PM, David Trudgian <
> David.Trudgian at utsouthwestern.edu> wrote:
>
> Hi,
>
>
>
> After working through the information on the web I’ve built Paraview 4.3.1
> with OSMesa support, osmesa 10.5.4 with llvmpipe, and libGlu 9.0.0. MPI is
> OpenMPI 1.8.3 and everything is compiled using the Intel 2015 compiler
> suite.
>
>
>
> A Paraview build using MPI and standard OpenGL is working nicely on our
> GPU compute nodes with NVIDIA Tesla cards. However, if I start the OSMesa
> pvserver (for our non-GPU nodes) and connect from the client I immediately
> receive an error:
>
>
>
> QWidget::repaint: Recursive repaint detected
>
>
>
> The client will exit. The server doesn’t exit, but is unresponsive to
> further connection attempts.
>
>
>
> Has anyone else come across this issue with an OSMesa build? As above, the
> build using a physical NVIDIA card works great, with MPI working nicely etc.
>
>
>
> Many Thanks,
>
>
>
> --
> David Trudgian Ph.D.
> Computational Scientist, BioHPC
> UT Southwestern Medical Center
> Dallas, TX 75390-9039
> Tel: (214) 648-4833
>
>
>
>
> ------------------------------
>
> *UT** Southwestern*
>
> Medical Center
>
> The future of medicine, today.
>
>
> _______________________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Please keep messages on-topic and check the ParaView Wiki at:
> http://paraview.org/Wiki/ParaView
>
> Search the list archives at: http://markmail.org/search/?q=ParaView
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview
>
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview/attachments/20150909/bee357fc/attachment.html>
More information about the ParaView
mailing list