[cgma-dev] [Fathom] cgm make check with occ

Iulian Grindeanu iulian at mcs.anl.gov
Fri Sep 16 17:02:46 CDT 2011


Hi Jane, 
Thank you very much. 
So 6.5 version is fine now, I patched the file and recompiled. There are no errors in make check, at least on my machine! 
It took a long time, somehow the compiler decided to compile everything in occ, although only one file was modified. 

Still, I had to apply the patch manually (in the sense that I edited the file manually); the patch file contains references to you home directory 

--- src/Extrema/Extrema_ExtPS.cxx 2011-04-11 13:06:46.000000000 -0500 
+++ /home/jhu/OpenCASCADE6.5.0/ros/src/Extrema/Extrema_ExtPS.cxx 2011-04 
-09 14:13:13.000000000 -0500 
... 

Anyway, it works now. 

I should have realized that you modified the patch file four months ago. I should have patched and rebuild occ at that time. Also, maybe your commit message should have suggested that occ needs to be recompiled. 

Have a good weekend! 
Iulian 






----- Original Message -----





	Hi, Iulian, 

I double checked /homes/fathom/3rdparty/occ/OpenCASCADE6.3.0/src/OpenCASCADE6.3.0/ros/src/BRepBuilderAPI/BRepBuilderAPI_Collect.cxx, it's the same problem: need to be patched. 

Except this one, trunk cgm on occ6.5 and occ6.3 should both pass make check. 

Please let me know how it turns out. 

Thanks. 

Jane 

--- On Thu, 9/15/11, Iulian Grindeanu <iulian at mcs.anl.gov> wrote: 



From: Iulian Grindeanu <iulian at mcs.anl.gov> 
Subject: Re: [Fathom] cgm make check with occ 
To: "Jiangtao Hu" <jiangtao_ma at yahoo.com> 
Cc: "CGMA Development" <cgma-dev at mcs.anl.gov>, "James Porter" <jvporter at wisc.edu> 
Date: Thursday, September 15, 2011, 2:02 PM 


Hi Jane, 

So maybe the occ build is wrong 
I built occ 6.5 on fathom space: 
/home/fathom/3rdparty/occ/OpenCASCADE6.5.0 

the configuration log file is here: 
home/fathom/3rdparty/occ/OpenCASCADE6.5.0/ros/config.log 

The installation is at 
/homes/fathom/3rdparty/occ/OpenCASCADE6.5.0/gcc 

It was built on my desktop, iulian-desktop 

gnep buildbot is using the same build 

Do you have an mcs account? Otherwise how can you access my machine ? Or gnep? 

Iulian 

----- Original Message ----- 
> Hi, Iulian, 
> 
> What is the build machine's name? I tried to use non-debug mode on my 
> machine, it still passed. I'd like to access your machine to debug 
> then. Thanks. 
> 
> Jane 
> 
> --- On Thu, 9/15/11, Iulian Grindeanu < iulian at mcs.anl.gov > wrote: 
> 
> 
> 
> From: Iulian Grindeanu < iulian at mcs.anl.gov > 
> Subject: Re: [Fathom] cgm make check with occ 
> To: "Jiangtao Hu" < jiangtao_ma at yahoo.com > 
> Cc: "CGMA Development" < cgma-dev at mcs.anl.gov >, "James Porter" 
> < jvporter at wisc.edu > 
> Date: Thursday, September 15, 2011, 1:01 PM 
> 
> 
> 
> 
> ----- Original Message ----- 
> > Hi, Iulian, 
> > 
> > Those missing files are only for 10.2 cgm, 12.2 cgm has already have 
> > them. Can you let me know where the modify failed? Thanks. 
> > 
> > Is the nightly build and check for cgm still on? Can James confirm 
> > it? 
> > Thanks. 
> > 
> > Jane 
> > 
> > --- On Thu, 9/15/11, Iulian Grindeanu < iulian at mcs.anl.gov > wrote: 
> > 
> 
> Hi Jane, 
> The backtrace is 
> 
> (gdb) backtrace 
> #0 0x00007fffee5c5a75 in *__GI_raise (sig=<value optimized out>) at 
> ../nptl/sysdeps/unix/sysv/linux/raise.c:64 
> #1 0x00007fffee5c95c0 in *__GI_abort () at abort.c:92 
> #2 0x00007ffff00378e5 in __gnu_cxx::__verbose_terminate_handler() () 
> from /usr/lib/libstdc++.so.6 
> #3 0x00007ffff0035d16 in ?? () from /usr/lib/libstdc++.so.6 
> #4 0x00007ffff0035d43 in std::terminate() () from 
> /usr/lib/libstdc++.so.6 
> #5 0x00007ffff0035e3e in __cxa_throw () from /usr/lib/libstdc++.so.6 
> #6 0x00007ffff03446ed in Standard_NoSuchObject::Throw() const () from 
> /homes/fathom/3rdparty/occ/OpenCASCADE6.5.0/gcc/lib/libTKernel.so.0 
> #7 0x00007ffff0344bc3 in Standard_NoSuchObject::Raise(char const*) () 
> from 
> /homes/fathom/3rdparty/occ/OpenCASCADE6.5.0/gcc/lib/libTKernel.so.0 
> #8 0x00007ffff232d78d in 
> BRepBuilderAPI_ModifyShape::ModifiedShape(TopoDS_Shape const&) const 
> () from 
> /homes/fathom/3rdparty/occ/OpenCASCADE6.5.0/gcc/lib/libTKTopAlgo.so.0 
> #9 0x00007ffff231cd43 in 
> BRepBuilderAPI_GTransform::ModifiedShape(TopoDS_Shape const&) const () 
> from 
> /homes/fathom/3rdparty/occ/OpenCASCADE6.5.0/gcc/lib/libTKTopAlgo.so.0 
> #10 0x000000000043c6aa in OCCLump::update_OCC_entity (this=0xcce470, 
> aBRepTrsf=0x7fffffffcc90, op=0x0) at OCCLump.cpp:378 
> #11 0x000000000042de01 in OCCBody::update_OCC_entity (this=0xcd40e0, 
> aBRepTrsf=0x7fffffffcc90, op=0x0) at OCCBody.cpp:511 
> #12 0x000000000042da4d in OCCBody::scale (this=0xcd40e0, 
> scale_factor_x=1, scale_factor_y=2, scale_factor_z=5) at 
> OCCBody.cpp:450 
> #13 0x00000000004556da in OCCQueryEngine::scale (this=0xcc14c0, 
> body=0xcd40e0, f=...) at OCCQueryEngine.cpp:3516 
> #14 0x00000000004870ce in OCCModifyEngine::scale (this=0xcc1600, 
> body=@0x7fffffffd0b0, factors=...) at OCCModifyEngine.cpp:8891 
> #15 0x00000000006ea8c2 in GeometryModifyTool::scale (this=0xcbefd0, 
> body=@0x7fffffffe1c8, factors=..., check_to_transform=true, 
> preview=false) at GeometryModifyTool.cpp:6779 
> #16 0x0000000000421345 in make_Point () at modify.cpp:146 
> #17 0x0000000000420b0f in main (argc=1, argv=0x7fffffffe3c8) at 
> modify.cpp:55 
> 
> 
> occ 6.5 is not build with debug option on our machines, so it is 
> harder to debug. 
> 
> Do you build occ with "enable-debug" on your machine? 
> 
> Iulian 


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/cgma-dev/attachments/20110916/5144bd3a/attachment.htm>


More information about the cgma-dev mailing list