- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi,
new at this CVF 6.6 and C integration. I'v got a pretty basic C-program (not C++) and I want to use the main routine as a wrapper to my F77 program.
I manage to compile my main.c to main.obj with .NET
I then include the main.obj as part of the resources in my Visual Fortran project, and I compile the .F files OK.
Linking crashes with the following spectacle:
As you can see it seems that the linker gives up at the time when it should be linking to the main.obj file.
My suspicion is that the C-code generated from the .NET thing is not compatible with the obj code required by the visual studio that CVF 6.6 uses. Am I right?
Any workarounds? I'm not able to get my C.NET to work with the .obj files from Fortran either ...
Cheers,
-Paal
new at this CVF 6.6 and C integration. I'v got a pretty basic C-program (not C++) and I want to use the main routine as a wrapper to my F77 program.
I manage to compile my main.c to main.obj with .NET
I then include the main.obj as part of the resources in my Visual Fortran project, and I compile the .F files OK.
Linking crashes with the following spectacle:
--------------------Configuration: cc - Win32 Debug-------------------- Compiling Fortran... D:homeworkmay2003fconappsrcseval.f D:homeworkmay2003fconappsrcinitpar.f D:homeworkmay2003fconappsrchydratetemp.f D:homeworkmay2003fconappsrchydrateeqs2subcooled.f D:homeworkmay2003fconappsrcFhydrate.f D:homeworkmay2003fconappsrcdlinpk.f D:homeworkmay2003fconappsrcddaspk.f D:homeworkmay2003fconappsrcdaux.f D:homeworkmay2003fconappsrcicubic2.f D:homeworkmay2003fconappsrcaxialeqs.f D:homeworkmay2003fconappsrcspline.f Linking... LINK : warning LNK4075: ignoring /EDITANDCONTINUE due to /INCREMENTAL:NO specification Creating library Debug/cc.lib and object Debug/cc.exp LINK : warning LNK4049: locally defined symbol "_TDPAR" imported LINK : warning LNK4049: locally defined symbol "_HYDRATKURVE" imported LINK : warning LNK4049: locally defined symbol "_AXIALGRID" imported LINK : warning LNK4049: locally defined symbol "_GRIDPAR" imported LINK : warning LNK4049: locally defined symbol "_PARAMS" imported LINK : warning LNK4049: locally defined symbol "_FLUXDATA" imported main.obj : error LNK2001: unresolved external symbol __RTC_CheckEsp main.obj : error LNK2001: unresolved external symbol @_RTC_CheckStackVars@8 main.obj : error LNK2001: unresolved external symbol __RTC_Shutdown main.obj : error LNK2001: unresolved external symbol __RTC_InitBase Debug/cc.exe : fatal error LNK1120: 4 unresolved externals Error executing link.exe. cc.exe - 5 error(s), 7 warning(s)
As you can see it seems that the linker gives up at the time when it should be linking to the main.obj file.
My suspicion is that the C-code generated from the .NET thing is not compatible with the obj code required by the visual studio that CVF 6.6 uses. Am I right?
Any workarounds? I'm not able to get my C.NET to work with the .obj files from Fortran either ...
Cheers,
-Paal
Link Copied
1 Reply
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Try the reverse,ie, use the .net linker instead of the VS 6 linker. Also, I believe that debug symbols for c.net differ from those used by VS 6 but I haven't confirmed this and in any event I think that having the latest dbghelp.dll (6.1.0017.2) should make this a non issue, maybe.
HTH,
Gerry T.
HTH,
Gerry T.

Reply
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page