A list for the developers of CellML tools

Text archives Help


[cellml-dev] Using the CellML API with Qt Creator (solved) was Re: CellML 1.10rc2


Chronological Thread 
  • From: alan.garny at dpag.ox.ac.uk (Alan Garny)
  • Subject: [cellml-dev] Using the CellML API with Qt Creator (solved) was Re: CellML 1.10rc2
  • Date: Wed, 14 Sep 2011 17:59:10 +0200

Hi Andrew,

> > I just tried the libstdc++-6.dll file you sent me and copied it to the
> > folder where my small CellML program gets generated (thus taking
> > precedence over my MinGW's copy of libstdc++-6.dll), but that doesn't
> > seem to be doing the trick for me. I am wondering whether there isn't
> > a conflict with mingwm10.dll and libgcc_s_dw2-1.dll which, in my case,
> > come from Qt's version of MinGW...?
>
> I'm not sure how your environment is set up (the MingW that comes with Qt
> SDK doesn't seem to include a make, so using CMake doesn't work out of the
> box unless you have some other form of make installed), so I decided to
> simplify everything and just write a shell script to set the right
environment
> variables and run the program.

It seems to me that you are overcomplicating things. The whole idea of the
Qt SDK is that you have *everything* you need to build a Qt application. On
Windows, this includes a patched version of MinGW which does come with a
make:

---------------------------------------
C:\QtSDK\mingw\bin>mingw32-make.exe --version
GNU Make 3.81
Copyright (C) 2006 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.
There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A
PARTICULAR PURPOSE.

This program built for i386-pc-mingw32

C:\QtSDK\mingw\bin>
---------------------------------------

> I did find one issue: after checking out the binaries from the github
> repository, the libcellml.dll file seems to be missing the execute bit -
which
> caused Windows to give a confusing warning about not being able to load
> libgcc_s_dw2-1.dll.

Well, for me (i.e. not a Cygwin environment, but a *pure* Windows
environment), the execute bit is not relevant at all (not least because it
doesn't exist in a *pure* Windows environment).

> After I set the execute bit, with this shell script, your test program
seems to
> compile and run fine. It therefore appears that this issue is completely
> unrelated to where the MinGW came from.
>
> Note in my case c:/build is the directory with libstdc++-6.dll - you could
> substitute it for a .
>
> I am using Cygwin purely to run the scripts - but you could equally well
write
> a .bat to do the same thing if you wanted.
>
> Anyway, all my hours for working on CellML for this week have been used up
> (and then some), and these issues look like build issues at your end more
> than genuine CellML SDK issues. I'd suggest that from here, you look into
> what CMake is actually invoking (it might be calling a different compiler
on
> your system somewhere), check the permissions on all DLLs, and use a
> dependency tracer to make sure that the correct DLLs are being found. If
> there are still problems you can't solve this way, I will be able to have
a look
> next week.

I believe you are wrong (see above), but nevermind for now. I will be in
Auckland from Monday and for a month, so we can have a look at it next week
before your allocated CellML time runs out...

Alan





Archive powered by MHonArc 2.6.18.

Top of page