FEAP User Forum
FEAP => Installation => Topic started by: Prof. S. Govindjee on March 03, 2015, 12:23:28 PM
-
To help others in installing FEAP, I assembled two videos showing a complete install of FEAP on:
(1) A Mac, which is also the same for a Linux/Unix machine as well as a Windows machine with Cygwin installed
and
(2) A Windows 7 Professional machine with Visual Studio 2010 and Intel XE Fortran 2011 -- 64 bit architecture.
Both videos can be found at: http://www.youtube.com/playlist?list=PLrMPyRXvlE9yx2bvWk0F2TX2FxfUafbfz
-
Dear Prof. S. Govindjee,
would you like to share to us how to install in debug mode? thank you very much for your kindly attention
-
One Linux is is easy just add -g to the compiler options in makefile.in.
On windows, I can not do this as my version of the Intel compiler is broken with respect to this feature.
-
Thank you for your tutorial videos, Prof. S. Govindjee! I am sure it will help a lot beginners. On the other hand, will you be considering posting videos on advanced topics like creating a user element, user material or a user command? By the way, I have learned a lot from your website.
-
Thanks alot for your video, it saved me alot of time!
If i use windows-32 i need to feel integer4, i understand right? Please sorry for my bad english
-
Yes. On Win32 you will need to use the include files in include\ and include\integer4. Do not use the files in \include\integer8.
-
Dear
I installed Feap8.4 on os x 10.9.1, during installation I think I followed the video carefully. However when I tried run one of example input file I got the following messages. is there any one has the same problem? thank you very much
regards,
herry
Program received signal SIGSEGV: Segmentation fault - invalid memory reference.
Backtrace for this error:
#0 0x103e41132
#1 0x103e418ee
#2 0x7fff90d805a9
#3 0x10357e340
#4 0x103706a76
#5 0x1035b4750
#6 0x103701ac8
#7 0x1034c9e89
#8 0x1035474ae
#9 0x1034d634f
#10 0x103758dca
Segmentation fault: 11
-
please redo your installation adding -g to the compiler options. then run dsymutil on your executable 'dsymutil feap'.
with this you should get a more informative error message.
you can also try running in the debugger (gdb or ldb) to get more information about the error.
-
Yes. On Win32 you will need to use the include files in include\ and include\integer4. Do not use the files in \include\integer8.
There a couple of possible reasons:
1) When you built the program you did tell the compiler that you wanted to build a 64bit executable. Note the default in the Intel compiler is 32-bits.
2) It could be that you have a 32-bit operating system installed (even if your computer has 64-bit hardware).
Hi FEAP_Admin,
I have used instructions from the following youtube video of Prof. Govindjee (I am using FEAP 8.2)
https://www.youtube.com/watch?v=7QAh6QvOT6s
I am using 64 bit Intel Visual Fortran 11.1.048 (with Visual Studio 2008 on Windows 7 64bit). I have checked the compiler version in Visual Studio menu: Tools -> Options -> Intel Fortran -> Compilers : x64
(https://s6.postimg.org/jb4g1s73l/Compilers.png)
ad the Configuration Manager also uses Release:x64 configuration
(https://s6.postimg.org/5ew7pwaup/configurations.png)
ad finally the include files are also integer8
(https://s6.postimg.org/f0ps6720h/properties.png)
despite all this, I am still getting this error during Build:
Error 1 error #6362: The data types of the argument(s) are invalid. [FREE] D:\feap82_test\ver82\compile\setmem.f 145
Error 2 error #6362: The data types of the argument(s) are invalid. [FREE] D:\feap82_test\ver82\compile\setmem.f 199
Error 3 Compilation Aborted (code 1) D:\feap82_test\ver82\compile\setmem.f 1
I will be very thankful for any hint about resolving the error.
Best regards,
Mubeen.
-
The error you see is the one that you get due to a mis-match in pointer type. Maybe when you created the compile directory, you copied in the include/integer4 files? The compiler would then probably see these before it sees the ones in the include/integer8 directory.
Also make sure that x64 is showing on the chosen configuration when you are in the main interface (at the top).
PS: I suggest you upgrade your code to the latest release (version 8.5).
-
Your version is very old and was developed before integer*8 was fully implemented. You need to change the array "adr(*)" in the routine setmem.f (windows directory) to and integer*8 variable. There are probably many other places that need to be changed so the recommendation to upgrade is the best option - or possibly stick to an integer*4 compile?
-
Hi ALL,
I followed instructions in youtube video of Prof. Govindjee (I am using FEAP 8.2)
I am using 64 bit Intel.Visual.Fortran.Composer.XE.2011.5.221 (with Visual Studio 2010 on Windows 7 64bit).
I have some errors, one of them is: PetscErrorCode ierr
The creation of the library is failed and the errors are showed in attached picture:
I will be very thankful for any hint about resolving the error.
Best regards,
Thanks for ALL.
-
parfeap is only supported on Linux/Mac OS X.
-
Thank you for the information.
-
Thanks for this helpful video!