Hi,
I'm using Composer XE 2018 with VS 2015 to create a Fortran DLL. I've added to the project a C source file but in the build process it is simply being ignored. My workaround was to create a C static library project and add the lib file to the Fortran DLL project but I prefer to have them all in one project. Have I missed something?
Thank you
Baruch
Sir,
I had installed the visual studio 2017 with desktop development package. I had installed the Intel parallel studio XE 2018 composer edition for Fortran offline. I tried creating a new project under Fortran in visual studio but when I debug the program it shows an error saying "win32" support not found and project can't be built. Sir, I want to know how to integrate Fortran with visual studio and get out the errors and run the program successfully
please help.
Dear All
I have encountered an alarming problem with v18, not present on v17, please have a look at the code and comments below
module MOD_A
implicit none; private
type, public :: MOD_A_data
real :: x
contains
procedure, pass :: new => MOD_A_new
end type
public MOD_A_new
interface
module subroutine MOD_A_new(d)
class(MOD_A_data) :: d
end subroutine MOD_A_new
end interface
end module MOD_A
submodule (MOD_A) MOD_A_subm
implicit none
contains
module subroutine MOD_A_new(d)
class(MOD_A_data) :: d
d%x = 10.
end subroutine MOD_A_new
end submodule MOD_A_subm
program TEST_CON_SUB
use, non_intrinsic :: MOD_A
implicit none
type(MOD_A_data) :: d
call d%new !This line causes message A below
!call MOD_A_new(d) !But this line is absolutely fine (and works)
write(*,*) 'd%x = ', d%x
end program TEST_CON_SUB
!Message A from v18.0.1...
!1>------ Build started: Project: Test_cont_sub, Configuration: Debug x64 ------
!1>Compiling with Intel(R) Visual Fortran Compiler 18.0.1.156 [Intel(R) 64]...
!1>Test_cont_sub.f90
!1>Linking...
!1>Test_cont_sub.obj : error LNK2019: unresolved external symbol MOD_A_mp_NEW referenced in function MAIN__
!1>x64\Debug\Test_cont_sub.exe : fatal error LNK1120: 1 unresolved externals
!1>
!1>Build log written to "blah"
!1>Test_cont_sub - 2 error(s), 0 warning(s)
!========== Build: 0 succeeded, 1 failed, 0 up-to-date, 0 skipped ==========
! but from v17.0.6...
!1>------ Build started: Project: Test_cont_sub, Configuration: Debug x64 ------
!1>Compiling with Intel(R) Visual Fortran Compiler 17.0.6.270 [Intel(R) 64]...
!1>Test_cont_sub.f90
!1>Linking...
!1>Embedding manifest...
!1>
!1>Build log written to "blah"
!1>Test_cont_sub - 0 error(s), 0 warning(s)
!========== Build: 1 succeeded, 0 failed, 0 up-to-date, 0 skipped ==========Law of conservation of bugs?
Thanks
N
I am trying to get Intel Visual Fortran 2016 Update 1 to work with Visual Studio 2015 Update 3.
I have already followed the instructions at https://software.intel.com/en-us/articles/troubleshooting-fortran-integration-issues-with-visual-studio
The compiler is working fine, and Fortran projects within visual studio exist, however the debugger does not seem to be working and the text editor does not apply any formatting of Fortran files.
Anything else that can be done to repair the installation? I've already tried uninstalling and reinstalling.
Hi,
I manage a large group of engineers maintaining C code that is relying on my Fortran core dll.
During debug session they need to inspect the values in the Fortran code but they seldom change and recompile it.
Can this be done while not holding the Fortran floating license token?
Thank you.
In a program I declare:
CHARACTER*4 PREVFILE
CHARACTER*60 PRINTFILE
In the program the following is checked:
IF(Printfile(1:4).NE.PREVFILE)LOWPAGE=0
During execution, the program fails with the message:
forrtl: severe (408): fort: (18) Dummy character variable ‘PRINTFILE has
length 60 which is greater than actual variable length 4.
The program is about to be migrated from CVF to IVF, I do not understand what is wrong.
Can anyone give me a tip ?
I got a single user Fortran XE Composer Edition license this fall (2017) using VS 2015 Shell.
Yesterday when I opened VS, edited a file and tried to compile I got the two line error:
Error A license for Comp-FW is not available (-9,57)
error #10052: could not checkout FLEXlm license
My license looks ok in the Intel Software Manager.
How can I get the compiler working again?
Thanks
I am trying Fortran to output to screen without making paragraphs - i.e. output several times but in a row. The code below outputs correctly to screen if I compile the program and run it from the command line. However, it does not output to screen if I am using the debug feature of Visual Studio. Is there any work around or a better solution to this?
program test
implicit none
integer :: ind
do ind = 1,20
write(*,'(1x,i0)',advance='no') ind
!Some other stuff here
end do
end programI am using the IMSL DIVPAG routine for numerical integration of an initial value problem (Gear option for stiff equations). I've hit a bit of a snag. I want to integrate for, say, 100 steps, and save the final system state vector. Then run the code again, read the system state vector from the first run to use as initial conditions,and integrate another 100 steps. I've done this and it seems to work, but results do not match a run of 200 steps in one go. The first 100 match, but not the second 100 (close, but not close enough). I'm guessing DIVPAG uses internally stored values that make the 1st step of the second 100 step run get calculated differently than the 101st step of the 200 step run. Unless I missed it, the IMSL documentation says nothing about how to smoothly restart an integration with a saved system state vector. Has this issue ever come up before?
I (lately) have a problem occurring when writing to direct-access files. Attached is the smallest code that reproduces the problem on my Win10 system.
In the 'Debug' mode for x64 the code runs as expected: data is written to the direct-access file. In the 'Release' mode for x64 the following error message is produced:
forrtl: severe (66): output statement overflows record, unit 12, file C:\arne\Visual Studio 2015\Projects\test_direct_access\test_da.dat
Image PC Routine Line Source
test_direct_acces 00007FF6624C7E07 Unknown Unknown Unknown
test_direct_acces 00007FF6624BB9C2 Unknown Unknown Unknown
test_direct_acces 00007FF6624B8AB8 Unknown Unknown Unknown
test_direct_acces 00007FF6624B12F5 Unknown Unknown Unknown
test_direct_acces 00007FF662511532 Unknown Unknown Unknown
test_direct_acces 00007FF6625118D5 Unknown Unknown Unknown
KERNEL32.DLL 00007FFF870B1FE4 Unknown Unknown Unknown
ntdll.dll 00007FFF89B1EFB1 Unknown Unknown Unknown
Press any key to continue . . .
The ‘Release’ code seems to assume that the record length is specified in ‘bytes’ and not in ‘4 byte units’ as specified in the project. Changing ‘LEN_R64 = 2’ to ‘LEN_R64 = 8’ makes that the ‘Release’ version runs without error.
Any idea about what causes this strange behavior?
Hi
I'm upgrading a CVF application that relies heavily on single precision floating point calculations.
I'm using the "Enable Floating Point Consistency" and "Extend Precision of Single Precision Constants".
Currently I create IA-32 Fortran DLL on XE.
From comparing the results of the CVF and XE versions I see that the CVF application is more precise.
It looks like in CVF there are more significant digits (or half digits).
I've read previous posts regarding Floating Point Consistency, including referred documents and presentations.
I've tried all possible combinations of the Fortran Floating Point Settings and still can not recreate precisely CVF results.
Will be grateful for any insight on this.
Baruch
hello
some day ago, i wrote a question about fortran
https://software.intel.com/en-us/comment/1918428#comment-1918428
so, i decide install intel parallel studio xe 2018 update 1 composer edition for fortran windows.
but, i have a problem.
Warning are detected. <The intel(r) extension for intel(r) MIC architecture cannot be installed.
so, i read the message
Installation can continue; however, Intel® Debugger Extension for Intel® MIC Architecture will not be installed because it requires that Microsoft Visual Studio* 2013, 2015 or 2017 be installed. Refer to the Release Notes for more details.
when i install the visual studio community 2017, what should I install from the list?
i attached the capture screenshot. thank you.
Hello,
I have downloaded parallel_studio_xe_2018_update1_cluster_edition, and got licence number but i am unable to use it because i am operating through a corporate proxy.
May I have a try-out licence file ?
Best regards,
Jean-Pierre
The attached fortran source file is from the NIST. I get the following compile error, and I can't figure out why.
Error 5 error #6634: The shape matching rules of actual arguments and dummy arguments have been violated. [F] F12854_stpv2a.for 2166
It appears that array F is dimensioned F(20) and passed to a routine where it is declared F(*). It that supposed to trigger this compile error? I wanted to ask about this before turning off /warn:interfaces which gets rid of the error.
| Attachment | Size |
|---|---|
Download F12854_stpv2a.zip | 22.4 KB |
I have several LIBIOMP5MD.DLL files whose file properties say:
Product Version 5.0, File Version 5.0.2013.227, Date modified 9/19/2013, Copyright (C) 2012 Product Version 5.0, File Version 5.0.2013.507, Date modified 6/8/2013, Copyright (C) 2013 Product Version 5.0, File Version 5.0.2011.606, Date modified 9/5/2014, Copyright (C) 2011
Is it just me, or is there something goofy about the numbering and dating?
Hi
The test-case below captures the gist of an error; where is the cause that the adjustable array is bigger than what has been passed in. It cannot be caught by the compiler; I understand. I am wondering is there a way (as in a flag) to catch it at run-time. Compiling with ifort /check:all Test.f90 won't do.
As you can imagine this is extracted from rather deep inside a combination of legacy and new program. In one of the test-cases of that program, it revealed itself as "access violation" but that was at a different location.
Abhi
---
Module OM
Implicit None
Integer :: M, N
Integer :: L
Contains
Subroutine Set()
M = 3; N = 5
End Subroutine Set
Subroutine Tech(K)
Integer :: K(N) ! Size is larger than what's passed; its incorrect.
Integer :: L
K = -1 ! any way to catch this error at run-time?
L = K(N) - 1 ! or here?
Write(*,"(2(A,X,I0,X))") "in Tech: K(N) =", K(N), " and L =", L
End Subroutine Tech
End Module OM
Program Test
Use OM
Implicit None
Integer, Allocatable :: I(:)
Call Set()
Allocate(I(M))
I = 0
Call Tech(I)
Write(*,"(A,X,I0,X)") "after the call L =", L
End Program Test
If I define a local target variable inside a subroutine, then associate a global pointer to that value, is it safe to use the pointer to access that value outside of subroutine or the data can be overwritten and changed unexpectedly?
I want to allocate an allocatable array of abstract class objects in a subroutine.
Subroutine adds a new object to the existing array of abstract objects by resizing the array:
subroutine apmArrayAddNewMember(array, member)
use class_absMaterial !! abstract class
use class_mat02 !! child class
implicit none
! class mat02 extends abstract class absMaterial
class(absMaterial), allocatable, dimension(:), &
intent(inout) :: array !! existing array
type(mat02), target, intent(in) :: member !! new object
class(absMaterial), allocatable, dimension(:) :: temp
integer intSize
type(mat02), pointer, dimension(:) :: molder
if (allocated(array)) then
intSize = size(array)
allocate(molder(intSize+1))
allocate(temp, mold=molder)
deallocate(molder)
temp(1:intSize) = array
temp(intSize+1) = member
deallocate(array)
call move_alloc(temp, array)
else
allocate(molder(1))
allocate(array(1), mold=molder)
deallocate(molder)
array(1) = member
end if
end subroutine apmArrayAddNewMember
I am not sure if this works correctly. As you can see I have to allocate molder so that I can specify 'mold ='
I could not find any other way, also I don't know what if all the objects are not mat02 class and be another objects that extends absMaterial.
Will this code steel works if the other objects need same or less memory as mat02?
I have a small subroutine transferring text from keyboard entries, etc from lower to uppercase letters. I used this routine for CVF and the version shown below I also apply for IVF Win32 applications. What happens is that the program crashes with the error message:Unhandled exception at 0x7790DF7C (ntdll.dll) in CaFeMS1-IVF.exe: 0xC0000005: Access violation writing location 0x002A0FFC... before the line ILENGTH = len_trim(STRING) - 1 where a breakpoint is set.
The subroutine is linked from a library file.
Can anyone give me a tip?
Program listing follows here:
! IF(I.GT.IEND)GOTO 2000 ! 25/5-2000
! ICOM is character*2
READ(IDEDIT,202)ICOM
CALL TRANS_TO_UPPERCASE(ICOM,2)
SUBROUTINE TRANS_TO_UPPERCASE(STRING,LENGTH)
use IFPORT
INTEGER*4 KAR,ILENGTH,LENGTH
CHARACTER*(LENGTH)STRING
c CHARACTER*(40)STRING
! ILENGTH = LNBLNK(STRING)
ILENGTH = len_trim(STRING) - 1
C....Transfer STRING to uppercase letters....................
DO 1000 I=1,ILENGTH
KAR=0
KAR=ICHAR(STRING(I:I))
IF(KAR.GE.97)STRING(I:I)=CHAR(KAR-32)
1000 CONTINUE
RETURN
END