Approximatrix Forums — Bug Reports

Profiler Bug

2025-10-24T07:36:36Z

Hi Jeff, For reasons that I can't determine, the profiler starts to misreport. It will produce an empty report containing just the headers with no data/routine names. Once it happens to your project - it never recovers. When I inspect the gmon file it is mostly filled with nuls. Do you know of this behavior and how to get it working again in a project?
Regards, Pete

SF Package manager bug - crashes!

2025-08-04T05:18:53Z

Hi Jeff,
I went to use the SFPackage manager and found that it immediately crashed with barely a flicker on the screen. I tried to run spfm from the command line and got a fortran error message as follows;
C:\Program Files (x86)\SF Package Manager>sfpm info
SF Package Manager - Fortran Command Line Client 2.2
Copyright 2021, 2022, 2023 Approximatrix, LLC

At line 223 of file .\config.F90
Fortran runtime error: Substring out of bounds: lower bound (0) of 'str' is less than one
Error termination.
I just changed machines and thought that maybe that had somehow caused the problem so i did an uninstall/reinstall but no change. Any help would be appreciated as I use it fairly extensively.
Regards,
Pete

I want utf-8

2025-07-25T02:38:21Z

Warning: command-line option '-finput-chartset=utf-8' is valid for C/C++/ObjC/ObjC++ but not for Fortran.
How to fix it?

undefined reference to a subroutine

2025-01-29T22:05:46Z

Just installed Simply Fortran on Windows11, tried to build and run a simple program

program test
implicit none

Double Precision s(10)
INTEGER IFAIL

Call G05FEF(0.5D0,0.5D0,10,s,IFAIL)

print *, "s is", s

end program test

that calls some functions and routines, which I added to the project outline. Project building looks fine, with message Generating Makefile ... Okay, * Complete*, but when I compile and run the program file, I get the message undefined reference to 'g05fef_'. Any help is greatly appreciated.

macOS Sequoia Debugger Issues

2024-10-31T20:28:51Z

A user has reported, and I've since confirmed, that our debugger is no longer functioning on macOS Sequoia on Intel Macs. There seems to be some serious issues, possibly related to but not limited to Apple's linker, causing the executables to be unusable by the GNU Debugger. The resulting executables seem to have DWARF debugging information that does not conform to the DWARF standard, though Apple's lldb does not seem bothered by the issue.

We're trying to find a workaround here, but the situation looks pretty bad right now.

compilation failure. Cannot read libgfortran.spec

2024-09-06T14:49:13Z

New 3.35.4225 release gives following error on compilation

==================
Generating Makefile... Okay
==================
Compiling .\hello.f90
Generating hello.exe
gfortran.exe: fatal error: cannot read spec file 'libgfortran.spec': No such file or directory
compilation terminated.
Error: Last command making (hello.exe) returned a bad status
Error: Make execution terminated

* Failed *

the use of the debugger

2024-07-13T09:35:27Z

Everytime I try to add a wacht sf exits from the program.
Any ideas?

Latest upgrade has broken LTO?

2024-06-19T05:58:05Z

Hi Jeff,
I just upgraded to the latest release and I've found that the LTO is erroring out with the following error
Generating target.exe
lto1.exe: fatal error: bytecode stream in file 'build\constants.o' generated with LTO version 13.0 instead of the expected 14.0
compilation terminated.
lto-wrapper.exe: fatal error: C:\Program Files (x86)\Simply Fortran 3\mingw-w64\bin\gfortran.exe returned 1 exit status
compilation terminated.
collect2.exe: fatal error: lto-wrapper returned 1 exit status
compilation terminated.
I did a full install when I did the install. Can you help me at all?
Regards,
Pete

Closing SF does not stop fwin.exe process

2024-04-04T15:08:59Z

Running SF version 3.34 4158 I observe strange behaviour when I'm closing the machine. Windows informs about still running SF. I investigated it a little deeper. I start Process Explorer and then start and close SF via clicking the Close button in the right/upper window's corner or taking File -> Exit. In both situations fwin.exe process still runs. In Temp folder there are leavings like SF_x.tmp.html and sfXx.tmp files/folders.

I recall somebody started the topic here and he saw fwin.exe still running that gave him surprising effects.

Outline bugs

2024-03-27T03:51:53Z

don't print whole methods

About settextboxdouble

2023-11-08T13:34:25Z

Sub settextboxdouble can’t work.

Smart indent file bugs

2023-11-08T03:01:35Z

It is not useful in if clause 、else if clause and end select

SF 3.31 3993 Profiler not working?

2023-09-28T22:44:55Z

Jeff,

I cannot get the SF 3.31 3993 Profiler to run after I've enabled Profiling and executed the project within SF environment. More specifically, when the project is executed and then a menu to open a data file is selected following by running a regression analysis and Aplot display of the data, I keep getting the message that "Profiling data does not yet exist" even after I've run the project with Profiling enabled. No gmon.out file is created in the working folder. However, if I execute the same project to display the window and menus but quickly quit the program without selecting a data file or running a regression of the data, a gmon.out file is created in the working folder. A gmon.out file is only created when the project is executed and then quit without performing any other operations or computations. FYI, I am using a multilinetextbox in the parent window.

Am I doing something incorrectly, or is there a bug in the Profiler of this build?

Frank

Odd optimizer bug

2023-08-18T04:42:47Z

I stumbled across an odd optimizer bug. I just though that I'd flag it here.
I was updating "Calpak" a collection of Calendar routines by John Burkardt
and I found that it generated a lot on nonsense at optimization level -O3
but worked perfectly with no optimizations. After a lot to sleuthing, I
tracked it down to this line: CALL Y_ASTRONOMICAL_TO_COMMON(Y , Y)
the left "Y" is the input and the right "Y" is the output. I am not sure exactly what
the optimizer was doing, but at -O3, it was returning just random junk.
doing this, fixed it;
temp=y
CALL Y_ASTRONOMICAL_TO_COMMON(Y , temp)
y=temp
So, if you are getting inexplicable results in your code, check that you don't have the same error.

Newunit problem(?)

2023-06-15T18:18:07Z

I use the newer method of opening files -- Open(newunit=lfn, etc)

I have a program where I did this but after a certain number of opening files, I got the fatal error "too many open files" even though I closed out each one after processing them.

I fixed it by using assignments for the fiiles:
file1=20 - Open(Unit=file1, etc)

These are processing fine on even longer lists of files.

Linda