bug-glpk
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Bug-glpk] Segfault in jday function


From: Andrew Makhorin
Subject: Re: [Bug-glpk] Segfault in jday function
Date: Thu, 27 Jun 2013 16:38:45 +0400

> I compiled glpk 4.51 with MinGW64 compiler. When I try to solve any 
> problem I get a segfault after problem is read in. I tried various 
> versions and toolchains of MinGW64 gcc with the same result.
> 
>  From GDB I have got:
> 
> (gdb) run xyz.mps
> Starting program: C:\mingw64\msys\1.0\local\bin\glpsol.exe xyz.mps
> [New Thread 5208.0x15c8]
> GLPSOL: GLPK LP/MIP Solver, v4.51
> Parameter(s) specified in the command line:
>   xyz.mps
> Reading problem data from `xyz.mps'...
> Problem: RUNTIME
> Objective: R0
> 36975 rows, 40653 columns, 240407 non-zeros
> 174637 records were read
> warning: Invalid parameter passed to C runtime function.
> 
> 
> Program received signal SIGSEGV, Segmentation fault.
> glp_time () at env/time.c:61
> 61            j = jday(tm->tm_mday, tm->tm_mon + 1, 1900 + tm->tm_year);
> 
> (gdb) bt
> #0  glp_time () at env/time.c:61
> #1  0x00000000004029e5 in glp_main (argc=<optimized out>,
>      argv=<optimized out>) at glpapi21.c:1160
> #2  0x00000000004013d9 in __tmainCRTStartup ()
> #3  0x000000000040113f in mainCRTStartup ()
> 

Thank you for your bug report.

The exception happens in a Posix version of glp_time function. It is
difficult to say what is wrong, most likely it is some MINGW64
inconsistency, because under GNU/Linux this code works.

To fix the bug you may simply to disable using the configuration header
(created by the configure script) by replacing the fragment in file
glpk-4.51/src/env/time.c, lines 24-26:

#ifdef HAVE_CONFIG_H
#include <config.h>
#endif

with the following one:

#if 0
#include <config.h>
#endif

In this case an ANSI portable version will be used, which should work on
any platform.

Hope this helps.

BTW, who displays the message "warning: Invalid parameter passed to C
runtime function." ? It is not a glpk message.






reply via email to

[Prev in Thread] Current Thread [Next in Thread]