Re: [Help-glpk] [bug] glpk cycling

2017-11-22 Thread DAVID MONNIAUX
> para.meth = GLP_PRIMAL; I know but on our examples where DUAL does not cycle, it runs quite faster than PRIMAL. ___ Help-glpk mailing list Help-glpk@gnu.org https://lists.gnu.org/mailman/listinfo/help-glpk

[Help-glpk] [bug] glpk cycling

2017-11-22 Thread David Monniaux
Dear GLPK friends, The attached example makes GLPK cycle infinitely (GLPK 4.63, x86-64). Is this a bug? Any way to avoid this? (The bug was found by my student Hang Yu. congratulations Hang!) Regards -- David Monniaux directeur de recherche au CNRS, laboratoire VERIMAG Université Grenoble

Re: [Help-glpk] Help-glpk Digest, Vol 170, Issue 34

2017-01-31 Thread David Monniaux
hread checkers Valgrind / drd and helgrind, no warnings. So I guess it works. -- David Monniaux directeur de recherche au CNRS, laboratoire VERIMAG Université Grenoble Alpes http://www-verimag.imag.fr/~monniaux/ <>___ Help-glpk mailing list Hel

Re: [Help-glpk] things I don't understand in src/env/time.c + bugs + suggested patch

2017-01-28 Thread David Monniaux
1-1-1970? -- David Monniaux directeur de recherche au CNRS, laboratoire VERIMAG Université Grenoble Alpes http://www-verimag.imag.fr/~monniaux/ <>___ Help-glpk mailing list Help-glpk@gnu.org https://lists.gnu.org/mailman/listinfo/help-glpk

Re: [Help-glpk] things I don't understand in src/env/time.c + bugs + suggested patch

2017-01-28 Thread David Monniaux
On 01/28/2017 02:35 PM, Heinrich Schuchardt wrote: > > 1E3 * 1E-3 = 1. > You wouldn't add seconds and microseconds without scaling. Typo (silly copying and pasting). It should have read: return tv.tv_sec * 1E3 + tv.tv_usec * 1E-3; -- David Monniaux directeur de recherche au CNR

[Help-glpk] things I don't understand in src/env/time.c + bugs + suggested patch

2017-01-28 Thread David Monniaux
((double)(j - EPOCH) * 24.0 + (double)tm->tm_hour) * 60.0 + - (double)tm->tm_min) * 60.0 + (double)tm->tm_sec) * 1000.0; - return t; +{ return time(NULL) * 1E3; } #endif Regards -- David Monniaux directeur de recherche au CNRS, laboratoire VERIMAG Université Grenoble Al

Re: [Help-glpk] Patch to configure a reentrant version of GLPK

2017-01-09 Thread David Monniaux
Hi > I think that running glpk in multiple threads is rather an exception > needed mainly in interactive programs and in some other exotic > applications, so in my opinion this option should be disabled by > default. I see the matter otherwise. 1) Putting this variable in thread-local storage usi

Re: [Help-glpk] Thread safety of GLPK

2016-12-27 Thread David Monniaux
platforms, since this seems a common request. -- David Monniaux Directeur de recherche au CNRS, laboratoire VERIMAG http://www-verimag.imag.fr/~monniaux/ ___ Help-glpk mailing list Help-glpk@gnu.org https://lists.gnu.org/mailman/listinfo/help-glpk