Hi, Just for the record. I found a workaround for my initial problem (cyl_bessel_i returning NaN in valgrind). It is possible to tell boost math not to use long doubles internally by adding
-DBOOST_MATH_PROMOTE_DOUBLE_POLICY=false to the compiler flags. This doesn't fix the real long double problem but makes cyl_bessel_i with doubles usable in valgrind. This allows me to use valgrind again on my software :) Cheers, Tom On Thu, Aug 13, 2009 at 12:09, Christoph Bartoschek<[email protected]> wrote: > Am Donnerstag, 13. August 2009 schrieb Tom Vercauteren: > >> In this case the difference seems critical to me as it breaks the >> semantics of numeric_limits... Maybe a note on numeric_limits<long >> double> could be added to the documentation. >> >> It's a pity I will not be able to use valgrind anymore to fix my >> software; valgrind had been serving me so good until now. > > You are right. In my opinion this is a valid testcase that shows that it is > critical to support 80bit precision. > > The code does not rely on 80bit precision and is therefore portable, however > it breaks within valgrind. > > Christoph > > ------------------------------------------------------------------------------ > Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day > trial. Simplify your report design, integration and deployment - and focus on > what you do best, core application coding. Discover what's new with > Crystal Reports now. http://p.sf.net/sfu/bobj-july > _______________________________________________ > Valgrind-users mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/valgrind-users > ------------------------------------------------------------------------------ Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july _______________________________________________ Valgrind-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/valgrind-users
