Hi, Alexandre, It seems to be relatively easy to have a configure option to disable the use of quiet_NaN. If you do disable quiet_NaN, the alternative is to use max(), which should be normally an obvious outlier. I will implement this option and do some testing in the next day or so. Will let you know when I have it checked into the SVN.
Thanks. John On 5/5/11 3:34 AM, Alexandre Maurel wrote: > Hello John, > > STLPORT is causing the problem. I found the following > > ? stlp_std::numeric_limits<double>::quiet_NaN returned > -1.#IND000000000000 double > > ? stlp_std::numeric_limits<double>::signaling_NaN returned > 1.#QNAN00000000000 double > > It appears that STLPORT has inverted the quiet NaN and the signaling NaN ! > > How to properly handle that bug in Fastbit code source ? > > Regards, > > Alex > > Le 03/05/2011 21:01, [email protected] a écrit : >> Send FastBit-users mailing list submissions to >> [email protected] >> >> To subscribe or unsubscribe via the World Wide Web, visit >> https://hpcrdm.lbl.gov/cgi-bin/mailman/listinfo/fastbit-users >> or, via email, send a message with subject or body 'help' to >> [email protected] >> >> You can reach the person managing the list at >> [email protected] >> >> When replying, please edit your Subject line so it is more specific >> than "Re: Contents of FastBit-users digest..." >> >> >> Today's Topics: >> >> 1. Floating point exception (Alexandre Maurel) >> 2. Re: Floating point exception (K. John Wu) >> >> >> ---------------------------------------------------------------------- >> >> Message: 1 >> Date: Tue, 03 May 2011 12:31:56 +0200 >> From: Alexandre Maurel<[email protected]> >> Subject: [FastBit-users] Floating point exception >> To: [email protected] >> Message-ID:<[email protected]> >> Content-Type: text/plain; charset=ISO-8859-1; format=flowed >> >> Hello John, >> >> Here is my setup : >> - Version ibis1.2.3 >> - Windows 7 >> - Micosoft Visual C++ 2010 Express >> - STLPORT >> >> Table : >> >> C1 C2 >> 6 6 >> 6 2 >> 55 42 >> >> When the query SELECT C1, (C1 + C2) AS C3 WHERE C2< 43 is executed a >> floating point exception is generated in file mensa.cpp, in function : >> >> ibis::table* ibis::table::select(const std::vector<const ibis::part*>& >> mylist, const char *sel, const ibis::qExpr *cond) >> >> at the following statement (line : 3928): >> >> ibis::util::addIncoreData(buff[i], tmp, >> nh,std::numeric_limits<double>::quiet_NaN()); >> >> Don't know why but the compiler doesn't like the quiet NaN. >> Any clues on what causes this exception ? >> >> Regards, >> >> Alexandre >> >> >> >> >> >> >> ------------------------------ >> >> Message: 2 >> Date: Tue, 03 May 2011 09:53:06 -0700 >> From: "K. John Wu"<[email protected]> >> Subject: Re: [FastBit-users] Floating point exception >> To: FastBit Users<[email protected]> >> Cc: Alexandre Maurel<[email protected]> >> Message-ID:<[email protected]> >> Content-Type: text/plain; charset=ISO-8859-1; format=flowed >> >> Hi, Alexandre, >> >> Thanks for reporting the problem. So far, I have tired querying your >> data on unix and windows, the command ibis runs to completion and >> produces expected results. In both cases, I am not using STLPORT. Is >> there any chance that you can compile your code without STLPORT? >> >> On Windows, I am using Visual Studio 2008. I am trying to install >> Visual Studio express 2010 and will see if it makes any difference. >> >> John >> >> >> On 5/3/11 3:31 AM, Alexandre Maurel wrote: >>> Hello John, >>> >>> Here is my setup : >>> - Version ibis1.2.3 >>> - Windows 7 >>> - Micosoft Visual C++ 2010 Express >>> - STLPORT >>> >>> Table : >>> >>> C1 C2 >>> 6 6 >>> 6 2 >>> 55 42 >>> >>> When the query SELECT C1, (C1 + C2) AS C3 WHERE C2< 43 is executed a >>> floating point exception is generated in file mensa.cpp, in function : >>> >>> ibis::table* ibis::table::select(const std::vector<const ibis::part*>& >>> mylist, const char *sel, const ibis::qExpr *cond) >>> >>> at the following statement (line : 3928): >>> >>> ibis::util::addIncoreData(buff[i], tmp, >>> nh,std::numeric_limits<double>::quiet_NaN()); >>> >>> Don't know why but the compiler doesn't like the quiet NaN. >>> Any clues on what causes this exception ? >>> >>> Regards, >>> >>> Alexandre >>> >>> >>> >>> >>> _______________________________________________ >>> FastBit-users mailing list >>> [email protected] >>> https://hpcrdm.lbl.gov/cgi-bin/mailman/listinfo/fastbit-users >> >> ------------------------------ >> >> _______________________________________________ >> FastBit-users mailing list >> [email protected] >> https://hpcrdm.lbl.gov/cgi-bin/mailman/listinfo/fastbit-users >> >> >> End of FastBit-users Digest, Vol 45, Issue 1 >> ******************************************** > _______________________________________________ > FastBit-users mailing list > [email protected] > https://hpcrdm.lbl.gov/cgi-bin/mailman/listinfo/fastbit-users _______________________________________________ FastBit-users mailing list [email protected] https://hpcrdm.lbl.gov/cgi-bin/mailman/listinfo/fastbit-users
