On 9 Oct 1997, Douglas Bates wrote: > > Another thing to try is explicitly setting the floating point control > word to the IEEE standard. Some older versions of the C libraries set > the floating point control word to a non-standard value. That can > affect the production and detection of NaN's. Checking a recently > installed Debian 1.3 system for the file /usr/include/i386/fpu_control.h > (included from /usr/include/fpu_control.h), it shows
Thanks. I tried your suggestion but I still keep getting the NANs. By the way the NaNs are not affecting my results because I do not use floats. I use integers for Hidden Markov Models (HMM) probabilities. I, however, need to occasionaly print this integers by converting them to floats and that is when they print as NANs. I have used electric-fence malloc to see if there is no accidental memory access but I have found none. I think the next thing is to learn some assembler code and see what is really taking place/ I want to thank all the Debianites for overwhelming comments and advices. I have send similar messages to the gcc newsgroups (.help etc) without receiving a single reply - but you guys have been fantastic. One more reason to stay with Debian. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
