From: root AT jacob DOT remcomp DOT fr (root) Subject: Re: rint function (continued) 16 Dec 1998 06:35:59 -0800 Message-ID: References: <4 DOT 1 DOT 19981214084527 DOT 035e9470 AT pop DOT ni DOT net> Content-Type: text To: jdeifik AT weasel DOT com (Jeff Deifik) Cc: gnu-win32 AT cygnus DOT com In gnu assembler language rint is: .globl _rint _rint: fldl 4(%esp) ; load double argument in the fpu push %eax ; make space in stack for result fistpl (%esp) ; store the rounded value there as 32 bit int popl %eax ; pop the result from the stack ret ; done This takes the current rounding mode into account, as rint should. Using this function makes your program pass all your tests. Enjoy! -- Jacob Navia Logiciels/Informatique 41 rue Maurice Ravel Tel 01 48.23.51.44 93430 Villetaneuse Fax 01 48.23.95.39 France - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request AT cygnus DOT com" with one line of text: "help".