From mboxrd@z Thu Jan 1 00:00:00 1970 From: Craig Burley To: jbuck@Synopsys.COM Cc: burley@gnu.org Subject: Re: /internet Date: Thu, 17 Dec 1998 10:26:00 -0000 Message-id: <199812171826.NAA29123@melange.gnu.org> References: <199812161958.LAA29146@atrus.synopsys.com> X-SW-Source: 1998-12/msg00652.html >If this is the rule, then a*b*c*d can't be rearranged in C9X. However, >it appears that it can be rearranged in Fortran (where the user must >use parentheses to force the order of evaluation) as well as in the >current ANSI/ISO C. Thus preserving Fortran as the language of choice for scientific/ numeric programming, at least over anything stemming from C, since the Fortran code can be implemented faster than the equivalent C code. So it strikes me as strange that they would impose these rules, given how they (same people, I *think*) have been trying so hard to wedge in new gimmicks to make C *potentially* as fast as Fortran, like `restrict'. (Which is much harder to use than writing ((a*b)*c)*d, AFAICT.) I wouldn't be surprised to see these rules dropped, if there's still time, so we probably shouldn't take them as the gospel according to C9X until it's reached the stage where substantial changes are no longer made to the draft standard (though I have no idea where the process is at this point, myself). tq vm, (burley)