Date: Thu, 14 Dec 1995 23:31:28 +1100 From: Bruce Evans <bde@zeta.org.au> To: CVS-committers@freefall.freebsd.org, cvs-user@freefall.freebsd.org, phk@freefall.freebsd.org Subject: Re: cvs commit: src/lkm/gnufpu Makefile Message-ID: <199512141231.XAA13391@godzilla.zeta.org.au>
next in thread | raw e-mail | index | archive | help
> Modified: lkm Makefile > Added: lkm/fpu Makefile > lkm/gnufpu Makefile > Log: > Two x87 emulators as LKMs. > If somebody with the right HW would make the change to /etc/rc to use > this, we could rip MATH_EMULATE from GENERIC... It can't work with the right h/w (no fpu) or the right testing procedure (fake no h/w) because at least /bin/sh uses the fpu for at least _setjmp() and _longjmp(). It's hard to tell which things in the library or in /etc/rc use floating point. In fact, fsck uses floating point just to print the fragmentation percentage, and csh uses floating point for its printf(1) builtin, and sh uses floating point for its printf(1) builtin (see /usr/src/usr.bin/printf/printf.c) although it uses a special broken version of printf(3) for builtins and screws up `printf %g 1.1'... Bruce
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199512141231.XAA13391>