This is the mail archive of the
mailing list for the glibc project.
Re: How do we test the IBM ieee754 math library?
On Sun, Nov 25, 2001 at 05:37:37PM -0800, Geoff Keating wrote:
> > Date: Sun, 25 Nov 2001 16:52:21 -0800
> > From: "H . J . Lu" <email@example.com>
> > Cc: firstname.lastname@example.org, email@example.com
> > Content-Disposition: inline
> > User-Agent: Mutt/1.2.5i
> > On Sun, Nov 25, 2001 at 04:26:25PM -0800, Geoff Keating wrote:
> > > >
> > > > This may not be worthwhile trying to fix except to ask what
> > > > computer these programs were developed on or what computers they have
> > > > actually been tested on. Evidently the code is intolerant of IEEE
> > > > flaws so there are going to be various systems on which it dosn't work
> > > > properly.
> > >
> > > Well, not that many. Powerpc, sparc, ia64, hppa, and alpha should all
> > > be fine, as should any target that uses soft-float; all those are
> > > strict IEEE. I think s/390 is also strict IEEE. That leaves m68k,
> > > which I understand is also OK for IEEE arithmetic, and x86.
> > >
> > glibc also supports arm, cris, mips and sh.
> I believe these are normally soft-float.
I don't know what you mean by soft-float. MIPS has FPU, but it doesn't
implement the strict IEEE754 in hardware. Linux/mips has quite a few
math test failures, some of which are common with those on x86 when
the IBM libm is used.
Maybe I should ask
1. Does the IBM libm pass "make check" in glibc on ANY platforms?
2. If there are failures, how many of them are kernel bugs and how
many of them are glibc/IBM libm bugs?