Re: [PATCH] math: Update i686 ulps

[Florian Weimer <fweimer at redhat dot com>]

On 06/01/2018 06:58 PM, Joseph Myers wrote:
> On Fri, 1 Jun 2018, Florian Weimer wrote:
>
> > On 06/01/2018 02:47 PM, Florian Weimer wrote:
> > >    Function: "y1_upward":
> > > -double: 1
> > > +double: 7
> > >    float: 3
> > >    float128: 5
> > > -idouble: 1
> > > +idouble: 7
> > >    ifloat: 3
> > >    ifloat128: 5
> > >    ildouble: 7
> >
> > I'm posting this for review because some of the changes (like the one above)
> > are fairly large.
>
> How do they compare to ulps for these functions on e.g. x86_64?  My
> presumption is that they arise from use of -mfpmath=sse (thus, meaning no
> excess precision, so results should mostly be the same as on other
> architectures).

Hmm.  Right.  sysdeps/x86_64/fpu/libm-test-ulps gas this:

Function: "y1_upward":
double: 7
float: 2
float128: 5
idouble: 7
ifloat: 2
ifloat128: 5
ildouble: 7
ldouble: 7

So they are not substantially worse.

(If I copy over the x86-64 ulp baseline to the i686 directory, I still 
get many testsuite failures, though.)

Thanks,
Florian