[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: floor() broken on EV4?



On Tue February 20 2007 00:31, prescott@phys.ufl.edu wrote:
> On Feb 11 2007, Falk Hueffner wrote:
> > The EV4 FPU cannot round to minus infinity, as glibc's floor wants to
> > do, so the kernel emulates this. So it's almost certainly a kernel
> > bug.
>
> <snip>
>
> Or pilot error :-(  I finally got a chance to look more into this,
> and I see that I have gaffed.  I did not have the math-emu module
> loaded.  Once loaded, floor() (and other stuff) works fine.

Well done!  I just tried loading the module on my PWS500 (EV56), and inf and 
denormalized numbers all works now.  That's an old problem solved.

> Maybe everyone except me knew to have math-emu loaded?  I was
> kind of surprised it was not built-in.  Is this intended?

I know I have never heard of it before.  Considering everything works out of 
the box on 2.4 kernels, I would guess it was split off sometime on the road 
to 2.6.  Maybe the config option didn't exist before, or maybe Debian kernels 
just used to compile it in by default.

-Tyson

-- 
 Tyson Whitehead  (-twhitehe at uwo.ca -- MC-)
 Computer Engineer                          Dept. of Applied Mathematics,
 Graduate Student- Applied Mathematics      University of Western Ontario,
 GnuPG Key ID# 0xF7666BFF                   London, Ontario, Canada

Attachment: pgpBuIrpeo1tL.pgp
Description: PGP signature


Reply to: