Re: [patch] PowerPC ieee float tests

[Franz Sirl <Franz dot Sirl-kernel at lauterbach dot com>]

On Saturday 02 February 2002 23:41, Tom Rix wrote:
> Dale Johannesen wrote:
> > On Saturday, February 2, 2002, at 12:11 PM, Tom Rix wrote:
> > > PowerPC uses 64 bit floating point registers which causes the ieee
> > > float tests to fail.  This patch appends -ffloat-store to the test
> > > flags.   It also cleans up other targets doing the same thing.
> >
> > I don't think this is right.  These tests have been passing on Darwin
> > (ppc32)
> > "forever", and it doesn't look like anybody has hacked the ieee.exp
> > file locally to make this happen.
>
> This is  a regression wrt  GnuPro branch that I work on.   In the dim past
> -ffloat-store was always appended to every target.  From ChangeLog this was
> 8/13/1998,  it was limited to only x86  7/22/2001.   You should check if
> your ieee.exp is up to date.
>
> The test I looked at specifically is 980619-1.  Fairly simple, count how
> may divides by two does it take to get a 1.0 to go to 0.   I have included
> a dump of the assembly.  Since this calculation is being done in   64 bit
> registers, it takes 53 iterations instead of the expected 24.
>
> > Perhaps you mean Power, which didn'thave the single-precision
> > instructions?
>
> This is powerpc-ibm-aix4.3.3.0.

Well, this is either a code-gen bug for AIX or mandated by the AIX ABI, the 
testcase works perfectly fine on powerpc-linux-gnu. If it's mandated by the 
AIX ABI, you can add -ffloat-store for powerpc-*-aix* I guess, but I would be 
really surprised by AIX violating IEEE math... (maybe AIX only does 64bit 
IEEE?).

Franz.

Attachment: 980619-1.s
Description: Text document