FloatTest

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

FloatTest

Thomas Koenig-2

Does anyone have an insight into how we ought to resolve these two FloatTest
failures.
FloatTest>>ttestNaN5  and >>testZero2 fail in 3.9a 7025.  Can they can be
fixed or do we think they're pointing to real problems in Float re ieee754?

> From: Paul McDonough wrote
> ... back to school ... in any case, I'm not 'ppm.'
> Although I'm sure I was at the same table where those
> tests were being worked on at CS4 ... hm, don't
> remember any ppm's.  Roger Whitney worked on Floats, I
> think?  And/or Jeff Hallman?


> --- Thomas Koenig <[hidden email]> wrote:
> > Paul,
> > Hi.  We're trying to get all s-unit tests to run clean
> > in 3.9.  I think you
> > added the following test cases (initials are ppm)
> > which is why I want to get
> > your opinion on how to proceed. FloatTest>>ttestNaN5
> > and >>testZero2 fail in
> > 3.9a 7025.  Can they can be fixed or do you think
> > they're pointing to real
> > problems in Float re ieee754? In the latter case
> > maybe they should be marked
> > "expected errors"?
> > Thanks for any help. Tom
> >
> >
> >
>
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com



Reply | Threaded
Open this post in threaded view
|

Re: FloatTest

stéphane ducasse-2
Tx thomas :)
Would be really cool to have them all green :)

Stef

On 5 mai 06, at 16:48, Thomas Koenig wrote:

>
> Does anyone have an insight into how we ought to resolve these two  
> FloatTest
> failures.
> FloatTest>>ttestNaN5  and >>testZero2 fail in 3.9a 7025.  Can they  
> can be
> fixed or do we think they're pointing to real problems in Float re  
> ieee754?
>
>> From: Paul McDonough wrote
>> ... back to school ... in any case, I'm not 'ppm.'
>> Although I'm sure I was at the same table where those
>> tests were being worked on at CS4 ... hm, don't
>> remember any ppm's.  Roger Whitney worked on Floats, I
>> think?  And/or Jeff Hallman?
>
>
>> --- Thomas Koenig <[hidden email]> wrote:
>>> Paul,
>>> Hi.  We're trying to get all s-unit tests to run clean
>>> in 3.9.  I think you
>>> added the following test cases (initials are ppm)
>>> which is why I want to get
>>> your opinion on how to proceed. FloatTest>>ttestNaN5
>>> and >>testZero2 fail in
>>> 3.9a 7025.  Can they can be fixed or do you think
>>> they're pointing to real
>>> problems in Float re ieee754? In the latter case
>>> maybe they should be marked
>>> "expected errors"?
>>> Thanks for any help. Tom
>>>
>>>
>>>
>>
>>
>> __________________________________________________
>> Do You Yahoo!?
>> Tired of spam?  Yahoo! Mail has the best spam protection around
>> http://mail.yahoo.com
>
>
>
>


Reply | Threaded
Open this post in threaded view
|

Re: FloatTest

Philippe Marschall
In reply to this post by Thomas Koenig-2
http://bugs.impara.de/view.php?id=3133

2006/5/5, Thomas Koenig <[hidden email]>:

>
> Does anyone have an insight into how we ought to resolve these two FloatTest
> failures.
> FloatTest>>ttestNaN5  and >>testZero2 fail in 3.9a 7025.  Can they can be
> fixed or do we think they're pointing to real problems in Float re ieee754?
>
> > From: Paul McDonough wrote
> > ... back to school ... in any case, I'm not 'ppm.'
> > Although I'm sure I was at the same table where those
> > tests were being worked on at CS4 ... hm, don't
> > remember any ppm's.  Roger Whitney worked on Floats, I
> > think?  And/or Jeff Hallman?
>
>
> > --- Thomas Koenig <[hidden email]> wrote:
> > > Paul,
> > > Hi.  We're trying to get all s-unit tests to run clean
> > > in 3.9.  I think you
> > > added the following test cases (initials are ppm)
> > > which is why I want to get
> > > your opinion on how to proceed. FloatTest>>ttestNaN5
> > > and >>testZero2 fail in
> > > 3.9a 7025.  Can they can be fixed or do you think
> > > they're pointing to real
> > > problems in Float re ieee754? In the latter case
> > > maybe they should be marked
> > > "expected errors"?
> > > Thanks for any help. Tom
> > >
> > >
> > >
> >
> >
> > __________________________________________________
> > Do You Yahoo!?
> > Tired of spam?  Yahoo! Mail has the best spam protection around
> > http://mail.yahoo.com
>
>
>
>

Reply | Threaded
Open this post in threaded view
|

RE: FloatTest

Thomas Koenig-2
So Philippe, we harvested you test but not your fix?

Philippe Marschall

> http://bugs.impara.de/view.php?id=3133
>
> 2006/5/5, Thomas Koenig <[hidden email]>:
> >
> > Does anyone have an insight into how we ought to resolve these two
> FloatTest
> > failures.
> > FloatTest>>ttestNaN5  and >>testZero2 fail in 3.9a 7025.  Can they can
> be
> > fixed or do we think they're pointing to real problems in Float re
> ieee754?