RE: [sv-bc] SV-BC #110 - 2-State Divide by 0 question

All bits set in the result is closer to infinity (the true answer of
x/y when y is zero) than is 0. The graph of x versus y will also look
better as y approaches zero

-mac

-- On Nov 23 2004 at 16:53, Maidment, Matthew R sent a message:
> To: doug_warmke@mentorg.com, cliffc@sunburst-design.com, sv-bc@eda.org
> Subject: "RE: [sv-bc] SV-BC #110 - 2-State Divide by 0 question"
>
> I disagree with Doug. It might be a warning, but even
> this I don't like. I would prefer that such decisions about
> warnings and errors be left to the users via assertions. Let
> the user write an assertion that the denominator is never 0
> and let the user react accordingly.
>
> Regardless, some default numerical result needs to be defined.
> 0 is a perfectly acceptable result to me. A simulator vendor
> could also provide other options to users such as a random return
> value or '1 or whatever.
>
> Matt
>
> >-----Original Message-----
> >From: owner-sv-bc@eda.org [mailto:owner-sv-bc@eda.org] On
> >Behalf Of Warmke, Doug
> >Sent: Tuesday, November 23, 2004 4:33 PM
> >To: Clifford E. Cummings; sv-bc@eda.org
> >Subject: RE: [sv-bc] SV-BC #110 - 2-State Divide by 0 question
> >
> >Cliff, others,
> >
> >How about saying something like:
> >
> > It shall be an error to perform a
> > divide-by-0 operation on a 2-state object.
> >
> >I don't think any numerical result is correct.
> >All 0's is too optimistic in my opinion.
> >
> >In light of the new datatypes work, my suggested wording
> >could probably use a little wordsmithing. Volunteers?
> >
> >Regards,
> >Doug
> >
> >> -----Original Message-----
> >> From: owner-sv-bc@eda.org [mailto:owner-sv-bc@eda.org] On
> >> Behalf Of Clifford E. Cummings
> >> Sent: Tuesday, November 23, 2004 4:06 PM
> >> To: sv-bc@eda.org
> >> Subject: [sv-bc] SV-BC #110 - 2-State Divide by 0 question
> >>
> >> Hi, All -
> >>
> >> Don Mills and I have been asked to make a proposal for Issue
> >> #110 on divide
> >> by 0 in 2-state.
> >>
> >> I believe Don and I will propose that divide by 0 in 2-state
> >> just gives
> >> zero as a result. The problem is that the errata refers to a
> >> 1364 section
> >> and I am not sure where we should add this in the
> >> SystemVerilog standard.
> >> There is no addition-operator related section in the SV
> >> standard to amend
> >> with this information.
> >>
> >> Any suggestions?
> >>
> >> Regards - Cliff
> >>
> >> ----------------------------------------------------
> >> Cliff Cummings - Sunburst Design, Inc.
> >> 14314 SW Allen Blvd., PMB 501, Beaverton, OR 97005
> >> Phone: 503-641-8446 / FAX: 503-641-8486
> >> cliffc@sunburst-design.com / www.sunburst-design.com
> >> Expert Verilog, SystemVerilog, Synthesis and Verification Training
> >>
> >>
> >>
> >
> >
>
Received on Wed Nov 24 08:33:04 2004