RE: [sv-ec] Effect of ignore/illegal bins - Mantis #1237

Swapnajit,

 

That line means that the distribution of values into the bins, i.e., the
space computation, must happen first, disregarding any ignored/illegal
bins specified. Then, after the bins have been computed, ignore values
are computed, which may result in some bins having no values (empty
bins). This is covered by the proposed example for 18.10.1:

 

bit [2:0] a,b;
covergroup ct;
    coverpoint b {
        option.auto_bin_max = 4;
        ignore_bins ig = { [0:1], [5:6] };
    }

endgroup

 

In this case, coverpoint b will have 4 auto bins: auto[0,1], auto[2,3],
auto[4,5], auto[6,7]. The ignore_bins declaration specifies that the
values 0,1,5,6 are ignored. After applying the ignore_bins, the bins
are: auto[], auto[2,3], auto[4], auto[7]. Since it no longer is
associated with any value, 'auto[]' does not contribute to coverage.

 

Without that clarification, someone may consider implementing ignored
bins in the reverse order, that is, first remove the ignored values and
the distribute the remaining (non-ignored) values onto bins.

 

            Arturo

 

________________________________

From: Swapnajit Chakraborti [mailto:swapnaj@cadence.com] 
Sent: Sunday, February 12, 2006 8:48 PM
To: Arturo Salz; Ryan, Ray; sv-ec@eda.org
Subject: RE: [sv-ec] Effect of ignore/illegal bins - Mantis #1237

 

Arturo & Ray, 

Could you please elaborate the following line added in this mantis: 

" The removal of ignored values shall occur after the distribution of
values to the specified bins." 

Thx, 
Swapnajit. 

>-----Original Message----- 
>From: owner-sv-ec@eda.org [mailto:owner-sv-ec@eda.org
<mailto:owner-sv-ec@eda.org> ] On 
>Behalf Of Arturo Salz 
>Sent: Sunday, February 12, 2006 1:04 AM 
>To: Ryan, Ray; sv-ec@eda.org 
>Subject: RE: [sv-ec] Effect of ignore/illegal bins - Mantis #1237 
> 
>Ray, 
> 
>Attached is a modified proposal with some minor friendly amendments. 
>The amendments do not change the proposal's semantics, only 
>the verbiage. 
> 
>       Arturo 
> 
>-----Original Message----- 
>From: owner-sv-ec@eda.org [mailto:owner-sv-ec@eda.org
<mailto:owner-sv-ec@eda.org> ] On 
>Behalf Of Ryan, Ray 
>Sent: Friday, February 10, 2006 4:29 PM 
>To: sv-ec@eda.org 
>Subject: [sv-ec] Effect of ignore/illegal bins - Mantis #1237 
> 
>Attached is a proposal for mantix #1237. 
> 
>- values are distributed, then ignore/illegal bins applied. 
>- bins with no values/transitions do not contribute to coverage 
>- coverpoint with no contributing bins does not contribute to coverage 
>- covergroup with no contributing items does not contribute to 
>overall coverage 
>- ignore/illegal VALUES do not effect transitions. 
> 
>- Ray 
>