Re: [sv-ec] dist operator and negative value used as weight

I've regiter :
http://www.verilog.org/mantis/view.php?id=3758

You are saying that there are legacy test which should still work -
question is how they should treat negative value - if this was not
described in LRM.
If such code really should pass then maybe warning should be issued and
all value should be treated as unsigned.

There is similar issue - LRM do not define how many bits should weight
be - what happen if I put there 128 bit variable is it ok?

DANiel

On 9/6/2011 7:06 PM, Arturo Salz wrote:
>
> I would prefer a new mantis as dist currently allows negative values.
> I agree with Daniel that distribution weights should be positive,
> however, I know legacy tests already exist with negative weights so
> the error may have to be optional.
>
> Arturo
>
> *From:*owner-sv-ec@eda.org [mailto:owner-sv-ec@eda.org] *On Behalf Of
> *Daniel Mlynek
> *Sent:* Tuesday, September 06, 2011 6:40 AM
> *To:* Bresticker, Shalom
> *Cc:* sv-ec@eda.org
> *Subject:* Re: [sv-ec] dist operator and negative value used as weight
>
> This is similar but for dist LRM even do not say that weigth should be
> positive valuei it just say that in can by any integral expression
> IMHO another mantis should be added.
> Should I add new mantis?
> Or better add a note to mantis 262?
>
>
> DANiel
>
> On 9/6/2011 3:16 PM, Bresticker, Shalom wrote:
>
> See a related issue in Mantis 262.
>
> Shalom
>
> *From:*owner-sv-ec@eda.org <mailto:owner-sv-ec@eda.org>
> [mailto:owner-sv-ec@eda.org] *On Behalf Of *Daniel Mlynek
> *Sent:* Tuesday, September 06, 2011 4:10 PM
> *To:* sv-ec@eda.org <mailto:sv-ec@eda.org>
> *Subject:* [sv-ec] dist operator and negative value used as weight
>
> LRM defines only how should dist behave is 0 is used as weigth.
> What should happen if wigth expression is negative?
> IMHO it should be an error
>
> module top;
> int a;
> bit b;
> int arr[int];
> initial begin
> repeat (100)begin
> b=std::randomize (a) with {
> a dist {
> 1:=2,
> 5:=-1, //<<negative
> 10:=0,
> 100:=1
> };
> };
> arr[a]++;
> end
> foreach (arr[i])
> $display( i, arr[i]);
> end
> endmodule
>
> There can be more complicated cases as weigth can be a variable or
> parameter.
>
>
> DANiel
>
> --
> This message has been scanned for viruses and
> dangerous content by *MailScanner* <http://www.mailscanner.info/>, and is
> believed to be clean.
>
> ---------------------------------------------------------------------
> Intel Israel (74) Limited
>
> This e-mail and any attachments may contain confidential material for
> the sole use of the intended recipient(s). Any review or distribution
> by others is strictly prohibited. If you are not the intended
> recipient, please contact the sender and delete all copies.
>
>
> --
> This message has been scanned for viruses and
> dangerous content by *MailScanner* <http://www.mailscanner.info/>, and is
> believed to be clean.
>

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.