Friday, November 21, 2008     Register | Login | Search | Contact Us
     

Many of you already received communications about the move of the Cadence user community into cadence.com. And many of you have already joined, with over 4000 registrations in the first two weeks.

The new Cadence Community enhances the ability of Cadence users to connect and collaborate. In addition to moving the community into cadence.com -- enabling single sign-on for community, Sourcelink and Cadence events -- the new site is organized around nine technology segments, giving you easy access to product information, training, forums and blogs. Some of the new features include:
  • Ability to respond to posts via e-mail
  • Technology-specific blogs
  • Latest Web 2.0 social networking capabilities
  • Public profile options
  • Private messaging
  • Friends lists
Visit the new Cadence Community today at www.cadence.com/community and join the discussions!

Registration note: Due to the scope of the enhancements and the new SSO registration system, we were not able to migrate existing cdnusers.org member accounts. So new registrations are required, but this enables a broader set of functionality we think you'll enjoy.

Forum note: Under the guidance of forum moderators, we have taken the 20+ cdnusers.org forums and consolidated them into 11 forums on the new site. Posts have been brought over so you can leverage that posting history. CDNusers forums will be set to read only starting 7/30, and cdnusers.org will be redirected to the new community on 8/4.

Best regards,
Mike and Tom

Michael A. Catrambone - Steering Committee Chairman
Distinguished Engineer
PCB/Mechanical
UTStarcom, Inc.

Tom Diederich
Cadence Community Manager
Home
Forums
Subject: N unique stable data vectors
Posting to forums is available to community members only.
Login or Register
Rate this topic:
   
Author Messages
weberrm
Posts: 0
Online: User is Offline
3/04/2007 3:34 PM  
Working off the concept that a property like the following causes a formal tool to try all values of the stable_data wire...

wire [W-1:0] stable_data;
//psl assume_stable_data: assume always(stable(stable_data))@(posedge clk);

Is there a way to generate N stable_data vectors and at the same time guarantee that at any one time the data in them is unique.

If I just wanted 2 vectors I might do something like the following:

wire [W-1:0] stable_data0,stable_data1;
//psl assume_stable_data0: assume always(stable(stable_data0))@(posedge clk);
//psl assume_stable_data1: assume always(stable(stable_data1))@(posedge clk);

//psl assume_unique_data: assume always(
// stable_data0!=stable_data1
//)@(posedge clk);

Extending this to N vectors seems troublsome.  Any ideas?

One might also ask how to fill a NxW array with unique stable data.

Just for clarification, the fact that the data is stable isn't really a necessary component of this question.  I am mainly just wondering how I could easily generate N unique vectors of data.  It's just that data like this is usually stable in proofs.

Thanks,

Ross
foster
Posts: 0
Online: User is Offline
3/05/2007 12:37 AM  
Hi Ross,

I don't have an answer to the question of repeating N-Times the unique constraint, but other remarks:

1) stable(stable_data))@(posedge clk) is only guaranteeing stability if the design is purely synchronous to the posedge, otherwise there may be arbitrary values at the clock high phase.

2) The unique vector set is a subset of the generic vector set. IFV will apply any vector set to FAIL an assertion, including the one with unique vectors, I don't really see a need to restrict it to the unique set. Potential reason might be

a) you don't like this CEX
b) you need it for a coverage trace
c) this is really a conatraint of the system and causes a false failure in an assertion

Can you elaborat eon your motivation for this conatraints?

Thanks,
Joerg.
weberrm
Posts: 0
Online: User is Offline
3/05/2007 6:57 AM  
Hi Joerg,

A couple uses may be as follows:

There are N requestors to a design and each requestor needs to be restricted such that none of the requestors make requests to the same address.  With a solution to this question I could set up address range limitations for each requestor.

A design needs N requestors to have unique requestor IDs.


Below is something I've come up with (I haven't tried it):

...
localparam W=4,N=8;
wire [W-1:0] stable_data [N-1:0]
genvar gi;
generate for (gi=0;gi
  if(gi==0) begin
    gt_last #(.W(W)) data(
      .clk(clk),
      .min({W{1'b0}}),
      .stable_data(stable_data[gi])
    );
  end else begin
    gt_last #(.W(W)) data(
      .clk(clk),
      .min(stable_data[gi-1]+{{W-1{1'b0}},1'b1}), // Next min 1 greater than last
      .stable_data(stable_data[gi])
    );
  end
end endgenerate
...

module gt_last
#(W=3)
(
input clk,
input [W-1:0] min,
output [W-1:0] stable_data
);

//psl assume_stable_data_ge_min: assume always(
//  stable_data>=min & stable(stable_data)
//)@(posedge clk);

endmodule

Thanks,

Ross
weberrm
Posts: 0
Online: User is Offline
3/23/2007 8:01 AM  
Below is a more elegant (and better) solution than what I posted above.

I just used some verilog to determine when the data slices were equal and wrote an assumption to prevent the signal from ever becoming true.  I also added a stable assumption that is only needed if you want the data to be stable throughout a sequence.

module unique_stable_data
#(
parameter
N=4,
DATA_W=5
)
(
input clk,
input [(N*DATA_W)-1:0] data
);

reg data_eq;
integer i,j;
always @* begin
  data_eq=1'b0;
  for(i=0;i < N-1;i=i+1) begin
    for(j=i+1;j < N;j=j+1) begin
      data_eq = data_eq | (data[i*DATA_W+:DATA_W]==data[j*DATA_W+:DATA_W]);
    end
  end
end

//psl assume_data_ne: assume never(data_eq)@(posedge clk);


// If the data needs to be stable
//psl assume_stable_data: assume always(
//  stable(data)
//)@(posedge clk);

endmodule

Posting to forums is available to community members only.
Login or Register



ActiveForums 3.6
     
Copyright 2006 Cadence Design Systems, Inc.