[sv-bc] RE: Q: [N] array bounds legal for unpacked but not packed arrays?

The [0:N] ordering is also safer for an array used with $readmem.  The original LRM specified that values would be read from leftmost index to rightmost index.  Verilog-XL read them from lowest to highest (apparently due to a bug), and other tools followed suit.  The LRM was eventually changed to match that as well.  With the [0:N] ordering, both rules give the same ordering, so it should work with tools matching either standard.


From: owner-sv-bc@eda.org [mailto:owner-sv-bc@eda.org] On Behalf Of Arturo Salz
Sent: Tuesday, September 30, 2014 5:06 PM
To: Rich, Dave; Brad Pierce; sv-bc@eda.org
Subject: [sv-bc] RE: Q: [N] array bounds legal for unpacked but not packed arrays?

Recall that the [0:N-1] ordering was also chosen because it matches the ordering of dynamic arrays and queues. Also compatibility with C/C++ is important when dealing with DPI.

                Arturo

From: owner-sv-bc@eda.org [mailto:owner-sv-bc@eda.org] On Behalf Of Rich, Dave
Sent: Tuesday, September 30, 2014 1:58 PM
To: Brad Pierce; sv-bc@eda.org
Subject: [sv-bc] RE: Q: [N] array bounds legal for unpacked but not packed arrays?

The current [0:N-1] ordering was chosen to match array index ordering in C/C++.

IMHO, [N] should never been allowed as a shortcut in a language that already supported ranges in either order because of the confusion it creates. Erik's workaround example is proof of that.

But it's too late to change the default range order now for an unpacked array now.

Dave


From: owner-sv-bc@eda.org<mailto:owner-sv-bc@eda.org> [mailto:owner-sv-bc@eda.org] On Behalf Of Brad Pierce
Sent: Tuesday, September 30, 2014 1:44 PM
To: sv-bc@eda.org<mailto:sv-bc@eda.org>
Subject: [sv-bc] RE: Q: [N] array bounds legal for unpacked but not packed arrays?

[In reply to http://www.eda.org/sv-bc/hm/11620.html .]

The current [0:N-1] default for unpacked array dimensions is also significant (and annoying). With the hardware order [N-1:0] you can split/merge dimensions fluidly, as in

typedef struct {int x, y;} T;

module test#(N=2,M=3)
(   output T A_2D[(2**N)-1:0][(2**M)-1:0]
,   output T A_1D[(2**(N+M))-1:0]
);


logic [N-1:0] idx_N;
logic [M-1:0] idx_M;

always_comb begin
   for (int i = 0; i < 2**N; i++) begin
     for (int j = 0; j < 2**M; j++) begin
       idx_N = i;
       idx_M = j;
       A_2D[idx_N][idx_M] = '{i,j};
       A_1D[{idx_N,idx_M}] = '{i,j};
       assert final (A_2D[idx_N][idx_M] == A_1D[{idx_N,idx_M}]);
     end
   end
end

endmodule

But with the [0:N-1] default you need to flip the order of dimensions in the indexes, as in

typedef struct {int x, y;} T;

module test#(N=2,M=3)
(   output T A_2D[0:(2**N)-1][0:(2**M)-1]
,   output T A_1D[0:(2**(N+M))-1]
);

logic [N-1:0] idx_N;
logic [M-1:0] idx_M;

always_comb begin
   for (int i = 0; i < 2**N; i++) begin
     for (int j = 0; j < 2**M; j++) begin
       idx_N = i;
       idx_M = j;
       A_2D[idx_N][idx_M] = '{i,j};
       A_1D[{idx_M,idx_N}] = '{i,j};
      assert final (A_2D[idx_N][idx_M] == A_1D[{idx_M,idx_N}]);
     end
   end
end

endmodule

-- Brad


--
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<http://www.mailscanner.info/>, and is
believed to be clean.

--
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.