[sv-bc] FW: Unsized parameter value propagation

This is what I had sent Surya.

Shalom

From: Bresticker, Shalom
Sent: Sunday, June 17, 2012 16:19
To: Surya Pratik Saha
Subject: RE: Unsized parameter value propagation

Hi,

It comes down to whether an unsized integer literal can be bigger than 32 bits. The LRM is ambiguous on it. NC left-truncates the big number in the statement instantiating A as instA. The rest follows from there.

It is an old issue, but unresolved. See Mantis 1107.

Regards,
Shalom

From: Bresticker, Shalom
Sent: Thursday, June 14, 2012 19:58
To: Surya Pratik Saha
Subject: RE: Unsized parameter value propagation

Hi, Surya.

Cadence NC gave me this:

    A #(36, 'b010001110111011101110111011101110111) instA();
                                                 |
ncvlog: *W,INTOVF (test.v,2|49): bit overflow during conversion from text [2.5(IEEE)] (32 bits).

parameter values :: instA.p_a 01110111011101110111011101110111 , instA. instB.p_b 000001110111011101110111011101110111

Synopsys VCS gives:

parameter values :: instA.p_a 010001110111011101110111011101110111 , instA. instB.p_b 010001110111011101110111011101110111

I will send a separate mail explaining what I think is happening.

Regards,
Shalom
---------------------------------------------------------------------
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.
Received on Thu Jun 21 21:41:07 2012