Ed-- in the paragraph above we have the statement "The action blocks are then executed with the same set of iterator values that caused the assertion to pass or fail." Doesn't this prevent the issue you mention? ________________________________ From: Eduard Cerny [mailto:Eduard.Cerny@synopsys.com] Sent: Friday, December 07, 2007 9:44 AM To: Seligman, Erik; Lisa Piper; sv-ac@eda.org Cc: Korchemny, Dmitry Subject: RE: [sv-ac] Updated proposal for 1995 (concurrent assertions in loops) posted Hi Erik, Just noticed something in your example: a1: assume property (foo[i]) $display("Good foo vector: %d", my_ints[i]); else $display("Bad foo vector: %d",my_ints[i]); The (minor) issue I see is that the assume uses the sampled value of foo[i], while $display of my_inst[i] ises the current value. If that one is being also updated elsewhere, even using NBA, this may display a different value that the one used in the sampled foo[i]. Perhaps use $sampled(my_ints[i]) ? Best regards, ed ________________________________ From: owner-sv-ac@eda.org [mailto:owner-sv-ac@eda.org] On Behalf Of Seligman, Erik Sent: Friday, December 07, 2007 12:12 PM To: Lisa Piper; sv-ac@eda.org Cc: Korchemny, Dmitry Subject: [sv-ac] Updated proposal for 1995 (concurrent assertions in loops) posted I added some comments in the coverage section, and a VPI change, in response to Lisa's concerns. Erik Seligman Formal Verification Architect Corporate Design Solutions Design Technology and Solutions Intel Corporation M.S. JF4-402 2111 NE 25th Ave Hillsboro, OR 97124 Phone: (503) 712-3134 -- 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.Received on Fri Dec 7 10:04:32 2007
This archive was generated by hypermail 2.1.8 : Fri Dec 07 2007 - 10:05:35 PST