Hi Byung, Using variables with in bin ranges may cause many conflicts. Taking an example of a fixed sized bin bins fixed [4] = {[1 : ddr_max]}; If at simulation time we are changing the value of "ddr_max" than the actual distribution of values according to size will get changed at the simulation time. This is certainly not intended. Regarding using covergroup arguments there are two rules which makes them constant as compared a simple variable. In section 18.3 :: Input arguments will not track the value of their arguments; they will use the value passed to the new operator. In section 18.5 :: The open_range_list used to specify the set of values associated with a bin shall be constant expressions , instance constants (for classes only), or non-ref arguments to the coverage group. As constant expressions and instance constants (for classes ) are allowed in bin ranges so run-time constants using const construct should also be allowed. Thanks Saurabh Sharma saurabhs@cadence.com Neil Korpusik wrote: > <forwarding email from new member> > > Questions such as this should go to the following email address. > > sv-ec@eda.org > > > > > -------- Original Message -------- > Subject: Defining range of bins in coverpoint > Date: Thu, 18 Dec 2008 09:28:22 -0700 > From: Yoo, Byung C <byung.c.yoo@intel.com> > > > Hi > > I just subscribed to the mail list as I was instructed by Maidment, > Matthew R, Levy, Yossef (DT) at Intel Corporation. > > I am not sure if I am sending email to the right address. Please let me > know what I should do if this it not he right process. > > There is a question/suggestion for enhancement as described below. > > When I define the range of a bin in coverpoint, I found that it is > strange that variables can be used as range of bins only if they are > covergroup argument > > > addr: coverpoint iT.address iff (iT.cmd == MEM_WRITE || iT.cmd == > MEM_READ) > { > bins memory = {[0:ddr_max]}; // Memory space > } > > > I could use neither any other variables nor const constant. > There are many randomly generated variables before the covergroup is > generated. > It will be convenient if I can use them directly without having to use > covergroup argument. > > Does this make sense? > > Thanks, > Byung Yoo > > DHG > Intel Corporation > -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.Received on Mon Dec 22 20:22:11 2008
This archive was generated by hypermail 2.1.8 : Mon Dec 22 2008 - 20:22:59 PST