Resolving PPA Issues with RTL Architect & Verdi Integration

The integration of Synopsys RTL Architect with the Synopsys Verdi debug system environment provides an easy way to get earlier insights into PPA issues, so these problems can be resolved for an overall better product and faster RTL closure.

Transcript:

So from RTLA, we can call Verdi. If we click on the Verdi icon at the top, we bring up the dialog to launch Verdi. We need to pass through kdb data and optionally, the PPA data from RTLA.

Now, if you already have a kdb, you can specify the path here, and that's what I've already done. I've actually written one out previously. And again, I've already written out the PPA data, so we can just read that.

So if I click OK, the tool is now going to start Verdi. It will load the kdb data, and it will also load the PPA data that has been exported from RTLA.

So here we see Verdi opening up. We've got the source code here, logical hierarchy browser. And then we've got the PPA data being shown here, just like you have the PPA data within RTLA. So I just changed this view to a summary view.

So like we can in RTLA, now you can, in Verdi, start to look-- you want to look at congestion. We can on sort on congestion down the hierarchy. So I can now click through and start to find the modules that have congestion.

So if I click down a little bit, we can see going to this crossbar here, and go all the way down to the local logic. And we can start to see case statements, et cetera, where congestion is being caused.

So here, we can see there's 578 cells that are causing - that are in this congested area. So I can simply click on this. And then in Verdi, that's going to go to the source code. You can see it's been highlighted here. This is where this case statement is, where this logical congestion is coming from.

And what I can do, if I want to see this data back in RTLA, I can simply click on here, drag across to the RTLA window, and then that will update the source code. And you see we've jumped to this case statement that we see here. This 578 for the logical congested cells is the same 578 that we have here.