Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Synopsys. 2007.03-SP*

Status
Not open for further replies.

shahal

Member level 2
Joined
Jan 11, 2007
Messages
50
Helped
7
Reputation
14
Reaction score
6
Trophy points
1,288
Activity points
1,638
Guys

any one have any experience with synopsys DC 2007.03-SP*? Any one have any experience in registers being optimized out when they are not suppose to be optimized out? I have been hearing about this through the garpe vine and was wondering about the validity of this statement.

Thanks
 

No just hearing rumours, so trying to verify it. Just got off a meeting with Synopsys, they deny it.
 

I really don't think something as simple as this would have gotten by the synopsys QA team. Having said this it is normal for a synthesis tool to optimize away registers. This happens especially in pipline designs as long as design functionality remains the same. Also, it is possible for the tool to combine and re-time different stages in the design to save registers. Of course the obvious ones such as un-connected outputs and multi-driven wires.......
I personally think this is a rumor spread by Magma people.
 

Yeah I think its rumor also. I checked with some other sources, no one has heard of it. Sorry if I alarmed anyone..
 

Dear rakko,
If it was so sure that synopsys QA would not let it go, then why synopsys will recommend 'formal verification' of the netlist produced by design compiler? The whole idea of formality is to catch bugs in DC.
kr,
Avi
http://www.vlsiip.com
 

I think the purpose of FV is to check the netlist aginst yout RTL code. It used to be that designs were simple and dynamic verification was used to catch post synthesis bugs. These days netlists are too complex and running all your verification suites on the netlist is too slow and just not practical anymore. So, one way to make sure the netlist is just as good as the RTL is to verify the RTL using all your verification tests and use FV to make sure netlist is identical to the RTL. So, I think the purpose is to gaurd against mistakes introduced by the designers such as accidentally deleting a few lines of the netlist while fixing timing violations by hand.

Don't get me wrong, I'm not saying synopsys is perfect but don't think they miss the very simple stuff.
 

yes you are true. I seen some post on deepchip on 07.03 DC has lot of bugs in seq. opto.
 

Of course the obvious ones such as un-connected outputs and multi-driven wires.......
I personally think this is a rumor spread by Magma people.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top