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.

RE : Timing violation where capture flop is being latched too early

Status
Not open for further replies.

S S Rayudu

Newbie level 4
Joined
Sep 24, 2014
Messages
7
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
58
Hello,

Im my design .. my launch flop's clock is 1.4ns and my capture flop clock is 10ns. The data is getting latched by capture flop in 4ps resulting in large negative slack and setup violation.

what would be the approach to solve this kind of situation.

Thanks,
S S Rayudu
 

I'm assuming the two values you provide are the clock periods of each clock.

These are asynchronous clock domains. I sure hope you correctly designed the interface between them and have syncrhonization logic between the two domains.

It is normal to add set_max_delay paths paths, or set_false_path (not my preferred method) for those cross clock domain crossing synchronizer paths.
 

Hello,

Im my design .. my launch flop's clock is 1.4ns and my capture flop clock is 10ns. The data is getting latched by capture flop in 4ps resulting in large negative slack and setup violation.

what would be the approach to solve this kind of situation.

Thanks,
S S Rayudu

if they are async, it means they have no relationship whatsoever. the tool has to consider the absolute worst scenario, and that is the 4ps case you are seeing.

refer to ads-ee answer for how to fix it.
 

Normally, clocks were defined by coverting the target frequencies.
But look like your clocks were added some margin, say, 8-10% which make the constraint faster than actual freqs.
Could you confirm ?

If no, take above comments.
 
Last edited:

Normally, clocks were defined by coverting the target frequencies.
But look like your clocks were added some margin, say, 8-10% which make the constraint faster than actual freqs.
Could you confirm ?

If no, take above comments.

No margin were added
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top