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.

Multiple clocks syntheis in design compiler

Status
Not open for further replies.

jkarthikr

Newbie level 4
Joined
Jun 12, 2010
Messages
5
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Location
bangalore
Activity points
1,324
Hi All ,

We have a digital block in which we use a divider internally to generated divided clocks.
There is logic between these divided clocks . We hard instantiated the divider flops to be able to clearly specify the generated clock pins .

However when I look at the timing on paths from clk1 to clk2 (where clk1, clk2 are generated clocks), the delays of the elements along the path are reported as zero ! Any idea on why this might be happening ? below , I've attached the results of a 'report timing ' command. We see the the clock to Q delay of the flop t1_0_reg_0 is reported as zero !! I've seen that for paths from clk1 to clk1, or for for paths from clk2 to clk2 this problem does not exist .

I though it might be a Design compiler version problem, but other versions also show the same behaviour . Any pointers would be very helpful

Thanks ,
Karthik

Startpoint: t1_0_reg_0_
(rising edge-triggered flip-flop clocked by clkby1)
Endpoint: t1_d0_reg_0_
(rising edge-triggered flip-flop clocked by clkby2)
Path Group: clkby2
Path Type: max

Des/Clust/Port Wire Load Model Library
------------------------------------------------
haar enG5K fsc0h_d_sc_wc

Point Incr Path
-----------------------------------------------------------
clock clkby1 (rise edge) 80000000.00
80000000.00
clock network delay (ideal) 0.00 80000000.00
t1_0_reg_0_/CK (QDFFRBEHD) 0.00 80000000.00 r
t1_0_reg_0_/Q (QDFFRBEHD) <- 0.00 80000000.00 f
U56/O (XNR2CHD) 0.00 80000000.00 r
t1_d0_reg_0_/D (QDFFRBEHD) 0.00 80000000.00 r
data arrival time 80000000.00

clock clkby2 (rise edge) 160000000.00
160000000.00
clock network delay (ideal) 0.00 160000000.00
clock uncertainty -0.20 160000000.00
t1_d0_reg_0_/CK (QDFFRBEHD) 0.00 160000000.00 r
library setup time -0.12 160000000.00
data required time 160000000.00
-----------------------------------------------------------
data required time 160000000.00
data arrival time -80000000.00
-----------------------------------------------------------
slack (MET) 80000000.00

Added after 1 hours 48 minutes:

Hi ,

We noticed that when we tried to synthesize the same design at a higher speed, this odd behaviour disappeared ! The cell delays were no longer being reported as zero !

Regards,
Karthik
 

HI jkarthikr:
I think you can ignore the problem, because the cell delay is so small ralated to the clock period.
what u puzzled is why timing calculator ignores the cell delay, please try to change the variable value :report_default_significant_digits, the default vaule is 2
good luck!
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top