Max Transition Violations in CTS report

Status
Not open for further replies.

bardia

Newbie level 5
Joined
Sep 14, 2010
Messages
8
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Visit site
Activity points
1,378
Hi,

I use Automatic gated CTS, and it is recommended by my foundry to set maxsinktran and maxbuftran to 2ns. Also I've found that the required maxtran is 4ns by my library.

After running CTS, there are some max tran violations in the CTS report for some of my clocks. like below:

Max. Rise Sink Tran : 2335.1(ps) 2000(ps)

However I do not see any violations in the postCTS timedesign report. Is this because this is still below 4ns which is the global target for max tran?

should I take these max tran violations in the CTS report seriously?

How can I remove this violations during CTS?

Also another interesting point is that after running CTS for one of my clocks, I got the following report:

###############################################################

# Complete Clock Tree Timing Report

#

# CLOCK: PCLK

#

# Mode: preRoute

#

# Delay Corner information

# Analysis View : func_max

# Delay Corner Name : corner_max

# RC Corner Name : ams_rc_corner_wc

# Analysis View : func_min

# Delay Corner Name : corner_min

# RC Corner Name : ams_rc_corner_bc

###############################################################





Nr. of Subtrees : 5

Nr. of Sinks : 672

Nr. of Buffer : 32

Nr. of Level (including gates) : 8

Root Rise Input Tran : 0.1(ps)

Root Fall Input Tran : 0.1(ps)

Max trig. edge delay at sink(R): CoeffMEM/Reg_in_reg[9]/CP 5064.6(ps)

Min trig. edge delay at sink(R): CoeffMEM/C2_mem_reg[17][6]/CP 4775.7(ps)



(Actual) (Required)

Rise Phase Delay : 4775.7~5064.6(ps) 1000~10000(ps)

Fall Phase Delay : 4329.3~4991.3(ps) 1000~10000(ps)

Trig. Edge Skew : 288.9(ps) 200(ps)

Rise Skew : 288.9(ps)

Fall Skew : 662(ps)

Max. Rise Buffer Tran : 1223.2(ps) 2000(ps)

Max. Fall Buffer Tran : 834.3(ps) 2000(ps)

Max. Rise Sink Tran : 1094.5(ps) 2000(ps)

Max. Fall Sink Tran : 776.9(ps) 2000(ps)

Min. Rise Buffer Tran : 113(ps) 0(ps)

Min. Fall Buffer Tran : 105.1(ps) 0(ps)

Min. Rise Sink Tran : 601.4(ps) 0(ps)

Min. Fall Sink Tran : 477.9(ps) 0(ps)



view func_max : skew = 288.9ps (required = 200ps)

view func_min : skew = 197.9ps (required = 200ps)









***** Max Transition Time Violation *****



Pin Name (Actual) (Required)

-------------------------------------------------------------------

PCLK__L1_I0/A [4008 2176.2](ps) 2000(ps)

PCLK__L1_I1/A [4008 2176.2](ps) 2000(ps)







***** NO Min Transition Time Violation *****



***** NO Max Capacitance Violation *****



***** NO Max_Fanout Violation *****



***** NO AC Irms Limit Violation *****



as you see above, in the Max Rise/Fall Buffer Tran the maximum transition is met, but at the end it reports two pin names which violate the transition time. I don't how this makes sense!

So again here this violation does not exist in the timedesign report of postCTS even though it is larger than 4ns!!!

So can you help me with these issues?

Thank you so much in advance,

Bardia
 

The max transition reported by CTS could be waived if the max transition defined for CTS is below or equal to the Std cell max transition.
If you want to "relax" CTS, you could used the max transition defined by the std cell library, if you want to be more restrictive and safer, stay with the current value.

In case you want to change the CTS parameters, you could use the following command:
Code:
specifyClockTree -update { AutoCTSRootPin  <clock_name> BufMaxTran  2000ps}
 
Reactions: bardia

    bardia

    Points: 2
    Helpful Answer Positive Rating
Status
Not open for further replies.
Cookies are required to use this site. You must accept them to continue using the site. Learn more…