[vc_row][vc_column][vc_column_text]

STA – Part1

Author : Nishant Lamani, Physical Design Engineer, SignOff Semiconductors,

Author : Abhishek Kumar, Physical Design Engineer, SignOff Semiconductors[/vc_column_text][vc_column_text]

What is STA ?

[/vc_column_text][vc_column_text]

Why we do STA ?

[/vc_column_text][vc_column_text]

Dynamic Timing Analysis (DTA) :

Example : If we have 100 inputs, then we need 2 to the power 100 simulation for complete the analysis. The amount of analysis is astronomical compared to static analysis.

[/vc_column_text][vc_column_text]

Difference between DTA and STA ?

[/vc_column_text][vc_column_text]

Input and Output Files of STA :

Figure 1 : Input & Output file for STA

[/vc_column_text][vc_column_text]

Basics of Setup and Hold Time :

To understand why setup and hold time requirement arises in a flip-flop, we have to look at the building blocks of a flip-flop which includes inverters and transmission gates. A transmission gate is a parallel connection of NMOS and PMOS with complementary control inputs to both. Whenever both NMOS and PMOS are turned on, any signal 1 or 0 passes equally well without degradation. Two back-to-back inverters form a latching circuit as it retains a logic value.

Figure 2(i) : Positive Edge Triggered D Flip-Flop circuit

[/vc_column_text][vc_column_text]

D flip-flop : Normal operation

Initially when D=0 and CLK is low, then TG1 & TG4 turn on and TG2 & TG3 are turned off. The input follows the path D-P1-P2-P3 and finally data at P3 is 0.

Figure 2(ii) : Operation of D Flip-Flop when CLK = 0

When CLK is high, then TG2 & TG3 turn on and TG1 & TG4 are turned off. The data 0 comes out of TG2 and follows the path TG2-P1-P2-TG3-P4-P5-Q . Finally we get Q = 0. The output arrives at positive edge of CLK so it is a positive edge triggered flip-flop.

Figure 2(iii) : Operation of D Flip-Flop when CLK = 1

When CLK is low there is no change in output. Any change in input appears at P3 but it is reflected at the output only when CLK turns high again.

Setup Time :It is the minimum amount of time before the active edge of clock for which the data must be stable for it to be latched correctly.

Figure 2(iv) : Setup requirement of a flip-flop

Setup requirement of a flip-flop :When D = 0 and CLK is low, input follows the path D-P1-P2-P3 and reches the input of TG2. The time taken by data D to reach the input of TG2 is called the setup time of a flop. If data changes in this time it will not be able to reach input of TG2 and when TG2 turns on, there might be two different data reaching node P1, which will cause meta-stability.

Hold Time :It is the minimum amount of time after the active edge of clock during which data must be stable for data to be latched correctly.

Figure 2(v) : Hold requirement of a flip-flop

Hold requirement of a flip-flop :It arises because of the finite delay in ramping up of the CLK and CLKb signals which controls the switching of transmission gates. TG takes some time to switch on or off, known as hold time of a flop. So it is necessary to maintain a stable value at D pin to ensure a stable value at node P1 which is translated to output.

[/vc_column_text][vc_column_text]

Basic Terminology in STA :

Setup Time :The minimum time before the active edge of the clock, the input data must remain stable is called the setup time.

Hold Time : The minimum time after the active edge of the clock, the input data must remain stable is called the hold time.

Figure 3 : Setup & Hold time

[/vc_column_text][vc_column_text]

Launch and Capture edge :

[/vc_column_text][vc_column_text]

Launch and Capture flip flop :

[/vc_column_text][vc_column_text]

Setup and Hold timing checks between launch and capture flip flop :

Setup timing checks :

Figure 4(i) : Setup check between two flops

[Tlaunch + TCK2Q + Tpd] < [Tcapture + TCLK – TSetup]

[/vc_column_text][vc_column_text]

Hold timing checks :


Figure 4(ii) : Hold check between two flops

[Tlaunch + TCK2Q + Tpd] > [Tcapture + Thold]

[/vc_column_text][vc_column_text]

Slack :

Slack is the difference between the required time and the time a signal arrives.

Setup Slack = Data Required Time – Data Arrival Time

Hold Slack = Data Arrival Time – Data Required Time

[/vc_column_text][vc_column_text css=””]

Slew :

Slew is the transition time of a signal, the time it takes for a signal to change between two specific levels.

Figure 5 : Slew / Transition Time

Propagation delay :Propagation delay is the time required for the input to be propagated to the output. In other words, the propagation delay of a logic gate is defined as the time it takes for the effect of change in input to be visible at the output.

Skew : Skew is the difference in the arrival of the clock at two consecutive clock pins of sequential elements.

Figure 6 : Positive & Negative Skew

Global Skew :It is defined as the difference between the maximum insertion delay and the minimum insertion delay between two flip flops.

Useful Skew :If the clock is skewed intentionally to resolve the violations, it is called as useful skew.

Clock Latency :Clock latency is the time it takes from clock source to end point.

Clock Latency = [ ( Source Latency ) + ( Network Latency ) ]

Figure 7 : Source & Network Latency

Clock Uncertainty :Clock uncertainty is the deviation of the actual arrival time of the clock edge with respect to ideal arrival time. The deviation happens mainly due to jitter and noise.

Jitter : Clock jitter refers to the temporal variation of the clock period at a given point — that is, the clock period can reduce or expand on a cycle-by-cycle basis.

Figure 8 : Jitter

Common sources of Jitter are,

False Path : A false path is a path existing in a design which should not be analysed for timing. For example,

Figure 9(i) : Falsepath between two MUX

Figure 9(ii) : Falsepath between two flops

Declaring a path to be false removes all timing constraints from the path and the advantage is that the analysis time and effort are reduced, thereby allowing the STA tool to focus only on the real paths.

Multicycle Path :A path that is designed to take more than one clock cycle for the data from launch flop to propagate to the capture flop. The combinational data path between two flip-flops can take more than one clock cycle to propagate the data. In such cases, the combinational path is declared as a multicycle path. For example,

The path from FF4 to FF5 is designed to take two clock cycles rather than one. However, by default, PrimeTime assumes single-cycle timing for all paths. Therefore, we need to specify a timing exception for this path.

Figure 10 : Single & Multicycle Path

Halfcycle Path : A path which requires only half cycle to capture the data. It is formed when data is launched on positive edge of the clock and captured on negative edge of the clock or when data gets launched on negative edge of the clock and gets captured on positive edge of the clock.

In such paths, setup check become more tight as setup gets only half cycle while hold constraint is relaxed by half cycle.

Figure 11: Half-cycle Path

The falling edge occurs at 5ns and the rising edge occurs at 10ns. Thus, the data gets only a

half-cycle, which is 5ns, to propagate to the capture flip-flop. While the data path gets only half-cycle for setup check, an extra half-cycle is available for the hold timing check.

The hold check always occurs one cycle prior to the capture edge. Since the capture edge occurs at 10ns, the previous capture edge is at 0ns, and hence the hold gets checked at 0ns. This effectively adds a half-cycle margin for hold checking and thus results in a large positive slack on hold.

Cross-talk Noise :It is undesired change in the output values of victim due to switching in the input of aggressor. If one net is switching and other is at a constant value , the switching net may cause voltage spikes on other net. This is called as cross talk noise. Cross talk noise is evolving as a key source in degrading performance and reliability of high speed integrated circuits.

Cross-talk Delay :When there is some delay in output transition of victim due to input transition of aggressor, it is called as cross talk delay. It occurs when some transition is happening in both the nets. Cross talk delay depends on the switching direction of the aggressor and victim nets too. If input transitions occur in same direction then output transition of victim becomes faster and if input transitions occur in opposite directions then output transition of victim becomes slower and delay is more which may violate setup time.

Read our blog on wire modelling and cross-talk for more information.

[/vc_column_text][vc_column_text]

Virtual Clock :

[/vc_column_text][vc_column_text css=””]

Minimum Pulse Width Violation :

Figure : Pulse Width Violation

Read our blog on pulse width reduction to know how and why pulse width violation occurs.[/vc_column_text][vc_column_text]

What is the difference between clock buffer and normal buffer ?

[/vc_column_text][/vc_column][/vc_row]