STATIC TIMING ANALYSIS

© Mirafra Technologies 1
 Introduction
 Effective methodology for verifying the timing characteristics of a design without the use of test vectors
 Conventional verification techniques are inadequate for complex designs
 Simulation time using conventional simulators
 Thousands of test vectors are required to test all timing paths using logic simulation
 Increasing design complexity & smaller process technologies
 Increases the number of iterations for STA

© Mirafra Technologies 2
 Simulation vs. Static timing

True timing paths False timing paths

Timing Simulation Static timing analysis

(adding vectors) (eliminating false paths)

0% 100%

STA approach typically takes a fraction of the time it takes to run

logic simulation on a large design and guarantees 100% coverage
of all true timing paths in the design without having to generate test
vectors

© Mirafra Technologies 3
 OVERVIEW

Previous Verification Flow

© Mirafra Technologies 4
 OVERVIEW

• Requires extensive vector creation

• Valid for FPGAs and smaller ASICs
• Falls apart on multi-million gate ASICs

© Mirafra Technologies 5
 What is Static Timing Analysis?

Static Timing Analysis is a method for determining if a circuit meets timing

constraints without having to simulate
Much faster than timing-driven, gate-level simulation
Proper circuit functionality is not checked
Vector generation NOT required

© Mirafra Technologies 6
 STA in ASIC Design Flow – Pre layout

Logic Synthesis
Constraints
(clocks, input drive,
output load) Design For test

Floor planning
Static Timing Analysis

Static Timing Analysis

(estimated parasitics)

© Mirafra Technologies 7
 STA in ASIC Design Flow – Post Layout

Floor planning
Constraints
(clocks, input drive,
output load) Clock Tree Synthesis

Place and Route

Static Timing Analysis
(estimated parasitics)
Parasitic Extraction

Static Timing Analysis SDF

(extracted parasitics) (extracted parasitics)

© Mirafra Technologies 8
 2 Types of Timing Verification
Dynamic Timing Simulation
Advantages
Can be very accurate (spice-level)
Disadvantages
Analysis quality depends on stimulus vectors
Non-exhaustive, slow
Examples:
VCS,Spice,ACE

© Mirafra Technologies 9
 2 Types of Timing Verification
Static Timing Analysis (STA)
Advantages
Fast, exhaustive
Better analysis checks against timing requirements
Disadvantage
Less accurate
Must define timing requirements/exceptions
Difficulty handling asynchronous designs, false paths

© Mirafra Technologies 10
Three Steps in Static Timing Analysis

 Circuit is broken down into sets of timing paths

 Delay of each path is calculated
 Path delays are checked to see if timing constraints
have been met

© Mirafra Technologies 11
 What is a Timing Path?

A Timing Path is a point-to-point path in a design which

can propagate data from one flip-flop to another
Each path has a start point and an endpoint
Start point:
Input ports Clock pins of flip-flops
Endpoints:
Output ports
© Mirafra Technologies Data input pins of flip-flops 12
Organizing Timing Paths Into Groups

Timing paths are grouped into path groups by the

clocks controlling their endpoints
Synthesis tools like PrimeTime and Design Compiler organize timing
reports by path groups

© Mirafra Technologies 13
 Net and Cell Timing Arcs
The actual path delay is the sum of net and cell delays along the
timing path

© Mirafra Technologies 14
 Net and Cell Delay
“Net Delay” refers to the total time needed to charge or discharge all of the
parasitics of a given net
Total net parasitics are affected by
net length
net fanout
Net delay and parasitics are typically
Back-Annotated (Post-Layout) from data obtained from
an extraction tool
Estimated (Pre-Layout)

© Mirafra Technologies 15
 Cell Delay

In ASICs, the delay of a cell is affected by:

The input transition time (or slew rate)
The total load “seen” by the output transistors
Net capacitance and “downstream” pin capacitances
These will affect how quickly the input and output transistors can “switch”
Inherent transistor delays and “internal” net delays

© Mirafra Technologies 16
 Clocked Storage Elements
Transparent Latch, Level Sensitive
– data passes through when clock high, latched when clock low

D-Type Register or Flip-Flop, Edge-Triggered

– data captured on rising edge of clock, held for rest of cycle

© Mirafra Technologies 17
 Flip-Flops

© Mirafra Technologies 18
 Basic terminologies
 Pulse Width  Recovery & Removal times
 Setup & Hold times
 False paths
 Signal slew
 Clock latency  Multi-cycle paths
 Clock Skew
 Input arrival time
 Output required time
 Slack and Critical path

© Mirafra Technologies 19
 Pulse Width

 Pulse width
 It is the time between the active and inactive states of the same signal

© Mirafra Technologies 20
 Setup and Hold time

 Setup time
 For an edge triggered sequential element, the setup time is the time interval before the active
clock edge during which the data should remain unchanged
 Hold time
 Time interval after the active clock edge during which the data should remain unchanged

Both the above 2 timing violations can occur in a design when

clock path delay > data path delay

© Mirafra Technologies 21
 Signal Slew
 Signal (Clock/Data) slew
 Amount of time it takes for a signal transition to occur
 Accounts for uncertainty in Rise and fall times of the signal
 Slew rate is measured in volts/sec

© Mirafra Technologies 22
 Clock Latency
 Clock Latency
 Difference between the reference (source) clock slew to the clock tree endpoint signal slew
values
 Rise latency and fall latency are specified

INV
INV INV INV INV CLKA
Rise=7 Rise=7 Rise=7 Rise=7
Fall=4 Fall=4 Fall=4 Fall=4

CLK INV BUF CLKB

Rise=7 Rise=7
Fall=4 Fall=4

BUF CLKC
Rise=7
Fall=4

© Mirafra Technologies 23
 Clock Latency

© Mirafra Technologies 24
 Clock Skew

 Clock Skew is a measure of the difference in latency between any two leaf pins in a clock tree.
 between CLKA and CLKB
rise = 22-8 = 14
fall = 22-14 = 8
 between CLKB and CLKC
rise = 8-7 = 1
fall = 14-4 = 10
 between CLKA and CLKC
rise = 22-7 = 15
fall = 22-4 = 18
It is also defined as the difference in time that a single clock signal takes to reach two different
registers

© Mirafra Technologies 25
 Input Arrival time
 Input Arrival time
 An arrival time defines the time interval during which a data signal can arrive at an input pin in
relation to the nearest edge of the clock signal that triggers the data transition

© Mirafra Technologies 26
 Output required time
 Output required time
 Specifies the data required time on output ports.

© Mirafra Technologies 27
 Slack and Critical path
 Slack
 It is the difference between the required (constraint) time and the arrival time (inputs and delays).
 Negative slack indicates that constraints have not been met, while positive slack indicates that
constraints have been met.
 Slack analysis is used to identify timing critical paths in a design by the static timing analysis tool

 Critical path
 Any logical path in the design that violates the timing constraints
 Path with a negative slack

© Mirafra Technologies 28
 Slack Analysis – Data Path types

© Mirafra Technologies 29
 Slack analysis – data path types
 Primary input-to-register paths
 Delays off-chip + Combinational logic delays up to the first sequential device.
 Register-to-primary output paths
 Start at a sequential device
 CLK-to-Q transition delay + the combinational logic delay + external delay requirements
 Register-to-register paths
 Delay and timing constraint (Setup and Hold) times between sequential devices for synchronous
clocks + source and destination clock propagation times.
 Primary input-to-primary output paths
 Delays off-chip + combinational logic delays + external delay requirements.

© Mirafra Technologies 30
 Hold Slack calculation

 Actual data arrival time definition

Data Input Arrival Timemin + Data path delaymin
If the data path starts in a primary input,
Data Input arrivalmin = Input arrival timemin
If the data path starts at a register,
(Source Clock Edgemin + Source Clock Path Delaymin) = Data Input Arrivalmin

 Required Stability time definition

(Destination Clock Edgemax + Destination Clock Path Delaymax) + Hold = Required Stability
Timemax

 Hold Slack definition

Actual Data Arrivalmin - Required Stability Timemax

© Mirafra Technologies 31
 Calculate the hold slack

Source Clock signal timing parameters: Min Data path delay = 0.802 ns
Min Edge = 8.002 ns Hold time constraint = 1.046 ns
Min clock path delay = 0.002 ns

Destination Clock signal timing parameters:

Max Edge = 2.020 ns
Max clock path delay = 0.500 ns

© Mirafra Technologies 32
 Hold slack calculation

© Mirafra Technologies 33
 Setup Slack calculation

 Actual data arrival time definition

Data Input Arrival Timemax + Data path delaymax
If the data path starts in a primary input,
Data Input arrivalmax = Input arrival timemax
If the data path starts at a register,
(Source Clock Edgemax + Source Clock Path Delaymax) = Data Input Arrivalmax

 Required Stability time definition

(Destination Clock Edgemin + Destination Clock Path Delaymin) - Setup = Required Stability
Timemin

 Setup slack definition

Required Stability Timemin - Actual Data Arrivalmax

© Mirafra Technologies 34
 Calculate the setup slack

Source Clock signal timing parameters: Min Data path delay = 13.002 ns
Max Edge = 2.002 ns Setup time constraint = 0.046 ns
Max clock path delay = 0.002 ns

Destination Clock signal timing parameters:

Min Edge = 20.02 ns
Min clock path delay = 0.500 ns

© Mirafra Technologies 35
 Setup slack calculation

© Mirafra Technologies 36
 Recovery and Removal time
 Recovery time
Like setup time for asynchronous port (set, reset)
 Removal time
Like hold time for asynchronous port (set, reset)

 Recovery time
It is the time available between the asynchronous signal going inactive to the active clock edge
 Removal time
It is the time between active clock edge and asynchronous signal going inactive

© Mirafra Technologies 37
 False Paths
 False paths
 Paths that physically exist in a design but are not logic/functional paths
 These paths never get sensitized under any input conditions

Mux 1 Mux 2

A C C1 C2 OUT
B1 B2
B

© Mirafra Technologies 38
 Multi-cycle paths
 Multi-cycle paths
 Data Paths that require more than one clock period for execution

2 clock period delay

© Mirafra Technologies 39
 Sequential Circuit Timing

Objectives
This section covers several timing considerations encountered in the design of synchronous sequential
circuits. It has the following objectives:
 Define the following global timing parameters and show how they can be derived from the basic
timing parameters of flip-flops and gates.
• Maximum Clock Frequency
• Maximum allowable clock skew
• Global Setup and Hold Times
 Discuss ways to control the loading of data into registers and show why gating the clock signal to do
this is a poor design practice.

© Mirafra Technologies 40
 Maximum Clock Frequency

 The clock frequency for a synchronous sequential circuit is limited by the timing parameters of its
flip-flops and gates. This limit is called the maximum clock frequency for the circuit. The minimum
clock period is the reciprocal of this frequency.
 Relevant timing parameters
 Gates:
• Propagation delays: min tPLH, min tPHL, max tPLH, max tPHL
 Flip-Flops:
• Propagation delays: min tPLH, min tPHL, max tPLH, max tPHL
• Setup time: tsu
• Hold time: th

© Mirafra Technologies 41
  Example

D Q Q

CK Q

TW ≥ max tPFF + tsu

For the 7474, max tPLH = 25ns, max tPHL = 40ns, tsu = 20ns
TW ≥ max (max tPLH + tsu, max tPHL + tsu)
TW ≥ max (25+20, 40+20) = 60

© Mirafra Technologies 42
  Example

D Q Q

CK

TW ≥ max tPFF + max tPINV + tsu

© Mirafra Technologies 43
  Example

0
Q0 D Q
Q1
D Q MUX
1

Q Q

CK

TW ≥ max tPFF + max tPMUX + tsu

© Mirafra Technologies 44
  Example

Paths from Q1 to Q1: None

Paths from Q1 to Q2: TW ≥ max tPDFF +tJKsu = 20 +10 = 30 ns
TW ≥ max tPDFF + max tAND + tJKsu = 20 + 12 + 10 = 42 ns
Paths from Q2 to Q1: TW ≥ max tPJKFF + tOR + TDsu = 25 + 10 + 5 = 40 ns
Paths from Q2 to Q2: TW ≥ max tPJKFF + max tAND + tJKsu = 25 + 12 + 10 = 47 ns

TW ≥ 47 ns
© Mirafra Technologies 45
 Clock Skew

 If a clock edge does not arrive at different flip-flops at exactly the same time, then the clock is said
to be skewed between these flip-flops. The difference between the times of arrival at the flip-flops is
said to be the amount of clock skew.
 Clock skew is due to different delays on different paths from the clock generator to the various flip-
flops.
• Different length wires (wires have delay)
• Gates (buffers) on the paths
• Flip-Flops that clock on different edges (need to invert clock for some flip-flops)
• Gating the clock to control loading of registers (a very bad idea)

© Mirafra Technologies 46
 • Example (Effect of clock skew on clock rate)
 Clock C2 skewed after C1

D2 Q2
D Q
Q1 D Q

C2
C1 Q
Q

CK

TW ≥ max TPFF + max tOR + tsu TW ≥ max TPFF + max tOR + tsu - min tINV
(if clock not skewed, i.e., tINV = 0) (if clock skewed, i.e., tINV > 0)

© Mirafra Technologies 47
  Clock C1 skewed after C2

D2 Q2
D Q
Q1 D Q

C2
C1 Q
Q

CK

TW ≥ max TPFF + max tOR + tsu

(if clock not skewed, i.e., tINV = 0)

TW ≥ max TPFF + max tOR + tsu + max tINV

(if clock skewed, i.e., tINV > 0)

© Mirafra Technologies 48
  Summary of maximum clock frequency calculations

Q1 D2
D Q D Q
Logic
C1 Network C2

TW
TW
C1
C1 tSK = tINV
tSK = tINV C2
C2
Q1
Q1
D2
D2
tPFF tOR
tPFF tOR ts u
ts u

C2 skewed after C1: TW ≥ max TPFF + max tNET + tsu - min tINV
C2 skewed before C1: TW ≥ max TPFF + max tNET + tsu + max tINV

© Mirafra Technologies 49
 Maximum Allowable Clock Skew
 How much skew between C1 and C2 can be tolerated in the following circuit?
D Q
Q1 D2 D Q

Q Q

C1 C2
– Case 1: C2 delayed after C1

tPFF > th + tSK

tSK < min tPFF - th

© Mirafra Technologies 50
  Case 2: C1 delayed from C2

D Q
Q1 D2 D Q

Q Q

C1 C2

© Mirafra Technologies 51
 How does additional delay between the flip-flops affect
the skew calculations?

tSK ≤ min tPFF - th

tsk ≤ min tPFF + min tMUX - th

© Mirafra Technologies 52
 Summary of allowable clock skew calculations

tSK + th ≤ tPFF + tNET

tSK ≤ min tPFF + min tNET - th

© Mirafra Technologies 53
  Example: What is the minimum clock period for the following circuit under the assumption that the clock
C2 is skewed after C1 (i.e., C2 is delayed from C1)?

N2

D1 Q1 D2 Q2
D Q N1 D Q

Q Q

C1 C2

© Mirafra Technologies 54
 N2

D1 Q1 D2 Q2
D Q N1 D Q

Q Q

C1 C2

 First calculate the maximum allowable clock skew.

tSK < min tPFF + min tN1 - th
 Next calculate the minimum clock period due to the path from Q1 to D2.
TW > max tPFF + max tN1 + tsu - min tSK
 Finally calculate the minimum clock period due to the path from Q2 to D1

TW > max tPFF + max tN1 + tsu + max tSK

TW > max tPFF + max tN2 + tsu + (min tPFF + min tN1 - th)

TW > max tPFF + min tPFF + max tN2 + min tN1 + tsu - th

© Mirafra Technologies 55
Global Setup Time, Hold Time and Propagation Delay

Global setup and hold times (data delayed)

X NET D D Q

CK
CLK Q

TSU = tsu + max tNET TH = th - min tNET

© Mirafra Technologies 56
  Global setup & hold time (clock delayed)

D D Q

CK
CLK Q

TSU = tsu - min tC TH = th + max tC

© Mirafra Technologies 57
  Global setup & hold time (data & clock delayed)

X NET D D Q

CK
CLK Q

TSU = + max =-0987654321\ - min . TH = th - min tNET + max tC

© Mirafra Technologies 58
  Global propagation delay

D Q
Q
NET Y
CK
CLK Q

TP = tC + tFF + tNET

© Mirafra Technologies 59
  Summary of global timing parameters

TSU = tsu + max tPN - min tPC ≤ tsu + max tPN

TH = th + max tPC - min tPN ≤ th + max tPC

TP = tPFF + tPN + tPC

© Mirafra Technologies 60
  Example

LD Q
D Q
D CK
CLK Q

 Find TSU and TH for input signal LD relative to CLK.

TSU = tsu +max tNET - min tC

= tsu + max tINV + max tNAND + max tNAND - min tINV

TH = th - min tNET + max tC

= th - min tNAND - min tNAND + max TINV

© Mirafra Technologies 61
 Register load control (gating the clock)
• A very bad way to add a load control signal LD to a register that does not have one is shown below
D D Q

LD CK
Q
CLK
• The reason this is such a bad idea is illustrated by the following timing diagram.

• The flip-flop sees two rising edges and will trigger twice. The only one we want is the second one.

© Mirafra Technologies 62
  If LD was constrained to only change when the clock was low, then the only problem would be the
clock skew.

© Mirafra Technologies 63
  If gating the clock is the only way to control the loading of registers, then use the following approach:

D D Q

CLK Q

LD

 There is still clock skew, but at least we only have one triggering edge.

© Mirafra Technologies 64
  The best way to add a LD control signal is as follows:

LD
D Q
D
CLK Q

© Mirafra Technologies 65
 Tips & Tricks

 Use timing diagrams to determine the timing properties of sequential circuits

 Using typical timing values from the data sheet (use only max and/or min values)
 Gating the clock

© Mirafra Technologies 66
 Detecting timing violations – CASE 1

(a) Hold time for clocks is 1.5 ns

Determine if there are any timing violations in this design

Delay (min) = 5 ns
DFF 1 DFF 2

Data

clk20Mhtzref clk10Mhtz

© Mirafra Technologies 67
 Detecting timing violations – CASE 2

(a) Hold time for clocks is 1.5 ns

(b) Clock skew of 3.72 ns between clk20mref and clk10mz

Determine if there are any timing violations in this design

Delay (min) = 5 ns DFF 2

DFF 1

Data

clk20Mhtzref clk10Mhtz

© Mirafra Technologies 68
 Detecting timing violations – CASE 3

(a) Hold time for clocks is 1.5 ns

(b) Clock skew of 3.72 ns between clk20mref and clk10mz

Delay (min) = 5 ns DFF 2

DFF 1

Data

clk20Mhtzref clk10Mhtz

© Mirafra Technologies 69
 Detecting timing violations – CASE 4

Consider
(a) Clock skew of 3.72 ns between clk20mref and clk10mz
(b) Clock network delays

Delay (min) = 5 ns DFF 2

DFF 1

Data

clk20Mhtzref clk10Mhtz

Propagation delay = 4 ns Propagation delay = 2 ns

(thru clock tree buffers) (thru clock tree buffers)

© Mirafra Technologies 70
 Thank you

© Mirafra Technologies 71