up load thử, không có gì xenm

1. ASIC Back-End Design
Lecture given by Saadat Khan, BAE Systems
Slides prepared by Jamie Bernard, BAE Systems

2. Agenda
• Introduction
• Design Flow
– Overview
– Floorplan
– Timing Driven Placement
– Clock Tree Synthesis
– Routing
• Verification
• Design Example

3. IntroductionIntroduction

4. Introduction
• Technological Advances
– 19th
Century - Steel
– 20th
Century – Silicon
• Growth in Microelectronic (Silicon) Technology
– Moore’s Law (# of transistors double/18 months)
– One Transistor
– Small Scale Integration (SSI)
• Multiple Devices (Transistor / Resistor / Diodes)
• Possibility to create more than one logic gate (Inverter, etc)
– Large Scale Integration (LSI)
• Systems with at least 1000 logic gates (Several thousand transistors)
– Very Large Scale Integration
• Millions to hundreds of millions of transistors (Microprocessors)
– Intel indicates that dual core processors will soon exist that
contain 1 billion transistors

5. Introduction
• Manual (Human) design can occur with small number of
transistors
• As number of transistors increase through SSI and VLSI,
the amount of evaluation and decision making would
become overwhelming (Trade-offs)
– Maintaining performance requirements (Power / Speed / Area)
– Design and implementation times become impractical
• How does one create a complex electronic design
consisting of millions of transistors?
Automate the Process using Computer-Aided Design (CAD) ToolsAutomate the Process using Computer-Aided Design (CAD) Tools

6. Introduction
• CAD tools provide several advantages
– Ability to evaluate complex conditions in which solving one
problem creates other problems
– Use analytical methods to assess the cost of a decision
– Use synthesis methods to help provide a solution
– Allows the process of proposing and analyzing solutions to occur
at the same time
• Electronic Design Automation
– Using CAD tools to create complex electronic designs (ECAD)
– Several companies who specialize in EDA
• Cadence® Design Systems
• Magma® Design Automation Inc.
• Synopsys®
CAD Tools Allow Large Problems to be SolvedCAD Tools Allow Large Problems to be Solved

7. Design FlowDesign Flow

8. Design Flow - Overview
• Generic VLSI Design Flow from System
Specification to Fabrication and Testing
• Steps prior to Circuit/Physical design are
part of the FRONT-END flow
• Physical Level Design is part of the BACK-
END flow
– Physical Design is also known as “Place and
Route”
• CAD tools are involved in all stages of VLSI
design flow
– Different tools can be used at different
stages due to EDA common data formats*
• Synopsys® CAD tool for Physical Design is
called Astro™

9. What does Astro™ do?

10. Where does the Gate Level Netlist come from?
1st
Input to Astro™

11. Standard Cell Library
2nd
Input to Astro™
• Pre-designed collection of
logic functions
– OR, AND, XOR, etc
• Contains both Layout and
Abstract views
– Layout (CEL) contains drawn
mask layers required for
fabrication
– Abstract (FRAM) contains
only minimal data needed for
Astro™
– Timing information
• Cell Delay / Pin Capacitance
• Common height for
placement purposes

12. • Integrated circuits are built out of active and passive components, also
called devices:
– Active devices
• Transistors
• Diodes
– Passive devices
• Resistors
• Capacitors
• Devices are connected together with polysilicon or metal interconnect:
– Interconnect can add unwanted or parasitic capacitance, resistance
and inductance effects
• Device types and sizes are process or technology specific:
– The focus here is on CMOS technology
Basic Devices and Interconnect
38

13. Transistor or Device
Representation
Gates are made up of active devices or transistors.Gates are made up of active devices or transistors.
CMOS Inverter Example
OUTIN
Gate Schematic
IN OUT
PMOS
NMOS
Transistor or Device View
VDD
GND
37

14. What is “Physical Layout”?
Physical Layout – Topography of devices and interconnects, made
up of polygons that represent different layers of material.
CMOS Inverter Example
NMOS
PMOS
OUT
VDD
GND
Physical or Layout View
ININ OUT
PMOS
NMOS
Transistor or Device View
VDD
GND
39

15. Layout or Mask (aerial) view
Silicon Substrate
Process of Device Fabrication
• Devices are fabricated vertically on a silicon substrate wafer by
layering different materials in specific locations and shapes on top of
each other
• Each of many process masks defines the shapes and locations of a
specific layer of material (diffusion, polysilicon, metal, contact, etc)
• Mask shapes, derived from the layout view, are transformed to
silicon via photolithographic and chemical processes
Wafer (cross-sectional) view
40

16. Wafer Representation of Layout Polygons
Example of complimentary devices in 0.25 um CMOS technology or
process.
Input
VDD
GND
Output
PMOS
NMOS
0.25
um
Aerial or Layout View Wafer Cross-sectional View
41

17. Contacts: Connecting Metal 1 to Poly/Diff’n
Diffusion, Poly and Metal layers are separated by insulating
oxide. Connecting from Poly or Diffusion to Metal 1 requires
a contact or cut.
Cut or
Contact
(a hole in
the oxide)
VDD
IN
GND
Diffusion Diffusion
Poly
Oxide insulation Metal 1
Metal 1
49

18. What is meant by “0.xx um Technology”?
- In CMOS Technology the um or nm dimension refers to the
channel length, a minimum dimension which is fixed for most
devices in the same library.
- Current flow or drive strength of the device is proportional to
W/L; Device size or area is proportional to W x L.
Gate or Channel Dimensions (L and W)
Narrow
er
Width
=
Lower
current
throug
h
channe
l
Length
Widt
h
G
A
T
E
W
L
L
Width
(W)
Wider
Width
=
Higher
current
throug
h
channe
l
G
A
T
E
Length
42

19. Comparing Technologies
The drive strength of both devices is the same: W/L = 6.
The diffusion area (5xLxW) of A is 4x that of B.
A: 0.5 um Technology
L = 0.5 um
2L 2L
W = 3 um
L = 0.25 um
W = 1.5 um
2L 2L
B: 0.25 um Technology Area Comparison
43

20. Relative Device Drive Strengths
To double the drive strength of a device, double the channel width
(W), or connect two 1X devices in parallel. The latter approach
keeps the height at a fixed or “standard” height.
“1X” NMOS (W/L = 6)
GND
OUT
L = 0.25 um
W = 1.5 um
IN
0.25 um
GND
3 um OUT
IN
“2X” NMOS (W/L = 12)
1.5 um
GND
0.25 um
OUT
IN
“2X” NMOS (W/L = 6 + 6)
44

21. Input Output
Gate Drive Strength Example
PMOS
transistor
1x
NMOS
transistor
Input Output
Parallel PMOS
transistors
2x
inv1 inv2
Parallel NMOS
transistors
Each gate in the library is represented by multiple cells with
different drive strengths for effective speed vs. area optimization.
45

22. Drive/Buffering Rules: Max Transition/Cap
1x 2x 1x
1x
1x
Maximum Transition
Rule Violation
Maximum Transition Rule
Met
Upsized Driver or Added Buffers
AfterOptimization
BeforeOptimization
46

23. Timing Constraints
3rd
Input to Astro™
• Derived from system specifications and implementation of
design
• Identical to timing constraints used during logic synthesis
• Common constraints in electronic designs
– Clock Speed/Frequency
– Input / Output Delays associated with I/O signals
– Multicycle Paths
– False Paths
• Astro™ uses these constraints to consider timing during
each stage of the place and route process

24. Concept of Place and Route
• Location of all standard cells is automatically chosen by the
tool during placement (Based upon routing and timing)
• Pins are physically connected during routing (Based upon
timing)

25. Concepts of Placement
• Standard cells are placed in “placement rows”
• Cells in a timing-critical path are placed close together to reduce routing related delays (Timing
Driven)
• Placement rows can be abutting or non-abutting

26. Concepts of Routing
• Connecting between metal layers
requires one or more “vias”
• Metal Layers have preferred routing
directions
– Metal 1 (Blue) Horizontal
– Metal 2 (Yellow) Vertical
– Metal 3 (Red) Horizontal

27. FloorplanFloorplan

28. Design Flow – Floorplan
• Layout design done at the chip level
– Defining layout hierarchy
– Estimation of required design area
• A blueprint showing the placement of major components in the
design (non-standard cell)
– Inputs / Output (I/O)
– RAMs / ROMs/
– Reusable Intellectual Property (IP) macros
• Approaches to Floorplanning (Automatic or Manual)
– Constructive
– Iterative
– Knowledge-Based

29. Design Must Be Floorplanned Before P&R
• Floorplan of design:
– Core area defined with large macros placed
– Periphery area defined with I/O macros placed
– Power and Ground Grid (Rings and Straps) established
• Utilization:
– The percentage of the core that is used by placed standard cells and
macros
– Goal of 100%, typically 80-85%

30. I/O Placement and Chip Package
Requirements
• Some Bond Wire
requirements:
– No Crossing
– Minimum Spacing
– Maximum Angle
– Maximum Length

31. Guidelines for a Good Floorplan
• A few quick iterations of place and route with timing checks
may reveal the need for a different floorplan

32. Defining the Power/Ground Grid and
Blockages
• Purpose of Grid is to
take the VDD and
VSS received from
the I/O area and
distribute it over the
core area
• Blockages can also
be added in the
floorplan to prohibit
standards cells from
being placed in those
areas

33. Timing Driven PlacementTiming Driven Placement

34. Design Flow – Timing Driven Placement
• Astro™ optimizes, places, and
routes the logic gates to meet
all timing constraints
• Balancing design requirements
– Timing
– Area
– Power
– Signal Integrity

35. Timing Constraints
• Astro™ needs constraints to
understand the timing
intentions
– Arrival time of inputs
– Required arrival time at outputs
– Clock period
• Constraints come from the
Logic Synthesis tool
– SDC (Synopsys Design
Constraints) format

36. Cell and Net Delays
• Astro™ calculates delay for every cell and every net
• To calculate delays, Astro™ needs to know the
resistance and capacitance of each net
– Uses geometry of net and Look Up Tables to estimate the
resistances and capacitances

37. Timing Driven Placement
• Timing Driven
Placement places
critical path cells close
together to reduce net
RC
• Prior to routing, RC
are based on Virtual
Routes
• What if critical paths
do not meet timing
constraints with
placement?

38. Logic Optimizations
• These optimizations can be done during pre-place, in-place,
or post-place stages of placement
• Each optimization can be done separately or all done
concurrently during placement (none – one – all)

39. Clock Tree SynthesisClock Tree Synthesis

40. Design Flow – Clock Tree Synthesis
• All clock pins are driven by a single clock source
• Large delay and transition time due to length of net
• Clock signal reach some registers before others (Skew)

41. Clock Tree Topologies
• Clock source is connected to center of the network
• Networks are distributed in a H or X shape until clock
pin of register is driven by a local buffer
H-Tree and X-Tree Topologies Solve Single Clock Pin ProblemH-Tree and X-Tree Topologies Solve Single Clock Pin Problem

42. After Clock Tree Synthesis
• A clock (buffer) tree is built to balance the output loads and
minimize the clock skew
• A delay line can be added to the network to meet the
minimum insertion delay (clock balancing)

43. Gated - CTS
• Clocks may not be generated directly from I/O
• Power saving techniques such as clock-gating are used to turn of
the clock to sections of the design
• Astro™ can interpret gated clocks and can build clock trees
“through” the logic to the registers

44. Effects of CTS
• Several (Hundreds/Thousands)
of clock buffers added to the
design
• Placement / Routing congestion
may increase
• Non-clock cells may have been
moved to less ideal locations
• Timing violations can be
introduced

45. RoutingRouting

46. Process of Routing Can Be Timing DrivenProcess of Routing Can Be Timing Driven
Design Flow – Routing
• Routing is a fundamental step in the place and route
process
• Create metal shapes that meet the requirements of a
fabrication process
– The physical connection between cells in the design
• Virtual routes used during placement and CTS need to
become reality
– Timing of design needs to be preserved
– Timing data such as signal transitions and clock skew needs to
match the virtual route estimates

47. Timing Driven Routing
• Routing along the timing-critical path is given priority
– Creates shorter, faster connections
• Non-critical paths are routed around critical areas
– Reduces routing congestion problems for critical paths
– Does not adversely impact timing of non-critical paths

48. Concept of Routing Tracks
• Metal routes must meet minimum width and spacing “design
rules” to prevent open and short circuits during fabrication
• In grid based routing systems, these design rules determine the
minimum center-to-center distance for each metal layer
(Track/Grid spacing)
• Congestion occurs if there are more wires to be routed than
available tracks

49. Grid-Based Routing System
• Metal traces (routes) are built
along and centered around
routing tracks
• Each metal layer has its own
tracks and preferred routing
direction
– Metal 1 – Horizontal
– Metal 2 – Vertical
• Track and pitch information can
be located in the technology file
– Design Rules

50. VerificationVerification

51. What Happens After Place and Route?
Verification

52. Formal Verification
• New standard cells have been added to the design
through timing optimizations and clock tree synthesis
• The final netlist created by Astro™ needs to be compared
to the original gate-level netlist
• Formal verification ensures the functional equivalency at
the logic level between the two implementations (original
vs. final) of the design
– The intended function was maintained throughout the physical
design process
Formality® is the Sign-Off Tool for Formal VerificationFormality® is the Sign-Off Tool for Formal Verification

53. Timing Verification
• Star-RCXT™ performs the layout parasitic extraction of
the resistances and capacitances of all routes in the
design
• Results in a format such as SPEF (Standard Parasitic
Extended Format)
– SPEF is an smaller, extended format of Standard Parasitic
Format (SPF), which enables the transfer of design specific
resistances and capacitances from physical design to timing
analysis and simulation tools
• Primetime® performs static timing analysis
– Detects timing violations by combining SPEF from Star-RCXT™
and netlist from Astro™ and checks against the design timing
constraints (clock frequencies)
Star-RCXT™ and Primetime®
are the Sign-Off Tools for Timing Verification
Star-RCXT™ and Primetime®
are the Sign-Off Tools for Timing Verification

54. Physical Verification
• Checks the design for fabrication feasibility and physical
defects that could result in the design to not function
properly
– 3 checks (DRC, ERC, and LVS)
• Design Rule Checks (DRC)
– Verifies that design does not violate any fabrication rules
associated with the target process technology (metal width/space,
antenna ratio, etc)
• Electrical Rules Checks (ERC)
– Verifies that there are no short or open circuits with power and
ground as well as resistors/capacitors/transistors with floating
nodes (part of LVS)
• Layout Versus Schematic (LVS)
– Final physical design matches the logical (schematic) version in
terms of correct connectivity and number of electrical devices
Hercules™ is the Sign-Off Tool for Physical VerificationHercules™ is the Sign-Off Tool for Physical Verification

55. Fabrication
• Physical Design process is complete
upon successful completion of timing,
functional, and physical verification
• The design can be “Taped-Out” and
GDSII created for the manufacturer
– GDSII (Graphic Design System II) is a
binary format containing the physical
geometry information of the design.
– The shapes are assigned numeric
attributes in the form of “Layer Number”
and “Data Type” (Metal 1 => 100:0)
• Fabrication and Test determine
which chips can be implemented into
the system (yield)

56. Physical
design
Data
GDSII (Stream)
Masks
Wafer
Mask Generation – GDSII /
Stream
30

57. Example Design – Cory Ellinger Independent Study
• 64x8 FIFO Block.
– Inputs:
• Direct input
• Input through 64-bit addition
• Read, Write, Enable, and Sum Control
– Able to be read and written simultaneously
– Outputs:
• 64-bit FIFO out
• Overflow flag
• Full, Empty flags

58. Block Diagram
register register
unsignedAdder
register bank
clk
rst
add_fifo
sum_cnt
64 x 8
FIFO
rd
wr
en
full
empty
overfl
6464
64
64
data_in_x data_in_y_fifo_in
data_out

59. Block Diagram – Critical Path
register register
unsignedAdder
register bank
clk
rst
add_fifo
sum_cnt
64 x 8
FIFO
rd
wr
en
full
empty
overfl
6464
64
64
data_in_x data_in_y_fifo_in
data_out
Critical Path

60. Major Physical Design Steps
• Floorplan
• Placement
• Clock Tree Synthesis
• Routing

61. Floorplanning
• Aspect Ratio
• Power Planning
• Utilization
• Pin Placement
• Macro Placement
• Define Core Rows and Routing Tracks
• Read in Netlist, Libraries, and SDC.
• Groups and Regions

62. Floorplan (Theoretical)
Aspect Ratio (2:1) W:H
FIFO
Input
Register
Data Flow
64
data_in
data
out
output
flags
64
data_in
sum_cnt clk reset

63. Floorplan

64. Floorplan Showing Logic Modules

65. Placement
• Timing Driven Standard Cell Placement
• Ignore Scan Chains ( if any )
• Timing
– First look at non-wire load model timing.
– Concentrate on any large setup violations.
– Ignore violations caused by design rule failures.

66. AutoPlace of Logic Modules

67. Design Placement

68. Reset Net

69. Pre-Clock Tree Synthesis

70. Clock Tree Synthesis
• Goals:
– Low Clock Skew
– Low Clock Insertion Delay
– Sharp Transitions
• Timing
– Setup violations clean
– Design Rules fixed
– Initial evaluation of real hold violations

71. Post Clock Tree Synthesis

72. Routed Design

73. Route (zoom)

74. QUESTIONS ?QUESTIONS ?