EE194-EE290C28nmSoCforIoT

Ref:Communica:onSystemsbyA.BruceCarlson,PaulB.CrillyandJanetC.RutledgeCMOSVLSIDesignbyNeilH.WesteandDavidMoneyHarrisTimingLibraryFormatReference,CadenceDesignSystemhSps://www.csee.umbc.edu/~cpatel2/links/641/slides/lect05_LIB.pdfSynopsysDesignCompilerUserGuide

Synchroniza:onBitSynchroniza:onCarrierFrequencyandPhaseFrameSynchroniza:on

Synchroniza:onPNsequencegenerator.

m1 m2 m3 m4 m51 1 1 1 1

+

Output sequence

111010...

Synchroniza:onAutocorrela:onofaPNsequence.

τNTc

Tc

Rs(τ)1

-Tc(N-1)Tc (N+1)Tc

1/N

Synchroniza:onFrameSynchroniza:on

Preamble

Start ofmessage

Message bitst

Synchroniza:onFrameSynchroniza:on/PreambleDetec:on

aK aK-1 aK-2 aK-N...c1 c2 cn...

Σ VK

vk = cii=1

N

∑ ak−i

MatlabExample

openExample('comm/MSKSignalRecoveryExample')MSKSignalRecoveryExample

TransmissionGateMuxq  Nonrestoring mux uses two transmission gates

-  Only 4 transistors

S

S

D0

D1YS

DLatchq When CLK = 1, latch is transparent

-  D flows through to Q like a buffer q When CLK = 0, the latch is opaque

-  Q holds its old value independent of D q  a.k.a. transparent latch or level-sensitive latch

CLK

D Q

Latch D

CLK

Q

DLatchDesignq Multiplexer chooses D or old Q

1

0

D

CLK

QCLK

CLKCLK

CLK

DQ Q

Q

DLatchOpera:on

CLK = 1

D Q

Q

CLK = 0

D Q

Q

D

CLK

Q

DFlip-flopFlop

CLK

D Q

D

CLK

Q

q When CLK rises, D is copied to Q q  At all other times, Q holds its value q  a.k.a. positive edge-triggered flip-flop, master-slave flip-flop

DFlip-flopDesignq  Built from master and slave D latches

QMCLK

CLKCLK

CLK

Q

CLK

CLK

CLK

CLK

D

Latch

Latch

D QQM

CLK

CLK

DFlip-flopOpera:on

CLK = 1

D

CLK = 0

Q

D

QM

QMQ

D

CLK

Q

RaceCondi:on

CLK1

D Q1

Flop

Flop

CLK2

Q2

CLK1

CLK2

Q1

Q2

q  Back-to-back flops can malfunction from clock skew -  Second flip-flop fires late -  Sees first flip-flop change and captures its result -  Called hold-time failure or race condition

SetupandHoldTime

Time

Volta

ge

Clock Setup Time

Hold Time

Non-overlappingClocksq  Non-overlapping clocks can prevent races

-  As long as non-overlap exceeds clock skew q We will use them in this class for safe design

-  Industry manages skew more carefully instead

φ1

φ1φ1

φ1

φ2

φ2φ2

φ2

φ2

φ1

QMQD

ScanInCell

D Q D Q

phi

phib

scan_i0o1

load

phi_out

phib_out

scan_i0o1_out

load_out

scan_in

chip_in_int

scan_i0o1

phi phib

01 scan_out

scan_out_int

D Qload chip_in

scan_out_int chip_in_int

ScanOutCell

D Q D Q

phi

phib

scan_i0o1

load

phi_out

phib_out

scan_i0o1_out

load_out

scan_in

chip_out

scan_i0o1

phi phib

01 scan_out

ScanIn-OutCell

D Q D Q

phi

phib

scan_i0o1

load

phi_out

phib_out

scan_i0o1_out

load_out

scan_in

chip_out

scan_i0o1

phi phib

01 scan_out

scan_out_int

D Qload chip_in

scan_out_int chip_in_int

GateLayoutq  Layout can be very time consuming

-  Design gates to fit together nicely -  Build a library of standard cells

q  Standard cell design methodology -  VDD and GND should abut (standard height) -  Adjacent gates should satisfy design rules -  nMOS at bottom and pMOS at top -  All gates include well and substrate contacts

StandardCellsq  Uniform cell height q  Uniform well height q M1 VDD and GND rails q M2 Access to I/Os q Well / substrate taps q  Exploits regularity

CopingWithComplexityq  How to design System-on-Chip?

-  Many millions (even billions!) of transistors -  Tens to hundreds of engineers

q  Structured Design q  Design Partitioning

StructuredDesignq  Hierarchy: Divide and Conquer

-  Recursively system into modules q  Regularity

-  Reuse modules wherever possible -  Ex: Standard cell library

q Modularity: well-formed interfaces -  Allows modules to be treated as black boxes

q  Locality -  Physical and temporal

DesignPar::oningq  Architecture: User’s perspective, what does it do?

-  Instruction set, registers -  MIPS, x86, Alpha, PIC, ARM, …

q Microarchitecture -  Single cycle, multi-cycle, pipelined, superscalar?

q  Logic: how are functional blocks constructed -  Ripple carry, carry look-ahead, carry select adders

q  Circuit: how are transistors used -  Complementary CMOS, pass transistors, domino

q  Physical: chip layout -  Datapaths, memories, random logic

HDLq  Hardware Description Languages

-  Widely used in logic design -  Verilog and VHDL

q  Describe hardware using code -  Document logic functions -  Simulate logic before building -  Synthesize code into gates and layout

•  Requires a library of standard cells

VerilogExamplemodule fulladder(input a, b, c, output s, cout); sum s1(a, b, c, s); carry c1(a, b, c, cout); endmodule module carry(input a, b, c, output cout) assign cout = (a&b) | (a&c) | (b&c); endmodule

a b

c

s

cout carrysum

s

a b c

cout

fulladder

CircuitDesignq  How should logic be implemented?

-  NANDs and NORs vs. ANDs and ORs? -  Fan-in and fan-out? -  How wide should transistors be?

q  These choices affect speed, area, power q  Logic synthesis makes these choices for you

-  Good enough for many applications -  Hand-crafted circuits are still better

Example:CarryLogic

ab

ac

bc

cout

x

y

z

g1

g2

g3

g4

a b

c

c

a b

b

a

a

b

coutcn

n1 n2

n3

n4

n5 n6

p6p5

p4

p3

p2p1

i1

i3

i2

i4

q  assign cout = (a&b) | (a&c) | (b&c);

Transistors? Gate Delays?

GateLevelNetlist

ab

ac

bc

cout

x

y

z

g1

g2

g3

g4

module carry(input a, b, c, output cout) wire x, y, z; and g1(x, a, b); and g2(y, a, c); and g3(z, b, c); or g4(cout, x, y, z);endmodule

TransistorLevelNetlist

a b

c

c

a b

b

a

a

b

coutcn

n1 n2

n3

n4

n5 n6

p6p5

p4

p3

p2p1

i1

i3

i2

i4

module carry(input a, b, c, output cout) wire i1, i2, i3, i4, cn; tranif1 n1(i1, 0, a); tranif1 n2(i1, 0, b); tranif1 n3(cn, i1, c); tranif1 n4(i2, 0, b); tranif1 n5(cn, i2, a); tranif0 p1(i3, 1, a); tranif0 p2(i3, 1, b); tranif0 p3(cn, i3, c); tranif0 p4(i4, 1, b); tranif0 p5(cn, i4, a); tranif1 n6(cout, 0, cn); tranif0 p6(cout, 1, cn);endmodule

SPICENetlist.SUBCKT CARRY A B C COUT VDD GND MN1 I1 A GND GND NMOS W=1U L=0.18U AD=0.3P AS=0.5P MN2 I1 B GND GND NMOS W=1U L=0.18U AD=0.3P AS=0.5P MN3 CN C I1 GND NMOS W=1U L=0.18U AD=0.5P AS=0.5P MN4 I2 B GND GND NMOS W=1U L=0.18U AD=0.15P AS=0.5P MN5 CN A I2 GND NMOS W=1U L=0.18U AD=0.5P AS=0.15P MP1 I3 A VDD VDD PMOS W=2U L=0.18U AD=0.6P AS=1 P MP2 I3 B VDD VDD PMOS W=2U L=0.18U AD=0.6P AS=1P MP3 CN C I3 VDD PMOS W=2U L=0.18U AD=1P AS=1P MP4 I4 B VDD VDD PMOS W=2U L=0.18U AD=0.3P AS=1P MP5 CN A I4 VDD PMOS W=2U L=0.18U AD=1P AS=0.3P MN6 COUT CN GND GND NMOS W=2U L=0.18U AD=1P AS=1P MP6 COUT CN VDD VDD PMOS W=4U L=0.18U AD=2P AS=2P CI1 I1 GND 2FF CI3 I3 GND 3FF CA A GND 4FF CB B GND 4FF CC C GND 2FF CCN CN GND 4FF CCOUT COUT GND 2FF .ENDS

PhysicalDesignq  Floorplan q  Standard cells

-  Place & route q  Datapaths

-  Slice planning q  Area estimation

MIPSFloorplan

MIPSLayout

SynthesizedControllerq  Synthesize HDL into gate-level netlist q  Place & Route using standard cell library

DesignVerifica:onq  Fabrication is slow & expensive q  Debugging chips is very hard

-  Limited visibility into operation q  Prove design is right before building!

-  Logic simulation -  Ckt. simulation / formal verification -  Layout vs. schematic comparison -  Design & electrical rule checks

q  Verification is > 50% of effort on most chips!

Specification

ArchitectureDesign

LogicDesign

CircuitDesign

PhysicalDesign

=

=

=

=

Function

Function

Function

FunctionTimingPower

VLSIFlowq  Design Compiler (DC) Synopsys tool

DesignCompilerop:mizesdesignstoprovidethesmallestandfastestlogicalrepresenta:onofagivenfunc:on.ItcomprisestoolsthatsynthesizeyourHDLdescrip:onsintoop:mized,technology-dependent,gate-leveldesigns.

DC

Convertsadesigndescrip:onwriSeninaHDL,intoanop:mizedgate-levelnetlistmappedtoaspecificlogiclibrary.Whenthesynthesizeddesignmeetsfunc:onality,:ming,power,andotherdesigngoals,youcanpassthedesigntoICCompilerforphysicalimplementa:on.

LibertyTimingFile(LIB)The.libfileisanASCIIrepresenta:onofthe:mingandpowerparametersassociatedwithanycellinapar:cularsemiconductortechnology.The:mingandpowerparametersareobtainedbysimula:ngthecellsunderavarietyofcondi:onsandthedataisrepresentedinthe.libformat.The.libfilecontains:mingmodelsanddatatocalculate:

•  �I/Odelaypaths•  �Timingcheckvalues•  �Interconnectdelays

I/Opathdelaysand:mingcheckvaluesarecomputedonaper-instancebasis.Pathdelaysinacircuitdependupontheelectricalbehaviorofinterconnectsbetweencells.Thisparasi:cinforma:oncanbebasedonthelayoutofthedesign,butmustbees:matedwhennolayoutinforma:onisavailable.Alsoitisnotpossibletopredicttheprocess,voltageandtemperaturevaria:onsanddera:ngfactorscanbeincludedtocompensateforthesevaria:ons.

LibertyTimingFile(LIB)Cell-baseddelaycalcula:onismodeledbycharacterizingcelldelayandoutputtransi:on:me(outputslew)asafunc:onofinputtransi:on:me(inputslew)andthecapaci:veloadontheoutputofthecell.Timingchecksarealsofunc:onsofinputslewandoutputcapaci:veload.Eachcellhasaspecificnumberofinput-to-outputpaths•  Pathdelayscanbedescribedforeachinputsignaltransi:onthataffectsanoutputsignal•  Thepathdelaycanalsodependonsignalsatotherinputs(statedependencies)•  Inmanysequen:alcells,thepathdelayfromaninputpintoanoutputpincandependonthepathdelayfromanotheroutputpintothisoutputpin

A

B

C

Z

LibertyTimingFile(LIB)Cell-baseddelaycalcula:onismodeledbycharacterizingcelldelayandoutputtransi:on:me(outputslew)asafunc:onofinputtransi:on:me(inputslew)andthecapaci:veloadontheoutputofthecell.Timingchecksarealsofunc:onsofinputslewandoutputcapaci:veload.Eachcellhasaspecificnumberofinput-to-outputpaths•  Pathdelayscanbedescribedforeachinputsignaltransi:onthataffectsanoutputsignal•  Thepathdelaycanalsodependonsignalsatotherinputs(statedependencies)•  Inmanysequen:alcells,thepathdelayfromaninputpintoanoutputpincandependonthepathdelayfromanotheroutputpintothisoutputpin

A

B

C

Z

LibertyTimingFile(LIB)

Input Slew

Delay,Power, Timing Checks

OutputCapacitance

Lookup-table(non-lineardelay)model.

LibertyTimingFile(LIB)

Input Slew

Delay,Power, Timing Checks

OutputCapacitance

Lookup-table(non-lineardelay)model.

Parameter Varia:onq  Transistors have uncertainty in parameters

Process: Leff, Vt, tox of nMOS and pMOS Vary around typical (T) values

q  Fast (F) Leff: short Vt: low tox: thin

q  Slow (S): opposite q  Not all parameters are independent

for nMOS and pMOS

nMOS

pMOS

fastslow

slow

fast

TT

FF

SSFS

SF

EnvironmentalVaria:onq  VDD and T also vary in time and space q  Fast:

VDD: high T: low

Corner Voltage Temperature F 1.98 0 C T 1.8 70 C S 1.62 125 C

ProcessCornersq  Process corners describe worst case variations

-  If a design works in all corners, it will probably work for any variation. q  Describe corner with four letters (T, F, S)

-  nMOS speed -  pMOS speed -  Voltage -  Temperature

ImportantCorners

Purpose nMOS pMOS VDD Temp

Cycle time S S S S

Power F F F F

Subthreshold leakage

F F F S

q  Some critical simulation corners include

DesignObjects

TopLevelPar::oning

DesignEnvironment

Beforeadesigncanbeop:mized,youmustdefinetheenvironmentinwhichthedesignisexpectedtooperate.Youdefinetheenvironmentbyspecifyingopera:ngcondi:ons,systeminterfacecharacteris:cs,andwireloadmodels.Opera:ngcondi:onsincludetemperature,voltage,andprocessvaria:ons.Systeminterfacecharacteris:csincludeinputdrivers,inputandoutputloads,andfanoutloads.Theenvironmentmodeldirectlyaffectsdesignsynthesisresults.

DriveCharacteris:cs

Todeterminethedelayandtransi:on:mecharacteris:csofincomingsignals,DesignCompilerneedsinforma:onabouttheexternaldrivestrengthandtheloadingateachinputport.Drivestrengthisthereciprocaloftheoutputdriveresistance,andthetransi:ondelayataninputportistheproductofthedriveresistanceandthecapacitanceloadoftheinputport.DesignCompilerusesdrivestrengthinforma:ontobuffernetsappropriatelyinthecaseofaweakdriver.Bydefault,DesignCompilerassumeszerodriveresistanceoninputports,meaninginfinitedrivestrength.

DriveCharacteris:cs

Bydefault,DesignCompilerassumeszerocapaci:veloadoninputandoutputports.

WireLoadModels

Wireloadmodelses:matetheeffectofwirelengthandfanoutontheresistance,capacitance,andareaofnets.DesignCompilerusesthesephysicalvaluestocalculatewiredelaysandcircuitspeeds.Semiconductorvendorsdevelopwireloadmodels,basedonsta:s:calinforma:onspecifictothevendors’process.Themodelsincludecoefficientsforarea,capacitance,andresistanceperunitlength,andafanout-to-lengthtablefores:ma:ngnetlengths(thenumberoffanoutsdeterminesanominallength).

DesignRuleConstraints

DesignRuleConstraints

Fabrica:on&Packagingq  Tapeout final layout q  Fabrication

-  6, 8, 12” wafers -  Optimized for throughput, -  not latency (10 weeks!) -  Cut into individual dice

q  Packaging -  Bond gold wires from die I/O pads to

package

Tes:ngq  Test that chip operates

-  Design errors -  Manufacturing errors

q  A single dust particle or wafer defect kills a die -  Yields from 90% to < 10% -  Depends on die size, maturity of process -  Test each part before shipping to customer

MIPSR3000Processorq  32-bit 2nd generation commercial processor (1988) q  Led by John Hennessy (Stanford, MIPS Founder) q  32-64 KB Caches q  1.2 µm process q  111K Transistors q  Up to 12-40 MHz q  66 mm2 die q  145 I/O Pins q  VDD = 5 V q  4 Watts q  SGI Workstations

http://gecko54000.free.fr/?documentations=1988_MIPS_R3000

TransistorasSwitches

gs

d

g = 0

s

d

g = 1

s

d

gs

d

s

d

s

d

nMOS

pMOS

OFF ON

ON OFF

q We can view MOS transistors as electrically controlled switches q  Voltage at gate controls path from source to drain

CMOSInverter

VDD

A Y

GND

A Y

0 1

1 0

A Y

CMOSNANDGate

A B Y

0 0 1

0 1 1

1 0 1

1 1 0 A

B

Y

CMOSNORGate

A B Y

0 0 1

0 1 0

1 0 0

1 1 0

A

BY

ComplementaryCMOSq  Complementary CMOS logic gates

-  nMOS pull-down network -  pMOS pull-up network -  a.k.a. static CMOS pMOS

pull-upnetwork

outputinputs

nMOSpull-downnetworkPull-up OFF Pull-up ON

Pull-down OFF Z (float) 1

Pull-down ON 0 X (crowbar)

SeriesandParallelq  nMOS: 1 = ON q  pMOS: 0 = ON q  Series: both must be ON q  Parallel: either can be ON (a)

a

b

a

b

g1

g2

0

0

a

b

0

1

a

b

1

0

a

b

1

1

OFF OFF OFF ON

(b)

a

b

a

b

g1

g2

0

0

a

b

0

1

a

b

1

0

a

b

1

1

ON OFF OFF OFF

(c)

a

b

a

b

g1 g2 0 0

OFF ON ON ON

(d) ON ON ON OFF

a

b

0

a

b

1

a

b

11 0 1

a

b

0 0

a

b

0

a

b

1

a

b

11 0 1

a

b

g1 g2

Conduc:onComplementq  Complementary CMOS gates always produce 0 or 1 q  Ex: NAND gate

-  Series nMOS: Y=0 when both inputs are 1 -  Thus Y=1 when either input is 0 -  Requires parallel pMOS

q  Rule of Conduction Complements -  Pull-up network is complement of pull-down -  Parallel -> series, series -> parallel

A

B

Y

SignalStrengthq  Strength of signal

-  How close it approximates ideal voltage source q  VDD and GND rails are strongest 1 and 0 q  nMOS pass strong 0

-  But degraded or weak 1 q  pMOS pass strong 1

-  But degraded or weak 0 q  Thus nMOS are best for pull-down network

PassTransistorsq  Transistors can be used as switches

g = 0s d

g = 1s d

0 strong 0Input Output

1 degraded 1

g = 0s d

g = 1s d

0 degraded 0Input Output

strong 1

g = 1

g = 1

g = 0

g = 01

g

s d

g

s d

TransmissionGatesq  Pass transistors produce degraded outputs q  Transmission gates pass both 0 and 1 well

g = 0, gb = 1a b

g = 1, gb = 0a b

0 strong 0

Input Output

1 strong 1

g

gb

a b

a bg

gb

a bg

gb

a bg

gb

g = 1, gb = 0

g = 1, gb = 0

Tristatesq  Tristate buffer produces Z when not enabled

EN A Y 0 0 Z 0 1 Z 1 0 0 1 1 1

A Y

EN

A Y

EN

EN

Non-restoringTristateq  Transmission gate acts as tristate buffer

-  Only two transistors -  But non-restoring

•  Noise on A is passed on to Y

A Y

EN

EN

TristateInverterq  Tristate inverter produces restored output

-  Violates conduction complement rule -  Because we want a Z output

A

YEN

A

Y

EN = 0Y = 'Z'

Y

EN = 1Y = A

A

EN

Mul:plexersq  2:1 multiplexer chooses between two inputs

S D1 D0 Y

0 X 0 0

0 X 1 1

1 0 X 0

1 1 X 1

0

1

S

D0

D1Y

Gate-LevelMuxDesignq  How many transistors are needed? 20

1 0 (too many transistors)Y SD SD= +