ADSP21020 AD [Analog Devices], ADSP21020 Datasheet

no-image

ADSP21020

Manufacturer Part Number
ADSP21020
Description
32/40-Bit IEEE Floating-Point DSP Microprocessor
Manufacturer
AD [Analog Devices]
Datasheet

Available stocks

Company
Part Number
Manufacturer
Quantity
Price
Part Number:
ADSP21020BG-100
Manufacturer:
ADI
Quantity:
163
Part Number:
ADSP21020BG-120
Manufacturer:
AD
Quantity:
15
Part Number:
ADSP21020BG-120
Manufacturer:
ADI
Quantity:
289
Part Number:
ADSP21020BG-80
Manufacturer:
AD
Quantity:
1
Part Number:
ADSP21020BG-80
Manufacturer:
ADI
Quantity:
202
Part Number:
ADSP21020KG-100
Manufacturer:
ADI
Quantity:
200
Part Number:
ADSP21020KG-133
Manufacturer:
ADI
Quantity:
168
Part Number:
ADSP21020KG-80
Manufacturer:
a
Quantity:
8
Part Number:
ADSP21020KG133
Manufacturer:
MAXIM
Quantity:
3 632
a
GENERAL DESCRIPTION
The ADSP-21020 is the first member of Analog Devices’ family
of single-chip IEEE floating-point processors optimized for
digital signal processing applications. Its architecture is similar
to that of Analog Devices’ ADSP-2100 family of fixed-point
DSP processors.
Fabricated in a high-speed, low-power CMOS process, the
ADSP-21020 has a 30 ns instruction cycle time. With a high-
performance on-chip instruction cache, the ADSP-21020 can
execute every instruction in a single cycle.
The ADSP-21020 features:
REV. C
Information furnished by Analog Devices is believed to be accurate and
reliable. However, no responsibility is assumed by Analog Devices for its
use, nor for any infringements of patents or other rights of third parties
which may result from its use. No license is granted by implication or
otherwise under any patent or patent rights of Analog Devices.
Independent Parallel Computation Units
The arithmetic/logic unit (ALU), multiplier and shifter
perform single-cycle instructions. The units are architecturally
arranged in parallel, maximizing computational throughput. A
single multifunction instruction executes parallel ALU and
FEATURES
Superscalar IEEE Floating-Point Processor
Off-Chip Harvard Architecture Maximizes Signal
30 ns, 33.3 MIPS Instruction Rate, Single-Cycle
100 MFLOPS Peak, 66 MFLOPS Sustained Performance
1024-Point Complex FFT Benchmark: 0.58 ms
Divide (y/x): 180 ns
Inverse Square Root (1/ x): 270 ns
32-Bit Single-Precision and 40-Bit Extended-Precision
32-Bit Fixed-Point Formats, Integer and Fractional,
IEEE Exception Handling with Interrupt on Exception
Three Independent Computation Units: Multiplier,
Dual Data Address Generators with Indirect, Immedi-
Two Off-Chip Memory Transfers in Parallel with
Multiply with Add & Subtract for FFT Butterfly
Efficient Program Sequencing with Zero-Overhead
Single-Cycle Register File Context Switch
15 (or 25) ns External RAM Access Time for Zero-Wait-
IEEE JTAG Standard 1149.1 Test Access Port and
223-Pin PGA Package (Ceramic)
Processing Performance
Execution
IEEE Floating-Point Data Formats
with 80-Bit Accumulators
ALU, and Barrel Shifter
ate, Modulo, and Bit Reverse Addressing Modes
Instruction Fetch and Single-Cycle Multiply & ALU
Operations
Computation
Looping: Single-Cycle Loop Setup
State, 30 (or 40) ns Instruction Execution
On-Chip Emulation Circuitry
One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A.
Tel: 617/329-4700
Data Register File
Single-Cycle Fetch of Instruction and Two Operands
Memory Interface
multiplier operations. These computation units support IEEE
32-bit single-precision floating-point, extended precision
40-bit floating-point, and 32-bit fixed-point data formats.
A general-purpose data register file is used for transferring
data between the computation units and the data buses, and
for storing intermediate results. This 10-port (16-register)
register file, combined with the ADSP-21020’s Harvard
architecture, allows unconstrained data flow between
computation units and off-chip memory.
The ADSP-21020 uses a modified Harvard architecture in
which data memory stores data and program memory stores
both instructions and data. Because of its separate program
and data memory buses and on-chip instruction cache, the
processor can simultaneously fetch an operand from data
memory, an operand from program memory, and an
instruction from the cache, all in a single cycle.
Addressing of external memory devices by the ADSP-21020 is
facilitated by on-chip decoding of high-order address lines to
generate memory bank select signals. Separate control lines
are also generated for simplified addressing of page-mode
DRAM.
The ADSP-21020 provides programmable memory wait
states, and external memory acknowledge controls allow
interfacing to peripheral devices with variable access times.
DAG 1
DATA ADDRESS
GENERATORS
ALU
32/40-Bit IEEE Floating-Point
DAG 2
ARITHMETIC UNITS
REGISTER FILE
MULTIPLIER
FUNCTIONAL BLOCK DIAGRAM
INSTRUCTION
SEQUENCER
PROGRAM
SHIFTER
CACHE
DSP Microprocessor
PROGRAM MEMORY ADDRESS
PROGRAM MEMORY DATA
DATA MEMORY ADDRESS
DATA MEMORY DATA
ADSP-21020
& EMULATION
JTAG TEST
TIMER
Fax: 617/326-8703
EXTERNAL
DATA
BUSES
EXTERNAL
ADDRESS
BUSES

Related parts for ADSP21020

ADSP21020 Summary of contents

Page 1

FEATURES Superscalar IEEE Floating-Point Processor Off-Chip Harvard Architecture Maximizes Signal Processing Performance 30 ns, 33.3 MIPS Instruction Rate, Single-Cycle Execution 100 MFLOPS Peak, 66 MFLOPS Sustained Performance 1024-Point Complex FFT Benchmark: 0.58 ms Divide (y/x): 180 ns Inverse Square ...

Page 2

ADSP-21020 • Instruction Cache The ADSP-21020 includes a high performance instruction cache that enables three-bus operation for fetching an instruction and two data values. The cache is selective—only the instructions whose fetches conflict with program memory data accesses are cached. ...

Page 3

DAG PMD BUS BUS CONNECT DMD BUS FLOATING & FIXED-POINT MULTIPLIER, FIXED-POINT ACCUMULATOR the standard IEEE format, whereas the 40-bit IEEE extended- precision format has eight additional LSBs of mantissa for greater accuracy. The ...

Page 4

ADSP-21020 in a specified register, either before (premodify) or after (postmodify) the access. To implement automatic modulo addressing for circular buffers, the ADSP-21020 provides buffer length registers that can be associated with each pointer. Base values for pointers allow circular ...

Page 5

SELECTS PROGRAM MEMORY The ADSP-21020 also implements on-chip emulation through the JTAG test access port. The processor’s eight sets of break- point range registers enable program execution at full speed until reaching a desired break-point address range. The processor can ...

Page 6

ADSP-21020 Pin Name Type Function DMPAGE O Data Memory Page Boundary. The ADSP- 21020 asserts this pin to signal that a data memory page boundary has been crossed. Memory pages must be defined in the memory control registers. DMTS I/S ...

Page 7

COMPUTE AND MOVE OR MODIFY INSTRUCTIONS 1. compute condition compute; 3a. IF condition compute, 3b. IF condition compute, 3c. IF condition compute, 3d. IF condition compute, 4a. IF condition compute, 4b. IF condition compute, 4c. IF condition compute, ...

Page 8

ADSP-21020 IMMEDIATE MOVE INSTRUCTIONS 14a. DM(<addr32>) = ureg ; PM(<addr24>) 14b. ureg = DM(<addr32>) ; PM(<addr24>) 15a. DM(<data32>, Ia) = ureg; PM(< data24>, Ic) 15b. ureg = DM(<data32>, Ia) ; PM(<data24>, Ic) 16. DM(Ia, Mb) = <data32>; PM(Ic, Md) 17. ...

Page 9

Table III. Universal Registers Name Function Register File R15–R0 Register file locations Program Sequencer PC* Program counter; address of instruction cur- rently executing PCSTK Top of PC stack PCSTKP PC stack pointer FADDR* Fetch address DADDR* Decode address LADDR Loop ...

Page 10

ADSP-21020 Table V. Multiplier Compute Operations MRF = MRB = MRF + Rx * ...

Page 11

Table Vll. Multifunction Compute Operations Fixed-Point Rm=R3-0 * R7-4 (SSFR), Ra=R11-8 + R15-12 Rm=R3-0 * R7-4 (SSFR), Ra=R11-8 – R15-12 Rm=R3-0 * R7-4 (SSFR), Ra=(R11-8 + R15-12)/2 MRF=MRF + R3-0 * R7-4 (SSF), Ra=R11-8 + R15-12 MRF=MRF + R3-0 * ...

Page 12

ADSP-21020–SPECIFICATIONS RECOMMENDED OPERATING CONDITIONS Parameter V Supply Voltage DD T Ambient Operating Temperature AMB Refer to Environmental Conditions for information on thermal specifications. ELECTRICAL CHARACTERISTICS Parameter 1 V Hi-Level Input Voltage Hi-Level Input Voltage IHCR 1, ...

Page 13

TIMING PARAMETERS General Notes See Figure 15 on page 24 for voltage reference levels. Use the exact timing information given. Do not attempt to derive parameters from the addition or subtraction of others. While addition or subtraction would yield meaningful ...

Page 14

ADSP-21020 Interrupts Parameter Timing Requirement: t IRQ3-0 Setup before CLKIN High 38 SIR t IRQ3-0 Hold after CLKIN High HIR t IRQ3-0 Pulse Width IPW NOTE * – Meeting setup and hold guarantees interrupts will ...

Page 15

Flags Parameter 1 Timing Requirement: t FLAG3-0 Setup before CLKIN High SFI IN t FLAG3-0 Hold after CLKIN High HFI IN t FLAG3-0 Delay from xRD, xWR Low DWRFI IN t FLAG3-0 Hold after xRD, xWR HFIWR IN Deasserted Switching ...

Page 16

ADSP-21020 Bus Request/Bus Grant Parameter Timing Requirement Hold after CLKIN High HBR t BR Setup before CLKIN High SBR Switching Characteristic: t Memory Interface Disable to BG Low –2 DMDBGL t CLKIN High to Memory Interface DME Enable ...

Page 17

External Memory Three-State Control Parameter Timing Requirement: t xTS, Setup before CLKIN High STS t xTS Delay after Address, Select DADTS t xTS Delay after XRD, XWR Low DSTS Switching Characteristic: t Memory Interface Disable before DTSD CLKIN High t ...

Page 18

ADSP-21020 Memory Read Parameter Timing Requirement: t Address, Select to Data Valid DAD t xRD Low to Data Valid DRLD t Data Hold from Address, Select HDA t Data Hold from xRD High HDRH t xACK Delay from Address DAAK ...

Page 19

CLKIN ADDRESS, SELECT t DAP DMPAGE, PMPAGE t DCKRL DMRD, PMRD DATA DMACK, PMACK DMWR, PMWR REV DARL RW t DRLD t DAD t DRAK t t SAK DAAK Figure 10. Memory Read –19– ADSP-21020 t HDA ...

Page 20

ADSP-21020 Memory Write Parameter Timing Requirement: t xACK Delay from Address, Select DAAK t xACK Delay from xWR Low DWAK t xACK Setup before CLKIN High SAK t xACK Hold after CLKIN High HAK Switching Characteristic: t Address, Select to ...

Page 21

CLKIN ADDRESS, SELECT DMPAGE, PMPAGE DMWR, PMWR t DCKWL DATA DMACK, PMACK DMRD, PMRD REV DAP t DAWH t t DAWL WW t WDE t DWAK t t DAAK SAK Figure 11. Memory Write –21– ADSP-21020 t DWHA ...

Page 22

ADSP-21020 IEEE 1149.1 Test Access Port Parameter Timing Requirement: t TCK Period TCK t TDI, TMS Setup before TCK High STAP t TDI, TMS Hold after TCK High HTAP t System Inputs Setup before TCK High 7 SSYS t System ...

Page 23

TCK t STAP TMS,TDI TDO t SSYS SYSTEM INPUTS SYSTEM OUTPUTS REV TCK t HTAP t DTDO t HSYS t DSYS Figure 12. IEEE 1149.1 Test Access Port –23– ADSP-21020 ...

Page 24

ADSP-21020 TEST CONDITIONS Output Disable Time Output pins are considered to be disabled when they stop driving, go into a high-impedance state, and start to decay from their output high or low voltage. The time for the voltage on the ...

Page 25

Capacitive Loading Output delays are based on standard capacitive loads: 100 pF on address, select, page and strobe pins, and all others (see Figure 14). For different loads, these timing parameters should be derated. See the Hardware ...

Page 26

ADSP-21020 ENVIRONMENTAL CONDITIONS The ADSP-21020 is available in a Ceramic Pin Grid Array (CPGA). The package uses a cavity-down configuration which gives it favorable thermal characteristics. The top surface of the package contains a raised copper slug from which much ...

Page 27

All GND pins should have a low impedance path to ground. A ground plane is required in ADSP-21020 systems to reduce this impedance, minimizing noise. The EVDD and IVDD pins should be bypassed to the ground plane using approximately 14 ...

Page 28

ADSP-21020 PMA17 PMA20 TMS EGND U TCK T EGND PMA19 PMA23 PMS1 TRST S PMA11 PMA14 PMA18 PMA22 PMPAGE R EGND PMA10 PMA15 PMA16 PMA21 P PMA8 PMA9 PMA13 PMA12 N EVDD PMA5 PMA6 PMA7 ...

Page 29

U PMD31 PMD35 PMD39 PMD40 EGND PMD27 PMD30 PMD32 PMD37 PMD21 PMD26 PMD28 PMD34 PMD36 R EGND PMD23 PMD25 PMD29 PMD33 P PMD18 PMD19 PMD22 PMD24 N EVDD PMD16 PMD17 PMD20 M ...

Page 30

ADSP-21020 PGA PIN LOCATION NAME G16 DMA0 G17 DMA1 F18 DMA2 F17 DMA3 F16 DMA4 F15 DMA5 E18 DMA6 E17 DMA7 E16 DMA8 D18 DMA9 E15 DMA10 D17 DMA11 D16 DMA12 C18 DMA13 C17 DMA14 D15 DMA15 B18 DMA16 B17 ...

Page 31

SYMBOL NOTE When socketing the CPGA package, use of a low insertion force socket is recommended. REV. C OUTLINE DIMENSIONS Dimensions shown ...

Page 32

ADSP-21020 Ambient Temperature Part Number* Range ADSP-21020KG- +70 C ADSP-21020KG-100 +70 C ADSP-21020KG-133 +70 C ADSP-21020BG-80 – +85 C ADSP-21020BG-100 – +85 C ADSP-21020BG-120 – ...

Related keywords