Lucent Technologies DSP1617, DSP1629, DSP1618, DSP1611, DSP1627, DSP1628 Information Manual
Specifications and Main Features
Frequently Asked Questions
User Manual
Information Manual
January 1998
DSP1611/17/18/27/28/29
Digital Signal Processor
For additional information, contact your Microelectronics Group Account Manager or the following:
INTERNET:
E-MAIL:
N. AMERICA: Microelectronics Group, Lucent Technologies Inc., 555 Union Boulevard, Room 30L-15P-BA, Allentown, PA 18103
ASIA PACIFIC: Microelectronics Group, Lucent Technologies Singapore Pte. Ltd., 77 Science Park Dr ive, #03-18 Cintech III, Singapore 118256
CHINA:Microelectronics Group, Lucent Technologies (China) Co., Ltd., A-F2, 23/F, Zao Fong Universe Buildi ng, 1800 Zhong Shan Xi Road,
JAPAN:Microelectronics Group, Lucent Technologies Japan Ltd., 7-18, Higashi-Gotanda 2-chome, Shinagawa-ku, Tokyo 141, Japan
EUROPE:Data Requests: MICROELECTRONICS GROUP DATALINE:
Lucent Technologies Inc. reserves the rig ht to mak e changes to the product(s) or i nformat ion cont ained herein without notice. No liability is ass um ed as a res ult of th eir use or a ppli catio n. N o
rights under any patent accompany the sale of any such produc t(s) or information.
The following Lucent Technologies Inc. trademarks are used in this manual:
Tapdance®FlashDSP
The following trademarks, owned by entities other than Lucent Technologies Inc., are used in this manual:
IEEE
is a registered trademark of The Institute of Electrical and Electronics Engineers, Inc.
Intel
is a registered trademark of Intel Corporation.
Motorola
MS-DOS
TI
UNIX
X-Windows
is a registered trademark of Motorola, Inc.
and
Windows
is a registered trademark of Texas Instruments, Inc.
is a registered trademark licensed exclusively through X/Open Company Ltd.
is a trademark of Massachusetts Institute of Technology.
®
are registered trademarks of Microsoft Corporation.
iiDRAFT COPYLucent Technologies Inc.
Foreword
This manual contains detailed information on the design and application of the DSP1611/17/18/27/28/29 Digital
Signal Processor family, which includes the
FlashDSP
DSP1628-ST, and DSP1629-ST support software libraries, the
numerous DSP1611/17/18/27/28/29-specific hardware support tools are also available to aid in developing software and integrating the devices into systems.
Additional information on the digital signal processor product line is available in the form of manuals, data sheets,
and application notes.
1629 development devices. The DSP1611-ST, DSP1618-ST, DSP1617-ST, DSP1627-ST,
FlashDSP
®
1618,
FlashDSP
FlashDSP
1627,
FlashDSP
1600-HDS Development System, and
1628, and
Conventions Used in this Manual
In general, all registers writable or readabl e by DSP instructions are low er case. D e vice flags , I/O pins, and nonprogram-accessible registers are generally upper case. For clarity, register names and DSP instructions are printed in
boldface
cized, such as
when used in written descriptions. Variable names that are to be replaced by specific names are itali-
filename
. Instruction set notation conventions are defined in Chapter 4.
➤ 11 The JTAG Test Access Port....................................................................................................................... 11-1
➤11.1 Overview of the JTAG Architecture ................................................................................................. 11-1
➤11.2 Overview of the JTAG Instructions.................................................................................................. 11-3
➤11.3 Elements of the JTAG Test Logic .................................................................................................... 11-4
➤11.3.1 The Test Access Port (TAP) ............................................................................................. 11-4
➤11.3.2 The TAP Controller ........................................................................................................... 11-5
➤11.3.3 The Instruction Register—JIR .......................................................................................... 11-7
➤11.3.4 The Boundar y-Scan Register—JBSR .............................................................................. 11-8
➤11.3.5 The Bypass R egister— JBPR ......................................................................................... 11-16
➤11.3.6 The Device Identification Register—JIDR ...................................................................... 11-16
➤11.3.7 The JTAG Data Register—jtag ...................................................................................... 11-19
viiLucent Technologies Inc.
➤11.3.8 The JTAG Control R egister— JCON ............................................................................... 11-19
➤11.3.9 The JTAG Output S tage—JOUT .................................................................................... 11-19
➤11.4 The JTAG Instruction Set.............................................................................................................. 11-19
➤11.4.1 The EXTEST Instruction ................................................................................................ 11-19
➤11.4.2 The INTEST Instruction ................................................................................................. 11-19
➤11.4.3 The SAMPLE Instruction ................................................................................................ 11-20
➤11.4.4 The BYPASS Instruction ................................................................................................ 11-20
➤11.4.5 The IDCOD E Ins truction ................................................................................................ 11-20
➤ 13 Bit Manipulation Unit (BMU ) ...................................................................................................................... 13-1
➤15.2 Signal Descriptions......................................................................................................................... 15-5
➤15.2.1 System Interface .............................................................................................................. 15-5
➤15.4.2 ROM Secur ity Options (DSP1617/18/27/28/29 Only) .................................................... 15-14
➤15.5 Additional Electrical Characteristics and Requirements for Crystal.............................................. 15-15
➤ A Instruction Encoding....................................................................................................................................A-1
➤ B Instruction Set Summary.............................................................................................................................B-1
➤if CON goto/call/return................................................................................................................................ B-3
➤do K {.......................................................................................................................................................... B-6
➤R = Y ........................................................................................................................................................ B-13
➤Y = R ........................................................................................................................................................ B-14
➤*(OFFSET) = DR ...................................................................................................................................... B-17
➤if CON F2 ................................................................................................................................................. B-18
➤ifc CON F2................................................................................................................................................ B-19
➤F1 Y ....................................................................................................................................................... B-20
➤F1 Y = a0[l] ............................................................................................................................................ B-22
➤F1 Y = a1[l] ............................................................................................................................................ B-22
➤F1 x = Y ................................................................................................................................................. B-24
➤F1 y = Y x = *pt++[i] ............................................................................................................................ B-28
➤F1 y = a0 x = *pt++[i]........................................................................................................................... B-30
➤F1 y = a1 x = *pt++[i]........................................................................................................................... B-30
➤F1 aT[l] = Y ............................................................................................................................................ B-32
➤F1 Y = y[l]............................................................................................................................................... B-34
➤F1 Z : y[l]................................................................................................................................................ B-36
ixLucent Technologies Inc.
Information Manual
April 1998
DSP160X DIGITAL SIGNAL PROCESSOR
➤F1 Z : aT[l] ............................................................................................................................................. B-38
➤F1 Z : y x = *pt++[i].............................................................................................................................. B-40
➤aD = aS OP aT......................................................................................................................................... B-42
➤aD = aS OP p........................................................................................................................................... B-43
➤aD = aS<h,l> OP IM16............................................................................................................................. B-44
➤aD = a SHIFT aS ................................................................................................................................... B-46
S
➤aD = aS SHIFT arM.................................................................................................................................. B-47
➤aD = aS SHIFT IM16................................................................................................................................ B-48
➤ Table A-6.DR Field ...................................................................................................................................... A-5
➤ Table A-7.F1 Field ....................................................................................................................................... A-6
➤ Table A-8.F2 Field ....................................................................................................................................... A-6
➤ Table A-9.F3 Field ....................................................................................................................................... A-7
➤ Table A-10. I Field .......................................................................................................................................... A-7
➤ Table A-11. R Field for DSP1617................................................................................................................... A-8
➤ Table A-12. R F ield for DSP1611/18/27/28/29 ............................................................................................... A-8
➤ Table A-13. S Field......................................................................................................................................... A-9
➤ Table A-14. SI Field........................................................................................................................................ A-9
➤ Table A-17. X Field....................................................................................................................................... A-10
➤ Table A-18. Y Field....................................................................................................................................... A-10
➤ Table A-19. Z Field ....................................................................................................................................... A-10
➤ Table B-1.CON Field Encoding ................................................................................................................... B-3
➤ Table B-2.R Field Replacement Values....................................................................................................... B-8
Information ManualDSP1611/17/18/27/28/29 DIGITAL SIGNAL PROCESSOR
April 1998
1 Introduction
Designed specifically for applications requiring low-power dissipation in digital cellular systems, the DSP1611,
DSP1617, DSP1618, DSP1618x24
4
DSP1629x16
are signal coding devices that can be programmed to perform a wide variety of fixed-point signal
1
, DSP1627, DSP1627x322, DSP1628x083, DSP1628x163, DSP1629x104, and
processing functions. The de vices are based on the DSP1600 core with a bit manipulation unit for enhanced signal
coding efficiency. The DSP1611/17/18/27/28/29 include a mix of peripherals specifically intended to support processing-intensive, but cost-sensitive, applications in the ar ea of digital mobile c ommunicat ions . The features of the
DSP1611/17/18/27/28/29 are as follows:
Optimized for digital cellular applications with a bit manipulation unit for higher signal coding efficiency
Multiple speed and operating voltage options
Low power consumption
Flexible power management modes
— Standard sleep
— Sleep with slow internal clock
— Hardware STOP pin halts DSP
Multiple packaging options available including low-profile TQFP and BQFP packaging
Multiple mask-programmable clock options
Single-cycle squaring
16 x 16-bit multiplication and 36-bit accumulation in one instruction cycle
Instruction cache for high-speed, program-efficient, zero-overhead looping
Memory sequencer for single-instruction access to both X and Y external memory space
Two external vectored interrupts and trap
Flexible internal ROM and internal dual-port RAM configurations
Dual serial I/O ports with multiprocessor capability—16-bit data channel, 8-bit protocol channel
8-bit parallel interface
8-bit control I/O interface
256 memory-mapped I/O por ts, one internally decoded for glueless device interfacing
Interrupt timer
CMOS I/O levels
IEEE
5 P1149.1 test port (JTAG with boundary-scan)
Full-speed in-circuit emulation hardware development system on-chip
Supported by DSP1611/17/18/27/28/29 software and hardware development tools
Each device also includes specific features for specialized applications
— Error correction coprocessor (ECCP) in DSP1618/28
— On-chip phase-lock loop (PLL) in DSP1627/28/29
— Bootstrap ROM in DSP1611
This manual is a user's reference guide for the DSP1611/17/18/27/28/29.
1.The DSP1618x24 is basically the same as the DSP1618. They differ in the amount of internal ROM memory and X-memory mapping (see
Table 3-11, Section 3.2.2, X-Memory Space). Discussion of the DSP1618 also refers to the DSP1618x24 except if noted otherwise.
2.The DSP1627x32 is basically the same as the DSP1627. They differ in the amount of internal ROM memory and X-memory mapping (see
Table 3-12, Section 3.2.2, X-Memory Space). Discussion of the DSP1627 also refers to the DSP1627x32 except if noted otherwise.
3.The DSP1628x08 and DSP1628x16 differ only in the size of internal dual-port RAM. Discussion of the DSP1628 refers to both the
DSP1628x08 and DSP1628x16 except if noted otherwise.
4.The DSP1629x10 and DSP1629x16 differ only in the size of internal dual-port RAM. Discussion of the DSP1629 refers to both the
DSP1629x10 and DSP1629x16 except if noted otherwise.
5. IEEE is a registered trademark of The Institute of Electr ical and Electronics Engineers, Inc.
Lucent Technologies Inc.
DRAFT COPY
1-1
DSP1611/17/18/27/28/29 DIGITAL SIGNAL PROCESSORInformation Manual
IntroductionApril 1998
1.1 General Description
1.1.1 Architecture
The DSP1611, DSP1617, DSP1618, DSP1627, DSP1628, and DSP1629 ar e made up of the DSP1600 core processor, a dual-port RAM, ROM, and several peripheral blocks. The core contains the data arithmetic unit, the
memory addressing units, the cache, and the control section.
The data arithmetic unit (DAU) is the main computational execution unit of the processor. It supports a 16-bit x
16-bit multiply, a 36-bit ALU operation, and two 16-bit data fetches from memory in a single instruction cycle. The
DA U is made up of two input data registers, the multiplier, two accumulators , the ALU, and various control registers.
The product from the multiplier can be acc umulated i n one of the two 36- bit accum ulators . The data in these accumulators can be directly loaded from or st ored to memory in 16-bit words . The ALU supports a full set of arithmetic
and logic operations on either 16- or 32-bit data. Because a standard set of ALU conditions can be tested to perform conditional branches and subroutine calls, the processor functions as a powerful 16-bit or 32-bit microprocessor for logical and control applications.
A bit manipulation unit (BMU) is provided to accelerate signal coding algorithms. It performs full 36-bit barrel shifting, normalization, and bit field extraction or insertion of data in the accumulators. Two alternate accumulators provide storage for 36-bit data.
An on-chip cache memory can selectively store repetitive operations like those found in an FIR or IIR filter section.
The code in the cache can repeat up to 127 times with no looping overhead. In addition, operations in the cache
that require an X-memory data access (for example, reading fixed coefficients) execute at twice the normal rate.
The cache greatly reduces the need for writing in-line repetitive code and, therefore, reduces program memory size
requirements. In addition, power consumption is reduced because use of the cache eliminates a memory access
for instruction fetches.
Two addressing units support high-speed, register-indirect memory addressing with postincrementing of the register. Four address pointer registers can be used for either read or write addresses to the RAM. One address register is dedicated to the instruction/coefficient memory space for table look-up. Direct data addressing is supported
for 16 k ey register s . A unique compound addressing mode that s w aps data between a r egister and memory in only
two instruction cycles is available. Immediate addressing can be done by using a 9-bit address in a one-cycle
instruction or a 16-bit address in a two-cycle instruction.
The DSP1611/17/18/27/28/29 on-chip memory includes both ROM and dual-port RAM. The RAM has separate
ports to the instruction/coefficient bus and the data bus, and it can write either bus. A program can be downloaded
from slow off-chip memory into the RAM and then e xecuted at full-speed without wait-states. The RAM can also be
downloaded through the JTAG interface for full-speed, remote, in-circuit emulation or for self-test.
The external memory interface (EMI) connects either the instruction/coefficient buses or the data buses to the
external memory buses. The bit input/output (BIO) unit has eight pins that can be individually selected as inputs or
outputs. The timer provides programmable peri odic interrupts. The JTAG interface is a four- wire standard test port
defined by
pointing and branch tracing in support of full-speed, in-circuit emulation with only the low-speed serial JTAG interface required off-chip.
The DSP1611/17/18/27/28/29 have both a paral lel I/O port (PIO or PHIF) and two serial I/O ports (SIO). The serial
I/O units are double-buff er ed and easily interf ac e to other DSP1600 f amily de vices , commerci ally available codecs,
and time-division multiple xed (TDM) channels with few, i f any, additional components. Both ports connect as many
as eight DSPs in multiprocessor operation. The parallel I/O unit is capable of interfacing to an 8-bit bus containing
other DSP1600 family devices, microprocessors, microprocessor peripherals, or other I/O devices.
IEEE
P1149.1. On-chip hardware development system (HDS) circuitry performs instruction break-
1-2
DRAFT COPY
Lucent Technologies Inc.
Information ManualDSP1611/17/18/27/28/29 DIGITAL SIGNAL PROCESSOR
April 1998Introduction
1.1 General Description (continued)
1.1.1 Architecture
(continued)
Many applications , suc h as portabl e c ellular terminals, requi re programmable sleep modes f or po w er management.
There are three different control mechanisms for achieving low-power operation: the
STOP pin, and the AWAIT bit in the
alf
register. The
powerc
register configures various power-saving modes by
powerc
control register
,
the
controlling internal clocks and peripheral I/O units. The STOP pin controls the internal processor clock. The
AWAIT bit in the
alf
register allows the processor to go into a power-saving standby mode until an interrupt occurs.
The external interrupts asynchronously restart the processor from a deep sleep power-saving mode, and program
execution continues without any loss of state. The various power management options are chosen based on
power consumption, wake-up latency, or both requirements.
The DSP1611/17/18/27/28/29 are implemented in low-power CMOS technology and are offered in a variety of
packaging options . For optimal matching to system requirements, sev eral opt ions for low -voltage po wer supply and
clock speeds are available. See the latest data sheet for the current offerings.
1.1.2 Instruction Set
The DSP1611/17/18/27/28/29 instructions fall into seven categories: multiply/ALU, special function, control, data
move , F3 ALU , BMU , and cache. All instructions are 16 bits wide and hav e a C-lik e assemb ler syntax . Instructions
typically execute in one or sometimes two cycles, and data-path latency effects have been eliminated. Very high
performance is achieved by the use of concurrent instructions in the DAU.
1.2 Typical Applications
The devices in the DSP16XX1 family of digital signal processors are used in many different application areas
including telecommunications, speech processing, image processing, graphics, array processors, robotics, studio
electronics, instrumentation, and military applications. Some of the possible applications follow:
TELECOMMUNICATIONS
Mobile CommunicationsSpeech coding, modulation/demodulation, channel coding/decoding
ModemsEcho cancellation, filtering, error correction and detection
PBXTone detection, tone generation, MF, DTMF
SwitchesTone detection, tone generation, line testing
Answering MachinesSpeech coding/decoding, system control
EntertainmentSpeech coding/decoding
Educational—
Many of these applications can use standard algorithms that have been designed to reduce computational and
data transfer requirements for these DSPs. These algorithms have been coded in DSP1600 assembly language
and are available to registered users via Lucent’s DSP tech support web page at
http://www.lucent.com/micro/wam/tse
.
1.3 Application Support
The use of the DSP1611/17/18/27/28/29-ST Support Tools and the DSP1600-HDS Hardware Development System aids application development.
1.3.1 Support Software Library
Software development tools to help create, test, and debug DSP1611/17/18/27/28/29 application programs are
available from the Lucent Technologies’ appropriate support software library for the particular device. Each support software library consists of an assembler, linker, and software simulator that run on
3
DOS
operating systems. The software includes a menu driven,
Windows
3
based, graphical user interface.
The assembler transforms DSP1611/17/18/27/28/29 source code into object code in a standard format (COFF)
that is then processed by the linker. The assembler contains a preprocessor similar to the C preprocessor and provides the features of a ful l macr o ass emb ler. The linker creates load modules f or the sim ulator b y combi ning objec t
files, performing relocation, resolving external references, and supporting symbol table information for symbolic
testing. The DSP1611/17/18/27/28/29 software simulator provides access to all registers and memory and allows
program breakpointing. The simulator also provides the user interface to the DSP1600 Hardware Development
System.
1.3.2 Hardware Development System
The DSP1600 JTAG communication system (JCS) supports application system hardware development and software testing.
Sun-4
1
,
UNIX
2
, or
MS-
1.
Sun, Sun Microsyste ms
States and other countries.
UNIX
is a reg i stered t radema rk licensed exclusively through X /Open Company Ltd.
2.
MS-DOS
3.
and
Windows
1-4
, the Su n l ogo,
are registered trademarks of the Microsoft Corporation.
SunOS
, and
Solaris
are trademarks or registered trademarks of Sun Microsystems, Inc. in the United
DRAFT COPY
Lucent Technologies Inc.
Information ManualDSP1611/17/18/27/28/29 DIGITAL SIGNAL PROCESSOR
April 1998Introduction
1.3 Application Support (continued)
1.3.2 Hardware Development System
(continued)
Figure 1-1 shows the components of the DSP1600 hardware development system for in-circuit emulation. The PC
is an
MS-DOS
386, 486-based, or better machine. The enhanced system controller card (ESCC) plugs into an
8-bit slot on the PC ISA I/O bus and connects to the enhanced target interface box (ETIB). The ETIB provides a
JTAG interface to the target DSP1611/17/18/27/28/29 device using a 9-pin connector cable. With this configuration, a program can be downloaded into the DSP on the user's board and executed at full speed. The emulation is
performed with the actual DSP located on the user's board, and not one separated from it by a performancelimiting cable. Program development with breakpointing, single-stepping, and branch tracing is available with the
simulator; it is aided by the hardware development system module on the DSP1611/17/18/27/28/29.
ESCC
ESCC – ENHANCED SYSTEM CONTROLLER CARD
ETIB – ENHANCED TARGET INTERFACE BOX
37-PIN CABLE
ETIB
9-PIN CABLE
(JTAG INTERFACE)
TARGET BOARD
POWER
SUPPLY
ac SUPPLY
TARGET BOARD
POWER CABLE
(12.0 V —15.0 V)
Figure 1-1. In-Circuit Emulation with the
FlashDSP
1600—JCS
Another development tool available is the demonstration board (DSP1611/17/18/27/28/29-DEMO). The demonstration board replaces the customer board in Figure 1-1 and provides a development platform with external memory (static RAM or PROM), a DSP1611/17/18/27/28/29 device, and access many DSP signals.
Lucent Technologies Inc.
DRAFT COPY
1-5
DSP1611/17/18/27/28/29 DIGITAL SIGNAL PROCESSORInformation Manual
IntroductionApril 1998
1.4 Manual Organization
This document is a ref erence guide for the DSP1611, DSP1617, DSP1618, DSP1627, DSP1628, and DSP1629. It
describes the architecture, instruction set, and interfacing requirements of the device. The remaining chapters of
this manual are outlined below:
Chapter 2.
Hardware Architecture
ing the major elements of the architecture and how they function.
: An overall description of the device including separate sections describ-
Included are a register view of the chip, arithmetic and precision of data, memory space description, and the interrupt structure.
Instruction Set
Notation and addressing modes are also discussed in detail. Appendix B lists the complete
instruction set and provides a description of each instruction including r estrictions and normal
uses.
Core Architecture
External Memory Interface
Serial I/O
clocking, interrupts, and multiprocessor operation.
ation of this port, including interrupt information.
Bit I/O Unit
JTAG Test Access Port
Timer
: Operation and programming.
Bit Manipulation Unit
Error Correction Coprocessor (DSP1618/28 Only)
: This section describes the general characteristics of the groups of instructions.
: A detailed analysis of the operation of the serial I/O ports including active and passive
: A functional description of the operation and programming of this port.
: A description of the topics associated with the software of the device.
: A detailed description of the DSP1600 core architecture.
: A description of the EMI port including functional timing.
: A detailed analysis of the operation of this parallel I/O port including
: A functional description of the oper-
: Functional description of the JTAG port.
: A detailed description of the bit manipulation unit.
: A detailed description of this coprocessor.
Chapter 15.
Appendix A.
Appendix B.
1-6
Interface Guide
Instruction Encoding
Instruction Set Summary
: A functional description of each category of pins with tables describing pins.
: Lists the hardware-level encoding of the instruction set.
: Each instruction is described in detail.
DRAFT COPY
Lucent Technologies Inc.
Information ManualDSP1611/17/18/27/28/29 DIGITAL SIGNAL PROCESSOR
April 1998Introduction
1.4 Manual Organization (continued)
1.4.1 Applicable Documentation
A variety of documents exists to provide specific information on various members of the DSP1600 product family.
Contact your Lucent Technologies Account Manager for the latest issue of any of the following documents. The
back cover lists contact numbers for customer assistance.
DSP1611/17/18/27/28/29 Digital Signal Processor Information Manual (this manual) is a reference guide for the
DSP1611/17/18/27/28/29. It describes the architecture, instruction set, and interfacing requirements.
DSP1611, DSP1617, DSP1618, DSP1627, DSP1628, and DSP1629 Digital Signal Processor data sheets provide
up-to-date timing requirements and specifications, electrical characteristics, and a summary of the instruction set
and device architecture for each device.
DSP1600 Support Tools Manual
includes the appropriate DSP1611/17/18/27/28/29 supplement that provides the inform ation necessary to install
and use the DSP1611/17/18/27/28/29 support software. The suppor t tools manual is also required if working with
the DSP1600 Hardware Development System because the support software provides an interface between the
host computer and the development system. Each hardware development tool is packed with a user manual and
schematics.
is an online document shipped with DSP1611/17/18/27/28/29 software tools. It
➤2.14 Power Management.........................................................................................................................2-23
Information ManualDSP1611/17/18/27/28/29 DIGITAL SIGNAL PROCESSOR
April 1998
2 Hardware Architecture
This chapter presents an overview of the hardware in the DSP1611, DSP1617, DSP1618, DSP1627, DSP1628,
and DSP1629. First, an overall view of the architecture is discussed; then, each major functional block is
described. The following chapters give full details on each block.
2.1 Device Architecture Overview
2.1.1 Harvard Architecture
Figure 2-1 shows a view of a simple operation in the DSP1611/17/18/27/28/29 architecture to demonstrate funda-
mentally how an instruction is processed. The architecture is a Harvard architecture defined as having two separate memory spaces. The first is the instruction/coefficient space or program space that is referred to in this
manual as the X-memory space. The second is the data memory space that is referred to as the Y-memor y
space. Each memory space has a corresponding address arithmetic unit. In the instruction/coefficient memory
space, the progr am addr essi ng unit (XAAU) places addresses on the program address bus (XAB). In this example ,
these addresses go to the internal ROM that, then, places instructions on the program data bus (XDB). The
instructions are decoded in the control block that, in turn, provides control signals to all of the processor sections.
The control signals respond to instructions that, in this example, call for arithmetic operations on data residing in
the RAM. The data addressing unit (YAAU) addresses the RAM over the data address bus (YAB), and data is
transferred between the RAM and data arithmetic unit (DAU) over the data bus (YDB). The power of the architecture lies in the parallel operations that are possible. In this case, instruction processing, data transfer, and arithmetic operations can all be done simultaneously.
ROM
INSTRUCTIONS
16
CONTROL
XAB
16
PROG. COUNTER
PROGRAM
ADDRESS
UNIT
DATA
ADDRESS
UNIT
YAB
16
Figure 2-1. Harvard Architecture
DATA
ARITHMETIC
UNIT
16
YDBXDB
RAM
5-4140
Lucent Technologies Inc.
DRAFT COPY
2-1
DSP1611/17/18/27/28/29 DIGITAL SIGNAL PROCESSORInformation Manual
Hardware ArchitectureApril 1998
2.1 Device Architecture Overview (continued)
2.1.2 Concurrent Operations
Figure 2-2 shows the hardware view of an example of concurrent operations in the de vice. It also demonstrates the
flexibility of the memory spaces. In this example, the progr am is e x ecuti ng from the instruction cache. Instructions
are fed directly to the contr ol section freeing the XAB. The program addressing unit (XAAU) is now addressing one
bank of the dual-port RAM (Bank 1) to transfer variable coefficients between the RAM and the DAU. It could alternatively have been addressing the ROM to transf er fixed coefficients to the DAU . The data addressing unit (YAAU)
is addressing another bank of the dual-port RAM (Bank 4) to transfer data between the RAM and the DAU. Thus,
in one instruction cycle, two words of data can be transferred to the DAU simultaneously during internal calculations in the DAU. In the DAU, a multiplication can occur at the same time as an accumulation of a previous
product. In fact, a multiplication can occur in parallel w i th a variety of ALU operations.
YAAU
YAB
DAU
CACHE
INSTRUCTIONS
DUAL-PORT
RAM
BANK 1
DATA
y REGISTER
YDB
MULTIPLIER
ACCUMULATOR
INTERNAL
BUS
DUAL-PORT
XDB
x REGISTER
RAM
BANK 4
VARIABLE
COEFFICIENTS
CONTROL
XAB
XAAU
5-4141.a
2-2
Figure 2-2. Concurrent Operations in the DSP1611/17/18/27/28/29
DRAFT COPY
Lucent Technologies Inc.
Loading...
+ 413 hidden pages
You need points to download manuals.
1 point = 1 manual.
You can buy points or you can get point for every manual you upload.