Processors used in System on chip

PROCESSORS
Mr. A. B. Shinde
Assistant Professor,
Electronics Engineering,
PVPIT, Budhgaon.
shindesir.pvp@gmail.com

CISC
 CISC stands for Complex Instruction Set Computer
 CISC is a instruction set architecture (ISA) in which each instruction
can execute several low-level operations, such as a load from memory,
an arithmetic operation, and a memory store, all in a single instruction.
 CISC are chips that are easy to program and which make efficient use of
memory.
 Examples of CISC processor families are
 System/360,
 PDP-11, VAX,
 68000, and
 x86.

Complex Instruction Set
Computer
 CISC History
 The first PC microprocessors developed were CISC chips, because
all the instructions the processor could execute were built into the
chip.
 Memory was expensive in the early days of PCs, and CISC chips
saved memory because their programming could be fed directly into
the processor.
 CISC was developed to make compiler development simpler. It shifts
most of the burden of generating machine instructions to the
processor.
For example, instead of having to make a compiler write long
machine instructions to calculate a square-root, a CISC processor
would have a built-in ability to do this.

Computer
 CISC Philosophy
 The three decisions that led to the CISC philosophy, which drove all
computer designs until the late 1980s, and is still in major use today
are the
 use Microcode,
 build rich instruction sets, and
 build high-level instruction sets.

Computer
 CISC Philosophy
 Use Microcode:
 simple logic to control the data paths between the various elements
of the processor.
 In a micro programmed system, the main processor has some built-
in memory (typically ROM) that contains groups of microcode
instructions which correspond with each machine-language
instruction.
 Since the microcode memory can be much faster than main
memory, an instruction set can be implemented in microcode
without losing much speed over a purely hard-wired
implementation.

Computer
 CISC Philosophy
 Build rich instruction sets:
 By using a micro programmed design, designers could build more
functionality into each instruction.
 This design cut down on the total number of instructions required to
implement a program, so it made more efficient use of a slow main
memory.
 Made the job for assembly-language programmer simpler
 The enhancements included string manipulation operations, special
looping constructs, and special addressing modes for indexing
through tables in memory.

Computer
 CISC Philosophy
 Build high-level instruction sets :
 After the programmer-friendly instruction sets were built, designers
started to build instruction sets which map directly from high-level
languages.
 Because micro program instruction sets can be written to match the
constructs of high-level languages, the compiler does not have to
be as complicated.
 Allows compilers to emit fewer instructions per line of source

Computer
 Characteristics
 CISC are Mostly Von Neumann Architecture
(There are few exceptions)
 Same bus for program memory, data memory, I/O, registers, etc
 Generally Micro-coded ,Variable length instructions
 Segmentation is possible with Segment Register s like DS, ES and an
offset which can be common to all segments.
 Many powerful instructions are supported, making the assembly
language programmer’s job much easier.
 Physical Memory Extension Possible

Computer
 Characteristics Of CISC Design
 Instruction sets : CISC instruction sets have some common
characteristics:
 A 2-operand format, where instructions have a source and a destination.
 Register to register, register to memory, and memory to register
commands.
 Multiple addressing modes for memory, including specialized modes for
indexing through arrays
 Variable length instructions where the length often varies according to the
addressing mode
 Instructions which require multiple clock cycles to execute.

Computer
 Characteristics Of CISC Design
 Hardware architectures: CISC hardware architectures have several
characteristics in common:
 Complex instruction-decoding logic, driven by the need for a single
instruction to support multiple addressing modes.
 A small number of general purpose registers. This is the direct result of
having instructions which can operate directly on memory and the limited
amount of chip space not dedicated to instruction decoding, execution, and
microcode storage.
 Several special purpose registers. Many CISC designs set aside special
registers for the stack pointer, interrupt handling, and so on. This can
simplify the hardware design.
 A "Condition code" register which is set as a side-effect of most
instructions.

Computer
 Characteristics of CISC Design
 CISC and the Classic Performance Equation
The equation for determining performance is
(the number of cycles per instruction * instruction cycle time) = execution time.
This allows you to speed up a processor in 3 different ways :
- use fewer instructions for a given task,
- reduce the number of cycles for some instructions, or
- speed up the clock (decrease the cycle time.)
 CISC tries to reduce the number of instructions for a program

Computer
 The Advantages of CISC
 Microprogramming is as easy as assembly language to implement,
and much less expensive than hardwiring a control unit.
 The ease of micro-coding new instructions allowed designers to make
CISC machines upwardly compatible: a new computer could run the
same programs as earlier computers because the new computer
would contain a superset of the instructions of the earlier computers.
 As each instruction became more capable, fewer instructions could
be used to implement a given task. This made more efficient use of
the relatively slow main memory.
 Because micro-program instruction sets can be written to match the
constructs of high-level languages, the compiler does not have to be
as complicated.

Computer
 The Disadvantages Of CISC
 As many instructions as possible could be stored in memory with the
least possible wasted space, individual instructions could be of almost
any length this means that different instructions will take different
amounts of clock time to execute, slowing down the overall
performance of the machine.
 Many specialized instructions aren't used frequently enough to justify
their existence --- approximately 20% of the available instructions are
used in a typical program.
 CISC instructions typically set the condition codes as a side effect of
the instruction. Setting the condition codes take time, and
programmers have to remember to examine the condition code bits
before a subsequent instruction changes them.

Computer
Intel 8086 Architecture, the 1st member of x86 family

Computer
 Addressing modes
 Register Addressing Mode
 Memory Addressing Modes
 Displacement Only Addressing Mode
 Register Indirect Addressing Modes
 Indexed Addressing Modes
 Based Indexed Addressing Modes
 Based Indexed Plus Displacement Addressing

RISC
 RISC Stands for Reduced Instruction Set Computer
 RISC is a type of microprocessor architecture that utilizes a small, highly-
optimized set of instructions, rather than a more specialized set of
instructions found in other types of architectures.
 RISC represents a CPU design to make instructions execute very quickly.
 Well known RISC families include
 Alpha,
 ARC,
 ARM,
 AVR,
 MIPS,
 PA-RISC,
 Power Architecture (including PowerPC),
 SuperH and
 SPARC.

CHARACTERISTICS OF RISC
 RISC chip will typically have far fewer transistors dedicated to the core
logic which originally allowed designers to increase the size of the
register set and increase internal parallelism.
 Other features, which are typically found in RISC architectures
are:
 Uniform instruction format
Using a single word with the opcode in the same bit positions in every
instruction, demanding less decoding;
 Identical general purpose registers
Any register can be used in any context, simplifying compiler design
(there are separate floating point registers)
 Simple addressing modes.
Complex addressing performed via sequences of arithmetic and/or load-store
operations;
 Few data types in hardware, some CISCs have byte string instructions.

RISC
 RISC designs are also more likely to feature a Harvard memory
model, where the instruction stream and the data stream are
conceptually separated;
this means that modifying the memory where code is held might
not have any effect on the instructions executed by the
processor.
 On the upside, this allows both caches to be accessed simultaneously,
which can often improve performance.
 Many early RISC designs also shared the characteristic of having
a branch delay slot. A branch delay slot is an instruction space
immediately following a jump or branch.

RISC
 Key features
 Large number of general purpose registers or use of compiler
technology to optimize register use
 Limited and simple instruction set
 Emphasis on optimizing the instruction pipeline

RISC
 History
 The first RISC projects came from IBM, Stanford, and UC-Berkeley in the late
70s and early 80s.
 The IBM 801, Stanford MIPS, and Berkeley RISC 1 and 2 were all designed with
a similar philosophy which has become known as RISC.
 Certain design features have been characteristic of most RISC processors:
 one cycle execution time:
 Pipelining:
 large number of registers:

CISC Vs RISC
CISC RISC
Emphasis on hardware Emphasis on software
Includes multi-clock complex instructions Single-clock, reduced instruction only
Memory-to-memory:
"LOAD" and "STORE"
incorporated in instructions
Register to register:
"LOAD" and "STORE"
are independent instructions
Small code sizes large code sizes
Transistors used for storing complex
instructions
Spends more transistors on memory
registers
High cycles per second Low cycles per second
Variable length Instructions Equal length instructions which make
pipelining possible
Primary goal is to complete a task in as
few lines of assembly as possible
Primary goal is to speedup individual
instruction

CISC Vs RISC
The CISC
Approach
 Instruction :
MULT 2:3, 5:2
Operations:
1. Loads the two operands into separate
registers
2. Multiplies the operands in the execution unit
3. Then stores the product in the some
temporary register
4. Stores value back to memory location 2:3
The RISC Approach
 Instructions :
LW A, 2:3
LW B, 5:2
MULT A, B
SW 2:3, A
Operations:
1. Load operand1 into register A
2. Load operand2 into register B
3. Multiply the operands in the execution unit
and store result in A
4. Store value of A back to memory location
2:3

VON NEUMANN
ARCHITECTURE
John Von Neumann

VON NEUMANN
ARCHITECTURE
 The Von Neumann architecture is a design
model for a stored-program digital
computer that uses a processing unit and a
single separate storage structure to hold both
instructions and data.
 It is named after the mathematician and early
computer scientist John Von Neumann.

VON NEUMANN BOTTLENECK
 The separation between the CPU and memory leads to the von
Neumann bottleneck, the limited throughput (data transfer rate)
between the CPU and memory compared to the amount of memory.
 In most modern computers, throughput is much smaller than the rate at
which the CPU can work.
 The performance problem can be alleviated (to some extent) by several
mechanisms. Providing a cache between the CPU and the main
memory, providing separate caches with separate access paths for data
and instructions.
 The problem can also be sidestepped somewhat by using parallel
computing, using for example the NUMA architecture—this approach
is commonly employed by supercomputers.

HARVARD ARCHITECTURE
 The Harvard architecture is a computer architecture with physically
separate storage and signal pathways for instructions and data.
 The term originated from the Harvard Mark: relay-based computer, which
stored instructions on punched tape (24 bits wide) and data in electro-
mechanical counters. These early machines had limited data storage,
entirely contained within the central processing unit, and provided no
access to the instruction storage as data.
 Today, most processors implement such separate signal pathways for
performance reasons but actually implement a Modified Harvard
architecture, so they can support tasks like loading a program from
disk storage as data and then executing it

HARVARD ARCHITECTURE MEMORY
DETAILS
 In a Harvard architecture, there is no need to make the two memories
share characteristics.
 In particular, the word width, timing, implementation technology, and
memory address structure can differ.
 In some systems, instructions can be stored in read-only memory while
data memory generally requires read- write memory.
 In some systems, there is much more instruction memory than data
memory so instruction addresses are wider than data addresses.

CONTRAST WITH VON NEUMANN
ARCHITECTURES
 In a computer with the contrasting von Neumann architecture, the
CPU can be either reading an instruction or reading/writing data
from/to the memory.
 Both cannot occur at the same time since the instructions and
data use the same bus system.
 In a computer using the Harvard architecture, the CPU can both read
an instruction and perform a data memory access at the same time,
even without a cache.
 A Harvard architecture computer can thus be faster for a given
circuit complexity because instruction fetches and data access do not
contend for a single memory pathway.
 Also, a Harvard architecture machine has distinct code and data address
spaces: instruction address zero is not the same as data address zero.
Instruction address zero might identify a twenty-four bit value, while data
address zero might indicate an eight bit byte that isn't part of that twenty-
four bit value.

Soft processors
 A soft processor is an Intellectual Property (IP) core that is
implemented using the logic primitives of the FPGA. Being soft allows it
to have a high degree of flexibility and configurability.
 Soft processor is a microprocessor core that can be wholly implemented
using logic synthesis.
 It can be implemented via different semiconductor devices containing
programmable logic (e.g., ASIC, FPGA, CPLD).
 Key benefits of using a soft processor include configurability to trade
between price and performance, faster time to market, easy integration
with the FPGA fabric, and avoiding obsolescence.

Soft processors
 Most systems, if they use a soft processor at all, only use a single soft
processor. However, a few designers tile as many soft cores onto an
FPGA as will fit
 While many people put exactly one soft microprocessor on a FPGA, a
sufficiently large FPGA can hold two or more soft microprocessors,
resulting in a multi-core processor. The number of soft processors on a
single FPGA is only limited by the size of the FPGA.
 Some people have put dozens or hundreds of soft microprocessors on a
single FPGA

Soft processors
 What are the key benefits of having a soft FPGA-based processing
system ?
 FPGA-based provides many key benefits.

IBM’s power PC
 PowerPC is acronym for Performance Optimization With Enhanced
RISC – Performance Computing,
 PowerPC sometimes abbreviated as PPC

PPC405Fx Embedded
Processor
 The IBM 405Fx 32-bit reduced instruction set computer (RISC)
processor core, referred to as the PPC405Fx core, implements the
PowerPC Architecture with extensions for embedded applications.
 PPC405Fx Features
 The PPC405Fx core provides high performance and low power
consumption.
 The PPC405Fx RISC CPU executes at sustained speeds
approaching one cycle per instruction.
 On-chip instruction and data cache arrays can be implemented to
reduce chip count and design complexity in systems and improve
system throughput.

PPC405Fx Embedded
Processor
 The PowerPC RISC fixed-point CPU features:
 PowerPC User Instruction Set Architecture (UISA) and extensions for
embedded applications
 Thirty-two 32-bit general purpose registers (GPRs)
 Five-stage pipeline with single-cycle execution of most instructions, including
loads/stores
 Unaligned load/store support to cache arrays, main memory, and on-chip
memory (OCM)
 Hardware multiply/divide for faster integer arithmetic (4-cycle multiply, 35-
cycle divide)
 Multiply-accumulate instructions
 Enhanced string and multiple-word handling
 True little endian operation
 Parity detection and reporting for the instruction cache, data cache, and
translation lookaside buffer (TLB)
 Programmable Interval Timer (PIT), Fixed Interval Timer (FIT), and watchdog
timer

PPC405Fx Embedded
Processor
 Storage control :
 Separate, configurable, two-way set-associative instruction and data cache
units;
 the instruction cache array is 16KB and
 the data cache array is 16KB
 Eight words (32 bytes) per cache line
 Support for any combination of 0KB, 4KB, 8KB, and 16KB, and 32KB
instruction and data cache arrays, depending on model
 Read and write line buffers
 Instruction fetch hits are supplied from line buffer
 Data load/store hits are supplied to line buffer
 Programmable ICU prefetching of next sequential line into line buffer
 Programmable ICU prefetching of non-cacheable instructions, full line (eight
words) or half line (four words)
 Write-back or write-through DCU write strategies
 Programmable allocation on loads and stores
 Operand forwarding during cache line fills

PPC405Fx Embedded
Processor
 Memory Management
 Translation of the 4GB logical address space into physical addresses
 Independent enabling of instruction and data translation/protection
 Page level access control using the translation mechanism
 Software control of page replacement strategy
 WIU0GE (write-through, cachability, compressed user-defined 0,
guarded, endian) storage attribute control for each virtual memory
region
 WIU0GE storage attribute control for thirty-two real 128MB regions in
real mode
 Support for OCM that provides memory access performance identical
to cache hits
 Full PowerPC floating-point unit (FPU) support using the auxiliary
processor unit (APU) interface
(the PPC405Fx does not include an FPU)

PPC405Fx Embedded
Processor
 PowerPC timer facilities
 64-bit time base
 PIT, FIT, and watchdog timers
 Synchronous external time base clock input
 Debug Support
 Enhanced debug support with logical operators
 Four instruction address compares (IACs)
 Two data address compares (DACs)
 Two data value compares (DVCs)
 JTAG instruction to write to ICU
 Forward or backward instruction tracing
 Minimized interrupt latency
 Advanced power management support

PPC405Fx Embedded
Processor
 PowerPC Architecture
 The PowerPC Architecture comprises three levels of standards:
 PowerPC User Instruction Set Architecture (UISA), including the base user-level
instruction set, user level registers, programming model, data types, and
addressing modes.
 PowerPC Virtual Environment Architecture, describing the memory model, cache
model, cache-control instructions, address aliasing, and related issues. While
accessible from the user level, these features are intended to be accessed from
within library routines provided by the system software.
 PowerPC Operating Environment Architecture, including the memory
management model, supervisor level registers, and the exception model. These
features are not accessible from the user level.

PPC405Fx Embedded
Processor
 Processor Core Organization

PPC405Fx Embedded
Processor
 The processor core consists of a 5-stage pipeline, separate instruction
and data cache units, virtual memory management unit (MMU), three
timers, debug, and interfaces to other functions.
 Instruction and Data Cache Controllers
 The instruction cache unit (ICU) and data cache unit (DCU) enable
concurrent accesses and minimize pipeline stalls.
 The storage capacity of the cache units, which can range from 0KB–32KB,
depends upon the implementation. Both cache units are two-way set-
associative, use a 32-byte line size.
 The instruction set provides a rich assortment of cache control instructions,
including instructions to read tag information and data arrays.

PPC405Fx Embedded
Processor
 Instruction Cache Unit
 The ICU provides one or two instructions per cycle to the execution
unit (EXU) over a 64-bit bus. A line buffer enables the ICU to be
accessed only once for every four instructions, to reduce power
consumption by the array.
 The ICU can forward any or all of the words of a line fill to the EXU to
minimize pipeline stalls caused by cache misses.
 The ICU aborts speculative fetches abandoned by the EXU,
eliminating unnecessary line fills and enabling the ICU to handle the
next EXU fetch.
 Aborting abandoned requests also eliminates unnecessary external
bus activity to increase external bus utilization.

PPC405Fx Embedded
Processor
 Data Cache Unit
 The DCU transfers 1, 2, 3, 4, or 8 bytes per cycle, depending on the
number of byte enables presented by the CPU.
 The DCU contains a single-element command and store data queue
to reduce pipeline stalls; this queue enables the DCU to
independently process load/store and cache control instructions.
 Dynamic PLB request prioritization reduces pipeline stalls even
further. When the DCU is busy with a low-priority request while a
subsequent storage operation requested by the CPU is stalled, the
DCU automatically increases the priority of the current request to the
PLB.

PPC405Fx Embedded
Processor
 Data Cache Unit
 The DCU uses a two-line flush queue to minimize pipeline stalls caused by
cache misses. Line flushes are postponed until after a line fill is completed.
Registers comprise the first position of the flush queue; the line buffer built
into the output of the array for manufacturing test serves as the second
position of the flush queue.
 Single queued flushes are non-blocking. When a flush operation is pending,
the DCU can continue to access the array to determine subsequent load or
store hits.
 Requests abandoned by the CPU can also be aborted by the cache
controller.
 Additional DCU features enable the programmer to tailor performance for a
given application. The DCU can function in write-back or write-through mode,
as controlled by the Data Cache Write-through Register (DCWR) or the
translation look-aside buffer (TLB).

PPC405Fx Embedded
Processor
 Memory Management Unit
 The 4GB address space of the PPC405Fx is presented as a flat address space.
 The MMU provides address translation, protection functions, and storage attribute
control for embedded applications.
 Working with appropriate system level software, the MMU provides the following
functions:
 Translation of the 4GB logical address space into physical addresses
 Independent enabling of instruction and data translation/protection
 Page level access control using the translation mechanism
 Software control of page replacement strategy
 Additional control over protection using zones
 Storage attributes for cache policy and speculative memory access control
 The MMU can be disabled under software control. If the MMU is not used, the
PPC405Fx core provides other storage control mechanisms.

PPC405Fx Embedded
Processor
 Timer Facilities
The processor core contains a time base and three timers:
 Programmable Interval Timer (PIT)
 Fixed Interval Timer (FIT)
 Watchdog timer
The time base is a 64-bit counter incremented either by an internal signal equal
to the CPU clock rate or by a separate external timer clock signal.
The PIT is a 32-bit register that is decremented at the same rate as the time
base is incremented. The user loads the PIT register with a value to create the
desired delay. When a decrement occurs on a PIT count of 1, the timer stops
decrementing, a bit is set in the Timer Status Register (TSR), and a PIT interrupt
is generated. Optionally, the PIT can be programmed to reload automatically the
last value written to the PIT register, after which the PIT begins decrementing
again.

PowerPC 7xx
 The PowerPC 7xx is a family of third generation 32-bit PowerPC
microprocessors designed and manufactured by IBM and Motorola.
 The 7xx family is also widely used in embedded devices like printers,
routers, storage devices, spacecraft and video game consoles.
 The 7xx family had its shortcomings, namely lack of SMP (Symmetric
multiprocessing) support and SIMD capabilities and a relatively weak
FPU (Floating-point unit).

IBM 750CL RISC
Microprocessor
 The IBM 750CL PowerPC® RISC microprocessor is an implementation
of the PowerPC Architecture with enhancements to improve the floating
point performance and the data transfer capability .

IBM 750CL RISC
Microprocessor
 Overview
 750CL implements the 32-bit portion of the PowerPC Architecture, which
provides 32-bit effective addresses, integer data types of 8, 16, and 32
bits, and floating-point data types of single and double-precision.
 750CL is a superscalar processor that can complete two instructions
simultaneously. It incorporates the following six execution units:
 Floating-point unit (FPU)
 Branch processing unit (BPU)
 System register unit (SRU)
 Load/store unit (LSU)
 Two integer units (IUs): IU1 executes all integer instructions. IU2
executes all integer instructions except multiply and divide
instructions.

IBM 750CL RISC
Microprocessor
 750CL Microprocessor Features
 High-performance, superscalar microprocessor.
 Six independent execution units and two register files.
 Rename buffers.
 Completion unit.
 Separate on-chip L1 instruction and data caches (Harvard architecture).
 On-chip 1:1 L2 cache.
 DMA engine.
 Write gather pipe.
 ECC error correction for most single-bit errors, detection of double-bit errors.
 Separate memory management units (MMUs) for instructions and data.
 Bus interface features include the following:
 Multiprocessing support features
 Power and thermal management
 Performance monitor can be used to help debug system designs
 In-system testability and debugging features through JTAG boundary-scan
capability.

IBM 750CL RISC
Microprocessor
 PowerPC Instruction Set
 Integer instructions — These include computational and logical
instructions.
 Floating-point instructions — These include floating-point
computational instructions, as well as instructions that affect the
FPSCR.
 Load/store instructions — These include integer and floating-point
load and store instructions.
 Flow control instructions — These include branching instructions,
condition register logical instructions, trap instructions, and other
instructions that affect the instruction flow.
 Processor control instructions — These instructions are used for
synchronizing memory accesses and management of caches, TLBs,
and the segment registers.
 Memory control instructions — To provide control of caches, TLBs,
and SRs.

IBM 750CL RISC : Block Diagram

Spartan-3 FPGA
 Spartan-3 family of FPGA is specifically designed to meet the needs of
high volume, cost-sensitive consumer electronic applications.
 The Spartan-3 family has increased amount of
 logic resources,
 capacity of internal RAM,
 total number of I/Os and
 overall level of performance by improved clock management functions.
 Spartan-3 FPGA enhancements, combined with advanced process
technology, deliver more functionality and bandwidth than was previously
possible.
 Spartan-3 FPGAs are ideally suited to a wide range of consumer
electronics applications, including broadband access, home networking,
display/projection and digital television equipment.
 The Spartan-3 family is a superior alternative to mask programmed
ASICs.

Spartan-3 FPGA
 Features
 Low-cost, high-performance logic solution for high-volume, consumer-
oriented applications
 Select IO interface signaling
 Up to 633 I/O pins
 622+ Mb/s data transfer rate per I/O
 DDR, DDR2 SDRAM support up to 333 Mb/s
 Logic resources
 logic cells with shift register capability
 Wide, fast multiplexers
 Dedicated 18 x 18 multipliers
 JTAG logic compatible with IEEE 1149.1/1532

Spartan-3 FPGA
 Features
 Select RAM hierarchical memory
 Up to 1,872 Kbits of total block RAM
 Up to 520 Kbits of total distributed RAM
 Digital Clock Manager (up to four DCMs)
 Clock skew elimination
 Frequency synthesis
 High resolution phase shifting
 Eight global clock lines and abundant routing
 Fully supported by Xilinx ISE and WebPACK Software development systems
 MicroBlaze and PicoBlaze processor, PCI, PCI Express PIPE Endpoint, and
other IP cores.

Spartan-3 FPGA Attributes
CLB: Configurable Logic Block
DCM: Digital Clock Manager
I/O: Input Output

Spartan-3 Family Architecture
 Architectural Overview
 Configurable Logic Blocks (CLBs) contains flexible Look-Up Tables (LUTs) that
implement logic plus storage elements used as flip-flops or latches. CLBs perform a wide
variety of logical functions as well as store data.
 Input/Output Blocks (IOBs) controls the flow of data between the I/O pins and the
internal logic of the device. IOBs support bidirectional data flow plus 3-state operation.
Supports a variety of signal standards, including several high-performance differential
standards. Double Data-Rate (DDR) registers are included.
 Block RAM provides data storage in the form of 18-Kbit dual-port blocks.
 Multiplier Blocks accept two 18-bit binary numbers as inputs and calculate the product.
The Spartan-3A DSP platform includes special DSP multiply-accumulate blocks.
 Digital Clock Manager (DCM) Blocks provide self-calibrating, fully digital solutions for
distributing, delaying, multiplying, dividing, and phase-shifting clock signals.
 Digitally Controlled Impedance (DCI) feature provides automatic on-chip
terminations, simplifying board designs

Simplified IOB Diagram
 IOB Overview
 The Input/Output Block (IOB) provides a programmable, bidirectional interface
between an I/O pin and the FPGA’s internal logic.
 There are three main signal paths within the IOB: the output path, input path,
and 3-state path. Each path has its own pair of storage elements that can act as
either registers or latches. The three main signal paths are as follows:
 The input path carries data from the pad, which is bonded to a package pin, through an
optional programmable delay element directly to the line. The IOB outputs IQ1, and IQ2
all lead to the FPGA’s internal logic.
 The output path, starting with the O1 and O2 lines, carries data from the FPGA’s
internal logic through a multiplexer and then a three-state driver to the IOB pad.
 The 3-state path determines when the output driver is high impedance. The T1 and T2
lines carry data from the FPGA’s internal logic through a multiplexer to the output
driver. The output driver is active-Low enabled.
 All signal paths entering the IOB, including those associated with the storage elements,
have an inverter option.

Simplified IOB Diagram
 Storage Element Functions
 There are three pairs of storage elements in each IOB, one pair for each
of the three paths.
 It is possible to configure each of these storage elements as an edge-
triggered D-type flip-flop (FD) or a level-sensitive latch (LD).
 The storage-element-pair on either the Output path or the Three-State
path can be used together with a special multiplexer to produce Double-
Data-Rate (DDR) transmission. This is accomplished by taking data
synchronized to the clock signal’s rising edge and converting them to
bits synchronized on both the rising and the falling edge.
 The combination of two registers and a multiplexer is referred to as a
Double-Data-Rate D-type flip-flop (FDDR).

Arrangement of Slices within the
CLB

CLB
 All slices have the following elements in common:
 Two logic function generators,
 Two storage elements,
 Wide-function multiplexers,
 Carry logic, and
 Arithmetic gates,
 The left-hand pair supports two additional functions:
 Storing data using Distributed RAM and
 Shifting data with 16-bit registers.
 The RAM-based function generator—also known as a Look-Up Table or
LUT—is the main resource for implementing logic functions.
 The LUTs in each left-hand slice pair can be configured as Distributed
RAM or a 16-bit shift register.
 The function generators located in the upper and lower portions of the
slice are referred to as the "G" and "F", respectively.

CLB
 The storage elements in the upper and lower portions of the slice are
called FFY and FFX, respectively.
 Wide-function multiplexers effectively combine LUTs in order to permit
more complex logic operations. Each slice has two of these multiplexers
with F5MUX in the lower portion of the slice and FiMUX in the upper
portion.
 The carry chain, together with various dedicated arithmetic logic gates,
support fast and efficient implementations of math operations.
 Five multiplexers control the chain: CYINIT, CY0F, and CYMUXF in the
lower portion as well as CY0G and CYMUXG in the upper portion.
 The dedicated arithmetic logic includes the exclusive-OR gates XORG
and XORF as well as the AND gates GAND and FAND.

PicoBlaze
 The PicoBlaze microcontroller is a compact, capable and cost-effective
fully embedded 8-bit RISC microcontroller core optimized for the
Spartan-3 family.
 It also provides support for the Virtex-5, Spartan-6, and Virtex-6 FPGA
families.
 The PicoBlaze microcontroller provides cost-efficient microcontroller-
based control and simple data processing.
 The PicoBlaze microcontroller is optimized for efficiency and low
deployment cost.
 It occupies just 96 FPGA slices, (only 12.5% of an XC3S50 FPGA).
 Typically a single FPGA block RAM stores up to 1024 program
instructions, which are automatically loaded during FPGA configuration.
 The PicoBlaze microcontroller performs a respectable 44 to 100 million
instructions per second (MIPS) depending on the target FPGA family
and speed grade.

PicoBlaze
 The PicoBlaze microcontroller core is totally embedded within the target
FPGA and requires no external resources.
 The PicoBlaze microcontroller is extremely flexible.
 The basic functionality is easily extended and enhanced by connecting
additional FPGA logic to the microcontroller’s input and output ports.
 The PicoBlaze peripheral set can be customized to meet the specific
features, function, and cost requirements of the target application.
 PicoBlaze microcontroller is delivered as synthesizable VHDL source
code, the core is future-proof and can be migrated to future FPGA
architectures.
 Being integrated within the FPGA, the PicoBlaze microcontroller reduces
board space, design cost, and inventory.

Why the PicoBlaze
Microcontroller
 There are literally dozens of 8-bit microcontroller architectures and
instruction sets.
 The PicoBlaze microcontroller is specifically designed and optimized for the
Spartan-3 family, and with support for Spartan-6, and Virtex-6 FPGA
architectures.
 It is compact, yet capable architecture consumes considerably less FPGA
resources than comparable 8-bit microcontroller architectures within an
FPGA.
 Furthermore, the PicoBlaze microcontroller is provided as a free, source-
level VHDL file with royalty-free re-use within Xilinx FPGAs.
 Because it is delivered as VHDL source, the PicoBlaze microcontroller is
immune to product obsolescence as the microcontroller can be retargeted to
future generations of Xilinx FPGAs, exploiting future cost reductions and
feature enhancements.
 Furthermore, the PicoBlaze microcontroller is expandable and extendable.

Why the PicoBlaze
Microcontroller
 Before the advent of the PicoBlaze and MicroBlaze embedded processors,
the microcontroller resided externally to the FPGA, limiting the connectivity
to other FPGA functions and restricting overall interface performance.
 By contrast, the PicoBlaze microcontroller is fully embedded in the FPGA
with flexible, extensive on-chip connectivity to other FPGA resources.
 Signals remain within the FPGA, improving overall performance.
 The PicoBlaze microcontroller reduces system cost because it is a single-
chip solution, integrated within the FPGA and sometimes only occupying
leftover FPGA resources.
 The PicoBlaze microcontroller is resource efficient. Consequently, complex
applications are sometimes best portioned across multiple PicoBlaze
microcontrollers with each controller implementing a particular function, for
example, keyboard and display control, or system management.

Why Use a Microcontroller within an
FPGA?
 Microcontrollers and FPGAs both successfully implement practically any
digital logic function. Each has unique advantages in cost, performance, and
ease of use.
 Microcontrollers are well suited to control applications, especially with widely
changing requirements.
 The FPGA resources required to implement the microcontroller are relatively
constant. The same FPGA logic is re-used by the various microcontroller
instructions, conserving resources.
 The program memory requirements grow with increasing complexity.
Programming control sequences or state machines in assembly code is
often easier than creating similar structures in FPGA logic.
 As an application increases in complexity, the number of instructions
required to implement the application grows and system performance
decreases accordingly.

FPGA?
 FPGA is more flexible than microcontroller.
For example, an algorithm can be implemented sequentially or completely in
parallel, depending on the performance requirements.
A completely parallel implementation is faster but consumes more FPGA
resources.
 A microcontroller embedded within the FPGA provides the best of both
worlds. The microcontroller implements non-timing crucial complex control
functions while timing critical or data path functions are best implemented
using FPGA logic.
 For example, a microcontroller cannot respond to events much faster than a few
microseconds. The FPGA logic can respond to multiple, simultaneous events in just a
few to tens of nanoseconds. Conversely, a microcontroller is cost-effective and simple
for performing format or protocol conversions.

FPGA?
PicoBlaze Microcontroller FPGA Logic
Strengths  Easy to program, excellent for
control and state machine
applications
 Resource requirements remain
constant with increasing
complexity
 Re-uses logic resources,
excellent for lower-performance
functions
 Significantly higher performance
 Excellent at parallel operations
 Sequential vs. parallel
implementation tradeoffs optimize
performance or cost
 Fast response to multiple,
simultaneous inputs
Weaknesses  Executes sequentially
 Performance degrades with
increasing complexity
 Program memory requirements
increase with increasing
complexity
 Slower response to simultaneous
inputs
 Control and state machine
applications more difficult to program
 Logic resources grow with increasing
 Complexity

PicoBlaze Microcontroller
Features
 16 byte-wide general-purpose data registers
 1K instructions of programmable on-chip program store, automatically
loaded during FPGA configuration
 Byte-wide Arithmetic Logic Unit (ALU) with CARRY and ZERO indicator
flags
 64-byte internal scratchpad RAM
 256 input and 256 output ports for easy expansion and enhancement
 Automatic 31-location CALL/RETURN stack
 Predictable performance, always two clock cycles per instruction, up to
200 MHz or 100 MIPS in a Virtex-II Pro FPGA
 Fast interrupt response; worst-case 5 clock cycles
 Optimized for Xilinx Spartan-3 architecture—just 96 slices and 0.5 to 1
block RAM
 Support in Spartan-6, and Virtex-6 FPGA architectures
 Assembler, instruction-set simulator support

PicoBlaze Microcontroller Functional
Blocks
 General-Purpose Register
 The PicoBlaze microcontroller includes 16 byte-wide general-purpose
registers, designated as registers s0 through sF. For better program
clarity, registers can be renamed using an assembler directive. All
register operations are completely interchangeable.
 There is no dedicated accumulator; each result is computed in a
specified register.
 1,024-Instruction Program Store
 The PicoBlaze microcontroller executes up to 1,024 instructions from
memory within the FPGA. Each PicoBlaze instruction is 18 bits wide.
The instructions are compiled within the FPGA design and automatically
loaded during the FPGA configuration process.
 Other memory organizations are possible to accommodate more
PicoBlaze controllers within a single FPGA or to enable interactive code
updates without recompiling the FPGA design.

Blocks
 Arithmetic Logic Unit (ALU)
 The byte-wide Arithmetic Logic Unit (ALU) performs all microcontroller
calculations, including:
 basic arithmetic operations such as addition and subtraction
 bitwise logic operations such as AND, OR, and XOR
 arithmetic compare and bitwise test operations
 comprehensive shift and rotate operations
 All operations are performed using an operand provided by any specified
register (sX). The result is returned to the same specified register (sX). If an
instruction requires a second operand, then the second operand is either a
second register (sY) or an 8-bit immediate constant (kk).
 Flags
 ALU operations affect the ZERO and CARRY flags.
 The ZERO flag indicates when the result of the last operation resulted in zero.
 The CARRY flag indicates various conditions, depending on the last
instruction executed.
 The INTERRUPT_ENABLE flag enables the INTERRUPT input.

Blocks
 64-Byte Scratchpad RAM
 The PicoBlaze microcontroller provides an internal general-purpose
64-byte scratchpad RAM, directly or indirectly addressable from the
register file using the STORE and FETCH instructions.
 The STORE instruction writes the contents of any of the 16 registers
to any of the 64 RAM locations.
 The complementary FETCH instruction reads any of the 64 memory
locations into any of the 16 registers.
 The six-bit scratchpad RAM address is specified either directly (ss)
with an immediate constant, or indirectly using the contents of any of
the 16 registers (sY).
 Only the lower six bits of the address are used; the address should
not exceed the 00 - 3F range of the available memory.

Blocks
 Input/Output
 The Input/Output ports extend the PicoBlaze microcontroller’s
capabilities and allow the microcontroller to connect to a custom
peripheral set or to other FPGA logic.
 The PicoBlaze microcontroller supports up to 256 input ports and 256
output ports or a combination of input/output ports.
 The PORT_ID output provides the port address.
 During an INPUT operation, the PicoBlaze microcontroller reads data from the
IN_PORT port to a specified register, sX.
 During an OUTPUT operation, the PicoBlaze microcontroller writes the
contents of a specified register, sX, to the OUT_PORT port.

Blocks
 Program Counter (PC)
 The Program Counter (PC) points to the next instruction to be executed. By default, the
PC automatically increments to the next instruction location when executing an
instruction.
 Only the JUMP, CALL, RETURN instructions and the Interrupt and Reset Events
modify the default behavior. The PC cannot be directly modified by the application
code. The 10-bit PC supports a maximum code space of 1,024 instructions (000 to 3FF
hex). If the PC reaches the top of the memory at 3FF hex, it rolls over to location 000.
 Program Flow Control
 The default execution sequence of the program can be modified using conditional and
non-conditional program flow control instructions.
 The JUMP instructions specify an absolute address anywhere in the 1,024-instruction
program space.
 CALL and RETURN instructions provide subroutine facilities for commonly used
sections of code.
 If the interrupt input is enabled, an Interrupt Event also preserves the address of the
preempted instruction on the CALL/RETURN stack while the PC is loaded with the
interrupt vector, 3FF hex.

Blocks
 CALL/RETURN Stack
 The CALL/RETURN hardware stack stores up to 31 instruction addresses, enabling
nested CALL sequences up to 31 levels deep.
 The stack is implemented as a separate cyclic buffer. When the stack is full, it
overwrites the oldest value. No program memory is required for the stack.
 Interrupts
 The optional INTERRUPT input, allows the PicoBlaze microcontroller to handle
asynchronous external events. “Asynchronous” relates to interrupts occurring at any
time during an instruction cycle.
 However, recommended design practice is to synchronize all inputs to the PicoBlaze
controller using the clock input.
 The PicoBlaze microcontroller responds to interrupts quickly in just five clock cycles.
 Reset
 The PicoBlaze microcontroller is automatically reset immediately after the FPGA
configuration process completes. After configuration, the RESET input forces the
processor into the initial state. The PC is reset to address 0, the flags are cleared,
interrupts are disabled, and the CALL/RETURN stack is reset.

MicroBlaze Processor
 The MicroBlaze embedded processor soft core is a reduced instruction set
computer (RISC) optimized for implementation in Xilinx Field Programmable
Gate Arrays (FPGAs).
 In terms of its instruction-set architecture, MicroBlaze is very similar to the
RISC-based DLX architecture.
 With few exceptions, the MicroBlaze can issue a new instruction every
cycle, maintaining single-cycle throughput under most circumstances.
 MicroBlaze's primary I/O bus, the CoreConnect PLB bus, is a traditional
system-memory mapped transaction bus with master/slave capability.
 For access to local-memory (FPGA BRAM), MicroBlaze uses a dedicated
LMB bus, which reduces loading on the other buses.
 User-defined coprocessors are supported through a dedicated FIFO-style
connection called FSL (Fast Simplex Link). The coprocessor(s) interface can
accelerate computationally intensive algorithms.

 Many aspects of the MicroBlaze can be user configured:
 cache size,
 pipeline depth (3-stage or 5-stage),
 embedded peripherals,
 memory management unit, and
 bus-interfaces can be customized.
 The area-optimized version of MicroBlaze, which uses a 3-stage pipeline,
sacrifices clock-frequency for reduced logic-area.
 The performance-optimized version expands the execution-pipeline to 5-
stages, allowing top speeds of 210 MHz
 Also, key processor instructions which are rarely used but more expensive
to implement in hardware can be selectively added/removed
 This customization enables a developer to make the appropriate design
tradeoffs for a specific set of host hardware and application software
requirements.

 With the memory management unit, MicroBlaze is capable of hosting
operating systems requiring hardware-based paging and protection,
such as the Linux kernel.
 Otherwise it is limited to operating systems with a simplified protection
and virtual memory-model: e.g. Free RTOS or Linux without MMU
support.
 MicroBlaze's overall throughput is substantially less than a comparable
hardened CPU-core (such as the PowerPC440 in the Virtex-5.)

MicroBlaze
 Features
 The MicroBlaze soft core processor is highly configurable, allowing you
to select a specific set of features required by your design.
 The fixed feature set of the processor includes:
 Thirty-two 32-bit general purpose registers
 32-bit instruction word with three operands and two addressing modes
 32-bit address bus
 Single issue pipeline
 In addition to these fixed features, the MicroBlaze processor is
parameterized to allow selective enabling of additional functionality.

MicroBlaze
 Data Types and Endianness
 MicroBlaze uses Big-Endian bit-reversed format to represent data.
The hardware supported data types for MicroBlaze are word, half
word, and byte.
Word Data Type
Half Word Data Type
Byte Data Type

MicroBlaze
 Instructions
 All MicroBlaze instructions are 32 bits and are defined as either Type A or Type
B.
 Type A instructions have up to two source register operands and one destination
register operand.
 Type B instructions have one source register and a 16-bit immediate operand
(which can be extended to 32 bits by preceding the Type B instruction with an
imm instruction).
 Type B instructions have a single destination register operand.
 Instructions are provided in the following functional categories:
 arithmetic,
 logical,
 branch,
 load/store, and
 special.

MicroBlaze
 Registers
 MicroBlaze has an orthogonal instruction set architecture. It has thirty-
two 32-bit general purpose registers and up to eighteen 32-bit special
purpose registers, depending on configured options.
1. General Purpose Registers
The thirty-two 32-bit General Purpose Registers are numbered
R0 through R31. The register file is reset on bit stream download
(reset value is 0x00000000).

MicroBlaze
2. Special Purpose Registers
Program Counter (PC)
The Program Counter (PC) is the 32-bit address of the execution
instruction.
When used with the MFS instruction the PC register is specified by
setting Sa = 0x0000.

MicroBlaze
Machine Status Register (MSR)
The Machine Status Register contains control and status bits for the
processor.
When reading the MSR, bit 29 is replicated in bit 0 as the carry copy.
When writing to the MSR, the Carry bit takes effect immediately
and the remaining bits take effect one clock cycle later.
The MSR is specified by setting Sx = 0x0001.

MicroBlaze
Exception Address Register (EAR)
The Exception Address Register stores the full load/store address that caused the
exception for the following:
The contents of this register is undefined for all other exceptions.
- - The EAR is specified by setting Sa = 0x0003.
- - An unaligned access exception that means the unaligned access address
- - A DPLB or DOPB exception that specifies the failing PLB or OPB data access address
- - A data storage exception that specifies the (virtual) effective address accessed
- - An instruction storage exception that specifies the (virtual) effective address read
- - A data TLB miss exception that specifies the (virtual) effective address accessed
- - An instruction TLB miss exception that specifies the (virtual) effective address read

MicroBlaze
Exception Status Register (ESR)
The Exception Status Register contains status bits for the processor.
The ESR is specified by setting Sa = 0x0005.
Branch Target Register (BTR)
The Branch Target Register only exists if the MicroBlaze processor is
configured to use exceptions.
The register stores the branch target address for all delay slot branch
instructions executed while MSR[EIP] = 0.
The BTR is specified by setting Sa = 0x000B.

MicroBlaze
Floating Point Status Register (FSR)
The Floating Point Status Register contains status bits for the floating
point unit.
The register is specified by setting Sa = 0x0007.
Exception Data Register (EDR)
The Exception Data Register stores data read on an FSL link that
caused an FSL exception.
The contents of this register is undefined for all other exceptions.
The EDR is specified by setting Sa = 0x000D.

MicroBlaze
Zone Protection Register (ZPR)
The Zone Protection Register is used to override MMU memory
protection defined in TLB entries.
Translation Look-Aside Buffer Low Register (TLBLO)
Translation Look-Aside Buffer High Register (TLBHI)
Translation Look-Aside Buffer Index Register (TLBX)
Translation Look-Aside Buffer Search Index Register (TLBSX)
Processor Version Register (PVR)

MicroBlaze
 Pipeline Architecture
 MicroBlaze instruction execution is pipelined. For most instructions, each
stage takes one clock cycle to complete.
 Consequently, the number of clock cycles necessary for a specific
instruction to complete is equal to the number of pipeline stages, and
one instruction is completed on every cycle.
 A few instructions require multiple clock cycles in the execute stage to
complete.
 When executing from slower memory, instruction fetches may take
multiple cycles.
 MicroBlaze implements an instruction prefetch buffer that reduces the
impact of such multi-cycle instruction memory latency.
 When the pipeline resumes execution, the fetch stage can load new
instructions directly from the prefetch buffer instead of waiting for the
instruction memory access to complete.

MicroBlaze
 Pipeline Architecture
Three Stage Pipeline
Five Stage Pipeline
Fetch (IF), Decode (OF), Execute (EX), Access Memory (MEM), and Writeback (WB).

MicroBlaze
 Memory Architecture
 MicroBlaze is implemented with a Harvard memory architecture;
instruction and data accesses are done in separate address spaces.
Each address space has a 32-bit range (that is, handles up to 4-GB of
instructions and data memory respectively).
 Both instruction and data interfaces of MicroBlaze are 32 bits wide and
use big endian, bit-reversed format.
 MicroBlaze supports word, halfword, and byte accesses to data memory.
 Data accesses must be aligned, unless the processor is configured to
support unaligned exceptions.
 All instruction accesses must be word aligned.

Processors used in System on chip

Recomendados

Recomendados

Mais conteúdo relacionado

Mais procurados

Mais procurados (20)

Destaque

Destaque (19)

Semelhante a Processors used in System on chip

Semelhante a Processors used in System on chip (20)

Mais de A B Shinde

Mais de A B Shinde (20)

Último

Último (20)

Processors used in System on chip