MSC8156SAG1000B - Freescale Semiconductor

Freescale Semiconductor

Product Brief Document Number: MSC8156PB

Rev. 3, 5/2011

Contents

The MSC8156 device is the fourth generation of

Freescale high-end multicore DSP devices that target the

broadband wireless base stations and other

communications infrastructure. It builds upon the proven

success of the previous multicore DSPs and is designed

to bolster the rapidly changing and expanding wireless

markets, such as 3G-LTE (FDD and TDD), WiMAX,

3GPP/3GPP2, and TD-SCDMA.

1 Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

1.1 Block Diagram. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

1.2 Crit i cal Perfor man ce Metrics. . . . . . . . . . . . . . . . . . 4

1.3 Devi c e Level Featu re s. . . . . . . . . . . . . . . . . . . . . . . 4

1.4 Module Level Features . . . . . . . . . . . . . . . . . . . . . . 5

2 Developer Environment. . . . . . . . . . . . . . . . . . . . . . . . . . 14

2.1 Tool s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

2.2 Appl i ca tion S o ft w a r e . . . . . . . . . . . . . . . . . . . . . . . 15

MSC8156 Product Brief

Broadband Wireless Access DSP

MSC8156 Product Brief, Rev. 3

Features

Freesca le Sem ico nd uctor2

1Features

The MSC8156 device targets high-bandwidth highly computational DSP applications and is optimized for

3GPP, TD-SCDMA, 3G-LTE, and WiMAX applications.

1.1 Block Diagra m

The MSC8156 devices are highly integrated DSP processors that contain six StarCore® SC3850 DSP

subsystems with 32 Kbyte L1 instruction cache per core, 32 Kbyte L1 data cache per core, 512 Kbyte L2

unified instruction/data cache per core (can be configured as M2 Memory); 1056 Kbyte of shared M3,

memory; two DDR memory controllers; a multi-accelerator platform engine (MAPLE-B) for Turbo

decoding, Viterbi decoding, Fast Fourier Transform and Discrete Fourier Transform acceleration; two

serial RapidIO interfaces; two Gigabit Ethernet controllers; a PCI-Express controller; four 256-channel

time-division multiplexing (TDM) interfaces; a 16 bidirectional channels DMA controller; an SPI

interface; a UART interface; and an I2C interface. Each SC3850 DSP core has four ALUs each with a dual

16 × 16 MAC per ALU and performs at 8000 million multiply accumulates per second (MMACS) at 1

GHz yielding a maximum total performance of 48000 MMACS per device.

In each SC3850 core subsystem, the SC3850 core connects to the following:

• 32 Kbyte 8-way level 1 instruction cache (L1 ICache)

• 32 Kbyte 8-way level 1 data cache (L1 DCache)

• 512 Kbyte 8-way level 2 unified instruction/data cache (L2 Cache/M2 Memory)

• Memory management unit (MMU)

• Enhanced programmable interrupt controller (EPIC)

• Debug and profiling unit (DPU)

• Two 32-bit timers

Features

MSC8156 Product Brief, Rev. 3

Freescale Semiconductor 3

A block diagram of the MSC8156 is shown in Figure 1. A separate block diagram for the SC3850 DSP

core platform is shown in Figure 2.

Figure 1. MSC8156 Block Diagram

Figure 2. StarCore SC3850 DSP Subsystem Block Diagram

JTAG

Note: The arrow direction indicates master or slave.

DDR Interface 64/32-bit 800 MHz

4 TDMs

DMA 32 ch

I/O-Interrupt

Concentrator

UART

Clocks

Timers

Reset

Semaphores

Other

DDR

CLASS

High-Speed

Modules

QUICC

Four TDMs each supporting 8 E1

Boot ROM

I2C

Virtual

Interrupts

Controller

SPI

MAPLE-B

Twwo Serial RapidIO 1x/4x 3.125 Gbaud

Six DSP Cores at 1 GHz

Turbo/ Two SGMII

Viterbi FFT/

IFFT

DFT/

IDFT

Two RGMII

M3 Memory

1056 Kbyte

PCI-Express 1x/2x/4x

Subsystem

SC3850

DSP Core

512 Kbyte

32 Kbyte 32 Kbyte

ICache L1

DCache

L2 Cache / M2 Memory

DDR Interface 64/32-bit 800 MHz

DDR

Controller

Two SGMII

Engine™ Serial

Interface

CRCU

Dual RISC Engine

Aggre ga ted ban dw idth of 240 Mbps

32 Kbyte Address

Translation

Task

Protection

32 Kbyte

(WTB) (WBB)

EPIC

Interrupts

P-bus 128 bit

Xa-bus 64 bit

Xb-bus 64-bit

DQBus

Debug Support

OCE30

512 Kbyte L2 Cache / M2 Memory

MMU

Timer

128 bits master

IQBus

DPU

SC3850

Core

TWB

Write-

Through

Buffer

Write-

Back

Buffer

Instruction

Cache Data

Cache

bus to CLASS 128 bits slave

bus from CLASS

MSC8156 Product Brief, Rev. 3

Features

Freesca le Sem ico nd uctor4

1.2 Critic al Performan ce Metric s

• Offered with a core frequency of 1 GHz, supports:

— Eight 16 × 16 or 8 × 8 multipliers, enabling up to 48000 MMACS at 1 GHz with six SC3850

cores. A multiply-accumulate operation includes a multiply-add instruction with the associated

data move and pointer update. This i s double MMACS versus the pervious generation SC3400

DSP core.

— The six cores deliver a performance equivalent to a single core running at 6 GHz.

• The MAPLE-B contains a highly flexible and programmable Turbo and V iterbi decoder supporting

various configurable decoding parameters (for exa mple, polynomia ls and rate). It can perform up

to 200 Mbps of Turbo decoding (6 iterations) or up to 115 Mbps of K = 9 (zero tail) Viterbi

decoding, while supporting various standards such as: 3GPP-R6, 3GPP2, WiMAX and 3G-LTE.

• The MAPLE-B also supports:

— FFT/iFFT for sizes 128, 256, 512, 1024, or 2048 points at up to 350 Million samples per second

— DFT/iDFT for sizes up to 1536 points at up to 175 Million samples per second (for 1200 points)

• Dual RISC core QUICC Engine subsystem operating at up to 500 MHz provides parallel packet

processing independent of the DSP cores, allowing the cores to process data while the RISC

engines manage the data flow and packetization.

• Power supplies:

— Core power: 1 V nominal

— I/O power: 1.0 V, 1.5 V, 1.8 V and 2.5 V nominal

• Flip Chip-Plastic Ball Grid Array (FC-PBGA), 783-ball, 1 mm pitch, 29 mm × 29 mm

1.3 Device Level Features

This multicore DSP delivers a high level of performance and integration, combining six

fully-programmable StarCore DSP cores, each running at up to 1 GHz with an architecture highly

optimized for wireless applications. The MAPLE-B supports hardware acceleration for T urbo and Viterbi

channel decoding and for DFT/iDFT and FFT/iFFT algorithms. An internal RISC-based QUICC Engine™

subsystem supports multiple networking protocols to guarantee reliable data transport over packet

networks while significantly off loading such processing from the DSP cores. The MSC8156 embeds large

internal memory and supports a variety of advanced interface types, including two RapidIO interfaces, two

gigabit Ethernet interfaces for network communications, a PCI-Express controller, two DDR controllers

for high-speed, industry-standard memory interface and a multi-channel TDM interfaces. The highly

flexible, fully-programmable and powerful MSC8156 broadband wireless access DSP of fers tremendous

processing power while maintaining a competitive price and power.

Features

MSC8156 Product Brief, Rev. 3

Freescale Semiconductor 5

1.4 Module Level Features

• StarCore DSP subsystem. The DSP subsystem includes:

— StarCore SC3850 core

– Running at up to 1 GHz

– Up to 8000 16-bit or 8-bit MMACS. A MAC operation includes a multiply-accumulate

command with the associated data moves and a pointer update.

– Backwards binary compatible with the SC140 and SC3400 architectures.

– Data Arithmetic a nd Logic Unit (DALU) contai ning 4 ALUs, each capable of performing 2

16 × 16 multiply accumulate operations, effectively doubling the performance of

convolution-based kernels relative to the SC3400 core

– New instructions double the performance of complex and extended precision multiplication.

– Address Generating Unit (AGU) containing 2 Address Arithmetic Units (AAU)

– Up to six instructions executed in a single clock cycle: 4 DALU and 2 AGU instructions

– Variable-length Execution Set (VLES) execution model.

– 16 data registers, 40 bits each; 27 address registers, 32 bits each.

– Hardware support for fractional and integer data types.

– Four hardware loops with near-zero cycle overhead

– Very rich 16-bit wide orthogonal instruction set.

– Application specific instructions for Viterbi and Multimedia.

– Special SIMD (Single instruction, mult iple data) instructions working on 2-word or 4-byte

operands packed in a register , enabl ing to perform 2 to 4 operations per instruction (8 to 16

operations per VLES)

– New dedicated instructions accelerate FFTs enabling a 40% cycle count reduction and

improved SNR

– User and Supervisor privilege levels, supporting a protected SW model

– New instructions and features to improve control code performance

– Precise exceptions for memory accesses enabling good RTOS support and Soft Error

corrections

– Branch Target Buffer (BTB) for acceleration of change of flow operations

— L1 ICache:

– 32 Kbytes

– 8 ways with 16 lines of 256 bytes per way

– Multi-task support

– Real-time support through locking flexible boundaries

– Line pre-fetch capability

– Software coherency support

– Software pre-fetch support by core instructions

— L1 DCache:

– 32 Kbytes

– 8 ways with 16 lines of 256 bytes per way

– Capable of serving two data accesses in parallel (XA, XB)

– Multi-task support

– Real-time support through locking flexible boundaries

– Software coherency support

MSC8156 Product Brief, Rev. 3

Features

Freesca le Sem ico nd uctor6

– Writing policy programmable per memory segment as either write-back or write-through

– 0.25 Kbytes Write-back Buffer (WBB)

– Six 64bit entry WTB

– Line pre-fetch capability

– Software pre-fetch, synchronize, and flush support by core instructions

— Unified L2 Cache/M2 Memory:

– 512 Kbyte

– 8 ways with 1024 indexes and a 64 byte line

– Physically addressed

– Dynamically configured as a DMA accessible M2 Memory

– Maximum user flexibility for real time support through address partitioning of the cache

– Support various write policies and methods to reduce cache inclusiveness

– Multi-channel, two dimensional software pre-fetch support

– Software coherency support with seamless transition from L1 cache coherency operation.

— Memory management unit (MMU):

– Highly flexible memory mapping capability

– Provides virtual to physical address translation

– Provides task protection

– Supports multi-tasking

– Supports precise interrupts. Enabling to have an open RTOS.

— Debug and Profiling Unit (DPU) block:

– Supports the debugging and profiling of the platform in cooperation with the OCE Block

– Supports various breakpoint and event counting options

– Supports real-time tracing to the main memory with the Trace Write Buffer (TWB)

— Extended programmable interrupt controller (EPIC)

– 256 interrupts

– 32 priority levels with NMI support

— Two general-purpose 32-bit timers

— Low-power design with the following modes of operation:

– Wait processing state for peripheral operation

– Stop processing state

– Power down processing state

— ECC/EDC support.

• Chip-level arbitration and switching system (CLASS)

— A full fabric that arbitrates between the DSP cores and other CLASS masters to the core M2

memory, shared M3 memory, DDR SDRAM controllers, MAPLE-B, and the device

configuration control and status registers (CCSRs).

— High bandwidth.

— Non-blocking allows parallel accesses from multiple initiators to multiple targets.

— Fully pipelined.

— Low latency.

— Per target arbitration highly optimized to the target characteristics using prioritized round-robin

arbitration.

— Reduces data flow bottlenecks and enables high-bandwidth internal data transfers.

Features

MSC8156 Product Brief, Rev. 3

Freescale Semiconductor 7

• Internal memory. The 4608 Kbyte for MSC8156 internal memory space includes:

— 32 Kbyte L1 ICache per core.

— 32 Kbyte L1 DCache per core.

— 512 Kbyte unified L2 Cache/M2 Memory per core.

— 1056 Kbyte shared M3 memory. 1024 Kbyte of M3 memory can be turned off to save power,

if necessary, which reduces the M3 memory size to 32 Kbyte.

— 96 Kbyte boot ROM accessible from the cores.

•Clocks

— Three input clocks:

– Global input clock.

– Two differential input clocks (one per each serial RapidIO PLL).

— Five PLLs:

– Three system PLLs

– Two serial RapidIO PLLs.

— Clock ratios selected during reset via reset configuration pins.

— Clock modes user-configurable after reset.

• Multi Accelerator Platform Engine for Baseband (MAPLE-B)

– Programmable System Interface

• Software friendly buffer descriptor based handshake and task assignment.

• Support for high priority and low priority tasks via multiple descriptor rings.

• Processing elements management and scheduling.

• Two master buses for data transfers from/to the system memory at total throughput up to

50 Gbps.

• One slave bus for accessing MAPLE-B internal memories and registers.

• Multi-Core Aware.

• Interrupt or RapidIO Door Bell generation and/or status bit i ndication on job or multiple

jobs completion.

• System memory utilized only for input/output data, all the internal calculations are

performed using MAPLE-B memories.

– Turbo Decoding

• Scalable architecture with 1, 2 or 4 Radix 4 dual-recursion engines.

• Supports 3GPP-R6 turbo decoding as specified in 3GPP TS 25.212, section 4.2.3.2.

• Supports 3GPP2 turbo decoding as specified in 3GPP2 C.S0002, section 2.1.3.1.4.2.

• Supports Wimax OFDMA turbo decoding as specified in IEEE® 802.16™-2004

standard and corrigendum IEEE P802.16-2004/Cor2/D2 standard, section 8.4.9.2.3.

• Supports 3GLTE (Evolved UTRA) turbo decoding as specified in 3GPP TS 36.212,

section 5.1.2.2.

• Programmable Number of Iterations

• Multiple stop conditions, including built in CRC check and A-Posteriori Quality

Indication.

MSC8156 Product Brief, Rev. 3

Features

Freesca le Sem ico nd uctor8

• Programmable de-puncturing schemes including support for rate-de-matching in 3GPP

(and 3G LTE).

• Binary and Duo-Binary turbo codes.

• Support for trellis termination bits and tail biting.

• Decoding techniques using: Max Log Map or Linear Log Map (MAX*) including

extrinsic factorization.

• 8-bit soft symbol inputs with Hard or Soft decision outputs.

– Viterbi Decoding

• Supports 3GPP-R6 viterbi decoding as specified in 3GPP TS 25.212, section 4.2.3.1.

• Supports 3GPP2 viterbi decoding as specified in 3GPP2 C.S002, section 2.1.3.1.4.1.

• Supports Wimax OFDMA channel decoding as specified in IEEE 802.16-2004 and

corrigendum IEEE P802.16-2004/Cor2/D2, section 8.4.9.2.1.

• Supports 3GLTE (Evolved UTRA) channel decoding as specified in 3GPP TS 36.212,

section 5.1.2.1.

• Fully programmable polynomials

• Programmable schemes for supporting various rates/puncturing cases (e.g. 1/2, 1/3).

• Support for zero tailing.

• Support for tail-biting with multiple-iteration or WAVA* approaches.

• 8-bit soft symbol inputs with Hard or Soft decision outputs.

– FFT/iFFT and DFT/iDFT processing

• Variable lengths FFT/iFFT processing of lengths 128, 256, 512, 1024 and 2048.

• Variable lengths DFT/iDFT processing of the form 2k·3m·5n·12, up to 1536 points.

• Mixed radix implementation using R2, R3, R4, R5 and R8 building blocks.

• 16-bit I, 16-bit Q input, output and twiddles resolutions.

• Input and output I/Q samples are interleaved in memory.

• Internal twiddles memory.

– CRC processing with the following features:

• Four possible polynomials

• CRC check or CRC calculation for block sizes of up to 128 Kb

• Optional byte reverse CRC processing

• CRC processing with throughput of up to 12 Gbps at 450 MHz

– When it is not required, the MAPLE-B power can be disabled internally to reduce overall

device power consumption.

• Two DDR Controllers, each supporting:

— Up to 400 MHz clock rate (800 MHz data rate).

— Supports both DDR2 and DDR3 devices

— Programmable timing supporting both DDR2 and DDR3 SDRAM (but not simultaneously)

— Support for a 64-bit data interface (72 bits including ECC), up to 800 MHz data rate, for DDR2

and DDR3

Features

MSC8156 Product Brief, Rev. 3

Freescale Semiconductor 9

— Support for a 32-bit data interface (40 bits including ECC), up to 800 MHz data rate, for DDR2

and DDR3

— Full ECC support for single-bit error correction and multi-bit error detection up to the

maximum specified data rates for DDR2 and DDR3

— Two banks of memory via two chip selects. Each chip select supports up to 1 Gbytes, the sum

of the memory cannot exceed 1 Gbyte total (2 Gbyte total for the two controllers).

— DRAM chip configurations from 64 Mbits to 4 Gbits with x8/x16 data ports

— Support burst lengths of 4 beats for DDR2 devices

— Support burst lengths of 4 (burst chop) and On the Fly for DDR3 devices

— Sleep mode support for self-refresh SDRAM

— On-die termination support

— Supports auto refreshing

— Support for SODIMMs

• High-Speed Serial Interface (HSSI)

— Serial RapidIO® Subsystem

– Two serial RapidIO ports supporting 1x/4x operation up to 3.125 Gbaud with a RapidIO

messaging unit and two RapidIO DMA units.

– Each 1x/4x serial RapidIO endpoint operates at 1.25/2.5/3.125 Gbaud and complies with the

following parts of Specification 1.2 of the RapidIO trade association interconnect

specification:

• Part I (input and output logical specifications)

• Part II (message passing logical specification)

• Part III (common transport specification)

• Part VI (physical layer 1x LP-serial specification)

• Part VIII (error management extension specification)

– Each serial RapidIO port supports read, write, messages, doorbells, and maintenance

accesses:

• Small and large transport information field only

• All priorities flow

• Pass-through between the two ports that allows cascading devices using the serial

RapidIO and enabling message/data path between the two serial RapidIO ports without

core intervention. A message/data that is not designated for the specific device passes

through it to the next device.

– RapidIO Messaging Unit supports:

• Two outbound message queues

• Two inbound message queues

• One outbound doorbell queue

• One inbound doorbell queue

• One inbound port-write queue

– Each RapidIO DMA unit supports:

• Four high-speed/high-bandwidth channels accessible by local and remote masters

• Basic DMA operation modes (direct, simple chaining)

MSC8156 Product Brief, Rev. 3

Features

Freesca le Sem ico nd uctor10

• Extended DMA operation modes (advanced chaining and stride capability)

• Programmable bandwidth control between channels

• Up to 256 bytes for DMA sub-block transfers to maximize performance over the

RapidIO interface

• Three priority levels supported for source and destination transactions

— PCI-Express Controller

– Complies with the PCI Express™ Base Specification, Revision 1.0a

– Supports root complex (RC) and endpoint (EP) configurations

– 32- and 64-bit address support

– x4, x2, and x1 link support

– Supports accesses to all PCI Express memory and I/O address spaces (requestor only)

– Supports posting of processor-to-PCI Express and PCI Express-to-memory write

– Supports strong and relaxed transaction ordering rules

– PCI Express configuration registers (type 0 in EP mode, type 1 in RC mode)

– Baseline and advanced error reporting support

– One virtual channel (VC0)

– 256-byte maximum payload size (MAX_PAYLOAD_SIZE)

– Supports three inbound general-purpose translation windows and one configuration window

– Supports four outbound translation windows and one default window

– Supports eight non-posted and four posted PCI Express transactions

– Supports up to six priority 0 internal platform reads and eight priority 0 to 2 internal

platform writes. (The maximum number of outstanding transactions at any given time is

eight.)

– Credit-based flow control management

– Supports PCI Express messages and interrupts

• DMA Controller

— 32 unidirectional channels, providing up to 16 memory-to-memory channels.

— Buffer descriptor programing model.

— Up to 1024 buffer descriptors per channel direction provide a total of 32 Kbyte buffer

descriptors. Buffer descriptors can reside in M2 or DDR memories.

— Priority-based time-multiplexing between channels, using four internal priority groups with

round-robin arbitration between channels on equal priority group.

— Earliest deadline first (EDF) priority scheme that assures task completion on time.

— Flexible channel configuration with all channels supporting all features.

— A flexible buffer configuration, including:

– Simple buffers

– Cyclic buffers

– Single address buffers (I/O device).

– Incremental address buffers

– Chained buffers

– 1D to 4D buffers, optimized for video applications

– 1D or 2–4D complex buffers, a combination of buffer types

— Two external DMA request (DREQ) and two DONE signal lin es that allo w an extern al devi ce

to trigger DMA transfers.

Features

MSC8156 Product Brief, Rev. 3

Freescale Semiconductor 11

— High bandwidth

— Optimized for DDR SDRAM

•TDM

— Backward-compatible with the MSC8102/MSC812x/MSC814x TDM interface

— All the four TDM modules toget her support up to 1K time-slots for receive and 1K time-slots

for transmit

— Up to four independent TDM modules:

–Independent receive and transmit mode. Independent transmitter and receiver. Transmitter

input clock, output data, and frame sync can be configured as either input or output. Up to

256 transmit channels and up to 256 receive channels. Receiver input clock, input data, and

input frame sync.

–Shared sync and clock mode. Two receive and two transmit links share the same clock and

frame sync. The sync can be configured as either input or output. Up to 128 transmit

channels and 128 receive channels.

–Shar ed dat a link. Up to four full-duplex data links can operate as either transmit or receive.

All links have the same clock and frame sync. Each link supports up to 128 channels.

— Word size of 2, 4, 8, or 16-bit. All the channels share the same size.

— Hardware A-law/μ-law conversion

— Up to 62.5 Mbps data rate per TDM module

— Up to 16 Mbyte per channel buffer (granularity 8 bytes), where A/μ law buffer size has double

size (16-byte granularity)

— Separate or shared interrupts for receive and transmit with two programmable receive and two

programmable transmit thresholds for double buffering

— Each channel can be programmed as active or inactive

— Support either 0.5 ms (4 frames) or 1 ms (8 frames) latency

— Glueless interface to E1/T1 framers

• The QUICC Engine subsystem includes dual RISC processors and 48-Kbyte multi-master RAM

to handle the Ethernet and SPI interfaces, thus off loading the tasks from the cores. The three

communication controllers support:

— Two Ethernet controllers supporting Gigabit operation

— SPI controller

• Ethernet Controllers

— Two Ethernet physical interfaces, each of which supports:

– 1000 Mbps SGMII protocol using a 4-pin SerDes interface

– 1000 Mbps RGMII protocol

— MAC-to-MAC connection in all modes

— Full-duplex operations

— Full-duplex flow control feature (IEEE Std. 802.3x™)

— Receive flow control frames

— Detection of all erroneous frames as defined by IEEE Std. 802.3®-2002

— Multi-buffer data structure

— Diagnostic modes: Internal and external loopback mode and echo mode

— Serial management interface MDC/MDIO

— Transmitter network management and diagnostics

MSC8156 Product Brief, Rev. 3

Features

Freesca le Sem ico nd uctor12

— Receiver network management and diagnostics

— VLAN Support

—IEEE Std. 802.1p/Q™ QoS

— Eight Tx/Rx queues

— Queuing decision for IP/MAC/UDP filtering based on MAC destination addresses, IP

destination address, and UDP destination port

— Programmable maximum frame length

— Enhanced MIB statistics

— Optional shift of data buffer by two bytes for L3 header alignments

— Extended features

– IP header checksum verification and calculation

– Parsing of frame headers and adding a frame control block at the frame head, containing L3

and L4 information for CPU acceleration

• Serial peripheral interface (SPI)

— Four-signal interface (SPIMOSI, SPIMISO, SPICLK and SPISEL)

— Full-duplex operation

— Works with 32-bit data characters, or with a range from 4-bit to 16-bit data characters

— Supports back-to-back character transmission and reception

— Supports master or slave SPI mode

— Supports multiple-master environment

— Continuous transfer mode for automatic scanning of a peripheral

— Maximum clock rate is (QUICC Engine clock)/8 in master mode and (QUICC Engine clock)/4

in slave mode (not in back-to-back operation)

— Independent programmable baud rate generator

— Programmable clock phase and polarity

— Local loopback capability for testing

— Open-drain outputs support multimaster configuration

— Communication with Ethernet PHY for configuration and status (MIIMCOM-MII

management communication protocol)

— Multi-MIIMCOM environment with up to 32 PHYs

— Programmable clock gap between two characters in master mode

— Controlled by the DSP cores and the QUICC Engine RISC processors according to user

configuration.

• I/O Interrupt Concentrator consolidates all chip maskable interrupt and non-maskable interrupt

sources and routes them to INT_OUT, NMI_OUT, and the cores.

• UART

— Bit rate up to 6.25 Mbps

— Two signals for transmit data and receive data

— Full-duplex operation

— Standard mark/space non-return-to-zero (NRZ) format

— 13-bit baud rate selection

— Programmable 8-bit or 9-bit data format

— Separately enabled transmitter and receiver

— Programmable transmitter output polarity

Features

MSC8156 Product Brief, Rev. 3

Freescale Semiconductor 13

— Separate receiver and transmitter interrupt requests

— Receiver framing error detection

— Hardware parity checking

— 1/16 bit-time noise detection

— Single-wire and loop operations

• Timers

— Two general-purpose 32-bit timers for RTOS support per SC3850 core

— Four TMR modules, each with four 16-bit timers; cascadable timers; count up/down;

programmable count modulo; count once or repeatedly; counters are preloadable; compare

registers can be preloaded; counters can share available inputs; separate prescaler for each

counter; each counter has capture and compare capability; any of the following clock sources:

system clock, TDM clock input, or external clock input

— Eight software watchdog timer (SWT) modules

• Eight programmable hardware semaphores, locked by simple write access without need for

read-modify-write operation by the DSP core.

• Virtual interrupts

— Generation of 32 virtual interrupts by a simple write access

— Generation of virtual NMI by a simple write access

•I

2C interface

— Two-wire interface

— Multi-master operational

— Calling address identification interrupt

— START and STOP signal generation/detection

— Acknowledge bit generation/detection

— Bus busy detection

— Programmable clock frequency

— On-chip filtering for spikes on the bus

• General-purpose input/output (GPIO) ports:

— 32 GPIO ports

— Each GPIO port can either serve the on-device peripherals or act as a programmable I/O pin

— Sixteen GPIO pins can be configured as external interrupt inputs

— All ports are bidirectional

— All ports are set as GPIO inputs at system reset

— All port values can be read while the pin is connected to an internal peripheral

— All ports have open-drain output capability

• Boot interface options:

— Ethernet

— Serial RapidIO interface

—I

—SPI

• JTAG. Test Access Port (TAP) and Boundary Scan Architecture designed to comply with IEEE

Std. 1149.1™.

• Reduced power dissipation

MSC8156 Product Brief, Rev. 3

Developer Environment

Freesca le Sem ico nd uctor14

— Very low power CMOS design

— Low-power standby modes

— Optimized power management circuitry (instruction-dependent, peripheral-dependent, and

mode-dependent)

• Technology: The MSC8156 device is manufactured using CMOS 45 nm SOI technology.

2 Developer Environment

Freescale supplies a complete set of DSP development tools for the MSC8156 device. The tools provide

easier and more robust ways for designers to develop optimized DSP systems. Whether the application

targets a 3G-LTE, TD-SCDMA, or WiMAX system, the development environment gives the designers

everything they need to exploit the advanced capabilities of the MSC8156 architecture.

2.1 Tools

The MSC8156 tool components include the following:

•Integrated development environment (IDE). Easy-to-use graphical user interface and project

manager for configuring and managing multiple build configurations.

•C and C++ compiler with in-line assembly. The developer can generate highly optimized DSP

code by exploiting the StarCore multiple-ALU architecture, with para llel fetch sets and high code

density.

•Librarian. The developer can create application-specific DSP libraries for modularity.

•Linker. The developer can efficiently produce executables from object code and partition memory

according to the application architecture; the linker supports code overlay.

•Multi-Core Debugger. Seamlessly integrated real-time, non-intrusive, multi-mode, multi-core, and

multi-DSP debugger handles highly optimized DSP algorithms. The developer can choose to

debug in source code, assembly code, or mixed mode. Supports RTOS-aware debugger.

•Royalty-free RTOS. Included with package and includes a graphical user interface (GUI) called

Kernel Aware that shows task information, interrupts, and other processing elements.

•Software Simulator. Full chip sim ulation (FC S) that allows the d eveloper to design an app lication

and run it on the simulator before running it on the silicon. FCS is integrated under integrator

developer environment (IDE), the simulator provides customers with tools to create projects and

debug them as they would on silicon (high speed simultaneous transfers). In addition, there is an

SC3850 subsystem performance accurate (PACC) simulator that is approximately 95% cycle

accurate.

•Profiler. The developer can analyze and identify program design inefficiencies.

•High Speed Run Control. USB TAP high speed host-target interface allows users to program in

Flash memory, ROM, and cache.

•Host Platform Support. Microsoft Windows and Solaris.

•Development Board. The application development system (ADS).

•Kit for MSC8156. A complete system for developing/debugging real-time hardware and software.

Developer Environment

MSC8156 Product Brief, Rev. 3

Freescale Semiconductor 15

2.2 Applicati on Software

Freescale offers a broad range of DSP applications through its third-party application software partners;

these applications target IP telephony, telephony modem, wireless and multimedia transcoding, and

wireless base stations. Applications and software modules are listed in Table 1.

Table 1. Application Software Modules

Application Modules

Baseband WiMAX solution supporting Wave 1 features with future extension to Wave 2 and beyond.

3G-LTE evolving kernels library.

Optimized FFT kernels, including matrix multiplication, and so forth.

Devi ce D riv er s and

Example Code MAPLE-B dri ver, DM A driver, s erial RapidIO driv er, TDM drive r, Ethernet d river, UART d river, sec urity

engine, memory allocation, and interrupt handling.

StarCore Libraries Rich set o f StarCore softw are l ibrarie s, inc ludin g: Ma th (Part 1 and 2), Sig nal, C omp lex vec tor, C ontrol

function, Frequency domain, Filter, Common, Image Processing, Communication, and Matrix.

Document Number: MSC8156PB

Rev. 3

5/2011

How to Reach Us:

Home Page:

www.freescale.com

Web Suppor t:

http://www.freescale.com/support

USA/Europ e or Locations Not Listed:

Freescale Semiconductor, Inc.

Technical Information Center, EL516

2100 East Elliot Road

Tempe, Arizona 85284

+1-800-521-6274 or

+1-480-768-2130

www.freescale.com/support

Europe, Middle East, and Africa:

Freescale Halbleiter Deutschland GmbH

Technical Information Center

Schatzbogen 7

81829 Muenchen, Germany

+44 1296 380 456 (English)

+46 8 52200080 (English)

+49 89 92103 559 (German)

+33 1 69 35 48 48 (French)

www.freescale.com/support

Japan:

Freescale Semiconductor Japan Ltd.

Headquarters

ARCO Tower 15F

1-8-1, Shimo-Meguro, Meguro-ku

Tokyo 153-0064

Japan

0120 191014 or

+81 3 5437 9125

support.japan@freescale.com

Asia/Pacific:

Freescale Semiconductor China Ltd.

Exchange Building 23F

No. 118 Jianguo Road

Chaoyang District

Beijing 100022

China

+86 010 5879 8000

support.asia@freescale.com

For Literature Requests Only:

Freescale Semiconductor

Literature Distribution Center

+1-800 441-2447 or

+1-303-675-2140

Fax: +1-303-675-2150

LDCForFreescaleSemiconductor

@hibbertgroup.com

Information in this document is provided solely to enable system and software

implementers to use Freescale Semi conductor products. There are no express or

implied copyright licenses granted hereunder to design or fabricate any integrated

circuits or integrated circuits based on the information in this document.

Freescale Semiconductor reserves the right to make changes without further notice to

any products herein. Freescale Semiconductor makes no warranty, representation or

guarantee regarding the suitability of its products for any particular purpose, nor does

Freescale Semiconductor assume any liability arising out of the application or use of any

product or circuit, and specifically disclaims any and all liability, including without

limitation consequential or incidental damages. “Typical” parameters that may be

provided in Freescale Semiconductor data sheets and/or specifications can and do vary

in different applications and actual performance may vary over time. All operating

parameters, including “Typicals”, must be validated for each customer application by

customer’s technical experts. Freescale Semiconductor does not convey any license

under its patent rights nor the rights of others. Freescale Semiconductor products are

not designed, intended, or authorized for use as components in systems intended for

surgical implant into the body, or other applications intended to support or sustain life,

or for any other application in which the failure of the Freescale Semiconductor product

could create a situation where personal injury or de ath may occur . Should Buyer

purchase or use Freescale Semiconductor products for any such unintended or

unauthorized application, Buyer shall indemnify and hold Freescale Semiconductor and

its officers, employees, subsidiaries, affiliates, and distributors harmless against all

claims, costs, damages, and expenses, and reasonable attorney fees arising out of,

directly or indirectly, any claim of personal injury or death associated with such

unintended or unauthorized use, even if such claim alleges that Freescale

Semiconductor was negligent regarding the design or manufacture of the part.

Freescale, the Freescale logo, CodeWarrior, and S tarCore are trademarks of Freescale

Semiconductor, Inc., Reg. U.S. Pat. & Tm. Off. QUICC Engine is a trademark of

Freescale Semiconductor, Inc. All other product or service names are the property of

their re spectiv e owners.