没有合适的资源?快使用搜索试试~ 我知道了~
RapidIO_Bring-Up_Procedure_on_PowerQUICC_III
需积分: 7 0 下载量 153 浏览量
2018-10-18
13:12:45
上传
评论
收藏 1.01MB PDF 举报
温馨提示
RapidIO_Bring-Up_Procedure_on_PowerQUICC_III,The MPC8540 and MPC8560 PowerQUICC III™ processors have an 8-bit parallel RapidIO interface. This document provides guidance in the basic use of this interface; detailing the setup of local and remote processors starting with basic verification and discovery, then describing booting over RapidIO and conducting simple memory tests
资源推荐
资源详情
资源评论
© Freescale Semiconductor, Inc., 2004. All rights reserved.
Freescale Semiconductor
The MPC8540 and MPC8560 PowerQUICC III™ processors
have an 8-bit parallel RapidIO interface. This document provides
guidance in the basic use of this interface; detailing the setup of
local and remote processors starting with basic verification and
discovery, then describing booting over RapidIO and conducting
simple memory tests.
Most of the guidance in this document applies to any RapidIO
enabled PowerQUICC III system. However, the concepts are
illustrated by a few specific examples.
1 Introduction
This document is designed to provide assistance to engineers
wishing to use the RapidIO interface on the MPC8540 and
MPC8560 PowerQUICC III processors. It summarizes some basic
aspects of the RapidIO specification, and the use of the RapidIO
interfaces on these processors. It also details a basic procedure to
bring up simple RapidIO systems; including the setup of the local
processor, simplified discovery, booting over RapidIO, and the
execution of simple memory reads/writes to the remote
processors.
Section 2, “Background Information,” provides background
information regarding the RapidIO specifications, the
implementation and usage of the RapidIO interface on the
MPC8540/60, and information regarding the example systems,
which are used to illustrate the steps of the procedure outlined in
Section 3, “Bring-up Procedure.”
AN2753
Rev. 1, 11/2004
Contents
1.Introduction 1
1.1. Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2
2.Background Information 3
2.1. RapidIO. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3
2.2. Using RapidIO on MPC8540 and MPC8560 . . . . . . .6
2.3. Example Target Hardware . . . . . . . . . . . . . . . . . . . . .8
3.Bring-up Procedure 11
3.1. Configure the Local Processor . . . . . . . . . . . . . . . . .12
3.2. Discover Other Devices in the System. . . . . . . . . . .20
3.3. Enable Access to Remote Configuration Space . . . .27
3.4. Boot over RapidIO . . . . . . . . . . . . . . . . . . . . . . . . . .30
3.5. Enable Memory Reads and Writes . . . . . . . . . . . . . .36
4.Output from Example Application 41
4.1. Point-to-Point Example. . . . . . . . . . . . . . . . . . . . . . .41
4.2. Fabric Example. . . . . . . . . . . . . . . . . . . . . . . . . . . . .45
5.References 54
6.Revision History 54
Appendix A. Notes on Maintenance Transactions . . . . . . .55
RapidIO Bring-Up Procedure on
PowerQUICC III™
by Lorraine McLuckie and Colin Cureton
NCSD Platforms, East Kilbride
RapidIO Bring-Up Procedure on PowerQUICC III™, Rev. 1
2 Freescale Semiconductor
Introduction
Section 3, “Bring-up Procedure,” describes a procedure for bringing up a basic MPC8540/60-based RapidIO
system. Both point-to-point and fabric-based systems are considered.
Section 4 provides example output from an application that follows the procedure outlined in Section 3, “Bring-up
Procedure,” executed on the example hardware systems described in Section 2, “Background Information.”
Appendix A contains additional information about maintenance transactions, paying particular attention to
the calculation of the maintenance offset used
1.1 Glossary
The following acronyms and terms are used throughout this application note.
Table 1. Glossary of Terms
Term Description
ATMU Address translation and mapping unit
BAR Base address register
CAR Capability attribute register
CCSR Configuration, control and status registers
CCSRBAR CCSR base address register
CSR Capability status register
DDR SDRAM Double data rate SDRAM
DIDCAR Device identity capability register
DMA Direct memory access
EEPROM Electrically erasable programmable read only memory
HBDIDLCSR Host base device lock ID CSR
I/O Input/Output
JTAG Joint test access group
LAW Local area window
LP-LVDS Link protocol, low voltage differential signaling
MAS MMU assist register
MMU Memory management unit
PowerQUICC III MPC85xx networking and communications processor
R/W Read/Write
RIO RapidIO
RIW RapidIO inbound window
ROW RapidIO outbound window
SBTG Software bring-up technical group
RapidIO Bring-Up Procedure on PowerQUICC III™, Rev. 1
Freescale Semiconductor 3
Background Information
2 Background Information
This section provides background information relevant to the bring-up procedure. Section 2.1, “RapidIO,” provides
a summary of the relevant aspects of the RapidIO specification. Section 2.2, “Using RapidIO on MPC8540 and
MPC8560,” describes the implementation of RapidIO on MPC8540/60 and considerations for its use. Section 2.3,
“Example Target Hardware,” provides information regarding the example systems which will be used to illustrate
the procedure.
2.1 RapidIO
The RapidIO interconnect architecture is a high-performance packet-switched interconnect technology. It addresses
the high-performance embedded industry's need for reliability, increased bandwidth, and faster bus speeds in an
intra-system interconnect. The RapidIO interconnect allows chip-to-chip and board-to-board communications at
performance levels scaling to 10 Gigabits per second and beyond.
This section summarizes some basic information about the RapidIO specification. Additional information can be
found in the RapidIO Interconnect Specification.
1
2.1.1 RapidIO Layers
The RapidIO specification is written in a layered manner, to ensure flexibility and that changes to one layer do not
necessarily affect other layers.
The layers of the RapidIO architecture are as follows:
• Logical layer—Defines the overall protocol and packet formats, the types of transaction that can be carried
out with RapidIO and how addressing is handled.
• Transport layer—Provides the necessary route information for a packet to move from one point to another.
• Physical layer—Contains the device level interface such as packet transport mechanisms, flow control,
electrical characteristics, and low level error management.
The logical layer is split into two categories depending on the system model in use. The first part is the input/output
logical specification which defines the basic system architecture of RapidIO. The other is the message passing
logical specification which enables distributed I/O processing. This note concentrates on the input/output logical
specification.
2.1.2 Processing Element Models
There are several types of devices that can be used on a RapidIO based system. This note describes only two device
model types: the integrated processor-memory processing element and the switch processing element.
SDRAM Synchronous dynamic random access memory
TLB Translation lookaside buffer
Table 1. Glossary of Terms (continued)
Term Description
RapidIO Bring-Up Procedure on PowerQUICC III™, Rev. 1
4 Freescale Semiconductor
Background Information
2.1.2.1 Integrated Processor-Memory Processing Element
One form of the processor-memory processing element is a fully integrated component that is designed to connect
to a RapidIO interconnect system as shown in
Figure 1. This type of device integrates a memory system and other
support logic with a processor core on the same piece of silicon, or within the same package, and is one example of
a RapidIO end-point. In this note the processing element is an MPC8540/60 device.
Figure 1. Integrated Processor-Memory Processing Element
2.1.2.2 Switch Processing Element
A switch processing element is a device that allows communication with other processing elements through a
switch. A switch may be used to connect a variety of RapidIO compliant processing elements. Behavior of the
switches, and the interconnect fabric in general, is addressed in the RapidIO Common Transport Specification.
1
In
this note the switch processing element used is the Tundra Tsi500™ 4-port switch.
3
Figure 2. Switch Processing Element Model
2.1.2.3 RapidIO Transactions
In the logical layer, RapidIO defines a broad range of transaction types, ranging from simple transactions to access
an agent device’s memory space, to user defined and implementation dependent transactions. Throughout this note,
three different classes of request transactions and their associated responses are used, where a request is a transaction
that is issued by a processing element to accomplish some activity on a remote processing element. The three
transactions that are used in this note are detailed below. These sections detail the logical layer aspects of these
transaction; no consideration is given to the transport or physical layer interactions.
2.1.2.3.1 Maintenance Transaction
A maintenance transaction is a special system support request. It is used by a processing element to read or write
data to capability attribute registers (CARs) and capability status registers (CSRs), defined in the RapidIO
Processor
Memory
RapidIO Based
System Interconnect
Switch
RapidIO Bring-Up Procedure on PowerQUICC III™, Rev. 1
Freescale Semiconductor 5
Background Information
specification.
1
Maintenance requests do not contain addresses, but use a maintenance offset into the CAR/CSR
block to specify which register is to be read or written.
Maintenance transactions are used for system initialization and discovery and for altering system configuration.
Figure 3. Maintenance Transaction
2.1.2.3.2 NREAD Transactions
A read transaction, consisting of the NREAD and RESPONSE transactions, is used by a processing element to read
data from a specified address. The data is returned in a response packet and is of the size requested.
Figure 4. NREAD Transaction
2.1.2.3.3 NWRITE/NWRITE_R Transactions
The write transaction is used by a processing element to write data to a specified address. The write transaction can
take several forms including NWRITE and NWRITE_R.
The NWRITE request allows data writes, with no expected response.
Figure 5. NWRITE Transaction
1. Maintenance Request
DestinationRequestor
2. Maintenance Response
DestinationRequestor
1. NREAD
2. Response, data
DestinationRequestor
1. NWRITE, data
剩余59页未读,继续阅读
资源评论
Gavin2099
- 粉丝: 11
- 资源: 12
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
最新资源
- formatted_task1121_alt_ja_khm_translation.json
- formatted_task1124_alt_ja_lo_translation.json
- formatted_task1122_alt_khm_ja_translation.json
- formatted_task1123_alt_ja_khm_answer_generation.json
- formatted_task1127_alt_ja_th_translation.json
- formatted_task1125_alt_lo_ja_translation.json
- formatted_task1126_alt_ja_lo_answer_generation.json
- formatted_task1130_xcsr_vi_commonsense_mc_classification.json
- formatted_task1128_alt_th_ja_translation.json
- formatted_task1129_alt_ja_th_answer_generation.json
- formatted_task1133_xcsr_nl_commonsense_mc_classification.json
- formatted_task1131_xcsr_es_commonsense_mc_classification.json
- formatted_task1132_xcsr_ur_commonsense_mc_classification.json
- formatted_task1136_xcsr_fr_commonsense_mc_classification.json
- formatted_task1135_xcsr_en_commonsense_mc_classification.json
- formatted_task1134_xcsr_hi_commonsense_mc_classification.json
资源上传下载、课程学习等过程中有任何疑问或建议,欢迎提出宝贵意见哦~我们会及时处理!
点击此处反馈
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功